,LLLLL, ,LL,LLL,LLLLLL,, , ~ LLLLLLLLL, ,, L,L,L,L,LLLL,LLLL,LMLH,, é? L,LLLLLLLLLLLL, 3L , ,LLL,LLLLLL, \M L L,L,LL,LLLL,L,L,, ,LHLLmLL, L,:LLL,HL,,, MLLLLHLL LHLWLLLLHWLWW L , Lam , L ,L HLL,LLLLLLHH,HL,LLL,LLLm,LL,LLL,L,L,LLL,LLL,LLL,LnLLL,,x,LH,L L L ,, L,LLLL, HL,LLLLL, ,,L m , ,L, L,,LLHLLLLLLHLLLH L, LLL LL ,, LL ,M , L ,L :4» LLL,L,L,LL,LLL,LL,LLLLLH x x-, L,LLHLL L,,LLLL,L,LLL,LLL,LLLLHLL,L,LL, ,LL,LLL,LLLLHWLL,LLLLL,LLL,LLLL,LLLLLLLLLLLLLHLn, ,LLLL, LHLLLHLL,» s ,LLL,LLLLH, L ,L,LL,L :5 “iii: L‘ A”? L, LL,L,L,LLLW,LWLL LLLLLLLLLLLLLL LL L, LLH,HLML L,L,LLMLLLLLL, LLLLLLLLLLLLL L LLMwLwWmeLHLLHmmeNL, L,LL,, sLstmxs L LéLL LLLLLLL ,,LLLHLsLLLLLL LLLLLLLLLm,LLLLmL, L ,LLHLHLLL,L LLL,,LLLW ,LLL,,LL,L,LL,L L, 2&5 “L L LHHLLLLLLL,LL ,L LL,LHLHLLL,LL,L,LLL w as L,LLML H, , L,LLLLL,,L,LL,,L,,L3§n LHLLLL,LL,LLwL, , L,LLLLLLLLLLHLLLLLL,,,LL,HLLLLLLLLL.L,LL,L, L,LLL L ,L ,LLL LLLLLHLLLLmfiL LLL, L,L,L ,HLLLLLLLLLLL LLLLLLLmLHL,LLLL,L,L,,L,,AwLLLL,MLLm L L ,HLL,LL L LL,,LL,,LL, LL,LLLLWLHLLLLL LLLLLLLHL,,, L, , L a a , LL,LLL,L LL LLL,H ,LHLHUL,L,L,L L ,,L,,LL,LMLLHL,L L, LuLHmLmWLLL , LL,L,MLLLHLLL , L,,LLLLHL,LLL, L , L,,: ,L LLLLHLLLLLHLLLL, L,LLL,LLLLLLLLHL,L,,,L, LLLL,L,L,LLwWL,LLLL L ,LLL LLLM LH,LLLLLL, 3L «LL LL ,LLL, L stLL LLL,LL,LLL,,LLLLLLLHHL,,L LLLLLLLLLLL LL ,L L LL La,LLLa3LsLLL,LLL,L,L,LLLL., LLLLLLLLLLW L LL ,L33-H LLx-Mxr LLLLLLLLLLL,LLH ,L ,L 2 LL LLLLH LL L A L 3» ,L,L, L,LLLWLLLLmLLmLMLLLmeL, Eu LL x , L L,HLLL,LLLL, , , LLLL,LLL L,L L; LLL,: L,,LLLLALLL, L, LL L, L ,LLL,LL,? L,,LLLELLLL L,L,LLLKLLLL, L LLL,LLHLLLL,,LH, LL L LL,,LTL, ,,,LLLLLHLLLLLL:,LL,,LLHw , ,LLL,LLLLLLL L , 15¢,LssmsL LL ,L,LLLLMLLLLL LL ,, LLLL, LL ,L .LLM LLLLLLMLLLLLLLLLL ,, HL, LL, L , ,L LLLLM 3 x LL ,HLL,LLM,LL m1 , s L LH,LL,LLHMLLLLL,LL » ,L, iv LLLL,LLm LLLLLL,LLLWLHL,,L, LL L,LLHLLLLLLLWLWLWLLL,LL,LHLH,LL,LLHL,, L LL ,,L LLLL,L,.LL,LLL,L,L LL , L,L,LW ,LHL,L,mH LL ,LLL LL ,LLuLLLLwaLLLL , LL,H,LL,LW_L»,&L, LLLLLLLLLLLLMLmLLmLLLL LLLLLH,L,LLLLLWLL,LL, L L: WWWLLLLmLLLLLL LM§L L x; 3&5 r L,,LL,LL,LL,xL,LWL,L,HL,L L,LLL,L L,LLLLLLLL,LL,LL, L, I: y L: ' "”171." LLLLLLLLL,LL,H L ,,LLL,,LLLLL , LLHHLLLHL, L Lav: , L,LL,L,L,LLLLML, , ,,,L,LLLLLWLL,LLHLL L,L,LLL LLLLLMLLLLLLLL L LLLLLuLLLHLLLLL, LL,L,LHLLLLLmLLWu,L LxLL,,LLLMyH L,L,,LLLLL,LLL L LL L,LLLHLLLL L,LLLELLLLLLL, LLMLLLL LLHLLLL,LLLL , L , LH,LL, L,L,L,LL,LHL,L,,LLL, L,L,XLLLL LL,LLL, LL ,LLLLLLLLLL, LL,LLL, LL,L,L,LLL,L,LL LLMLmiwL, L,LLLLLLLLHLLLLLL,“ ,L ,LLLLLH, , LLLL,L,,L, LL,L L,L,LWWLLLLLLL, L,LL,L ,LL,L,L,L L i L,L,LLL GEOLOGICAL SURVEY PROFESSIONAL PAPER 1015 E? s 5) L, L , LLLL L LMLLLLLLL L LLL ,, L , LL,L L, L,, LLLLLLLLLLLL. LLLL,LLL3 LLULLLLMLLL , ,L L LLLLLLL LuL,,LLLLL, , HLLWL LLL,,,LLx,L ,,L,LLHL,L ,,LLLLLLHL,,,, L,LLLHLLLLLLmLMLLz LL ,L, , LL , LL: PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS,SOUTH DAKOTA WILLIAM T. PECORA PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS,SOUTH DAKOTA P.W. WOLL and W. A. FISCHER, Editors GEOLOGICAL SURVEY PROFESSIONAL PAPER 1015 Sponsored by American Mining Congress in concert with US. Geological Survey American Association of Petroleum Geologists American Society of Photogrammetry The Association of American Geographers The Geological Society of America Society of Economic Geologists UNITED STATES GOVERNMENT PRINTING OFFICE: 1977 UNITED STATES DEPARTMENT OF THE INTERIOR Cecil D. Andrus, Secretary GEOLOGICAL SURVEY V. E. McKelvey, Director The US. Geological Survey agreed to publish the proceeding of the first annual William T. Pecora Memorial Symposium in its Professional Paper series because the subject material is related to the mission of the Survey. The usual standards for this series have been modified to accommo- date the variety of styles used by the participants in this symposium. All color illustrations are placed at the front of the book for economy in printing. They are identified by the names of the authors of the papers from “which they are extracted. Library of Congress Cataloging in Publication Data William T. Pecora Memorial Symposium, 1st, Sioux Falls, S. D., 1975. Proceedings of the First Annual William T. Pecora Memorial Symposium, October 1975, Sioux Falls, South Dakota. (U. 5. Geological Survey professional paper; 1015) Supt. of Docs. no.: I 19.1621015 1. Prospecting—Remote sensing—Congresses. l. Woll, Priscilla W. M. Fischer, William A., 1919— III. American Mining Congress. IV. Series: United States. Geological Survey. Professional paper; 1015. TN270.A1W54 1975 622'.15 76-608272 COLOR ILLUSTRATIONS FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM V| Zena-non Iowa-05H uzmvmm‘m iron .mm mmmmmmOOO>AOo .35 .5250 Kmhyfi 5:3 wEnmmE 22.88le 550: a»: 1.5.. . coto.mmc>. .30) 83:5. 2 :2 5 m.vm:._.?l . y . a; FDVB_$_VO§Z&__JVB_J\.N.O¢2 u 2.53. .x v13. Raw 9 73160.: .8.B.: .8.B_J 8.$_:_ bismoao. voucozcm . :ozoficofn . 3: 100m . mm .5: S u.‘ ‘u‘gmNLmA Eat NNnfi Ed D m.” WW‘_VQ,J\8‘SV2 z Nymo, 7ch u whiz. .8‘85 2 8,83: m 8,123. 8.2.3: _8.mm_1 .818.) ’mum‘mml lava-61ml vzov...m- VII COLOR ILLUSTRATIONS sex: so. Don/pXL manna KEYS ZWEB'M SW ELOZ “2‘30 Pficax 1-07 ”MI” - SYIEYCN 1584-15190 45 7 IEB '24‘38/hm [RYVIX 1975 we fl. IPL QYS MY 6- Florida Keys (Landsat). FIGURE 11. CO LVOCORESS ES FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM VIII Z<_>_._.DG DZ< NHme AR; \w ESE. s880:3: Ea ‘33 .8 £52 Kmmmnouomur 892:. .5»? B 258:. 86935 Tummvcflla “~50: COLOR ILLUSTRATIONS IX FIGURE 8.—Residual total magnetic intensity map with Landsat interpretation superimposed. The +25 gamma contour is at the division between yellow and green; values increase toward the warmer colors and decrease toward the cooler colors in 25-gamma increments. BENTZ AND GUTMAN X FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM ///GE8EL ‘ MAGHARIEM 2 FIGURE 9.—-Bouguer anomaly map with Landsat interpretation superimposed. The zero milligal contour is at the division between yellow and yellow-ochre; values increase toward the warmer colors and decrease toward the cooler colors in 2.5-milligal increments. BENTZ AND GUTMAN COLOR ILLUSTRATIONS FIGURE 11.—Landsat—1 false-color mosaic of Yemen. Images 1082—07011, October 13, 1972; 1082—07013, October 13, 1972; 1117—06562, November 17, 1972; and 1117—06565, No- vember 17, 1972. BENTZ AND GUTMAN XI FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 6.—Great Kavir, September 20, 1972; false-color composite of ERTS—1 image EDC—010015. KRINSLEY COLOR ILLUSTRATIONS FIGURE 7.—Creat Kavir, May 12, 1973; false-color composite of ERTS—1 image EDC—010019. KRINSLEY XIV FIGURE 9.—Machine-aided enhancement of thermal-infrared imagery. (Daedalus Enterprises, Inc.) EICH EN AND PASCUCCI FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 10.—Mineral occurrence overlay—synthesized overlay of analyses from Landsat aeromagnetic and gravimetric data. EICHEN AND PASCUCCI EXPLANATION BIack.—Swaths are known faults or coincidents of ERTS linea— ments with aeromagnetic and gravity Iineaments. Circular areas denote intersections. Blue.—-Swaths are coincidents of ERTS lineaments with aeromag- netic or gravity lineaments. Circular areas denote intersections. Major through-going lineament system. Light dots—Areas of known mines. COLOR ILLUSTRATIONS FIGURE 12.—Composite map showing Iineaments interpreted from Landsat (yellow), EICHEN AND PASCUCCI SLAR (red), and U—2 (black) photography. XVI FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM m ‘ i FIGURE 6A (above) and B (below).—The upper scene includes the area of the Ewaso N’Giro depression, the southern half of Landsat image 1190—07054. The Yamicha lava plateau lies at the extreme northwest corner. The lower scene is the area east of Waiir, including Anomaly A, the southern half of image 1189—07000. These color composites were made by combining a blue image of band 4, a green image of band 5, and red images of bands 6 and 7 on an additive-color viewer and copying the scenes on Ektachrome film. Loca- tion and true orientation shown on figure 8 (p. 150). Scale about 121,000,000. MILLER COLOR ILLUSTRATIONS XVII INIIS-BB Wild-3BI Ulla-BBI H113-3OI , I I 30 NIIS~22 4 26\?¢ -2? ~72? 14-29 N N4l-48/NII4-23 HSS D SUN EL39 92149 lSI-BSSB-G-I-N‘D-ZL NHSR ERTS E-IB7I-l7531~5 l “115* BZOCT72 C N4I-51/NI FIGURE 12.—False-color composite image of a Landsat scene of the region of the Great Basin south of Twin Falls, Idaho. Image 1071—17531, ()ctober 2, 1972. HODGSON I08 I N982 IOU I —L®Z I IDS I —J)DZ XVIII FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 13.—Landsat lineament mapped to scale on ”The Tectonic Map of the United States.” (Modified from USGS and AAPC, 1961.) HODGSON COLOR ILLUSTRATIONS XIX o 100 KM I I FIGURE 6.—Simulated natural color Landsat image of the Nabesna quadrangle. Image made from mosaic of scenes 1692—20150 and 1692—20152 taken Iune 15, 1974. ALBERT AND CHAVEZ hu> ow~E_ .o_0u-wm_£II.n 550: 53. cos COLOR ILLUSTRATIONS FIGURE 8.—Color combination of HH (green) and HV (yellow) imagery near Freeport, Texas. Marshy areas which produce high HV returns appear yellow. DELLWIG AND MOORE Koh-LSU lton APPLlCATlON AREA o 10 20 30 4o 50 MILES L l l l l J FIGURE 2.—Enhanced false-color composite of part of Landsat image 1125—05545. Location of the Saindak porphyry copper deposit is indicated by “S”, new prospects located as a result of the digital processing experiment are shown at 5—c, 5—d, 6—d, 6—e, and B—a. The light—toned patches at the crest of the volcano Koh-i-Sultan are altered rock resulting from fumarolic action. SCHMIDT AND BERNSTEIN XXII FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM II|7°W |||6°W l|5°W l|4 W I|3°W 37°N— —37°N 36°N— —36°N 35°N— —-35°N 34°N- - ' i ‘ mun“: 32‘: 7. . 7 I " M‘I 34°N I I l n7°w ||e°w |15°W H4°W “3°W FIGURE 2.—False-color mosaic of 11 Landsat—1 MSS composites covering the area of study outlined in figure 1. Identifica- tion numbers for the frames used in this mosaic are shown in figure 3 (p. 255). See text for discussion of the M55 bands and printing filters used in producing the composites. LIGGETT AND CHILDS COLOR ILLUSTRATIONS XXIII FIGURE 7.—Comparison of aerial photography (A), Landsat color image (B), Landsat color ratio image (D), and geological map (C), for Yerington, Nevada. The satellite materials are prepared directly from the CCT’s. 7A, Ektachrome color print, taken from 20 km (RB 57, MX 248, August 11, 1973). 78, Color infrared image made from CCT, 1397—48051, August 24, 1973, 13 days later than the aerial photograph; Dicomed print using digital data in an enhanced mode (note increased con- trast). 7C, Published geological map (Moore, 1969). 70,- Color ratio image made from the same CCT data as for 78, Dicomed print, using ratios R74, R64, and R54. Note the similarity between the color patches on 7D and the geological patterns on the map, 7C, especially the areas of Ta (Alta volcanics)—open arrow, and Th (Hartford Hill volcanics)—solid arrow. Also Kg (granodiorite) and the pbrphyritic granodiorite (Kpg). LYON (rm Opening remarks, by V. E. McKelvey __________________________ International implications of Landsat data from a geological view- point, by J. A. Reinemund ______________________________ Table 1. Other countries for which Landsat data have been provided by the EROS Data Center __________________________________ 2. Other countries having existing or planned facilities using the Landsat system _______________________________________ 3. Landsat regional seminars and workshops in which the U. 8. Geological Survey has been involved _____________________ NASA plans for future earth resources missions, by William Nordberg Table 1. Earth resources and environmental survey satellites—op- erating and in preparation _______________________________ 2. Earth resources and environmental survey satellites—planned and in prospect Application of remote sensing (Landsat data) to petroleum ex- ploration (abs), by M. T. Halbouty ________________________ 13 14 16 XXV Landsat application to resource exploration and gas planning, by C. E. Brockmann ___________________________________ Figure l. 2 3. 4, 5 6 Table 1. Geological provinces of Bolivia ____________________________ Geologic and geomorphological map from ERTS images _______ Drainage interpretation from ERTS images _____________________ Geological interpretations from Skylab image _________________ Mineralogical map _________________________________________ Gasline selective routes ____________________________________ Comparison of alternative gasline route ______________________ An overview of Canadian progress in the use of Landsat data in geology, by A. F. Gregory and L. W. Morley _______________ Table 1. 2. 3. 4. 5 Relative interest in remote sensing by user discipline, August 1975 Relative interest in remote sensing by economic sector, August 1975 Net sales of standard Landsat products for 3 years ending September 15, 1975 _____________________________________ Geological users of Landsat data, 1975 ______________________ Net sales of standard Landsat products to recognized geological users ___________________________________________________ Mapping and charting from Landsat, by A. P. Colvocoresses _______ Figure 1, SDWNQFJ‘PWN 10, 11. 12. 13. 14. 15. 16. 17. 18. Map of Copper Creek, Australia, and ERTS imagery of the creek in flood __________________________________________ Map accuracy improved by Landsat imagery _________________ Thin cloud penetration capability of ERTS IR sensor __________ Water boundary delineation from ERTS _____________________ Space Oblique Mercator projection __________________________ Nominal ERTS scenes, Florida ______________________________ Seasonal mapping with ERTS ________________________________ Lac Manicouagan, Canada __________________________________ Comparison of nautical chart and ERTS image ________________ ERTS correlation to nautical chart __________________________ Florida Keys (Landsat) _____________________________________ Gas flares in Saudi Arabia (Landsat) _______________________ DMSP image ______________________________________________ Enlargement of a portion of ERTS image showing mirror response The National Capital Region, Ontario—Quebec, Canada ___. _____ Phoenix __________________________________________________ Arizona __________________________________________________ Florida ___________________________________________________ Relationship of mineral resources to linear features in Mexico as determined from Landsat data, by G. P. Salas ____________ Figure 1. 2. 10. XXVI Composite of Mexico from Landsat—1 imagery showing the Neovolcanic Axis Metallogenetic Province _________________ Structural interpretation of Landsat—1 imagery showing incidence of faults, fractures, mines, and adits _______________________ East Pacific rise and Nevolcanic Axis structure _______________ Bouguer anomalies showing coincidence of strong gradient with Sierra Madre Occidental _________________________________ Trend of orientation of silver deposits in Mexico showing ap- parent displacement of subprovinces ______________________ Schematic geologic map of the Neovolvanic Axis Metallogenetic Province of Mexico _____________________________________ Landsat—1 imagery of the Neovolcanic Axis of Mexico showing predominance of NW—SE and NE—SW trending faults and shorter fractures ________________________________________ Structural map of Pachuca showing E—W and NW—SE faults and veins __________________________________________________ Taxco mining district, State of Guerrero, showing vein control by faults and fractures __________________________________ El Oro-Tlalpujahua mining district ___________________________ 21 22 24 25 26 28 30 31 33 35 36 36 37 37 43 44 45 45 46 47 48 VI 49 50 5 1 VII 52 53 54 55 56 57 58 61 62 63 64 65 66 67 68 71 72 73 Landsat contributions to studies of plate tectonics, by Jan Kutina and W. D. Carter ______________________________________ Figure 1. 2. Tectolinear interpretation of a 1:5,000,000«scale ERTS—l mosaic of the United States ____________________________________ Tectolinear interpretation of the eastern part of the United States, set in context with some structural and metallogenic features of the Canadian Shield ___________________________ Landsat data contributions to hydrocarbon exploration in foreign regions, by F. P. Bentz and S. I. Gutman __________________ Figure 1. .OONQ‘Q‘PWN 5° 11, 12. 13. 14. Index map of northern Africa showing Santa Fe Minerals, Inc, exploration area ________________________________________ Geology of Cairo—Suez District, after Said __________________ Landsat—1 false-color mosaic of Egypt _______________________ Lineament interpretation of Landsat mosaic of Egypt ___________ Star diagram illustrating relationships between mapped lineaments Wrench fault tectonics as related to lineament interpretation ____ Regional tectonics of Northeast Africa _______________________ Residual total magnetic intensity map with Landsat interpretation superimposed ___________________________________________ Bouguer anomaly map with Landsat interpretation superimposed Index map showing Sante Fe Minerals, Inc, exploration area in Yemen ________________________________________________ Landsat—1 false»color mosaic of Yemen ______________________ Interpretation of lineaments, circular and tonal anomalies from Landsat data ___________________________________________ Circular anomalies and well locations _______________________ Detail of streaks and tonal anomalies on Yemen coastal plain __ Measurement of luminescence of geochemically stressed trees and other materials, by W. R. Hemphill, R. D. Watson, R. C. Bigelow, and T. D. Hessen ______________________________ Figure 1. 2 10. Photograph of laboratory fluorescence spectrometer ____________ Photograph of Perkin—Elmer FLD, showing the optical head, electronic console, and light collector _____________________ Photograph of FLD optical head and light collector installed in a basket mid~ship of a Bell Jet Ranger helicopter ___________ Map showing location of Malachite Mine area, Jefferson County, Colorado _______________________________________________ Map showing distribution of copper in soil and location of sample trees growing in background and anomalous soils ___________ Graph showing excitation spectra of Pinus ponderosa needles from geochemically stressed and background trees, Malachite Mine, Jefferson County, Colorado, as measured with a laboratory fluorescence spectrometer ________________________________ Graphs of temporal luminescence of Pinus ponderosa and meteor— ological parameters. Maximum luminescence contrast between background and anomalous trees tends to occur during periods of minimum cloud cover _________________________________ Graphs of temporal luminescence of Pinus ponderosa and meteor- ological parameters. Minimum luminescence contrast between background and anomalous trees tends to occur during periods of maximum cloud cover _________________________________ Graph of temporal luminescence of Pinus ponderosa and meteor- ological parameters. Luminescence contrast of background and anomalous trees do not correlate with cloud cover conditions cited in figures 7 and 8 __________________________________ Map showing location of geochemically stressed and background trees in the Alpine Mill area, Douglas County, Nevada _______ 75 77 78 83 84 85 VIII 86 87 87 87 IX 88 XI 89 91 93 94 95 96 98 98 99 100 101 102 104 XXVII Measurement of luminescence of geochemically stressed trees and other materials—Continued Figure 11. 12. 13. 14. Table 1. Linear regressive analysis correlating FLD luminescence counts for specific trees with molybdenum concentrations from 20 to 300 ppm _______________________________________________ Geologic map of part of the Sespe Creek area showing loca- tion of a helicopter traverse and helicopter hovers __________ Statistical plot showing correlation between relative luminescence and specific gravity of 29 crude oils at the following Fraunhofer lines: A, 396.8 nm; B, 422.7 nm; C, 486.1 nm; D, 518.4 nm; E, 5890 nm; and F, 656.3 nm _____________________________ Map showing: (1) Dispersal of oil from natural seep off Coal Oil Point, Santa Barbara Channel, and (2) FLD luminescence re sponses from clear water, thin oil film, and heavy crude layer__ Luminescence (source—detector, solar, and depth corrected) of 10 bean plants expressed in terms of rhodamine WT equivalency Mean and standard deviation of copper and zinc content of needle ash from 13 Pinus ponderosa growing in geochemically anoma— lous and background soils ________________________________ Location, molybdenum content in ash, and luminescence, expressed in rhodamine WT equivalence, of geochemically stressed and background juniper trees in the Alpine Mill area _____________ Luminescence of 10 phosphate samples measured with the labor- atory fluorescence spectrometer at 486.1 nm in terms of rhoda— mine WT equivalence (source—detector, solar, and depth cor- rected) _________________________________________________ Luminescence of phosphate and gypsum samples, collected from the Santa Margarita Formation near Pine Mountain, California, and measured with the MPF—3 at 486.1 nm in terms of rhoda» mine WT equivalence (source—detector, solar, and depth cor— rected) _________________________________________________ Reflectance, FLD counts, and luminescence in terms of rhoda- mine WT equivalency (source-detector, solar, and depth cor- rected) of phosphate, gypsum, and background reference ma- terials _________________________________________________ Reflectance, FLD counts, and luminescence in terms of rhoda» mine WT equivalency of phosphate, gypsum, and background materials _______________________________________________ FLD counts and luminescence in terms of rhodamine WT equiva- lency of associated phosphate and gypsum materials in the Lakeland, Florida, area __________________________________ Integrated excitation intensity of crude oils at specific Fraunhofer wavelengths in terms of equivalency with rhodamine WT _____ Use of ERTS—l (Landsat—1) images for engineering geologic applica- tions in north-central Iran, by D. B. Krinsley ________________ Figure 1. XXVIII Map showing the relationship of the desert basin sumps (playas) and associated features to relief in Iran ___________________ Map showing interior watersheds of Iran: the Great Kavir and the superposed boundary of the ERTS—l (Landsat—1) images__ Map showing elements of the Great Kavir ___________________ Photograph of rough black salt ridges and pinnacles adjacent to the dry—season road across the Great Kavir ________________ Map showing existing Iranian road net and the proposed Great Kavir road _____________________________________________ Great Kavir, September 20, 1972; false—color composite of ERTS- 1 image _______________________________________________ Great Kavir, May 12, 1973; false—color composite of ERTS—I image __________________________________________________ 105 106 110 111 97 99 103 105 106 107 107 108 109 113 114 115 116 118 119 XII XIII Use of ERTS—l (Landsat—1) images for engineering geologic applications in north-central Iran—Continued Table 1. Hydrologic conditions along critical segments of the dry-season road across the Great Kavir as inferred from ERTS—l images from September 2, 1972, to May 12, 1973 _________________ Applications of remote-sensing technology for powerplant siting, by Leo Eichen and R. F. Pascucci __________________________ Figure 1. >4 H... N!“ H W owmsowaww Landsat satellite __________________________________________ Landsat image of New York City and vicinity _________________ Thermal infrared image recording of surface heat emissions __-_ Sketch diagram, typical side»looking airborne radar system _____ Side—looking airborne radar image ___________________________ Typical aeromagnetic map, Mineral, Virginia __________________ Lineament analysis of Landsat image _________________________ Lineament analysis of SLAR imagery __________________________ Machine-aided enhancement of thermal-infrared imagery ________ Mineral occurrence overlay—synthesized structural overlay of anal- ysis from Landsat, aeromagnetics, and gravimetric data ______ Digitally processed land use data from Landsat observation ____ Composite map showing lineaments interpretated from Landsat (yellow), SLAR (red), and U»2 (black) photography ___________ Multispectral computer processed classification map of an area west of Koh«i—Dalil, Pakistan, showing five areas classed as ”mineralized” ___________________________________________ Landsat—1 image studies as applied to petroleum exploration in Kenya, by J. B. Miller ________________________________________ Figure 1, 2. Index map showing regional features and areas of the Chevron exploration license and option ____________________________ Landsat image mosaic, Garissa-Wajir'El Wak area, Kenya, band 7 _____________________________________________________ Landsat image mosaic, Garissa-Wajir-El Wak area, Kenya, band 5 _____________________________________________________ Lineaments, rock unit boundaries, and other features as inter— preted from the Landsat imagery _________________________ Rock unit distribution, broadly defined from Landsat imagery and other sources _______________________________________ Landsat false—color composites, The upper scene includes the area of the Ewaso N’Giro depression. The lower scene is the area east of Wajir, including Anomaly A ___________________ Comparison of Landsat imagery structural features with crudely represented geophysical data _____________________________ Structural pattern derived from geophysical data over the north— ern part of the initial Chevron license _____________________ Regional linear analysis as a guide to mineral resource exploration using Landsat data, by R. A. Hodgson ____________________ Figure 1. Chart showing sequence of Landsat data flow from acquisition to application in minerals exploration _____________________ Primary factors considered in use of Landsat data for analysis and interpretation of regional linear geologic features _______ Main features of the Landsat—1 MSS subsystem _______________ Diagram showing configuration of the Initial Image Generating Subsystem ______________________________________________ Diagram showing the main elements comprising a Landsat image 120 123 124 125 126 127 128 129 130 131 XIV XIV 133 XV 134 137 138 141 142 143 144 XVI 149 150 155 156 157 158 158 159 XXIX Regional linear analysis as a guide to mineral resource exploration using Landsat data—Continued Figure 6. Schematic diagrams showing the composite structural and physio« graphic aspects of lineaments ____________________________ Basic types of geologic structures ___________ . _______________ Orders of systematic linear features _________________________ Sketch and structure contour map of Florence Pass lineament ___ Smaller orders of linear geologic structures __________________ Smallest order of linear geologic structure seen on Landsat imagery by direct observation or by inference ____________________ 12. False-color composite image of a Landsat scene of the region in the Great Basin south of Twin Falls, Idaho ________________ 13. Landsat lineament mapped to scale on the tectonic map of the United States __________________________________________ 14. Sets of major lineaments visible on the 1974 band 7 USDA Land- sat mosaic of the United States at scale 15,000,000 ________ 15. First~order lineaments of the United States taken from the band 5 Mosaic of the United States as published in the July 1974 issue of Geotimes ____________________________________________ 16. Spurr’s Silver Line and the platinum-nickel line of Thamm ______ 17. Zonal distribution of platinum~nickel deposits along a Great Circle in eastern Africa _______________________________________ 18. Sectional diagram showing various orders of fractures which may be genetically related and which may control the location of mineral deposits _________________________________________ i—w—n f-‘QSQQON The application of remote sensing technology to assess the effects of and monitor changes in coal mining in eastern Tennessee, by A. E. Coker, A. L. Higer, and R. L. Rogers ______________ Exploration by petroleum independents using imagery and photos from EROS and manned space surveys, by R. W. Worthing ______ Tectonic deductions from Alaskan space imagery, by E. H. Lathram and R. G. H. Raynolds _________________________________ Figure 1. Space image lineaments in Alaska __________________________ 2. Space image lineament pattern in North American Cordillera _____ 3. Lineaments and crustal fractures suggested in selected inductive analyses _______________________________________________ Computer-enhanced Landsat imagery as a tool for mineral exploration in Alaska, by N. R. D. Albert and P. S. Chavez _____________ Figure 1. Map of Alaska showing location of the Nabesna and McCarthy quadrangles ____________________________________________ 2. Map of the Nabesna and McCarthy quadrangles showing gen- eralized geologic terranes ________________________________ 3. Map of the Nabesna and McCarthy quadrangles showing linear features and mine locations _______________________________ 4. Map of the Nabesna quadrangle showing linear and circular fea— tures and color anomalies observed on computer-enhanced Land- sat imagery _____________________________________________ 5. Map of the McCarthy quadrangle showing linear and circular features ________________________________________________ Simulated natural color Landsat image of the Nabesna quadrangle False—color Landsat image with sinusoidal stretch of the Mc— Carthy quadrangle. ______________________________________ 8. Breakdown of color anomalies seen on computer~enhanced Land- sat imagery and their association with mineralized areas and geochemical anomalies in the Nebesna quadrangle __________ >10 XXX 159 160 161 161 163 164 XVII XVIII 165 166 167 168 169 173 175 179 182 184 186 193 194 194 195 196 197 XIX XX 198 Computer-enhanced Landsat imagery as a tool for mineral exploration in Alaska—Continued Figure 9. Map of the McCarthy quadrangle showing known faults and their possible extensions as determined by linear and curvilinear fea- tures visible in the sinusoidally stretched false-color Landsat image of the McCarthy quadrangle ________________________ Table 1. Number of significant mineral occurrences in the various linear directions in the McCarthy quadrangle, Alaska ______________ Evaluation of improved digital processing techniques of Landsat data for sulfide mineral prospecting, by R. G. Schmidt and Ralph Bernstein ____________________________________________ Figure 1. Index map showing the Chagai District and Saindak, Pakistan____ Enhanced false-color composite of part of Landsat image 1125— 05545 _________________________________________________ 3. Graphic summary of digital processing and data analysis per— formed in the experiment _______________________________ 4. The Saindak—Mashki Chah area of the western Chagai District, Pakistan, showing the control area near Saindak and the area where the digital-classification method was applied __________ 5. Classification map of an area west of Koh-i‘Dalil showing five areas classed as ”mineralized" ___________________________ 6. Comparison of digital-classification map made by using table “B” and geology mapped in field, Saindak porphyry copper de— posit, Pakistan __________________________________________ Table 1. Digital classification table “B" used in 1974 mineral evaluation in the Chagai District, Pakistan __________________________ 2. Revised classification table “SSD” prepared in April 1975 ______ 3. Revised classification table “S” prepared in September 1975 ____ A deeper look at Landsat—1 images of Umiat, Alaska, by A. F. Maurin and E. H. Lathram ____________________________________ Figure 1. Index map of northern Alaska showing physiographic provinces, the area covered by Landsat image 1004—21395, and the area of oriented lakes previously studied by Carson and Hussey____ Landsat image 1004—21395, band 7, of the Umiat area ________ Ozalid print of positive transparency of Landsat image 1004— 21395, band 7, showing contrast enhancement and location of detailed figures _________________________________________ 4. Direction diagram computed from the three heavy small circles of figure 3 ____________________________________________ Steps in erosion and opening process _______________________ Another area studied using the erosion and opening process ____ Structural interpretation by conventional geomorphologic study us- ing outcropping anticline as a standard and the elliptical “path" concept ________________________________________________ 8. Two selected areas of Landsat image 1345—21344, band 7, with common orientation showing mushroom ice cakes in black lakes __________________________________________________ 9. Experimental processing of part of the Northern Foothills shown on Landsat image 1004—21395 ___________________________ WN N09“ Why Landsat? A management view, by J. R. Porter, Jr _____________ Regional and global geological studies using satellite magnetometer data, by R. D. Regan ____________________________________ 199 196 201 202 XXI 204 208 209 211 206 207 207 213 214 215 216 217 218 219 220 221 221 225 229 XXXI AIRTRACET’I—An airborne geochemical exploration technique, by A. R. Barringer _______________________________________ 231 Figure 1. Nickel and copper in surface dust, Agnew, Western Australia -- 238 2. Two-dimensional model of the convective plume _______________ 239 3. lnfiight analog record of particulate count, temperature, and ver— tical g, Arizona, February 6, 1975 _________________________ 240 4. Contributions of global, continental, regional, and local back— grounds to near-surface particulate loadings ________________ 241 5. Effect of mixing activity on particulate loading ______________ 241 6. AIRTRACET'“ equipment mounted in a 2063 Jet Ranger _______ 242 7. A test of linearity of laser analytical system for variations in load» ings of USGS soil (XRG—6), silicon versus titanium _________ 244 8. A test of linearity of laser analytical system for variations in loadings of U865 soil (XRG—6), zinc versus titanium ________ 245 9. Limerick prospect, near Bancroft, Ontario, AIRTRACE Mk III— copper ________________________________________________ 246 10. Flight A AIRTRACET“ activity contours Cu+Zn regression ______ 247 11. Flight B AIRTRACETM activity contours Cu+Zn regression ______ 247 12. Cement comparison of AlRTRACETM data and ground truth ______ 248 Table 1. Analyses of soils, plants, and vegetative particulates __________ 234 An application of satellite imagery to mineral exploration, by M. A. Liggett and J. F. Childs ________________________________ 253 Figure 1. Location map for the area of study showing generalized positions of the Black Mountains and Nye County volcanic provinces--- 254 2. False-color mosaic of 11 Landsat—1 MSS composites covering the area of study ______________________________________ XXII 3. Index map showing Landsat—1 MSS frames used in the false»color mosaic _________________________________________________ 255 4. Generalized map of the Cenozoic fault systems visible in the Landsat—1 imagery ______________________________________ 256 5. Generalized map of the Cenozoic volcanic and plutonic rocks in the study area _________________________________________ 257 6. Diagrammatic model relating right-lateral strike—slip movement on the Las Vegas Shear Zone to crustal extension in the two volcanic provinces _______________________________________ 259 7. Generalized chronology of strike-slip movement on the Las Vegas Shear Zone and normal faulting, igneous activity, and related mineralization in the Black Mountains and Nye County volcanic provinces _______________________________________________ 262 8. Late Cenozoic mineralization in relation to Cenozoic faulting in the Back Mountains volcanic province ______________________ 264 9. Diagrammatic model illustrating the interrelationship of strike— slip and normal faulting in the Black Mountains volcanic prov— ince ___________________________________________________ 266 Mineral exploration applications of digitally processed Landsat imagery, by R. J. P. Lyon ______________________________________ 271 Figure 1. Channel-by—channel spectral plots for picture elements along a raster line ____________________________________________ 272 2. Time series of averaged spectra and radiance and normalized reflectance _____________________________________________ 273 3. Yerington pit area, Nevada, aerial photograph ________________ 276 4. Raw lineprinter output for channels 5 and 7, Yerington pit area, Nevada _______________________________________________ 277 5. Effect of varying the tolerance level in STANSORT/CLUSTER 278 6. Clustering analysis using STANSORT/SEARCH _______________ 278 7. Comparison of aerial photography, Landsat color image, color ratio image, and geological map of Yerington, Nevada ______ XXIII XXXII Mineral exploration applications of digitally processed Landsat imagery— Continued Figure 8. Views of Goldfield, Nevada, site and Pine Nuts Mountains, Ne— vada, comparing vegetation cover ________________________ 280 9, Relation between regional structure, alteration zones, and Land- sat R54 ratio image, Goldfield, Nevada ___________________ 281 10. Locality map of Landsat—detected anomaly and the Mo-anomaly in the vegetation, Mt. Siegel quadrangle, Nevada ____________ 285 11. Enhanced Landsat images of Karasjok, northern Norway ______ 287 12. Landsat analysis of Tifalmin, New Guinea, area ______________ 288 Table 1. Localities and results of case histories ________________________ 275 2. Typical ratio R54 values for Goldfield _______________________ 282 3. Ratio comparisons for Goldfield, Nevada _____________________ 282 4. Means and interband ratios for some altered rocks and some un- altered rocks, Goldfield, Nevada _________________________ 283 5. Published ratio values for some alteration minerals ___________ 285 6 Correlation between variables in data sets with increasing vege» tation content __________________________________________ 289 7. Means and variability in data sets with increasing vegetation con- tent ___________________________________________________ 290 Tradeoff considerations in utilization of SLAR for terrain analysis, by L. F. Dellwig and R. K. Moore ________________________ 293 Figure 1. Two radar images of Phoenix, Arizona ______________________ 295 2. Interpretability vs. equivalent resolution for hard targets, roads, ’ and other transportation _________________________________ 297 3. Comparison of square and rectangular cells __________________ 298 41 Mono Craters, California, AN/APQ—97 SLAR imagery __________ 299 5. Bountiful, Utah, AN/APQ—97 SLAR imagery __________________ 300 6. Atchafalaya River basin, Louisiana, AN/APQ—97 SLAR imagery__ 301 7. Freeport, Texas, AN/APQ—97 SLAR imagery _________________ 302 8. Color combination of HH and HV imagery near Freeport, Texas__ XXI 9. Phoenix, Arizona, AN/APS—94D SLAR imagery _______________ 303 10. Denver—Colorado Springs corridor, Colorado, SC-01 SAR imagery 304 11. Price comparisons—photos vs. radar, spring 1975 ______________ 305 Worldwide indexing and retrieval of Landsat images, by F. F. Sabins, Jr. ___________________________________________ 307 Figure 1. Location of COFRC Landsat index maps _____________________ 308 2 COFRC Landsat index map plotting routine __________________ 309 3. Portion of COFRC printout of Landsat—1 index tape ___________ 311 4. Landsat—1 index map of southern Africa, 0—73 days ___________ 312 5 Landsat—1 index map of southern Africa, 74—144 days _________ 312 6 Landsat—1 index map of southern Africa, 145—248 days ________ 313 7 Retroactive index map for first 248 days plotted from second in- dex tape _______________________________________________ 313 8. Updated Landsat—1 index map, 249—478 days, plotted from second index tape _____________________________________________ 314 9, Composite of 0% cloud cover index maps to indicate areas lacking cloud—free images ________________________________ 314 10. Portion of COFRC world zone map for southwest quadrant (zone 3) _____________________________________________________ 315 11. Landsat mosaic of southern part of Sudan ___________________ 316 Intercrossing crustal structure and the problem of manifestation of its deep-seated elements on the surface, by V. I. Makarov and L. I. Solov’yeva ___________________________________ 319 Figure 1. Scheme of intercrossing structure of the Turan platform and Tien Shan based on the results of interpretation of space images, structural analysis of the relief, and geologic data ___________ 322 XXXIII Intercrossing crustal structure and the problem of manifestation of its deep- seated elements on the surface—Continued Figure 2. Scheme showing location of zones of higher seismic activity in Northern and Central Tien Shan __________________________ 3. Some examples of interference between the elevation zones of various trends __________________________________________ 4. Scheme of echelon—like structure of elevation zones __________ Combined formalized processing of space image and geologic-geo— physical data in connection with the study of deep structure of petroliferous platform regions, by P. V. Florensky, A. S. Petrenko, and B. P. Shorin-Konstantinov __________________ Figure 1. Southern Russian plain. TV image obtained by the Soviet Earth- orbiting satellite “Meteor” _______________________________ 2. Lower Volga region. TV image transmitted from the American satellite ERTS—l on 16 July 1973 __________________________ 3. Example of a composite profile of geologic-geophysical and pho- tometric information ____________________________________ 4. Photometric scheme of the image shown in figure 2 __________ 5. Geologic map for the Lower Volga region ___________________ 6. Scheme of neotectonic movements of the Lower Volga region __ 7. Scheme of isolines showing the correlation coefficients for the gravity and magnetic fields _______________________________ 8. Basement surface topography ______________________________ 9 Scheme of isolines showing the coefficients of correlation of the gravity field with the optical density _______________________ 10. Scheme of lineaments ____________________________________ 11. Zoning scheme for the crystalline basement of the Lower Volga region compiled from a complex of geologic and geophysical data and taking into account the results of deciphering and pho— tometry of TV-space images _______________________________ Table 1. Quantitative characteristics of the identified blocks ___________ Geological studies by space means in the U.S.S.R., by V. G. Trifonov, V. I. Makarov, V. M. Panin, S. F. Scobelev, P. V. Florensky, and B P. Shorin-Konstantinov ___________________________ Figure 1. Large-scale space photo of Babadag-Karatau region of the central part of the Tadjik depression ___________________________ 2. Structural—geologic map of the central part of the Tadjik depres— sion compiled using space photos _________________________ 3. Scanner image of Eastern Fergana, Landsat—1, June 24, 1973, band 5 ______________________________________________________ 4. Geological interpretation of the image figure 3 ______________ 5. Scanner image of Eastern Fergana, Landsat—1, June 24, 1973, band 6 ________________________________________________ 6. Geological interpretation of the image figure 5 ______________ 7. Middle-scale photo of the lssik~Kul region ___________________ 8. Geological—geomorphological interpretation of figure 7 __________ 9. Large-scale photo of the eastern coast of Caspian Sea, Manghy— shlak Peninsula, and Ustjurt ______________________________ 10. Structural map of Manghyshlak and Ustjurt compiled using photos from “Sojuz—S” and “Sojuz-12” ___________________________ 11. Middle-scale space photo of Eastern Caucasus and Lower Kura lowland ________________________________________________ 12. Correlation of lineaments deciphered on “Sojuz-9” photos and zones of deep seismic deformations of Eastern Caucasus ____ 13. Small-scale scanner image of Caucasus, 18 satellite “Meteor," August 21, 1974, band 0.6—0.7 mkm _______________________ XXXlV 323 326 326 339 340 341 342 342 343 343 347 348 349 35 1 352 344 355 356 357 358 359 360 36 1 362 363 364 365 366 367 368 Geological studies by space means in the USSR—Continued Figure 14. 15. 16. 17‘ Correlation of lineaments deciphered on TV and scanner “Meteor" images of Eastern Caucasus and epicenters of earthquakes 1911— 1966 ___________________________________________________ Scanner image of Lower Volga region, Landsat—1, July 16, 1973 Photometrical map of figure 15 in conditional figures without clouds Schematic structural-geologic map of crystalline basement of Lower Volga region using geological and geophysical data and space photo interpretation _______________________________ 369 369 370 370 XXXV PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Opening Remarks to the First William T. Pecora Memorial Symposium By Vincent E. McKelvey, Director, US. Geological Survey Governor Kneip, Dr. Overton, Ladies, and Gentle- men: It is a great privilege for me to be able to open this First William T. Pecora Memorial Symposium. Bill Pecora was a remarkable man—a great scien- tist, a philosopher in the broadest sense of that term, and an inspiring leader. He considered himself to be a “field-boot” geologist and so characterized himself, as a matter of fact, in those terms‘ in a speech here in 1970 when the EROS Data Center was still on the drawing board. But while he was, indeed, an outstanding “field— boot” geologist, by taking the lead in forming the EROS Program and pressing hard for its develop- ment, he showed himself to be a visionary, both in his recognition of the possible applications of space technology to the study of the Earth, and of the need for it to help satisfy the requirement of a world pressing hard in the means of assessment. Bill saw the rapidly accelerating needs for re- sources and for the preservation of environmental quality as well, and he saw the great potential in satellite observations for helping to provide the worldwide information needed to achieve both of those objectives. Much of Bill Pecora’s vision has al- ready been realized. NASA has provided the tech- nology to acquire valuable satellite data, and the peo- ple of Sioux Falls, the Congress, and the Federal Administration have made it possible for the full community of scientists, technologists, and environ- mentalists to obtain and interpret and use the data through the Sioux Falls EROS Data Center. As pa- pers here to be presented will show, new uses and applications of satellite imagery are being devel- oped and formatted. Indeed, we have already had a glimpse of some of those applications in the descrip- tion which Governor Kneip gave us of the activities already in progress here in the State of South Dakota. Many people and organizations have contributed to these achievements, but naturally I am proud of the part which the Geological Survey and the De- partment of the Interior have played in bringing about a program that has fostered international rela- tions and improved interorganizational and interdis- ciplinary exchange and communication and is adding to our understanding of the Earth and its resources. These contributions flowed from Bill Pecora’s imagi- native leadership, and it is fitting, indeed, that this Symposium be named in his honor. A word, now, about the William T. Pecora Sym- posium. This is, of course, the first of what we hope will be a long series devoted to the exchange of experience and knowledge in the application of re- motely sensed data. Because of Bill Pecora’s great interest in mineral resources and because of the exciting progress being made in the applications of Landsat data to mineral exploration, it seemed ap- propiate to focus this First Symposium on mineral and mineral fuel exploration—and we are glad that the American Mining Congress was able to take the lead in sponsoring it. The next Symposium will be primarily sponsored by the American Society of Photogrammetry and will be held here in Sioux Falls next August 23—27. The theme will be “Mapping with Remote Sensing,” a topic fundamental to all resource and environ- mental efforts. We anticipate that subsequent sym- posia will address other themes, such as wildlife preservation, forestry, water resources, and others. The Survey has no desire to direct the content of these symposia, but we do pledge continued support to the interested scientific societies to make this a forum for the development and exchange of knowl- edge in keeping with the vision of a truly great man, William T. Pecora. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA International Implications of Landsat Data From a Geological Viewpoint By John A. Reinemund, US. Geological Survey, Reston, Virginia 22092 INTRODUCTION It is indeed a privilege and an honor for me to have a part in this William T. Pecora Memorial Sym‘ posium on the applications of remote sensing to min- eral and mineral fuel exploration. Although I am cer- tainly no expert on remote sensing, I have been continuously involved in the international program of the US. Geological Survey for about 20 years and have been a fascinated observer of the impact re- mote-sensing technology has had, both in our own program and in overall cooperation between the nations of the world, It seems especially fitting that we should discuss the international implications of Landsat data at a symposium honoring Dr. Pecora. Not only was he primarily responsible for the Survey’s strong effort to develop an earth resources satellite system, in co- operation with the National Aeronautics and Space Administration (NASA), but he was also vitally inter— ested in the potential applications of that system to international resources problems. His interest in the international applications was a logical outgrowth of his participation in the Survey’s international program more than 30 years ago, when he assisted in surveys for strategic minerals in Latin America (Pecora, 1944; Pecora and Fahey, 1949, 1950; Pecora, Klepper, and others, 1950; Pecora, Switzer, and others, 1950). His enthusiasm for international cooperation, and his recognition of the mutual benefits to be gained from joint studies of geological phenomena, in cooperation with scientists abroad, were to a considerable extent responsible for the steady growth in the Survey’s pro- gram in international geology and the strong em- phasis on international applications of Landsat data which has characterized the Survey’s efforts in re- mote sensing. SIGNIFICANCE AND SCOPE OF THIS DISCUSSION It is interesting to note that the problems with which Dr. Pecora and other Survey geologists (Dorr, 1944; Johnston, 1947) were primarily involved in their Latin American work 30 years ago-the inter- national supply of critically needed mineral raw ma- terials—is once again a subject of intense interna- tional concern. This concern has been partly respon- sible for the worldwide interest in Landsat applica- tions. Geologists in the industrialized countries recog— nize the need to greatly accelerate the exploration and assessment of world resources to help meet their nations’ future import requirements, and they look to Landsat data as an aid in this process. Their interest was reflected at the meeting of the European Geo- logical Societies in Reading, England, last month when 10 percent of the 130 presentations dealt with Land- sat applications to mineral investigations on five continents. We here in the United States are also increasingly concerned, as discussed recently in a Conference on Requirements for Fulfilling a National Materials Policy (Promisel, 1974), for in the last 3 years our annual imports of energy and mineral raw materials have risen from about $4 billion or 11 percent of total needs to about $20 billion or 27 percent of total needs, and last year we imported more than half our requirements of 23 mineral commodities (Secretary of the Interior, 1975). But most developing countries are even more concerned with the need to increase 3 4 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM TABLE 1.—Other countries for which Landsat data have been provided by the EROS Data Center Afghanistan Denmark Israel Algeria Dominican Republic Italy Argentina Ecuador Jamaica Australia El Savador Japan Austria Ethiopia Jordan Bangladesh Finland Kenya Barbados France Khmer Republic Belgium Gabon Korea Bolivia Gambia Kuwait Brazil Germany, East Lebanon Brunei Germany, West Lesotho Bulgaria Ghana Liberia Burma Greece Libya Canada Guatemala Luxembourg Central African Guinea Madagascar Republic Guyana Malawi Chile Honduras Malaysia China Hungary Mali Colombia Iceland Mauritania Costa Rica India Mauritius Cyprus Indonesia Mexico Czechoslovakia Iran Morocco Dahomey Ireland Nepal Spain Netherlands Sudan New Zealand Surinam Nicaragua Swaziland Niger Sweden Nigeria Switzerland Norway Taiwan Pakistan Tanzania Panama Thailand Paraguay Trinidad Peru Tunisia Philippines Turkey Poland United Arab Republic Portgual United Kingdom Rhodesia Upper Volta Romania Uruguay Saudi Arabia U.S.S.R. Senegal Venezuela Sierra Leone Yemen Singapore Yugoslavia South Africa Zaire South Vietnam Zambia worldwide resources exploration and assessment, be- cause they not only compete for access to raw materi- als for their own internal needs but are faced with the necessity of increasing raw—material exports to pay for imports of food, manufactured goods, and technology. And it is in the developing countries, where mineral surveys are generally not well advanced, that the use of Landsat data can be especially helpful to accel- erate mapping, improve existing maps, and identify geologic conditions favorable for mineralization. So it is that at this critical time, when industrial and and developing countries alike are faced with the urgent need for intensified resources studies, that the Landsat system, along with other satellite and aircraft systems, offers a new, powerful, and challenging tool, the full dimensions of which have not yet been real- ized. The worldwide interest in this tool is evident from the number of articles, meetings, and projects that deal with it. In our Geological Survey program last year, for example, 58 scientists from 35 countries attended the two international courses given here at the EROS Data Center, and 127 scientists from 24 countries participated in seminars or workshops we conducted abroad. About 55 governments are cur— rently sponsoring experimental applications of Land- sat imagery, and most of these involve cooperation with one or more organizations in the United States. The EROS Data Center has supplied Landsat ma- terials to 111 countries (see table 1), and already this year data for more than 50,000 Landsat frames have been purchased by private and governmental users for areas outside the United States. In this discussion I do not propose to describe or illustrate in detail the specific applications of Landsat data that are, or may become, internationally impor- tant. Many excellent presentations to follow in this symposium will do this far better than I could. How- ever, I would like to comment briefly on some signi- ficant applications of Landsat data in other countries, on the basis of the rapid growth in the international use of Landsat data, on the more significant benefits of participation in the Landsat program, and on con- straints that could prevent the most effective use of this new technology. SIGNIFICANT INTERNATIONAL APPLICATIONS OF LANDSAT DATA Geologists around the world are focusing most of their attention, in one way or another, on three major problems: Accelerating and intensifying the explora- tion and assessment of resources, as mentioned pre- viously; protecting and efficiently utilizing the environ- ment; and minimizing the effects of natural disasters. Landsat data have important applications in each of these problem areas, and the papers that follow illus- trate these applications. Many of the recent applica- tions were reviewed in the Tenth International Symposium on Remote Sensing of Environment held in Ann Arbor, Michigan (Environmental Research In- INTERNATIONAL IMPLICATIONS OF LANDSAT DATA 5 stitute of Michigan, 1975), earlier this month. An initial evaluation of economic value of these applications was made last year by Lietzke (1974). With regard to resources exploration and assess- ment, the principal applications of Landsat data have been in accelerating the mapping of unexplored terri- tory and in improving existing maps. The cooperative mapping of high-relief mountain provinces such as the Himalaya Mountains, the Alps (Bodechtel and Lammerer, 1973) and the Andean Cordillera (Carter, 1975), which will no doubt be illustrated later this morning by Dr. Carlos Brockmann, is especially sig- nificant because we have not previously been able to obtain synoptic and essentially distortion-free images of these provinces, which are so important for an understanding of mineral genesis. Equally significant is the application of Landsat data to study of the Precambrian shield areas, as has been described re- cently by Weecksteen (1974) for West Africa, Corréa (1975) for Brazil, and Blodget, Brown, and Moik (1975) for Arabia. The use of Landsat multispectral techniques permits the identification and definition of contacts that are not readily located without the use of this new tool. And in studies currently underway, it has been possible to correlate the geology in the shield areas on the opposite sides of the Red Sea (Abdul—Gawad and Tubbesing, 1975) and to demon- strate the value of Landsat data characteristics in long—distance correlation of geologic units. Extensive and highly significant revisions of struc- tural maps are underway in many countries using Landsat imagery, and these are especially significant in prospecting for minerals and hydrocarbons. For example, a new map of Thailand is nearing comple- tion as a joint effort by Prayong Angsuwathana of the Thai Department of Mineral Resources and S. J. Gawarecki of the US. Geological Survey, and a new map of Pakistan has been prepared by Kazmi1 in which it has been possible to classify faults on age of movement. A closely related and highly significant use of Landsat data in East Asia by Maurice Terman of the Geological Survey, as a contribution to a 35—nation cooperative Circum-Pacific Map Project (International Union of Geological Sciences, 1975), is in the evalua- tion of the relative accuracy of existing geologic maps. This demonstrates the value of Landsat data for checking geologic map accuracy as well as for im- proving and augmenting map detail. ‘ Kazmi, A. H., 1975, Application of ERTS—1 imagery to recent tectonic studies in Pakistan: Unpub. rept. prepared for the Geological Survey of Pakistan. Perhaps no use of Landsat data has created more interest recently than the identification of major structural lineaments, many of which seem to be re- lated to the distribution of minerals and hydrocarbons (Lathram and others, 1973; Saunders and Thomas, 1973). Several papers to be presented at this Sym- posium deal with this subject, and important work has been done in Australia and in several countries in Asia, South America, and North America. Some im- portant known ore bodies, such as Broken Hill in New South Wales, seem clearly related to major linea- ments, at least some of which reflect through—going structures of major dimensions in the continental crust. Agah (1975) has reported on lineaments across the Zagros Mountains in Iran that seem related both to facies changes in the sedimentary sequence and to hydrocarbon distribution. This line of research is of such widespread interest and importance that an international research project on correlation of struc- tural data derived from satellite and other sources with distribution of known mineral deposits is now being considered under the International Geological Correlation Program jointly sponsored by the Inter- national Union of Geological Sciences and UNESCO. An equally challenging use of Landsat data in the detection of ore deposits through digital-processing techniques is discussed in six presentations to follow, here in this Symposium. The work carried out by Schmidt in Pakistan is especially indicative of the impact and potential of Landsat data, when applied to a well-studied terrain, in which the extent and significance of the mineralization were much more apparent after digital processing of Landsat data. The technique of digital analysis to detect alteration in rocks and soil, and also changes in vegetation related to mineralization, may prove to be one of the most important applications for accelerating mineral ex— ploration. Geologists concerned with environmental studies and surficial phenomena are having spectacular suc- cess with Landsat data. The work of McKee and others (McKee and others, 1973; McKee, 1975), for example, in worldwide studies of sand seas, provides a basis for classification and evaluation of environ- mental changes in the principal desert areas that could not readily be made without Landsat data. In a recent study of Landsat applications in the Sahelian zone in Africa, Cooley and Turner (1975) demon- strated the usefulness in identifying and mapping laterite duricrust, locating areas favorable for ground water, classifying land for agricultural potential,\ 6i : lxxxxx 0° WA —. :5} I \ “ v . . e \2;\\ _: _ * x x X )6 \ : E— \ x x 0 ‘ — — CALAVERA-l x ‘ if kt} \ x x [J O- STA.ROS’—ll \ ‘. \ x I) / / x); 90 4 x V a f g o A . PALOMETAS ’ x x f a \\ X ' .— __ ’S_’L 2 / /’.2: ’1’ / ‘\ unreno \\ \ 44 l" X \LI MONTECRISTO k 50 KM 1—“; MENORA «1/ EXPLANATION 0 Capital of province 'A‘ . Highway \I/ Subsurface probable anticline H—O—t—l- Railroad a)”; Wedging out probable; arrow f’qé; Rivers indicates direction of wedging 5:; \V Dips —:—:— Flooded area V’I/ {1* Gas well Lineamonts C) Dry well I? 7 TI Crystalline baumont Mixes (Brazilian Shield) FIGURE 4.—Ceo|ogical interpretations from Skylab image SL—2—A—339 of the Santa Cruz-Montero-Buena Vista area of Bolivia. Approximate scale 1:800,000. possible to correlate structural and geologic data which are basic to petroleum exploration. MINING EXPLORATION The Mining Exploration Subprogram is conducted by the Mining Corporation of Bolivia using Landsat imagery interpretation aimed at locating new favor- able areas for exploration and exploitation of mining resources. Because of the scale limitations of Landsat imagery for visual interpretation, the conventional mineral re- source evaluation methods were modified. Generally, LANDSAT APPLICATION TO RESOURCE EXPLORATION AND GAS PLANNING 27 mining exploration methods are conducted at large scales ranging from 1:1,000 to 125,000, and generally close to known exploration and mining sites. How- ever, new maps at 1:250,000 scale are being prepared for the purpose of establishing regional genetic rela~ tions of the mineralized bodies, their host rocks, structures and, where present, alteration zones. Results of the Landsat imagery interpretation are important because it was possible to identify linea- ments related to faults, tonal anomalies, and geo- morphic anomalies which, when correlated to the existing information, let us delimit new potential mining areas. To establish a methodology which could let us evaluate the mineralized zones from different points of view, the geomorphic characteristics, tonal differ- ences, and drainage patterns were studied and con- sidered to be fundamental criteria. Tonal anomalies are defined as being the delimita- tion of zones with tonal characteristics that are different from the surrounding area and which may be related to intense zones of weathering, contact aureoles, vegetation concentration, facies variations of certain formations, and hydrothermal alteration. Geomorphic anomalies are shapes that vary from the general tendency of the landscape and which are generally related to endogenic processes, such as magmatism and diapirism. Tectonic-Structural anomalies are zones affected by the intersection of faults and lineaments, generally of regional character, and related to the emplacement of igneous bodies. They are generally represented by straight courses of streams and valleys. FIELD RESULTS With the purpose of proving the relationship of existing mineralization with anomalies defined by Landsat interpretation, fieldwork was performed in the area covered by Landsat image 1010-14033, se- lecting the best defined zones. The field results can be summarized as follows (fig. 5): 1. The Structural Anomaly of Patacamaya is located 15 km to the northeast of the town of the same name on the Tapacari lineament at its inter- section with the Achuta lineament. There are several mineralized subvolcanic bodies along the Tapacari lineament. Field verification tests were negative at the lineaments intersection, due possibly to thick cover of Quaternary sediments that exist in the zone. 2. Structural Anomaly of Huayllamarca is located on the ridge of the same name. It corresponds to a fault zone where a direct relation between the anomalies, longitudinal faults, and diapirous rocks were determined. There we found copper minerals in the form of sulfates and carbonates. 3. Geomorphic Anomaly of Laurani is located close to the town of Sica Sica. It coincides geological- ly with a subvolcanic body, where dacites and andesites are known; this igneous body ap- pears to be related to the Colquencha—Laurani and Achiri-Umala lineament intersection which strike northwest and southeast, respectively. Fieldwork showed a direct relation between the anomaly and igneous activity with strong hydrothermal alteration and mineralization con- sisting of copper, silver, and lead minerals that are actually being exploited. 4. Tonal Anomalies of Sica Sica are located 10 km from the town, where fieldwork has determined the presence of sedimentary rocks with dis- cordant red tones in contact with Paleozoic rocks with gray tones. The objective of this reconnaissance was to identify a possible altera- tion zone as the zone is located along the inter- section of two regional faults. No evidence of mineralization was found. 5. Geomorphic Anomalies of Antaquira are located southwest of the town of Caquiaviri, where the direct relation between a longitudinal fault, a synclinal structure and subvolcanic intrusive rocks was found. Copper mineralization was found for the first time in the fault and joints that comprise the zone. CONCLUSIONS It was proven that mineralization can be found associated with longitudinal faults in the structural anomalies (Huayllamarca and others) identified on Landsat images. The geomorphic anomalies are important as they always coincide with intrusive bodies and/or sub- volcanic bodies, which are generally mineralized. When systematic and sufficient detailed fieldwork were accomplished on major lineaments, these areas became interesting for mineral and petroleum ex- ploration. ' Even though the present example of tonal anomaly did not provide the expected positive result, there is evidence in other studied zones that tone has a rela- tion with mineralization processes. 28 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM wao'w WN’W \Ay “ "4c", ”0.“va %l A, L)? \(w: J “V SICA 1T;\\\ \\ TONAL womb} \\ I \ \\ \ f % \ LAunAm \\ EN u'h 39 u touovmucu \ ‘ ANOMALY -y L" 'RIL’NE" “5339’ ,cvslBl— \ 18°00’S \ 4 \ HUAYLLAMARCA \ srnucrunAIU \ \ \ \ ANOIALY EXPLANATION — Lineamonts—Faults ___ Probable lineaments r’ a \ \fibfi/ Subvolcanic bodies z-\ 1 <_ ,’ Geomorphic anomalies FIGURE 5.—-—Mineraloglcal map. Approximate scale 121,250,000. PRELIMINARY SELECTION OF THE SANTA location of the most direct route of the Santa Cruz- CRUZ_PUERTO SUAREZ GASLINE ROUTE Puerto Suarez gasline in relation to the actual existing railroad and to design several alternative routes. OBJECTIVES Methodology and materials used. The principal objective of the present work was to In [order to comply with the proposed objectives study the applicability of the Landsat imagery for the and due to the scanty cartographic information of the LANDSAT APPLICATION TO RESOURCE EXPLORATION AND GAS PLANNING 29 area, the Landsat—1 imagery that covers the zone with a maximum of (SO-percent cloud cover was selected and supported by a Skylab space photograph. We also used a photoindex of available aerial photo- graphs that cover the study area. Images 1240—13415, 1240—13421, 1005—13344, 1005—13350, 1111—13243, bands 5 and 7, photograph SL—2—19OA~339, bands 2 and 6, and the photoindex of the South Roboré area, were used due to the lack of satellite images of this section at the time the work was begun. The study was undertaken at a 1:250,000 scale, using the Santa Cruz-Corumba railroad as a reference to prepare the corresponding map. As the study area contains peculiar geomorphologic features, the first general physiographic map of the zone was prepared. In it we attempted to delimit the boundary of the Brazilian Shield in relation to the Chaco-Beni Plain, the Chiquitos Ridges, and other important geomorphologic features that could affect the selection of the preliminary route. Lagoons, rivers, creeks, hills, sand dunes, and especially the “Izozog Bafiados (swamps)” and the “Otuquis Bafiados (head- waters)”, lowland areas that are subject to temporary flooding, were all decisive in the selection of the route that the gasline should have. STUDY AREA DESCRIPTION Generally, the Brazilian Shield in this zone is ex- pressed as a peneplane where isolated granitic rocks and metamorphic rock outcrops exist and are separated from the San Jose, Roboré, and Quimone Ridges by a poorly drained depression filled with Quaternary sediments (fig. 6). The Chaco-Beni Plain corresponds to a poorly drained alluvial plain covered by low vegetation in the area of the Izozog Banados and Otuquis Bafiados. In these areas there are a great number of small braided rivers and scattered temporary lagoons, which are characterized by being located in topogra- phic depressions with poor drainage. The interpretation work was performed with images obtained during the period of the lowest stage of the rivers. The use of images, taken mainly at the end of the rainy season, is considered necessary for the accurate mapping of the flooding areas. PRELIMINARY ALTERNATIVES OF THE ROUTE Considering the locations of Santa Cruz city and the gasfields in exploitation as a starting point for the gasline to the town of Puerto Suarez and taking as basic parameters the longitude route and physio- graphic conditions of the area for the construction and maintenance of the route, the following alterna- tives were designed: 1. Colpa Field—Puerto Suarez alternative Two routes [subalternatives] were considered for the portion of this alternative from the Colpa Field to Pozo del Tigre station. The alternative continues then as a single route to Puerto Suarez. North Alternative Approximate distance: 604.4 km. Average altitude: Colpa Field 345 m above mean sea level. Striking directly south-southeast from the Colpa Field for a distance of 138 km we can arrive at the P020 del Tigre station. Field conditions along this route are good and characterized by gentle topogra- phy with adequate drainage where soils are generally limy-clayey sand. However, it is necessary to indicate that in the first stage of it, the Piray and Grande Rivers would have to be crossed. Their meandering courses must be studied in detail from existing sup- plementary information. South Alternative Approximate distance: 607.1 km. Average altitude: Pozo del Tigre 272 m above mean sea level. Colpa Field 345 m above mean sea level. This alternative was designed with a southeast strike and an approximate distance of 140.7 km be- tween the Colpa Field and P020 del Tigre station, where the gasline crossing the Grande River would parallel the actual railroad bridge. This subalternative has the advantages of being located farther away from the town of Warnes and of crossing the Grande River in an area where geo- logic information obtained for the bridge construction is available. Another positive characteristic for this section would be the fact that the route would parallel the railroad, facilitating by this way tube material transportation and would avoid new path construc- tion. However, problems can be foreseen on the route right-of-way, as there is intense agricultural activity in the area. Common Point From Pozo del Tigre station, the route is common for both subalternatives and parallels the railroad with an east-southeast strike displaced from it an approximate distance of 2 km to the south. The pur- pose for this was to reduce the distance to the sur- roundings of the Musuruqui station (74.8 km distance, 358 m altitude) where the north edge of the Izozog Bafiados is flooded and marshy during the rainy sea- son. Total lengths of these proposed sections of gas- 30 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM MONTERO . } .J« o I E (I. 5 \v. RIO GRAN DE OIL- FIELD Ridges Hills Lakes Rivers 1M“ Railroad _. ._ EXPLANATION Alternative South, Colpa gas field Pto. Suarez 607.1 km. Alternative North, Sta. Cruz—Pto. Suarez 591.9 km Direct Route Santa Cruz—Pto. Suarez 574.1 km. Rio Grande Oil Field Pto. Suarez 539.5 km. FIGURE 6.—-Cas|ine selective routes. Approximate scale 123,500,000. line are 215.8 and 218.8 km for the north and south subalternatives, respectively. As it can be noticed, the proposed designs to the Musuruqui station do not present much variation in relation to the railroad. It is from this point that we propose the route to go south of the San José Ridge in a direct line to near the town of Roboré. The chosen route in this section is supported by the fact that the area is well drained; there are few, if any, rock outcrops and the soils are generally sandy having formed by the erosion of the San José Ridge. The route will go through a smooth plateau with a uniform altitude in the east section where it is pre- sumed that sufficient soil thickness would be found for the excavation of the gasline trench. Making a comparison of this route with the actual railroad in the same section that lies between Musuru- qui and Roboré, there appear to be several negative factors: (1) the construction of the bridge over the Quimome River indicated that there is insufficient soil thickness for the excavation of the trench, a factor that would increase the construction cost; (2) this zone is characterized by bad drainage in which temporary lagoons appear during the rainy season; and (3) the most important factor, if this route paralleling the railroad is selected, is related to the distance as it is 13 km more than the distance proposed south of the bridge. 2. Santa Cruz city-Puerto Suarez route This route was designed to collect the gas from the different production fields surrounding Santa Cruz city. LANDSAT APPLICATION TO RESOURCE EXPLORATION AND GAS PLANNING North Alternative Santa Cruz city-P020 del Tigre station-Puerto Suarez. Distance: 591.9 km. Altitude: Santa Cruz 465 m above mean sea level. Puerto Suarez 220 m above mean sea level. There is a distance of 125.5 km from Santa Cruz city to the common point of the P020 del Tigre sta- tion where an alluvial plain is the main landscape. It consists mainly of limy sand that extends along the main route already designed from Puerto Suarez. As a restriction for the selection of this alternative, it must be indicated that a suitable crossing of the Grande River must be chosen where its course is only slightly meandering. Also, to be considered is the construction of the feeder gaslines from the produc- ing areas to the common point in the vicinity of Santa Cruz city. Direct Alternative Santa Cruz city-Puerto Suarez Distance: 574.1 km. Generally, it must be indicated that this alternative would go first through the similar terrain described previously but with a significant problem. The Izozog Bafiados is considered to be a potential problem area, both for the construction and maintenance of the line, especially during the rainy season. 3. Grande River Field-Puerto Suarez alternative Distance: 539.5 km. Altitude: Rio Grande Field 338 m above mean sea level. Puerto Suarez 220 m above mean sea level. This possible route was delineated because the Rio Grande Field is the biggest deposit of gas reserves in the country, and the route would be the most direct to TABLE 1.-——Comparison of 31 the town of Puerto Suarez. The design route, how— ever, would cross the Izozog Banados, a total distance of 92 km. This potentially difficult terrain constitutes the most serious restriction that must be considered. CONCLUSIONS The area maps prepared for this study are the most up-to-date information available for the study of the selection of the route for the gasline. Landsat imagery, due to its regional coverage and geometric accuracy along with the multispectral factors, has enabled us to obtain the information of extensive areas both quickly and economically in comparison with the more conventional aerial methods. RECOMMENDATIONS 0 To make an aerial reconnaissance at a low altitude, combining the existing aerial photographs with Landsat imagery. 0 To make a new study with Landsat—2 imagery to obtain more information with the multitemporal coverage in order to delineate more accurately the area influenced by temporary flooding. 0 To study in detail the course of all the rivers that would have to be crossed in order to select the most suitable points for the construction of the bridges. 0 To make an economic study on each one of the alternatives detailed before, relating distance to construction and maintenance. These factors will help the ultimate selection of the route. alternative gasline routes Comparative sheet Construction Distance Railroad km Distance Physiographic and mainte- Alternatives (km) difference evaluation evaluation nance Total Colpa Field-Puerto Suarez: > North Alternative ___________________________ 604.400 -——25.600 x x x x x x x x x x x x 12 South Alternative _____________________________ 607.100 —22.900 x x x x x x x x x x x 11 Santa Cruz city-Puerto Suarez: North [Route] Alternative ____________________ 591.900 —38.100 x x x x x x x x x x 11 Direct [Route] Alternative _____________________ 574.100 —55.900 x x x x x 6 Rio Grande [River] Field—Puerto Suarez __________ 539.500 —90.500 x x x x x x 7 Railroad Route: Santa Cruz-Puerto Suarez _____________________ 630 ______ x x x x x x 6 x x x x x Excellent x x x x Good x x x Regular x x Bad x Very Bad 32 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM SELECTED REFERENCES Anonymous, 1975, Technical report of the Thematic Cartographic Project of the Department of La Paz: La Paz, ERTS Program/Bolivia, Servicio Geologico de Bolivia, Office rept., July. Brockmann, Q. Carlos, 1974, Preliminary alternatives of the Santa Cruz-Puerto Suarez gasline route: La Paz, ERTS Program/Bolivia, Servicio Geo— logico de Bolivia, Office rept., October. Claure, H., 1974, ERTS imagery research for mining exploration: La Paz, Mining Corporation of Bolivia for ERTS Program, Servicio Geologico de Bolivia, Office rept., June. Luizaga, S., 1975, Information on the revised Anomal- ous Zones detected through the technological satellite ERTS~1: La Paz, Servicio Geologico de Bolivia, Office rept., March. Oblitas 6., Jaime, Salinas E., Carlos, Davila E., Juan, Cabrera V., Arturo, and Hidalgo F., Manuel. 1973, Summary of the petroleum geology of Bolivia: La Paz, Yacimientos Petroliferos Fiscales Boliv- ianos, 92 p. Vargas, Carlos, 1973a, Drainage interpretation of the ERTS images of the Cobija- Puerto Heath-Madre de Dios River—Ixiamas area: Yacimientos Petroli- feros Fiscales Bolivianos: La Paz, ERTS Pro- gram/Bolivia, Servicio Geologico de Bolivia, Office rept., September. 1973b, Results of the geologic-geomorphical study of two ERTS images of the San Borja Zone—Mamore River-Rogaguado Lake-West of Bolivia: La Paz, Yacimientos Petroliferos Fiscales Bolivianos. Office rept., August. 1975, Geologic—physiographical interpretation of the Skylab photograph corresponding to the Santa Cruz-Montero-Buena Vista area: Yacimi- entos Petroliferos Fiscales Bolivianos: La Paz, ERTS Program Bolivia, Servicio Geologico de Bolivia, Office rept., February. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA An Overview of Canadian Progress in the Use of Landsat Data in Geology By A. F. Gregory, President Gregory Geoscience Ltd., Ottawa, Canada, and L. W. Morley, Director Canada Centre for Remote Sensing, Ottawa, Canada ABSTRACT During the past 3 years, Canadian geologists have been assessing the geological use of Landsat data in the context of a vast but relatively well known hinter- land. Although few in number, these geologists comprise the largest single class of users of Canadian Landsat data. In particular, a few oil and mineral exploration companies and consultants appear to use as many images as the other 90 percent of the geo- logical users, excluding reference collections. Most Canadian geological users have a relatively long and broad experience with Landsat data. Visual interpretation of black and white prints is the princi- pal method of analysis. There has been little research into automated processing; however, a major advance in digital classification of arctic terrain is imminent. Practical applications are related primarily to re— connaissance for geological structure and disposition of surficial materials. Half of the users claim a modest to large benefit from use of Landsat data but cannot identify specific dollar benefits. The level of development of practical geologic applications in Canada is greater than predicted in a pre-ERTS forecast. Visual interpretation of Landsat images will soon become a prime tool, in conjunction with others, for geological reconnaissance. The tech- nique will be especially valuable in poorly explored arid areas where vegetational cover is sparse. Auto- mated processing will remain in the research stage until specific low-cost methodologies are developed. LANDSAT IN A CANADIAN CONTEXT Canada, with a territory covering nearly 3.9 million square miles, is the largest country in the Western Hemisphere and the second largest in the world. Canadian terrain is extremely diverse, comprising almost semitropical areas in southern Ontario and southern British Columbia, great mountain arcs, wide fertile prairies, a vast lake-strewn interior, and seem- ingly endless stretches of northern wilderness and tundra. Eighty—nine percent of this area is not settled; indeed 95 percent of all Canadians live in 10 percent of the area comprising an irregular fringe along the border with the United States of America. From a geologist’s viewpoint, Canada is composed of 17 geological provinces in four main categories: continental shelf, platform, orogen, and shield. Geo- logically, the youngest provinces are the Atlantic, ‘ Pacific, and Arctic Continental Shelves. They com- prise slightly deformed sediments and volcanics, mainly of Mesozoic and Cenozoic age. The St. Law- rence, Interior, Arctic, and Hudson Platforms com- prise thick, fiat-lying Phanerozoic strata over crystalline basement. The Appalachian, Cordilleran, and Innuitian Orogens are mountain belts of deformed and metamorphosed sedimentary and volcanic rocks, mainly of Phanerozoic and Proterozoic age, and granitic intrusions. The remaining seven geological provinces comprise the Canadian Shield which is com- posed of Precambrian rocks. The Grenville, Churchill, Southern, and Bear provinces embrace Proterozoic orogenic belts. The Superior, Slave, and Nain provinces comprise metasedimentary and metavol- canic rocks of older Archean age, including the oldest continental crust known in Canada. The mineral industry is one of the most important sectors of the Canadian economy. The value of min- eral production in 1974 was $11.6 billion, of which fossil fuels represented 44 percent, metals 42 percent, 33 34 FIRST ANNUAL PECORA nonmetallics 8 percent, and structural materials 6 per- cent. Much was exported, mainly to the United States, Britain, and Japan. Principal mineral products were oil, natural gas, nickel, copper, iron ore, zinc, as- bestos, cement, and sand and gravel. Of the major geological provinces, the Interior Platform yields in value about 40 percent of all mineral production, mainly fossil fuels, and the Canadian Shield about 35 percent, mainly metals. Recorded Canadian history began with the quest for these and other riches in the New World. The search for economic gain was a particular stimulus, and indeed is a continuing theme, throughout Canadian history although the nation became a reality just over 100 years ago. Despite its vast area and relatively young age as a nation, Canada has been well ex— plored and mapped at reconnaissance scales. For example, an updated coverage of black and white aerial photographs is maintained for all of Canada at scales between 163,360 and 1:15,000. The whole country has been surveyed and mapped with high standards at a scale of 1:250,000, thus presenting a relatively detailed depiction of relief, river systems, transportation facilities, forest cover, and centres of population. Bedrock geology has now been mapped over 95 percent of the nation at reconnaissance scales of 12250000 and smaller, although only about 20 per- cent of the overlying Quaternary sediments have been mapped at similar scales. A vast but relatively well-known hinterland, thus, is the context in which Canadian geologists have as— sessed the geological use of Landsat data. The experi— ences reported here reflect involvement in the ERTS— Landsat program from early planning in the United States through direct readout of data at our own receiving station to current research and applications (Morley, 1971). The information comes from sources too numerous to acknowledge here, including indus- trial, academic, and government members of the Working Group on Geoscience of the Canadian Ad- visory Committee on Remote Sensing, staff and files of the Canada Centre for Remote Sensing, and con— tributions from private companies. DEVELOPMENT OF CANADIAN GEOLOGICAL INTEREST IN THE LANDSAT CONCEPT HISTORY Over the past 60 years, but especially since 1945, photogeology has made significant contributions to the science and practice of geological mapping. In the early 1960’s, the term “remote sensing” was adopted MEMORIAL SYMPOSIUM when other parts of the electromagnetic spectrum were used. Of course, geophysics had been developed earlier for other wave bands but, in this paper, such techniques are not considered as remote sensing (Gregory, 1972). During the past few years, space platforms have added new dimensions to the remote sensing of geology (Gregory and Moore, 1976). The late Professor H. L. Cameron was probably the first Canadian geologist to demonstrate the pos- sibilities of synoptic, small-scale photography and other novel methods of remote sensing. By the early 1960‘s, he was using stereoscopic, time—lapse methods to study early space photography for the National Aeronautics and Space Administration (NASA). In 1963, he briefly reviewed his work in a paper to the First Seminar on Air Photo Interpretation in the De- velopment of Canada (Cameron. 1964). At the same seminar, W. A. Fischer of the USGS reviewed his pioneer work in geological remote sensing and empha- sized the broadening spectral capabilities (Fischer, 1964). The enthusiastic but practical presentations of these two specialists were responsible for awakening the interest of Canadian geoscientists to the potential of remote sensing in all its aspects. The combination of that seminar with personal experiences in geophysics and photogeology lead the authors of this paper to initiate remote-sensing projects in 1964 and eventually to establish a remote-sensing section within the Geo- physics Division of the Geological Survey of Canada. By early 1967, serious consideration was being given to possible Canadian involvement with the re- source satellites that were being planned in the United States by the Department of the Interior and NASA. In 1968, one of us (LWM) initiated planning for Ca- nadian participation in remote sensing while the other (AFG) reviewed contemporary planning in the United States. It is interesting to note that at that time, geo- logical applications of satellite data were expected to be significant, especially for structure and geomorpho- logy, although of lesser importance than applications related to land use, forestry, surface water, agricul- ture, etc. Training, research, and development of applications were considered to be prime needs (Gregory and Morley, 1969). Following a national review and recommendations presented in 1971, the Canada Centre for Remote Sensing was established with Dr. Morley as founding director. In 1972, he also became chairman of the Canadian Advisory Committee on Remote Sensing while Dr. Gregory was appointed chairman of that committee’s Working Group on Geoscience. AN OVERVIEW OF CANADIAN PROGRESS 35 CANADIAN EXPECTATIONS FOR REMOTE SENSING OF GEOLOGY BY SATELLITES Timely acquisition and interpretation of data from Landsat—1 was recommended by the previous Work- ing Group on Geology (1971, p. 10). However, that working group reported the main geological need to be for airborne sensing, particularly aerial photogra- phy. It was concluded that for use in Canada, satellite data “will provide valuable new information about the surficial geology of Canada, which is incompletely mapped at any scale. However, these data will add only supplementary information on gross structure and bedrock geology, most of which has been mapped at scales smaller than 1:250,000 and much at larger scales” (ibid, p. 11). It was further concluded that “there are few geologic requirements for continuous repetitive surveys” (ibid, p. 11), atlhough their use was foreseen for seasonal and temporal analyses, e.g., studies of terrain, erosion, and sedimentation. Because of the limited availability of Landsat data during the year following launch of the satellite on July 23, 1972, expectations of Canadian geoscientists changed little. Most geoscientists continued to view Landsat as very experimental. However, a few geo- logists used standard photogeologic techniques on Landsat images and reported potential, low-cost ap— plications for mapping structural geology, particularly linears, and for planning exploration projects. Season- al enhancements of geological features suggested that the original expectation of little requirement for repeti- tive data was incorrect (Working Group on Geo- science, 1972 and 1973). The advent of Landsat fired unrealistic geological expectations, especially among nongeoscientists, both in Canada and abroad (Gregory, 1973, p. 88). How- ever, practicality grew over the following year as relevant results of Landsat analyses became avail- able. USERS OF CANADIAN LANDSAT DATA BASIS OF ANALYSIS The following analyses of Canadian users of Landsat data are based on three sets of data: (1) The current Canada Centre for Remote Sensing (National Air Photo Library (CCRS/NAPL) computer listing of sales of standard Landsat products for the 3 years ending September 15, 1975; (2) returns from a ques- tionnaire on the geological use of Landsat data which was mailed in early August 1975; and (3) the current CCRS computer listing of scientists and engineers with an interest in remote sensing, as of August 22, 1975. NUMBER OF USERS As of August 1975, 3307 separate persons and or- ganizations had registered an interest in Canadian remote sensing by returning a report to the Canada Centre for Remote Sensing. While there is, undoubted- ly, some overlap in representation between individuals and organizations, this number reflects a solid base of participation in remote sensing in Canada. Among the 3307 listings, there was much greater interest in remote sensing applied to the geosciences (1078) than to any other single scientific discipline or practice. A breakdown of user interest (table 1) shows that range of interests as well as the overlap in them. Contacts (i.e., letters, visits, and telephoned enquiries) at the User Services Section of the Canada Centre for Remote Sensing confirm a continuing major in- terest in remote sensing applied to the geosciences (30—40 percent of all contacts each month). Major and more-or-less equal interest was ex- pressed by representatives of companies, federal and provincial governments, and educational institutions (table 2). SALES OF CANADIAN LANDSAT DATA Interest is not a very satisfactory way of defining users of Landsat data. Another current indication of use, which is hardly more precise, is a cumulative list of customers who purchased Landsat data through the National Air Photo Library (NAPL). On this list of 536 customers, 205 (38 percent) could be identified as having a major geological base, i.e., a geological TABLE 1.—Relative interest in remote sensing by user discipline, August 1975 Category Number Percent Total Interest ________________________ 3307 100 Agriculture (including Soils) ____________ 694 20 Atmospheric Sciences _________________ 348 10 Cartography/Photogrammetry ___________ 643 19 Fisheries _____________________________ 373 11 Forestry _____________________________ 697 21 Geography ___________________________ 674 20 Geology _____________________________ 1078 32 Glaciology/Ice Reconnaissance __________ 389 11 Hydrology ___________________________ 616 18 Oceanography/Marine Science __________ 414 12 Wildlife/Livestock _____________________ 461 13 NOTE: Sums exceed listed totals because of multiple interests expressed by a single individual or organization. 36 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM TABLE 2.—Relative interest in remote sensing by economic sector, August 1975 Category Number Percent Companies ___________________________ 752 26 Educational Institutions ________________ 692 24 Federal Government Agencies __________ 664 23 Provincial Government Agencies ________ 539 18 Foreign Government Agencies _________ 124 4 Private Individuals ____________________ 110 3 Municipal Government Agencies _______ 29 1 Regional Government Agencies ________ 20 1 International Organizations ____________ 7 0 Total __________________________ 2937 100 TABLE 3.—Net sales 0) standard Landsat products for 3 years ending September 15, 1975 Product Total net Net geological Percent sales sales geol. B & W prints ____________ 60,016 32,506 54.2 Colour prints ____________ 7,209 1,925 26.7 B & W transparencies __s 18,354 1,471 8.0 Colour transparencies ____ 3,185 531 16.7 Total _____________ 88,764 36,433 41.0 NOTE: These figures represent net sales external to the National Air Photo Library and the Canada Centre for Remote Sensing. They do not include images of Canadian terrain sold by EROS (US. Geological Survey) in Sioux Falls, S.Dak. or by ISIS (a private company) in Prince Albert, Saskatche- wan. It is possible that these latter sales are equal to or greater than sales reported here. survey, an oil or mining company, or a consultant. The next largest identifiable groups of users were for— esters (45), library and information services (41), and geographers (39), each with 7 to 8 percent of the number of customers. However, there is a large group of individuals and otherwise unidentifiable purchasers (155 or 29 percent), which was not amenable to classi- fication. A semiquantitative estimate of the number of geo- logical users was made from a computer printout of net sales (i.e., excluding images generated for internal use at the CCRS and NAPL) of standard Landsat products* in Canada up to September 15, 1975 (table 3). This analysis indicates that 41 percent of net sales were made to identified geological users. How- ever, there were uncertainties in classifying customers as nongeological; hence the statistics are minimal although the trends are real. * Standard Landsat products are 9 X 9 inch prints and trans- parencies, both colour and black and white. CLASSES OF GEOLOGICAL USERS A precise breakdown of purchasers by subdiscipline within the geosciences is difficult to achieve. How- ever, the results of a questionnaire prepared in sup- port of this paper permit a semiquantitative analysis. Of the 173 questionnaires sent to selected purchasers with an identifiable geoscience interest, 70 (41 per- cent) were returned by December 31, 1975. These returns (table 4) show that mineral exploration com- panies (45 or 64 percent) are the largest users of Landsat data among Canadian geological users, excluding the major collection in the Geological Survey of Canada. The industrial sector was represented by returns from 52 companies and consultants. Of these, 4 com- panies concerned with oil and gas exploration, 1 company concerned with mining exploration, and 3 consulting companies reported that they each analyze over 100 Landsat scenes per year. One company in each class reported using over 500 scenes per year. Indeed, these 8 major users, by their own reports, appear to use more images annually than all other 62 users combined. The computer listing of sales by NAPL (table 3) provides another means of assessing the classes of purchasers within the geoscience community. Of the 36,433 standard products sold to recognized geo- logical users (table 5), about half (18,022) comprise black and white prints in a national collection at the Geological Survey of Canada. The balance (18,411 standard products) was sold to the various users noted previously. The figures in table 5 confirm the conclu- sion that in terms of number of users, number of images purchased, and number of scenes analyzed, the mineral exploration companies are by far the largest users of Canadian Landsat data for geological applications. Similar companies also appear to be major users of Landsat data in the United States (Sabins, 1974) and Australia (UK. Remote Sensing Society Newsletter No. 7, p. 14, Nov. 1975). RANGE OF EXPERIENCE Of the 70 users returning questionnaires, 31 per- cent (22) started using satellite data in 1972, immedi- ately after the launch of Landsat—1; 40 percent (28) began their use in 1973; 23 percent (16) in 1974; and only 6 percent (4) started in 1975. Of these same users with experience in Canada, 24 percent (17) have experience with Landsat data for other parts of North America, and 21 percent (15) have similar experience for other parts of the world. AN OVERVIEW OF CANADIAN PROGRESS 37 TABLE 4.—Geo/ogical users of Landsat data, 1975 Annual use (scenes per year) Un- Class 10 10—50 50—100 100—500 500 known Total Geological survey and research: (a) Gov’ts ______________ 1 3 -_ __ -_ 7 (b) Universities ________ 4 4 1 __ __ __ 9 Mineral exploration: (a) Oil and gas ________ 3 3 3 3 1 1 14 (b) Ore _______________ 7 12 __ 1 1 28 (c) Combined __________ __ 2 1 __ __ __ 3 Consultants and contractors _____________ 1 3 __ 2 1 _ 7 Not specified __________ __ 1 1 __ _ _ 2 Total _________________ 16 28 16 5 3 2 70 TABLE 5.——Net sales of standard Landsat products to recognized geological users [Excluding a national collection at C.S.C.] Percent of total net Class BWPRT" CPRT" BWTRA" CTRA" Total products Oil and gas exploration companies _______________ 3448 164 414 200 4226 22.9 Mining exploration companies _______________ 4020 486 229 85 4820 26.2 Consulting and contracting companies _______________ 2192 154 187 95 2628 14.3 Federal agencies ____________ 2463 755 244 72 3534 19.2 Provincial agencies _________ 1373 323 309 65 2070 11.2 Academic institutions _______ 941 37 88 12 1078 5.9 Other _____________________ 47 6 ____ 2 55 0.3 Net Products _______________ 14,484 1925 1471 531 18,411 100 National collection, C.S.C. _____________________________________ 18,022 Total products sold to recognized geological users: ________________ 36,433— * BWPRT: black and white prints. CPRT= colour prints. BWTRA: black and white transparencies. CTRA=co|our transparencies. LEVEL OF CANADIAN TECHNOLOGY FOR GEOLOGICAL INTERPRETATION Most users (93 percent of the 70 respondents) com- monly use visual interpretation based on standard photogeologic techniques. The remainder reported rare or nil use of such techniques. Enlargements and projections are commonly used for interpretation at scales as great as 12,250,000. Two major users and seven smaller users reported working at scales as large as 150,000. Thirty-six percent (25) of the users reported that they had tried seasonal enhancements, but only 13 percent (9) noted that they did so com- monly. Ten users (14 percent) reported using machine assistance, 6 did so commonly. Comparable figures for digital processing are 6 and 1. Of the seven users reporting common use of machine and digital assist- ance, only one purchases more than 100 scenes per year. 38 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM Most users (62 or 88 percent) use black and white prints while about half have used colour prints (37), enlarged prints (38), or black and white transparencies (32). The strong preference by geological users for black and white prints extends to all classes of such users (table 5). Few users (<1 percent) have worked with computer compatible tapes (CCT’s) or enlarged transparencies, although 70 mm transparencies have been more widely used (21 percent). All four MSS bands and colour composites are the preferred ma- terials for 40 percent (28) of the users while the re- mainder prefer band 6 either alone or in conjunction with band 5 and/or band 7. Band 4 receives less use although it is ordered frequently. In summary, the analytical technique in common use among geologists at this time appears to be visual interpretation of black and white prints. Very little research seems to be directed to machine and com- puter assistance for geological applications, although there are a few notable exceptions. CURRENT GEOLOGICAL USES OF LANDSAT DATA GENERAL Interpretation, or the extraction of information about the surface of the Earth, is the prime use of Landsat data. Such use may be in support of re- search to increase scientific knowledge or in support of practical applications to meet economic and social goals (Gregory and Moore, 1976). Practical applica- tions are commonly assumed to have a high probability for imminent return on investment. The distinction is judgmental and lacks clarity. Other uses, such as reference collections, illustrations, educational dis— plays, and works of art have not been assessed, although the first, in particular, requires large num- bers of images. During the past 3 years, much of the use of Landsat data fell into the category of research as Canadian geologists strove to define potential applications, par- ticularly for use in geological mapping. However, only 13 percent of the respondents to the 1975 question- naire considered that they were conducting research currently. Hence there has been a steady growth in the practical application of Landsat data. Accordingly, we have attempted to sort and outline the various Canadian uses of Landsat data under the two cate- gories of research and practical application. RESEARCH Currently available results of research show a primary concern with the development of techniques to identify and map geological structures and the dis- position of major classes of rock. Standard photogeo- logic techniques were extended to the specific geometry and data for Landsat, with emphasis on texture and pattern. Thus, early visual studies estab- lished that many Canadian geological provinces have characteristic bulk textures and that geological linea- ments and structures may be well expressed in Landsat data (Gregory and Moore, 1975 and 1976; Palabekliroglu, 1974a and b; Moore and Gregory, 1974; and Slaney, 1974). Similar analyses suggested practi- cal applications in hydrogeology (Ozoray, 1974). The utility of temporal (or seasonal) analyses was demon- strated by Moore and Gregory (1974). Optical Fourier spectra were used by Arsenault and others (1974) to select and enhance linear features. In addition, they showed that holographic filtering could be used to detect simple characteristic shapes. Spectral discrimination for geological purposes does not appear to have received adequate research in Canada, at least outside the bounds of commercial proprietary knowledge. Spectral discrimination of vegetational anomalies and possible applications in mineral exploration have been considered (see, for example, Working Group on Geoscience, 1974), but to date no success or systematic use has been con- firmed. However, a most significant development has been reported in the spectral discrimination of terrain classes in the Canadian Arctic (Boydell, 1974, and subsequent personal communication). This evaluation shows that multispectral analysis of Landsat data, using CCT‘s and the Bendix MAD unit, could comprise a potentially more rapid and relatively accurate al- ternative to the conventional methods of preparing preliminary terrain maps by airphoto analysis. At the present stage of development, the validity of terrain classes is dependent upon detailed knowledge of the terrain and is not absolute, thus hindering general extrapolation and regional mapping. This research, however, is being continued by the Terrain Sciences Division of the Geological Survey of Canada. A strong interest in ice reconnaissance was indi- cated by major users of Landsat data in the oil and gas industry. Attendant research has focussed on breakup and freezeup patterns in the Beaufort Sea and on the measurement of fioe size and rates of movement during those periods. PRACTICAL APPLICATIONS No practical applications of Canadian Landsat data are currently documented in the literature. This re- fiects, in part, the newness of such applications in that case histories are still subject to field investigation AN OVERVIEW OF CANADIAN PROGRESS 39 and, in part, the common commercial requirement for protection of proprietary rights with respect to meth- odology and results. The questionnaire previously mentioned, however, shows that Landsat images are being widely, intensively, and systematically used by mineral exploration companies and consultants. The practical objectives of such applications are relatively simple, as revealed by respondents to that question- naire. A regional overview of geology is the prime requisite of 86 percent of the users, 67 percent are particularly interested in analyzing linears, and 41 percent in the selection of targets for more detailed investigation (c.f., Sabins, 1974; Collins and others, 1975, for US.) Secondary objectives included spectral discrimination of rock types and alterations (27 per- cent), which surely was mainly experimental, and the use of images as base maps for planning and man- aging exploration projects (26 percent). About 13 per- cent of the respondents reported using Landsat images for each of the following: recording environment in the vicinity of producing sites, regional overview of logistical factors, and classification of materials. The principal information derived from the Landsat data comprises: geological structures (66 percent), dispo- sition and classification of rock units (46 percent), and surficial geology (34 percent). Other types of informa- tion, each reported by less than 10 percent of the respondents, include information related to soil type, erosion and sedimentation, terrain classes, surface water, land use, construction aggregate, ground water, natural disasters, and waste disposal. Other interest- ing uses, each reported by one respondent, were: spectral discrimination of zones of subsidence and correlation with aeromagnetic data. While published case histories are lacking at pres- ent, there are several Canadian projects to which Landsat data are known to have made major or prime contributions. For example, they have been used to assess the structural framework of metallogenic provinces (linear analyses), to define structural and lithologic targets for diamond exploration in Africa and Canada (linear and spectral analyses), to plan alternate routes for a pipeline (mapping of well- drained ridges formed by glacial fiuting), to classify mine wastes at a scale of 1:50,000 (spectral analyses), and for geological mapping to locate rock types, folds, and faults of importance in oil and gas exploration. In addition, as noted previously, there are active projects for mapping ice conditions related to oil and gas ex- ploration in the Beaufort Sea and for forecasting snow and ice cover to assist in planning mineral ex— ploration in northern Canada. BENEFITS FROM USE OF LANDSAT DATA Respondents to the questionnaire reported that the major benefits which they recognized from their use of Landsat data were: 1. Acquisition of geological information that is not readily available otherwise (64 percent); 2. Acquisition of supporting information to be inte- grated with other data (53 percent); and 3. Saving in time in assessing the regional geology (48 percent). About 20 percent of the respondents reported that their use of Landsat data had resulted in a lower cost for assessing regional geology, had assisted in reduc- tion of area for detailed study, and had assisted in mineral prospecting. Unfortunately, from a scientific viewpoint, a successful application of Landsat data in mineral exploration is least likely to be disclosed publicly. About 14 percent of the respondents reported receiving additional nongeologic information that was not readily available otherwise. Ninety-three percent of the respondents reported that their use of Landsat data had contributed to their geoscience programs; 43 percent (30) considered the contribution to have been modest; 8 percent (5) con- sidered it large, while 43 percent (30) considered it minimal; and 7 percent did not answer. Specific cost savings could not be identified by 41 percent (29), while 13 percent (9) consider there was no measure- able saving, and 46 percent (32) did not answer the question. These responses about benefits confirm the view that Landsat data, in conjunction with other types of data, are employed extensively in the regional reconnaissance which precedes more expensive, de- tailed methods of exploration. As is well known, many different geological, geophysical, and geochemical techniques are employed in compiling the information that eventually leads to a mineral discovery. Thus, it is simplistic and unrealistic to credit a discovery to a single methodology. It is equally unrealistic to expect an identifiable cost benefit to be assigned to a specific use of Landsat data. However, it is significant that 50 percent of the respondents, including 6 of the 8 major users, claim a modest to large benefit from their use of Landsat data. FORECAST OF TRENDS The utilization of Landsat data will continue to fol- low the typical use curve that describes the temporal development of new technology. With respect to visual photogeologic interpretation, the use of Landsat data has rapidly risen to “Panacea Peak," fallen into 40 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM “Sanity Slump,” and is now beginning to climb “Reality Rise” toward the “Plateau of Practicality.” However, the geological goals of practical spectral discrimination and automated processing have yet to be defined, and regular low-cost production has not been achieved. Further research in these latter fields will be hindered by the fundamental limitations of surficial obscuration (lichens, vegetation, and overburden) and the lack of the third dimension, depth. Such limitations suggest that major support for this research must be obtained from organizations concerned with surficial mapping (e.g., governments and universities) rather than from mineral exploration companies which have equal or greater concerns with depth. Nevertheless, there are compositional contrasts of geoscience in— terest (e.g., gossans, alteration zones, and vegetational anomalies) that can serve to focus further research by means of spectral and automated analyses. The subsequent development of practical applications will be guided by the fact that these techniques, like all other methods of exploration, will not be specific but, rather, will undoubtedly present a multiplicity of anomalies for field checking. Visual interpretation of Landsat images will con- tinue to receive increasing acceptance as a tool, in conjunction with others, for developing concepts to guide mineral exploration and geological mapping, both in Canada and abroad. Landsat images will soon become a prime tool for regional geological recon- naissance, especially for poorly explored, arid and semiarid areas. This will probably become true even if Landsat were to fail in the near future. However, much additional experience must be accumulated be- fore geologists can confidently associate mineral de- posits with features on Landsat images. In particular, the ubiquitous linears will require classification and analysis before their geological significance can be fully understood. In this growth of practical applica- tions, transparencies will be substituted for prints in laboratory analyses because of the inherently greater resolution; however, prints will be preferred for anno— tation and field studies. Improved instrumentation will be developed to assist such analyses. Spectral discrimination and automated processing will evolve very slowly until substantial funding and dedicated programs are developed in support of geo- logical applications. A major and practical advance in terrain classification by automated methods appears imminent. However, because of complex interrela— tionships between bedrock, soil, water, and vegeta- tion, automated pattern recognition will remain unattainable for all but the simplest geological goals (Gregory, 1973, p. 89; Sabins, 1974) and even there may be too expensive for practical application. While Landsat data have the advantage of uniform raked illumination, minimum geometric distortion, synoptic view, and worldwide coverage, it is primarily the related low cost of purchase and analysis that has led to their widespread adoption for regional reconnais- sance. Spectral discrimination and automated process- ing will not be adopted for regular use until their current high costs are reduced. Indeed, as Kite (1975) has noted, it remains to be shown that such methods will be significant for mineral exploration and geo- logical mapping. Perhaps the most useful future development for geoscientists will simply be an in- crease in resolution to 2 or 3 times that currently attained with Landsat. CONCLUSIONS Canadian interest in Landsat data is widely spread and more—or—less equally divided among companies, educational institutions, and government agencies. Geoscientists have had a long, continuing, and ma— jor involvement with the Canadian Landsat program. However, at present only about 10 percent of the geoscientists have a strong commitment to the activity. Current data defining sales and use of Landsat data show that identified geoscience organizations are major, if not the principal, users of Landsat data. Excluding reference collections, mineral exploration companies and consultants are the principal users. Most (71 percent) geological users of Landsat data have been using such data since 1973 at least; the remainder started more recently but primarily in 1974. Nearly a quarter of the Canadian users have experience with Landsat data for other parts of the world. Visual photogeologic interpretation of black and white prints comprises the principal Canadian method of analysis. Digital processing and machine assistance have not really been assessed for geological applica- tions, and little relevant research has been reported. Most Canadian research has been directed to the development of visual techniques to meet specific requirements for the analysis of texture and patterns in Landsat images. However, there has been a major advance in the digital classification of Arctic terrain. Practical applications are not yet documented in the literature. However, known examples are related primarily to reconnaissance for geological structure and disposition of surficial materials. Half of the geo- logical users of Landsat data claim a modest to large benefit from such use but cannot identify a specific dollar benefit. REFERENCES 41 In view of the major geoscience involvement, in- creased support is warranted for research into digital and automated methods of processing Landsat data to meet geological objectives. In particular, this should include temporal change detection, spectral dis— crimination and ratioing, and directional filtering, if warranted by developments in linear analysis. The level of development of practical geologic ap- plications in Canada is greater than predicted in a pre—Landsat forecast. Visual interpretation of Landsat images will soon become a prime tool, in conjunction with others for regional exploration, particularly in arid regions. Automated processing will remain in the research stage until low-cost methodologies are de- veloped to meet specific geological objectives. Because the specific Canadian context comprises a large, well-mapped terrain, the conclusions reached here should be extrapolated with caution to other different and less-explored terrains. REFERENCES Arsenault, H. H., Sequin, M. K., Brousseau, N., and April, G., 1974, Le traitment optique des Images ERTS: Canadian Symposium on Remote Sensing, 2d, Guelph 1975, Proc., v. 2, pp. 488—493. Boydell, A. N., 1974, Evaluation of the potential uses of Earth Resources Technology Satellite (ERTS—1) data for small scale terrain mapping in Canada’s North: Internat. Soc. Photogramm. Comm. VII, Symposium on Remote Sensing of Environment, Alberta 1974, Proc., pp. 329—340. Cameron, H. L., 1964, Photo interpretation in geo— technical investigations: Canada Interdept. Comm. on Aerial Surveys, Seminar on Airphoto Interpretation in the Development of Canada, Proc., pt. 4, pp. 10—20. Collins, R. J., Petzel, G. J, and Everett, J. R., 1975, An evaluation of ERTS data for petroleum ex« ploration (abs): University of Kansas Space Technology Centre, Case History Research Con- ference on Remote Sensing, Lawrence, Kansas, pp. 8-10. Fischer, W. A., 1964, Geological interpretation from airphotos: Canada Interdept. Comm. on Aerial Surveys, Seminar on Airphoto Interpretation in the Development of Canada, Proc., pt. 4, pp. 21—31. Gregory, A. F., 1972, What do we mean by remote sensing?: Canadian Symposium on Remote Sens- ing, 1st, Ottawa 1972, Proc., pp. 33—37. 1973, A possible Canadian role in future global remote sensing: Can. Aeronautics and Space Jour., v. 19, no. 3, pp. 85—92. Gregory, A. F., and Moore, H. D., 1975, The role of remote sensing in mineral exploration with spe- cial reference to ERTS—1: Canadian Inst. of Mining and Metallurgy Bull, v. 68, no. 757, pp. 67—72. 1976, Recent advances in geologic applications of remote sensing from space: Internat. Astro- nautical Fed. Astronautical Congress, 24th, Baku 1973, Astronautical Research: 1973, pp. 1—18. Gregory, A. F., and Morley, L. W., 1969, Remote sensing and the solid-earth science, in Back— ground papers on the Earth Sciences in Canada: Geol. Survey of Canada Paper 69—56, pp. 170-— 179. Kite, R. L., 1975, Research on techniques for photo- graphic enhancement of ERTS imagery for geo- logic applications (abs.): University of Kansas Space Technology Centre, Case History Confer- ence on Remote Sensing, Lawrence, Kansas, pp. 28—30. Moore, H. D., and Gregory, A. F., 1974, Temporal analyses of ERTS—1 images for forest and tundra and their significance in visual interpretation of geology: Canadian Symposium on Remote Sens- ing, 2d, Guelph 1975, Proc., v. 1, pp. 47—58. Morley, L. W., 1971, Canada’s approach to remote sensing: Internat. Symposium on Remote Sensing of the Environment, 7th, Ann Arbor, Mich. 1971, Proc., v. 1, pp. 3—18. Ozoray, G., 1974, Lineament analysis using ERTS—1 images of Alberta: Internat. Water Resources Assoc. Internat. Seminar and Exposition on Water Resources Instrumentation, Chicago 1974, Water Resources Instrumentation, pp. 456—466. Palabekiroglu, S., 1974a, The value of ERTS-1 imagery for mineral exploration: Canadian Sym- posium on Remote Sensing, 2d, Guelph 1975, Proc., v. 2, pp. 464—470. 1974b, ERTS—1 imagery: The valuable tool of geoscientists: Internat. Soc. Photogramm. Comm. IV Symposium on Remote Sensing and Photo Interpretation, Proc., pp. 597—609. Sabins, F. F., Jr., 1974, Oil exploration needs for digital processing of imagery: Photogramm. En- gineering, v. 40, no. 10, pp. 1197—1200. Slaney, V. R., 1974, Satellite imagery applied to earth science in Canada: Internat. Soc. Photogramm. Comm. IV Symposium on Remote Sensing and Photo Interpretation, Proc., pp. 555—572. 42 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM Woolnough, D. F., 1975, Theoretical accuracies of space mapping for Canada: The Canadian Sur- veyor, v. 29, no. 3, pp. 267—276. Working Group on Cartographic and Photogrammetry, 1974, Report of the Working Group on Cartogra— phy and Photogrammetry, in Canadian Advisory Committee on Remote Sensing, 1974 Report: pp. 48—49. Working Group on Geology, 1971, Resource satellites and remote airborne sensing for Canada: Report No. 6, Geology: Ottawa, Dept. of Energy, Mines and Resources. Working Group on Geoscience, 1972, Report of the Working Group on Geoscience, in Canadian Ad- visory Committee on Remote Sensing, 1972 Re- port: Ottawa, Dept. of Energy, Mines and Resources, pp. 77—80. 1973, Report of the Working Group on Geo- science, in Canadian Advisory Committee on Remote Sensing, 1973 Report: Ottawa, Dept. of Energy, Mines and Resources, pp. 72—83. 1974, Report of the Working Group on Geo- science, in Canadian Advisory Committee on Remote Sensing, 1974 Report: p. 85. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Mapping and Charting from Landsat By Alden P. Colvocoresses, US. Geological Survey, Reston, Virginia 22092 INTRODUCTION Although I once was a mining engineer, I have been a mapmaker for the past 25 years, and my work is no more related to the oil and mining industry than it is to any other map-using group. Mapmakers, as well as others, now have a satellite in orbit known as Landsat—previously known as ERTS (Earth Resources Technology Satellitel—by which we can map the Earth at the smaller scales in image form. Since there is a wealth of written data available, I have no intention of trying to cover Landsat mapping in any detail. Those of you who want more information are invited to drop me a note, or see me personally, and I will be happy to respond. A few of the more pertinent references are appended to this paper. To me, mapping is fundamental to man’s efforts to understand and live on this planet, and as persons concerned with exploration for petroleum and other mineral resources, I am sure you will agree. But what and how well can we map with Landsat? By the use of photographs I will illustrate some of the things that can be done and leave it to you to decide to what extent Landsat mapping can be applied to your spe- cific problems. LIMITATIONS First, let me cover what we cannot do with Landsat. Objects too small to be recorded by the 79-m (260-ft) Landsat picture element cannot be seen on the imagery. Although the capture of specific detail is affected by contrast and shape, most features smaller than 8 or 10 acres, unless they are very high contrast or linear features such as roads and canals, will be lost. Second, we cannot use Landsat for topographic mapping because the system is basically orthographic, precluding the determination of elevations. Where suitable topographic data exist, a stereopair portray- ing topography can be processed from Landsat, and such a pair is, in fact, available to you as a handout. However, remember that the elevation data used to form the model did not come from Landsat. UNIQUE CHARACTERISTICS Most of you are familiar with aerial photography, so let’s look at the unique advantages of Landsat imagery as compared with aerial photographs. Continuity—Can you imagine covering the Earth every 18 days with photography? I can’t, but—subject to cloud cover, priorities, and transmission capabilities —Landsat does this, and moreover Landsat—1 has done it for more than 3 years. Just look at the pub- lished index maps showing only 2 years of Landsat coverage. Even so, we have a long way to go to ade- quately image the Earth for mapping and related purposes. Near real time—Most of the things you are looking for are fixed, but sooner or later you are going to want to see the results of a significant event. If it is local, an air photo is the way to go, but if it is wide- spread or in an isolated area, Landsat can help as figure 1 illustrates. This huge, seasonally dry area is shown here inundated, and the effects of this inun- dation will remain for several months. I would hate to have to map this changing condition with aerial photographs. Geometric fidelity—Landsat produces near-ortho- graphic imagery of high geometric fidelity, and image maps with accuracy compatible with the 1:250,000 scale can be produced. Relief displacement is min- imal, and the mapping is relatively cheap and simple as compared with mapping from aerial photographs. Figure 2 shows where we have actually corrected the standard 1:250,000-scale line map in the Phoenix area. 43 44 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM O ERT IMAGERY or m. s, 1974 50 MILES 80 KILOMETERS FIGURE ”Ir—Map of Cooper Creek, Australia, and ERTS imagery of the creek in flood. Landsat is revealing many small— and medium—scale map errors and omissions in various parts of the world, and corrections and revisions based on Landsat are being applied by several mapping agencies. The Canadians, in particular, utilize Landsat for map revi- sion—even up to the scale of 1:50.000 in provisional form. The World Bank is finding Landsat imagery to be far more accurate than available maps in many areas of the world where they have projects. Multispectral compatibility—On aerial film we re- cord a broad waveband in black and white or at the most three narrower discrete wavebands in color. Landsat now records four discrete wavebands as digital radiometric values, including a band beyond the range of film cameras. Band 7 of the multispectral scanner (MSS) records 0.8 to 1.1 [um whereas aerial film cuts off at 0.8 ,im. Look at the advantages of this near-infrared waveband (figs. 3 and 4). Landsat-C will also have a thermal band, which again is beyond the range of aerial films. The combining of waveband re- sponses either by digital or photographic methods is, of course, spectacular, and Larry Rowan (USGS) will show how combinations can be applied to bring out or emphasize features of specific geologic interest. Suitability for automation—Landsat output is digit— al as well as continuous. Coupled with its geometric fidelity it has the potential of becoming an automated mapping system. The imagery is now cast on a defined map projection which we caII the Space Oblique Mercator (fig. 5). As mentioned, Landsat also covers the same basic area every 18 days and thus forms a map series based on the nominal scenes covered (fig. MAPPING AND CHARTING FROM LANDSAT 45 FIGURE 2.—Map accuracy improved by Landsat imagery. Map detail (planimetry) was moved about 1 mm (0.04 in.) on this 1:250,000—scale map of Phoenix to conform with the detail on the Landsat image. The image was found to be correct. Visible spectrum Near Infrared MSS Band 5 (0.6 . 0.7 pm) MSS Band 7 (0.8 - Hum) FIGURE 3.—Thin cIoud penetration capability of ERTS IR sensor, southeast Pennsylvania, November 16, 1972‘ 46 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM MSS BAND 4 IOJ -0.6pm) MSS BAND 6 0.7-0.8 pm FIGURE 4.-——~Water MSS BAND 5 inc-0.7,...” MSS BAND ‘03-1‘1 ”mi boundary delineation from ERTS, Laguna Salada, northern Mexico. 7 MAPPING AND CHARTING FROM ‘LANDSAT 47 SCANNER SATELLITE 3.‘ y‘ \ YANGENT ‘ EARTH ORBIT PRECESSION Images the Earth Iram N 82" lo S 32" every la days 0 Scanner sweep 0 Earth rotation MOTIONS INVOLVED 0 Satellite orbit o Orbit precession FIGURE 5.—Space Oblique Mercator projection. 6). These characteristics mean that Landsat image maps can be made in a few days time when needed. Making maps from aerial photographs, as you know, is a matter of months or even years. Currently, total time involved is about 5 years per standard map in the Geological Survey. CARTOGRAPHIC APPLICATIONS Now let’s look at what Landsat can do cartographi— cally. Obviously, it can portray topography even though it can’t define the contours. Moreover, it portrays topography under different conditions of seasons. Let’s look at the Denver area (fig. 7, p. VI). Note how differently the topography, as well as other features, are recorded. Or let’s look at Lac Manicouagan, Canada (fig. 8). Note how the lower Sun angle en- hances the topography and how different the same features appear. Seasonal differences are obvious, but even in the same season and with the same lighting we should look at any given area at least two or three times so that fixed features can be separated from temporal phenomena and anomalies in image processing. Under the existing Landsat program a minimum of 10 years is needed to obtain suitable. coverage for the definitive mapping of even the Earth‘s fixed features. Of course, if long-term temporal and seasonal changes are to be mapped, a remote-sensing system such as Landsat must be extended indefinitely—and I trust that it will. The word Landsat implies land coverage, and the satellite was not designed to tell us much about the oceans. But how about coastal areas and particularly the shallow seas? I know two things about these shallow seas—they are extensive and they are very poorly mapped. From what I hear, they are also of ever increasing economic importance. Landsat’s water-penetration capability is 10 to 20 m (33 to 66 ft) even under ideal conditions, but we have asked the National Aeronautics and Space Administration to adjust one MSS channel on Landsat—C which should increase this capability. I realize that most of the con- tinental shelves are covered by water deeper than 20 m, but providing data even down to this depth would, I believe, be highly significant. Let’s look at what can now be done with Landsat in the shallow seas (fig 9). The published hydrographic chart obviously needs revision. In the Caribbean, figure 10 shows Landsat response related to depth over a well-charted bank, and nearer to home (fig. 11, p. VII). For a variety of reasons this application is pretty well limited to the mid and lower latitudes. In the high latitudes, I suspect that the monitoring and mapping of sea ice may be the most important application (other than land area analysis) of concern to the oil industry. In Canada, a quick-look facility which trans- mits Landsat data directly to users in a matter of a day or so after reception has been developed and put to use. On Landsat—C, the thermal channel could double the frequency of observations as compared to Landsat—1 and -2 because useful observations can also be obtained at night. Also note that the meteoro- logical satellites of National Oceanic and Atmos— pheric Administration (NCAA) and the Department of Defense (DOD) Defense Meteorological Satellite Program provides some excellent data on sea ice. Once we coordinate data from the three satellite sys- tems of NASA, NCAA, and DOD, effective surveil- lance of sea ice can become a reality. Landsat is not designed for atmospheric studies, but it will certainly record sizable smoke plumes from gas flares. The DOD weather satellite I previous- ly mentioned can clearly record nighttime illumination, which includes the gas flares themselves (figs. 12 and 13). Note what appear to be flares in Northwest Canada. If they are not gas flares, I hope one of you may be able. to offer some other explanation. The bright spots certainly correlate with the oil and gas fields of Alberta and British Columbia. A recent Canadian report indicates over 2.5 billion cubic feet of gas per month are being flared in Alberta alone. According to one authority (Library of Congress) more petroleum products are being burned up in 48 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM ALABAMA min: I O‘KA A I!” I APPROXIMATE SCALE 50 0 50 Li I I l I l I . H I I I I 50 0 50 100 KILOMETERS 100 MILES l H‘SSAU ’14 «b ‘5‘ ,,, 9K ALACHgfi LEVY J- O ~2848W07952 guns EL 2 é’RQWARD , \ FIGURE 6.—Nomina| ERTS scenes—Florida. flares on a worldwide basis, than are being used pro— ductively. Landsat—1 and —2 cannot record flare illumi- nation, but the thermal channel of Landsat—C should provide some meaningful signatures. Again, by com— bining the data from Landsat and the meteorological satellites, gas flaring can be monitored in both quanti- tative and positional terms on a worldwide basis. Another application of Landsat that I wish to men- tion is the recording of artificially induced signals. The easiest signal to use is nothing more than a solar beam reflected to the satellite by a mirror. The mirror may be as small as half a metre (20 inches) on a side, but it can saturate a Landsat picture element (pixel). Figure 14 shows how three mirrors located in the vicinity of our Reston building were recorded by Landsat. Similar mirrors can be set up at various re- mote stations, including offshore drilling platforms, and we plan to set up such mirrors in.Antarctica dur- ing the next few months. The technique provides a visible position mark for an otherwise nondefined point. Incidentally, using this technique requires con- siderable expertise and close coordination with NASA, MAPPING AND CHARTING FROM LANDSAT 49 OCTOBER JANUARY SUN ElEVATION 31° >4 vascxsicm Piocssssn mack ‘3 ft: H24 ms 22:: Posnmu salon 16.32an mace mm: caimsn 25M74 SUN ELEVATION 15° FIGURE 8.—Lac Manicouagan, Canada. Landsat imagery of a huge flooded impact (meteoritic) crater. but its practical application should be further tested. You may want to contact Mr. Evans (1974) of Stan- ford Research Institute for further information. He is the one who originated this novel technique. The last application is the actual image mapping. Landsat maps are being made for various parts of the world in ever increasing numbers, and in a wide variety of forms as figures 15, 16, 17, and 18 indicate. Florida and Arizona are on public sale by the USGS. The entire conterminous United States has been printed (USGS) in small-scale mosaic form (also on public sale) and anyone who has the need and the money, can now make such a map of practically any land area in the world. Once we justify it, this cover- age can also include the shallow seas. Technical data relative to such map compilation are contained in Chapman (1974), Colvocoresses (1975), McEwen and Asbeck (1975), and McEwen and Schoonmaker (1974). SUMMARY AND CONCLUSION I’ve covered technical aspects of Landsat as it ap- plies to cartography. I feel sure that such a satellite system can in fact map the land and shallow seas of the Earth at scales as large as 1:250,000. However, I ask that you do not take the system for granted. It is a program that has lots of competition for the avail- able space dollars. If you, individually or collectively, feel that the program is worthwhile and should con- tinue, you must say so. My immediate concern is that Landsat may be converted into a specialized satellite to meet some particular problem of immediate con- cern. To me, that would be a mistake. We need basic information about this Earth in a continuous, uniform, and readily usable form. That is what Landsat is pro- viding, and I strongly recommend that it be continued in the same basic form for the foreseeable future. Before we concentrate on a successor to Landsat, we should define the satellite that will immediately follow Landsat—C which may expire by 1979 or 1980. I sug- gest we now work on Landsat~D and perhaps —E, and if they fly in the form established by Landsat—1, —2, and ~C, I for one will not be disappointed. We have got a winner, so let‘s stay with it until we are sure we have something better. FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 50 .oooxoooéur >_SmE_xc.QQm Em imEmcc St 3 373m .m. mbms< 2mm $500 .zQBfiOfi «0:02;; pz ’“V'M‘WLMW "$19" * (9M5?- méash ‘ a ,‘h 1 FIGURE 13.——DMSP image. 54 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE TIL—Enlargement of a portion of ERTS image showingmirror response. Arrows point to pixels illuminated by small mirrors on the ground. 55 MAPPING AND CHARTING FROM LANDSAT _,_L,_ i, #7:”me w0\\/’ / \///‘ W\' '. 4/ L IS POTOSI 00 / 0/ i / I fig X n / / aflL o: suknALuv: FIGURE 2.—StructuraI interpretation of Landsat-1 imagery showing incidence of faults, fractures, mines, and adits. -\' // 0 EL PIIOS ”IA DE IIGELES O BAHIA PETACALCO I/ LIA/«Z . % \f ,- wé IO I BAHIA DE BANDERAS 63 64 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM EPRIETO \ \fi CLARION GULF OF MEXICO F“ 'f FRACTURING Q. t * + \« ”' “a Q ' ewes?“ ‘5 D 02% ‘ v ? Lu (J \ TRENC ‘ ‘ \ c ’7 + 0:] a: 55> vs + K it) v‘fiQ’ ‘2‘ + t , Q, cLipPERTSLN 5 ° 9“ \" ‘ G FRAcTUR‘N . + “4 fa k 9° '\ 3r) 0° ‘45 CRATONIC BLOCKS Lu J“ I JALISCO BLOCK n MICHOACAN BLOCK m GUERRERO BLOCK IV OAXACA BLOCK 0 200400600 Km v CHIAPAS BLOCK + BIGGEST VOLCANOES OF THE QUATERNARY 0 2“ T/ FIGURE 3.——East Pacific Rise and Neovolcanic Axis structure. (After Mooser, 1972.) and Gonzalez, 1967) provides further evidence of a major right lateral transverse displacement along the Neovolcanic Axis. Displacement of the silver provinces of the Sierra Madre Occidental, eastward to match silver provinces along the Guanajuato-Pachuca province agree with Mooser’s 160-km right lateral displacement (fig. 5). The author’s study of fracture trends and evidence from the Landsat imagery, how- ever, does not show continuity of large fractures which could sustain Gastil’s proposal. Undoubtedly, Mooser’s and Gastil’s research are the best efforts to get first—hand information about the Mexican Neovolcanic Axis’ modern geology. Un- fortunately, the observations were apparently based on isolated visits to different Neovolcanic Axis locali- ties with lack of continuity in time or space, so that they may yet require a thorough revision. There are discrepancies in the above-mentioned authors’ results. For instance, within the Neovolcanic Axis Metallogenetic Province, andesite surface flows predominate on its western end, and andesites and rhyolites predominantly appear on the eastern part >f the province. This is not explained by Mooser or Gastil. There is a large metamorphic basement com— plex of pre—Cretaceous age between Cabo Corrientes and Barra de Navidad, which extends inland almost to Talpa, Jalisco. The origin of this intrusive massif has not been explained by either of the authors nor is its importance noted within the Neovolcanic Axis Metallogenetic Province. Apparently, this metamorphic basement complex underlies the volcanic sequence of andesites and outcrops between Talpa, Jalisco, and Bahia de Banderas, Jalisco, eastward to Ameca and southward to the town of Autlan, Jalisco. Thus it has large bearing on tectonic surface evidence. On the other hand, the metamorphic basement com— plex is itself intruded by a granodiorite pluton, prob— ably of Cretaceous age, but certainly pre-Tertiary. These phenomena can be observed south of Puerto Vallarta and west of Talpa, Jalisco. Another wide area of intrusive acid igneous rocks is at Purificacion and Armeria Rivers west of the city of Colima (fig. 6). In studying the tectonic lineaments of the western part of the Neovolcanic Axis, consideration must be given to extensive outcrops of sedimentary rocks of Cretaceous age in the eastern part of the State of Colima and the southern part of the Neovolcanic Axis in the States of Jalisco and Michoacén. In this case, most of the Cretaceous section is made up of a thick limestone sequence interbedded with sandstone and shale formations. From the meridian of the city of Morelia eastward (fig. 6), the NeOVOlcanic Axis is bordered on the south by another large area of metamorphic rocks which is RELATIONSHIP OF MINERAL RESOURCES TO LINEAR FEATURES IN MEXICO 65 0100200300400me EIZEZE FIGURE 4,—Bouguer anomaliesshowing coincidence of strong gradient with Sierra Madre Occidental (Castil, 1973). clearly visible immediately west of Valle de Bravo in the State of Mexico, as well as by a thick sedimentary section in the States of Michoacan and Guerrero west of Taxco. At this latter place the basement metamor— phic complex underlying a thick sedimentary section outcrops extensively north of the city of Taxco. East- ward from the Valley of Mexico to the Gulf of Mexico, a sequence of volcanic rocks of late and middle Cenozoic age outcrop extensively. This volcanic se— quence overlies the thick geosynclinal facies of the Sierra Madre Oriental section of thick limestone se- quences and thinner shale and sandstone formations. This can be noticed around the town of Oriente to the east of the city of Puebla eastward to the city of Orizaba, Veracruz. The northern border of the Neo- volcanic Axis is made up of the same volcanic se- quence previously mentioned, to the town of Mizantla, Veracruz, just south of Poza Rica. This massif, the Teziutlén Massif, made up of the volcanic sequence and some intrusives, extends southward almost to the city and port of Veracruz on the Gulf of Mexico. Because of the heterogeneity of the section that has been affected by diastrophic distortions, .it is well to keep in mind in the interpretation of lineaments from Landsat imagery, that the different type of rock and their different ages of emplacement will of course behave differently to tectonic stresses. Also, lineaments that appear in older sections, like the basement complex, will not necessarily show in sur- rounding country which may be capped by a more recent volcanic sequence. It is also well to keep in mind the way lineaments and evidence of distortion will appear in the imagery studied, if one considers that there are important intrusives such as those at Tamasula, Jalisco, at Mascota on the Rio Ameca south of Zapotlan, Jalisco, and along mountains of igneous rocks which lie to the west of Ameca and Etsatlan in the State of Jalisco. There are a large number of stocks at Rosales and Tacambaro in the State of Michoacan, that can readily be seen in the Geologic Map of Mexico (Mejorada, 1968). There are intrusives which may not outcrop, as is the case of those that underlie a volcanic andesitic and rhyolitic sequence at the mining district of El Oro and Tlalpujahua. In this case, mining activity has re- vealed that these intrusive stocks and dikes are post- 66 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM ....: 1;. 2.: .‘: .' ° ':.': V v.11"; .‘-..’- . . .....'- . ° {£2.- \ :- -:-. -' - J G U L F (f/ .. . - - : ' . . :_ . : . F v 94 ' . ‘ ///////// .'- ' 0 .° "1'01? 0 . I o 0 ' O . //// .' / yéafff/fl o %. o /’ ’I III. . . / . EXPLANATION A /// . . /////////,, . ,// 1.1,,”mflnmm Silver deposit / //i// ' . .l/lfl % ' . Silver province 9 BOOK-n. Ill 0 III'I’I' . 0.. FIGURE 5.—Trend of orientation of silver deposits in Mexico showing apparent displacement of subprovinces (Castil, 1973). volcanic sequence in age and may have caused the mineralization (fig. 6). If the lithologic heterogeneity of the stratigraphic column and the age of the volcanic sequence which covers most of the Neovolcanic Axis is studied, per- haps an explanation will be found for the variety of orientation of fracture and faulting which may be observed in the Landsat—1 imagery. Gastil (1973) suggested that a zone of cortical weak- ness which extended northwestward and formed the Gulf of California, existed in the area that is now oc— cupied by the Neovolcanic Axis. He advocates the thesis that the zone of weakness might have existed during the Mesozoic and that at that time the lateral and longitudinal displacement started to separate the peninsula of Baja California from the continent. He concludes also that the rupture movement was in- tensified during Oligocene and that later, during Miocene time, coincided with the eruption of ignim— brites and igneous extrusives which eventually formed the Basin and Range province to the north in the areas of Arizona, New Mexico, Sonora, and Chihuahua. According to Gastil (1973), the real dislodging of the peninsula may have started approximately 14 mil- lion years ago. This may have originated the Neovol- canic Axis displacement of the andesite-rhyolite sequence which constitutes the volcanic sequence outcropping there. Finally, the larger features, alineaments of frac- tures and faults, observed from Landsat—1 imagery do not result from one or two diastrophic movements but are the result of any number of diastrophic move- ments that are coincident with the emplacement of the volcanic sequence and the intrusive rocks. IMAGERY INTERPRETATION Except for a few major fractures, such as those mentioned at the mining district of El Oro and Tlalpujahua, State of Mexico, the rest of the features in figure 7 have not been checked extensively in the field. Consequently, it is obvious that a large part of these structural features are inferred. They are highly interpretational, and they should be so considered when studying the interpreted figures. In spite of the quality of the photos, which were not the best, and the possibility of obtaining better band mixes with more contrast, the lineaments of RELATIONSHIP OF MINERAL RESOURCES TO LINEAR FEATURES IN MEXICO A§IENTos 2%" ‘€\\ ASCOTA" ' Q? 3'gESMCIRONADO \/ -AYUTL§A / AutLAN; LIIPDH ll K ‘x!‘»« / /O \\ tartar/u \ .COLORADA “EMA/1m *4 C hr Cii> a ‘ V W 3!. ,, APA ZINGAN. L E Y E L E G PLEISTOCENO Y RECIENTE PLEISTOCENE AND RECENT a 20 so IIILa- KILDI z [mm-93%” I'v IMORELIA URUAPAN £ATZCUAR“ \ TACAMBARO' / I Am DEIROSAL / , .C . ' ' M Q zITAgIARo- CENOZOICO INFERIOR LOWER CENOZOIC :vno vau 67 . /'I TULANCINGO 'TEZIUTLAN ATLALPOJAHUA Cii V GANGUE ' 0 . I I TL'XAciflLAzA/H ”LAP ”um 'TOLU C 0 ”Half/mum " ram” Pm © 3K I pU EB L A \I’ [mum \ g; uEIIIIAVACA' pUAUTLA . AgTCONK SDE TAMqIIos Ci a? N /: TEHUACA M “ ‘* IRE \o “W \ TIM ” I Ci UALA\ %2 £3 Ci I CHILPANCINGO 175?:4“; N A E N D INTRUSIVOS DEL CENOZOICO INFERIOR LOWER CENOZOIC INTRUSIVE mo Iuo FIGURE 6.—Schematic geologic map of the Neovolcanic Axis Metallogcnctic Province of Mexico. mountains, valleys, visible fracture lines. and intru- sives, are easy to detect as Iineaments of diastrophic origin. There are mountainous areas with extensive valleys and other features such as straight line ori- ented vegetation in arid valleys which would indicate moisture from fractures under the alluvium. Also the tectonic or structural control of river courses is ob- served in drainage patterns in the northwestern part of the Neovolcanic Axis, north of Bahia de Banderas and in the Rio Acaponeta, Rio San Pedro, Rio Santia- go, and other rivers in the State of Jalisco. The La— guna de Chapala seems to be the result of a graben. This is evidence of Mooser’s proposed anticline along these features. The imagery studied shows a number of east-west trending lakes in the States of Michoacan and Queré- taro. These, however, are due to dammed rivers that may not necessarily be controlled by structural fea- tures. The east-west orientation of the river basins may be coincidental. For instance, the Lago Cuitzeo may be caused by a water table suspended by lava flows damming a river basin. On the other hand. in the northeast part of the Neo— volcanic Axis, Iineaments are controlled by the Sierra Madre Oriental folds which run definitely NW-SE. In some areas a NE-SW displacement of these folds and faults may be observed on transverse faults. It is well to emphasize that fractures and faults that have been observed and marked by the author definitely tend to show that there is no continuity of east-west orientation to substantiate the thesis advo- cated by other authors, that the origin of the Mexican Neovolcanic Axis could have been an old east-west cortical line of weakness. It is well to emphasize also that there are fault lines longer than 100 km, such as the parallel faults immediately south of Lake Atotonilco, which run north and south. Notice also the north-south lOO-km-long fault which occurs between the town of Yesca, Nayarit, and the Ceboruco volcano (northwestern area, fig. 2), The latter may be connected with the faulting and structural control shown by the emplace- ment of the igneous massif which crops out close to the town of Yesca. There are in connection with this FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 68 .mwSUm: Lwtocm . UCN mzsmw mgbcwt quELHJOmmewfitoc DEM «meLHJOmlumogfitor. wO OUCNEEOUGLQ wc_>>Or_m 02x02 *0 m_x< U_CNU_O>O®Z up: *0 \fome: ruwmeEmJIN wMDU: v .3 «W ‘- ,.6aazmmo%muo ,. ,, ., , fiatanwxmi RELATIONSHIP OF MINERAL RESOURCES TO LINEAR FEATURES IN MEXICO 69 igneous massif a cluster of quite noticeable faults, some 20 to 30 km long, that cross the north-south faults. This cluster of smaller faults shows a marked tendency of orientation NE-SW. This phenomenon can be observed to the north and east of Rio Coahuayana, southeast and east of the city of Colima, 70 km east of the city (southwestern area, fig. 2). A similar phe- nomenon can be observed as NE—SW faults cross NW-SE faults 35 km to the north and southeast of the city of Zimapan, State of Hidalgo (east-central area, fig. 2). In this case, these crossed faults are usually short, less than 40 km long. In the central part of the area, in the State of Guerrero, and southwest of the city of Iguala, a cluster of faults trends NE—SW. The same NE-SW orientation predominates in practically all the small faults (30—40 km long). Again, within 100 km radius of Ciudad Altamirano, Guerrero (south—central area, fig. 2), there are any number of areas in which clusters of faults several tens of kilometres long appear, all of them oriented NE-SW. Examples are the area north and south of Lago Cuitzeo. and between this lake and Laguna Yuriria, in the State of Michoacén (central area, fig. 2), and also east of Lake Chapala and along Rio Huizcata northeast of the city of Guanajuato. Tectonics related to intrusions are shown by a number of faults which appear in an area called Cerro Desmoronado, some 30 km east of Puerto Vallarta on the Pacific side of the Neovolcanic Axis. Here faults are multidirectional, and the author believes that these are caused by the later intrusives to the west of the town of Talpa. It looks like the stock pierced and dis- located the volcanic sequence, also giving origin to the mineralization. In this Cerro Desmoronado area, which shows extensive mineralization, are the famous Amaltea and Cuale mines of the Compania Minera Fresnillo, S. A. A long east-west fault which appears extensively in the literature of the Neovolcanic Axis, is the so-called Falla de Acambay. The town of Acambay lies some 20 km north of the mining district of El Oro and Tlalpujahua. East-west faulting appears at the mining district and not necessarily at the town of Acambay. This fault is some 100 km long, but does not connect with those of Chapala, and, as formerly mentioned, it is more likely that this faulting occurred due to the emplacement of intrusives which underlie the El Oro and Tlalpujahua mining district. Emphasis must be made on the lack of continuity of many of these tectonic features as observed in Landsat—1 imagery. There are extensive lava flows of recent age, which may not reflect structural distortions of large magnitude and of previous geological ages. That is, Quaternary extrusives and pyroclastics may disguise evidence of diastrophic or tectonic move- ments. This may be the case of the mining district of Pachuca. Serious efforts have been made to decipher structural control therein. Both to the north and south of Pachuca City, Quaternary volcanics and pyro- clastics of great thicknesses completely disguise any structural features that might have been observed otherwise. Forty or 50 km north of the Pachuca min- ing district, evidence of the Sierra Madre Oriental folding may be observed in high mountains where thick limestone sequences outcrop. These show a per- fect lineament oriented NW—SE, but Quaternary vol- canics and pyroclastics cover older rocks which might otherwise have shown tectonic features. TECTONIC FEATURES AS RELATED TO METALLOGENETIC PROCESSES A general study of the principal structural features in Mexico revealed an intimate relationship of tec- tonics of continental scale with metallogenetic proc- esses which have given rise to the principal mining districts in the country. The author has subdivided the country into six metallogenetic provinces. The Neovolcanic Axis Metallogenetic Province, which is the main object of this exercise, is one of them. It is bound on the north, from west to east, by the southern end of Sierra Madre Occidental, the Central Plateau Province, and the Sierra Madre Oriental. The southern boundary is made up, from west to east, by the Sierra Madre del Sur and the Sierra Madre Oriental‘s southern con- tinuation. As previously stated, it is the author’s contention that metallogenetic processes are controlled by oro- genic movements along lines of diastrophic weakness. There has been some previous work done along this line by Russian economic geologists and metalloge- necists. In 1940, a group of Russian geologists, under the direction of Y. A. Bilikin, studied the factors that rule the distribution of minerals that constitute ore on the Earth’s surface. Shatalov reported on this work in 1972, and the Canadian geologists, McCarthy and Potter (1962), studied the Soviet’s results with regard to the origin of mineral deposits. Bilikin’s group prepared a thorough statistical compilation of the Soviet Union’s different geologic districts, mineral deposits, their geologic environment, the country rock types, absolute age of mineralization of the country rock itself, and the age of tectonisms that gave rise to the accumulation. Bilikin (1955) re- ports to have found an intimate relation in time and 70 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM a recurrent relationship regarding composition of ore deposits between the complex specific minerals and the specific intrusive rocks. These in turn were all related to a tectonic cycle in a zone of mobile belts. The Canadian geologists thought this interesting enough to attempt such a study of the eastern part of Canada and the Rocky Mountains province of British Columbia. There are a number of results of research on metallogenetic processes which relates ore deposits to intrusive and/or extrusive igneous rocks. Kraus- kopf’s (1967) study, for example, is a very well docu- mented study of orogenies with regard to the metallogenesis of ore deposits and the relation that these have to different igneous rock emplacements. Thus, it is reasonable to assume that ore deposit genetics are intimately related to the emplacement of those igneous rocks mentioned, and they in turn are related to orogenic epochs. The origin of ore deposits is not within the scope of this exercise. Rather, their occurrence with respect to movements along mobile belts is under study. It is important to demonstrate that the magma which is originated in the upper part of the mantle, possibly just below the Mohorovicic Discontinuity or immedi- ately above it, finds its way upward through the litho- sphere by using large fractures and faults to rise toward the surface. Consequently, the author’s Land- sat—1 imagery study to relate large faulted areas to metallogenetic processes has used that concept as a basis for his contention. To substantiate this thesis, the author points out the occurrence of large mineral deposits along the large fractured lines in the Neovolcanic Axis Metallogenetic Province such as the area in the east—central part of it, including the Pachuca silver deposits, the Zimapan lead-zinc-silver deposits, the Taxco leadszinc-silver and Fluorite deposits, the El Oro and Tlalpujahua gold and silver deposits and the Angangueo silver and gold de- posits. Other somewhat smaller deposits lie along the extreme western part of the Neovolcanic Axis Metal- logenetic Province, such as the lead-zinc-copper-silver deposits of Cuale and Amaltea in the Cerro Desmo- ronado area, immediately southeast of Puerto Vallarta on the Pacific seaboard (fig. 7). All of these mineralized areas occur along very extensively faulted areas. Faulting may occur oriented east-west, as at El Oro and Tlalpujahua; or NE—SW, as at Zimapan; or as at Taxco where faulting occurs oriented mainly WNW-ESE; or multidirectional fault— ing and fracturing as at Cuale and Amaltea in Cerro Desmoronado, south of Puerto Vallarta. From all these observations, the author surmises that if the tectonic feature is large enough, that is, if faulting occurs along more than 50 to 100 km, this by necessity must have also affected a large thickness of the lithosphere reaching into the asthenosphere, or the upper part of the mantle, where magmatic ma- terial exists. This magma, along with its constituent fluids, finds its way through the large faulted area to- wards the surface. The longer it has to migrate through the lithosphere the better opportunity it has to dissolve additional chemicals which eventually will form ore deposits useful to man at the nearest surface part of the lithosphere and/or outcropping. Figure 2 shows incidence of mineral occurrences along and in, or about, intensively disturbed areas. DESCRIPTION OF SOME MINERAL DEPOSITS ALONG THE MEXICAN NEOVOLCANIC AXIS AS RELATED TO TECTONIC FEATURES Specialized literature described adequately the geology of ore deposits of importance along the Neovolcanic Axis Metallogenetic Province. Conse- quently, and congruently with the thesis herein sus- tained, the author described only in general those tectonic features that have controlled mineralization in certain important mining districts. PACHUCA Y REAL DEL MONTE MINING DISTRICT, STATE OF HIDALGO This is the most important mining district in this province (fig. 8). Geyne and others (1963), in their monumental work have an excellent description of this district, which is located 100 km north-northeast of Mexico City, at the eastern edge of the Central Plateau Metallogenetic Province or the northeastern edge of the Neovolcanic Axis Metallogenetic Province. This mining district occurs precisely in the area where the folded belt of the Sierra Madre Oriental Metallo- genetic Province plunges under an andesitic volcanic sequence. East-west faults, 20-30 km long, are vaguely shown in the Landsat—1 imagery, but may be mapped on the ground. The volcanic sequence is of Pliocene- Pleistocene age and has not been eroded intensely enough to reveal more clearly structural features be- neath the strata. Some NW-SE, N-S, and NE-SW fault- ing may be observed in the satellite photos but not as clearly as on the ground. Valley fill and Quaternary rhyolitic lava flows mask the largest structures. The rock in this district consists of volcanoclastic material interstratified with low dipping andesite and rhyolite flows. The lowermost beds, at 1,000-m depth, RELATIONSHIP OF MINERAL RESOURCES TO LINEAR FEATURES IN MEXICO 71 Y PRES/1 u/A/MAI/L L0 N _ O STA.JUL|A ALLUVIAL PLAINS —— 00 T2 M Y PLEISTOCENE AND RECENT BASALTS 5 C A L 0 400 300 1200 ISOO 2000 E 5 411K 21%.MITLAN -:.:_'""""":-:5bafz' T o‘GUERRERO I)? MIOCENE VIZCAINA TV FORMATION RHYOLITE E E R FIGURE 8.—Structural map of Pachuca (Geyne and others, 1963) showing east-west and northwest-southeast faults and veins. Notice Quaternary rhyolite flows mask surface fault traces. are probably of early Oligocene age and the upper- most are probably of late Pliocene and/or Pleistocene (Geyne and others, 1963). The whole volcanic se— quence lies unconformably over the eroded, down— faulted, and intensively folded marine limestone sequence of the upper Mesozoic which forms the Sierra Madre Oriental. Immediately to the north of the city of Pachuca, this folded sequence of the Sierra Madre Oriental shows a perfect NW-SE Iineament. The Tertiary rocks have been deformed by several oscillating movements of varied intensities. These structural features have developed numerous faults with steep dips. The faults are usually of the normal tension type. However, this faulting shows well marked east-west strikes, but also some north-south and WNW-ESE orientations. It seems that the structural movement occurred at the end of the Oligocene and during the Miocene and was followed by the emplace- ment of dacitic-porphyry dikes and quartziferous dikes and sills. These dikes show a definite east—west orien- tation and undoubtedly were emplaced through faults which have the same orientation. Their age is post- Laramide. Unfortunately, most of this information has been obtained underground and surface evidence though relatively abundant is certainly not clearly observable from Landsat—1 imagery in this particular case. TAXCO MINING DISTRICT, STATE OF GUERRERO This mining district (fig. 9) lies approximately 120 km southwest of Mexico City. The region suffered tectonic distortion apparently in two different geologic epochs. The Cretaceous rocks outcropping show large folding in the shape of anticlines with axes locally oriented NE-SW. North— south folding is also occasionally observable. Mineral- ization occurs both in sedimentary and metamorphic rocks of the basement complex. On the other hand, the area suffered distortion caused by the early 72 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM ,, l" \_\ ‘ / ‘,\ ___ AL \\ /l , ::: —_ + ‘ i :1 "twig—M .. TAX 0/4 i“l“\ K) iE—G—QIEE-i—J'fll / " _ { ‘—:—: _ :55:— :—_{q?; 40; __— ‘Ij‘x—y EGRA _—_—. I'E' \_ oz :1 _l / —— ‘3—:; A‘IUEL .; 'L‘ "—5 J4" —— : Tflé—ELEX; 5-;ILGggCY'E " AL vE-V' r. x. 1‘ / :_:::c- ":U [TEN— I—Efi- _REMEDIO _ “F; ‘J‘ A:‘ LA PEI/.K—l'" —- A l.- ‘. I . a _,_ , JLIANTLA“ PCHAHUA -37. '— sr RNA_DO‘ ammo .' A?“ I - ' ocgezum— T 0 250 750 |250 3%: M E T R 0 S A T I O \_. VEINS AND FAULTS _ » LIMESTONES______ BASIC INTRUSIVES VOLCANIC ROCKS.— SHALES FIGURE 9.—Taxco mining district, State of Guerrero, sh DIORITE___ ______ L SCHIST_________ _ owing vein control by faults and fractures. (After Fowler and others, 1948.) Eocene orogeny (Laramide), and NW-SE trending large faulting predominates and may be observed in Landsat—1 imagery. There was another very clear tectonic orogenic movement which occurred at the end of the Pliocene. Fries (1960), has also found evidence of intense de— formation at the beginning of the Pleistocene. This seems to correlate well with orogenic movements in the mining district of Pachuca. The Taxco mining district’s main ore production is lead, zinc, silver, and gold. The ore occurs as vein filling of fractures and faults of the Taxco schists at Mina El Pedregal. Ore bodies also occur in fractured limestones of Cretaceous age at Mina El Pedregal, La Concha, and San Antonio. Again mineralization occurs in clastic beds of the Mezcala Formation at the Mina Jesus. Finally, a very weak mineralization process reached the lowermost part of the overlying Tertiary volcanic sequence. All of this indicates that the min- eralization was contemporary with the fracturing sys- tems of the late Pliocene and early Pleistocene much as it happens in Pachuca. However, the main metallo— genetic processes seem to have taken place in the late Oligocene or early Miocene and probably re— newed in more recent geologic epochs HUAHUAXTLA MINING DISTRICT, STATE OF GUERRERO This ore body is located approximately 15 km northwest of the city of Iguala, Guerrero. The ore bodies occur in faulted Morelos Formation as well as the Cuautla and Mezcala section of the Middle and Upper Cretaceous. The Huitzuco mercury district, 20 RELATIONSHIP OF MINERAL RESOURCES TO LINEAR FEATURES IN MEXICO km east to Iguala, also contains mercury and antimony which occur principally in faulted limestone of the Morelos Formation of Cretaceous age as well as in dolomite and anhydrite. Here again, there is indirect evidence of post—Cretaceous mineralization. It is thus possible to correlate mineralization metallogenetic processes to the same orogenic movements that gave origin to the Pachuca and Taxco ore deposits. EL ORO AND TLALPUJAHUA MINING DISTRICT, STATE OF MEXICO This is a district which has been developed since 1521 and shut down partially until recently. It pro- duces epithermal cavity and vein-filling gold and silver deposits. Stockworks are also common. The ground rock is made up of black slate (fig. 10). The veins fill cavities in great and small faults which run nearly east and west. They are covered by ande- sitic rocks of the volcanic sequence of post-mineral- ization age. Again, in this district, it is estimated that 73 metallogenesis took place during early Miocene time, as shown in the principal veins called San Rafael, La Verde, and San José de la Borda. The El Oro and Tlalpujahua mining district is con- trolled by large, parallel faults over 100 km long. Here NE-SW trending faults of less extension, and NW-SE short faulting also occur, some of which is rather re- cent much as it happens in Taxco and Pachuca. In the author's opinion this faulting may be the reason for the enrichment of the pre-Miocene and Miocene min- eralization. The author believes that volcanic action of Pliocene-Pleistocene age caused the reworking of the original ore and gave rise to enrichment. OTHER LESS IMPORTANT MINING DISTRICTS There are any number of smaller mining districts which occur where clusters of large and small frac- tures show in the Landsat—1 imagery. For instance, the intrusive that gave rise to the Tetela gold mine in the northern part of the State of Puebla, approximate— \'\ :2“; 354! All (/51 G+J wlk G +J ®J wAA/ACAHAA/ ' ’KV . AGUA CA olENTEm‘ PRESA .. \L A fr“ EJIDO BR 4.727534 G 23' in I as WIV ' ‘ ‘ '\ an NVA. ESTR LL % “viii“ ‘QK”:\\‘\ D . \ 4y EXPLANATION Carbonaceous slate Clays III Andesites III Ignimbrites Breccias and Tuffs_ Basaltic Andesites *- Lacustrine Sediments 2:8‘é‘fieltia‘: Alluvium Milonites, Talus Slidings Contact / Fault // Syncline /\/ Anticline X Volcano @ Mine shaft X‘ 0 1 2 5 Km 1 '—‘ K / xG+J G FIGURE 10.—E| Oro—Tlalpujahua mining district. 74 FIRST ANNUAL PECORA 1y 80 km northeast of the city of Tlaxcala, is one such result of the tectonic distortion. Again, in the immediate vicinity of the town of Teziutlan in the northeastern part of the State of Puebla and at the edge of the Gulf of Mexico water— shed, there are copper mines which originated at the contact of the intrusive rocks and limestones of post- Cretaceous age. In the immediate vicinity and farther north, the abandoned iron mines of Tatatila and the copper and gold deposit called Las Minas, 30 km northwest of the capital city of Jalapa, occur in intensely folded and faulted areas which show both the typical volcanic sequence structures of the Neo- volcanic Axis Metallogenetic Province and the NW-SE trending controlling axis of folds and faults of the Sierra Madre Oriental orogeny (Laramide). Along the western part of the area, smaller faults show a marked tendency of orientation NE-SW. This phenomenon can be observed to the north and east of Rio Cohaguayana 27 km southeast and east of the city of Colima. In the State of Hidalgo, a similar phe— nomenon can be observed as NE-SW faults cross NW-SE faults, immediately to the north and southeast of the city of Zimapan, Hidalgo. In this case, these cross faults are usually short, less than 40 km long. In the central part of the area, in the State of Guerrero, immediately to the east and northeast of Ciudad Altamirano, a cluster of NE-SW short faults, less than 30 km long, intersect long NW-SE trending faults which show up predominantly along the Rio Cutzamala from the area of Ciudad Altamirano north— ward to the town of Tzitzio. It seems that mineralization follows the NW-SE preting Landsat imagery for mineral or hydrocarbon NE-SW trending short faults intersect the longer ones. These are controlled by the early Eocene orogeny (Laramide). The author wishes to stress the fact that in inter- preting Landsat imagery for mineral or hydrocarbon exploration purposes in areas such as the Neovolcanic Axis of Mexico where very recent volcanic sequences mask older tectonic features, one must extrapolate into the area structural lineament from more eroded or uplifted areas where they can be observed clearly. It certainly enhances the possibilities of selecting prospective areas by helping to pinpoint targets. MEMORIAL SYMPOSIUM REFERENCES Bilikin, Y. A., 1955, Metallogenetic province and epochs: Moscow, Gosgcoldekhizdat. Fowler, (3. M, Hernon, R. M., and Stone, E. A., 1948, The Taxco mining district, Guerrero, Mexico: In- ternational Geological Congress, 18th, London 1948, Report of the Eighteenth Session, pt. 7, p. 2—12. Fries, Carl, 1960, Geologia del Estado de Morelos y de partes adyacentes de Mexico y Guerrero, region central meridional de Mexico: Mexico Instituto de Geologia, Bull. 60. Gastil, G. R., 1973, Evidence for strike-slip displace- ment beneath the Trans-Mexican Volcanic Belt: Stanford University School of Earth Sciences, Conference in Tectonic Problems of the San Andreas Fault System, Proc. Geyne, A. R., Fries, Carl, Segerstrom, Kenneth, Black, R. F., and Wilson, E. I. F., 1963, Geology and mineral deposits of the Pachuca-Real del Monte district, State of Hidalgo: Mexico, Consejo de Recursos Naturales no Renovables Extra—series publication 5E, p. 203—222, illust., 7 tables. Krauskopf, K. B., 1967, Source rocks for metal-bear— ing fluids, in Barnes, H. L, ed., Geochemistry of hydrothermal ore deposits: New York, Holt, Rine- hart and Winston, p. 1—33. McCarthy, W. D. A. I., and Potter, R. R., 1962, Min- eralization as related to structural deformation igneous activity and sedimentation in folded geosynclines: Canadian Mining Journal, v. 83 no. 4, p. 83—87. Mejorada, S. H. 8., ed., 1968, Carta Geologica de la Republica Mexicana: Comite de la Carta Geo- logica de Mexico, 1 sh., 122,000,000. Mooser, Federico, 1972, The Mexican Volcanic Belt, structure and tectonics: Geofisica Internacional, v. 12, no. 2. Salas, G. P., and Gonzalez, R. G, 1967, Yacimientos de Plata en México: Instituto de Geologia, U.N.A.M., Mexico. Shatalov, Ye. T., ed, 1972, Metallogenic analysis of ore-controlling factors in ore regions: Moscow, Izd. Nedra, 295 p., illust. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Landsat Contributions to Studies of Plate Tectonics By Jan Kutina. Laboratory of Global Tectonics and Metallogeny, Washington Technical Institute, Washington, DC. 20008 and William D. Carter. US. Geological Survey, Reston. Virginia 22090 ABSTRACT Tectolinear interpretation of satellite image mosaics has proven to be a powerful method for tracing the course of basement fracture zones which have been propagated upward through a platform cover and even through orogenic belts. The method has also been applied to locate fracture zones, the orientations of which have been changed by rotation of litho- spheric plates. A tectolinear interpretation map of the United States by W. D. Carter, set in context with structural geological knowledge of the Canadian Shield, sug- gests that the Hudson Bay paleolineament (HBP) of J. Kutina may extend into the Precambrian basement of the Eastern United States. Spatial correlation of sig- nificant ore districts with the HBP and associated structures in Canada (both in the Archean greenstone belt and outside) suggests that important ore concen- trations may also exist along the probable extension of the HBP beneath the sedimentary cover of the East- ern United States, where it may intersect with an east-west trending fracture zone, distinguished by A. V. Heyl (1972), between 38° and 40°N. in the Cen- tral and Eastern United States. A number of Missis- sippi Valley-type ore deposits are spatially associated with this east-west fracture zone. Three major northwest-trending lineaments have also been delineated in South America, and their ex- tension into the Andean Region has been traced by statistical treatment of linears in the Landsat mosaics. These northwest-trending lineaments may very well have been E—W fracture zones in the paleogeographic orientation of South America before it was separated from Africa. INTRODUCTION The investigations carried out in the last decades by Ewing. Heezen, Hess. their coworkers, students, and followers have brought new insight in the knowl- edge of the structure of the upper parts of the Earth. One of the outcomes of this scientific trend was the concept of the “new global tectonics” as introduced by Isacks, Oliver, and Sykes (1968). Morgan (1968), Le Pichon (1968) and others. This concept. referred to also as “plate tectonics." has revived the old con- tinental drift hypothesis of Alfred Wegener, giving it a new geophysical interpretation. Since the time of the above papers, an enormous literature on plate tectonics has accumulated, with some papers accepting it as a proven model, and others expressing different degrees of agreement or disagreement with the new concept. The controversy which exists in the world literature is reflected well in the volume of papers “Plate Tectonics’Assess- ments and Reassessments,” edited by Kahle (1974). There are a number of other critical papers that should be mentioned, especially that by Sheinmann (1973). The basic postulate of the new theory shows that the upper parts of the Earth’s body are composed of a number of rigid plates that are not in a fixed posi- tion but which move and the boundaries of which have changed in the course of geological time. These changes involved processes of different orders of magnitude which are mutually interrelated and are of basic importance for certain geochemical processes. Metallogenic studies have shown that there exists a relationship between mineralization of various types and ages and the different types of plate boundaries 75 76 FIRST ANNUAL PECORA (Guild, 1974). One of the most important discoveries was recognition of the role of subduction or Benioff zones on metallogenic processes connected with par- tial melting of the subducted oceanic crust and gen- eration of a special type of magma. The origin of porphyry copper deposits especially of the Andean orogenic belt has been explained in these terms (Sillitoe, 1972a, b). Many deposits, however, are located in the interior of the present plates, far from their boundaries. Some of them may be related, as Guild (1973) and others have shown, to former plate boundaries, but many ore deposits apparently do not exhibit such a relationship. Several authors, who studied metallogeny of broad regions such as the Western United States or Trans- baikalia have recognized that the distribution of major endogenic ore deposits is controlled, beside other fac- tors, by major fracture zones, and often by their intersections (Kutina, 1969; Favorskaya and others, 1974). As a follow up of these investigations, Kutina (1974) carried out a systematic study of structural control of ore deposition in selected regions of four continents. Correlating his results he concluded that large endo- genic ore deposits and ore districts are usually located in the vicinity of, or above the intersection of, deep- seated fractures, fracture zones or zones of weakness that belong to four sets, N—S, E—W, NW, and NE, or to some of these four sets, propagating upward from deeper levels. The upward propagation of these frac- tures proceeds differently in different regions depend- ing on their geological evolution. As a consequence, not all of these fracture sets are developed or clearly expressed in the surface or subsurface portion of each region and the detection of some of them requires detailed studies by different methods. Reactivation of movements proceeded along these fracture zones at different geological times. There exists a relationship between the intraplate tectonics and the course of the plate boundaries. For instance, one of the ore-controlling fracture sets of continental Central America (the NW—SE one) is parallel to the Middle America Arc, the boundary be- tween the Cocos Plate and the Caribbean Plate. Ore deposits of a post-Laramide age lie closer to the plate boundary than the majority of the older ore deposits (Kutina, 1974). The reconstruction of the pre-Cretaceous (“pre- drift”) configuration of the present plates has mostly been done by analyzing magnetic anomalies of the sea floor, by applying paleomagnetic data, and by geometric fitting of the continental margins. There- MEMORIAL SYMPOSIUM fore, King (1970) correctly notes, when referring to Bullard’s reconstruction of the continents on opposite sides of the Atlantic, that the reconstruction has been done without regard for the inner constitution of the blocks or plates. Correlation of the geology of the matching sides of continents, which are supposed to have been sepa- rated from each other, has been done in several in- stances. Also, the extension of metallogenic provinces across such boundaries has similarly been tested by several authors with reasonable success. These correlations, though of principal importance, are usually not sufficient to estimate the angle of rotation of the plates and are not sufficient geological evidence to provide an independent check of the amount of rotation deduced by geophysical methods. Also the question of possible disappearance of a part of a continental mass by disintegration and subsidence often remains unanswered. Detection and plotting of major, deep-seated frac- ture zones, especially those which are ore-controlling, provide a new, additional tool in fitting crustal plates back together. This kind of study is in initial stages because our knowledge of the pattern of deep~seated fractures within the individual plates is insufficient. The results of Kutina’s (1974) study in different parts of four continents and of an independent investi- gation by Favorskaya and others (1974) show a real possibility of comparing the positions and orientations of ore—controlling fracture zones of the individual plates. The use of the above metallogenic data in testing the validity of plates’ rotation as derived by geophysi- cal methods presents, however, a number of prob- lems. One of them is distinguishing whether or not some of the ore-controlling fracture zones represent new tectonic elements, superimposed on the old frac- ture pattern. Another question is whether zones of tectonic weakness may or may not exist in the sub- stratum over which the lithospheric plates are sup- posed to be moving. And, if they exist, how far they can influence the tectonics and metallogenic processes in marginal parts and in the interior of the plates. Some of these questions are discussed separately (Kutina, 1976). The introduction of Landsat (ERTS) imagery and the construction of image mosaics provides a new, common base for systematic study of linear features in broad territories. Swarms of lineaments extending sometimes for several hundreds of miles and crossing boundaries of different geological units are likely to reflect the course of major fracture zones in deeper LANDSAT CONTRIBUTIONS TO STUDIES OF PLATE TECTONICS 77 o 100 200 300 MILES 0 100 200300 KILOMETERS EXPLANATION . Missouri Lead/Zinc district FIGURE i.——Tect0linear interpretation of a 125,000,000-scale ERTS—i mosaic of the United States by W. D. Carter, 1974. Lineament At essentially corresponds to the 38th Parallel Lineament of Heyl (1968, 1972) and represents its inde- pendent discovery by satellite imagery. ln view of the metallogenic importance which the A1 lineament plays, the lineaments Ag—A'g, revealed by the same ERTS—i mosaic, deserve special attention. levels of the Earth’s crust which have been propa- gated upward due to movement of crustal blocks. The detection of such lineaments and their correla- tion with the distribution of ore deposits may very effectively help to solve long-standing controversial questions and will help to map the pattern of major deep-seated fractures which is needed for correlation of the intraplate tectonics and between individual plates. The relationships between the location of ore deposits and the course and intersection of the linea- ments coming from these studies may be of great assistance in mineral exploration. In the present paper, two examples are presented. They come from a joint study of the authors, using some of Carter’s tectolinear interpretations of Land— sat mosaics. The paper, originally presented at the Pecora Sym— posium at Sioux Falls, October 1975, has been updated by references to papers by Regan and others (1975), Stewart (1976), and Callahan (1976). CONTRIBUTION TO THE INTRAPLATE TECTONICS OF THE EASTERN UNITED STATES Figure 1 is the first tectolinear interpretation of a mosaic of satellite images of the United States as compiled by Carter in 1974. The illustration is based on the analysis of a mosaic of the conterminous United States consisting of 595 Landsat—1 images published at the scale of 1:5,000,000 (US. Geological Survey, 1974). The illustration is presented here in its original form without any additional changes in order to serve as a base not influenced by genetic concepts. Some of the longer lineaments have been marked by heavy lines by Carter, but severalvothers can sim- ilarly be distinguished. For instance, two broad north- south trending swarms of linears can be recognized in the Western United States, one at the longitudes enclosing the Great Salt Lake and the other extending 78 FIRST ANNUAL PECORA through Colorado and southward. The first one es- sentially extends along the north-south fracture trajec- tory +6 and the other along +9 of the model by Kutina (1969). Two of the most prominent features of figure 1 are the major east-northeast-trending lineaments in the Eastern United States. One of them is located be- tween latitudes 38° and 40°N. and is marked Al in figure 1, and the other extends between 32° and 36°N., and is marked Ag. A.J has a prominent counter- part in the Western United States, marked A’g. Lineament A, of figure 1 intersects the Appalachian orogen in the easternmost part of the country and apparently displaces the Precambrian of eastern Penn— sylvania and New Jersey in a right-lateral sense and in places, as a pronounced curvature of the Appala- chians. This apparent displacement has been recog- nized by Drake and Woodward (1963) who interpreted it as a right-lateral wrench fault extending seaward (Cornwall-Kelvin displacement). The continental part of this displacement has been doubted later by Root (1970). The course of the lineament AI (fig. 1) as recognized on the 1:5,000,000—scale mosaic should be understood as its first approximation; It is therefore not surprising that the east-northeast trend of this lineament as it appears in the easternmost part of the United States differs from the east—west strike of the Cornwall- Kelvin displacement of Drake and Woodward (1963) who applied geological and geophysical criteria on a more detailed scale. The main power and contributiOn of the ERTS—l mosaic lies in its synoptic view cover- ing a broad territory in which the respective linea— ment, though less accurate in detail, could have been traced for a long distance. Overlaying the pattern of linear features of figure 1 with the distribution of the main ore deposits of endo- genic and controversial origin shows (fig. 2) that most of the major Mississippi Valley-type deposits of lead, zinc, fluorite, and barite are spatially associated with the east—northeast-trending lineament or fracture zone A,. This regularity with further structural and litho- logical control was recognized as early as 1968 and 1972 by Heyl, who described the respective lineament as the 38th Parallel Lineament. The 38th Parallel Lineament is probably plotted more accurately by Heyl than in figure 2 as the geo- logical and metallogenic interpretation by Heyl is based on more detailed maps. Nevertheless, the 1:5,000,000-scale mosaic of Landsat—1 images proved useful in detecting the same regularity. Moreover, it is probable that more detailed information available from Landsat images at larger scales and the applica- MEMORIAL SYMPOSIUM FIGURE 2.—Tectolinear interpretation of the eastern part of the United States, set in context with some structural and metallogenic features of the Canadian Shield. Note especially the distribution of the Mississippi Valley— type deposits of lead, zinc, fluorite, and barite, many of which concentrate—in accordance with Heyl (1968, 1972)—in a belt following the 38th Parallel Lineament. The east-northeast-trending 38th Parallel Lineament traver- ses the Appalachian orogen at a latitude close to 40“N., with an apparent right-lateral displacement of the Precam- brian, thus giving support to the Cornwall-Kelvin wrench fault of Drake and Woodward (1963). EXPLANATION I. Precambrian of the Canadian Shield and the outcropping Precambrian of the eastern part of the United States (simpli- fied from King and Beikman, 1974). II. Younger rocks, mostly covering Precambrian basement. Ill. Canada: major ore fields in the Precambrian south of Hudson Bay, both in the Archean greenstone belts (1—4) and outside (5, 6). United States: two ore deposits in the Precambrian of the northeastern United States (20, 21) and the Ducktown deposit in Tennessee (14). IV. Ore districts and major deposits of the Mississippi Valley- type and a few smaller lead—zinc deposits of the Eastern United States. V. Deposits of other types (17, 18). ORE DEPOSITS AND DISTRICTS Canada: 1. Val d’Or district. 2 Noranda district. 3. Timmins district. 4. Chibougamau district. 5. Cobalt district and other deposits. 6. Sudbury district. United States: 7. Upper Mississippi Valley district. 8. Illinois-Kentucky district. 9. Tri-State district of Missouri, Kansas, and Oklahoma. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. Central Missouri district. Southeast Missouri lead district. Ducktown district, Tennessee. The Mascot-Jefferson City zinc district, Tennessee. Austinville-Ivanhoe district, Virginia. Hamme district, North Carolina. Cornwall deposit, Pennsylvania. Friedensville deposit, Pennsylvania. Franklin Furnace and Sterling Hill, New Jersey. Balmat-Edwards, New York. Central Kentucky district. Cumberland River deposit. Shenandoah Valley (Timberville), Virginia. Northern Arkansas district. Northeast Arkansas district. Central Tennessee district. LANDSAT CONTRIBUTIONS TO STUDIES OF PLATE TECTONICS w‘ ”Q 2/70 . = O =‘: .. / / é . HUDSON BAY \v @0131 (Nos.14.2o.21) 0-H o I 0 400 MILES 0 600 KILOMETERS 80 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM tion of enhancement techniques would provide a more complete geological plot of the lineament, especially at places where its course was interpolated due to lack of structural information. Thus, the 38th Parallel Lineament (A1 in fig. 1) is (1) spatially associated with the distribution of major de- posits of the Mississippi Valley—type and (2) may displace the Precambrian at the intersection with the Appalachian orogen. Therefore, the above lineament appears to be a significant structural element of the Central and Eastern United States or its basement. The experience from global metallogeny shows that the chances for economic concentration of metals increase along intersection of orogens by transversal deep-seated fracture zones, especially if ore deposits have been detected along them at other places. Ac- cordingly, the area of the intersection of the Appa- lachians by the 38th Parallel Lineament may be postulated as a target area for mineral exploration. A number of ore deposits and occurrences have al- ready been found in this area, among them the mag— netite deposits of Cornwall and Grace connected with the diabase magmatism in the Triassic basin (Lapham, 1970; Sims, 1970) and the sphalerite deposit of Friedensville located in the Cambrian and Ordovician carbonate sediments (Callahan, 1970), both in Penn- sylvania (18 and 19 in fig. 2). The unique character of these deposits (including the deposits of Franklin Furnace and Ogdensburg in rocks of Precambrian age) within the metallogeny of the Appalachians has been stressed already by Guild (written comm., 1968), who assumed their possible relation to deep basement offsets along an intersecting lineament. In view of the above significance which the 38th Parallel Lineament plays, special attention should also be given to the other east-northeast-trending linea- ment (A2 and A’:, fig. 1) running parallel to the former and located between the latitudes 32° and 36°N. in the Eastern United States and close to 34°N. in the Western United States. The courses of lineaments Agand A’2 essentially parallel the outline of the North American craton (with the basement rocks as old as 1,700 to 850 my.) against the encircling marginal miogeosynclines, as these two units have recently been distinguished by Stewart (1976). According to Stewart, the miogeosyn- clines started to form at approximately 850 my. indi- cating the initial stages of rifting. If the above relation is true, these lineaments may contribute to the defini- tion of the southern boundary of the North American craton. Another very interesting correlation may exist be- tween lineaments AI and A._,, A’._. of figure 1, and the northern and southern boundaries; respectively, of a newly detected belt of magnetic highs in the global magnetic anomaly map (Regan and others, 1975). The belt extends in an essentially east-west direction with some tendencies to turn east—northeasterly to north- easterly between latitudes 30° and 40°N. through nearly the whole United States. Lineaments A._. and A’2 of figure 1 follow very closely the southern bound- ary of the belt of magnetic highs. Since lineaments A2 and -A’._. also parallel, to some degree, the southern boundary of the North American craton, it is possible that the three features are interrelated. Figure 2 sets the tectolinear interpretation of the ERTS—l mosaic of the Eastern United States in con- text with some structural features of the Canadian Shield. The illustration shows the course of the Hudson Bay paleolineament (HBP) as postulated by Kutina (1971) as well as the main ore fields south of Hudson Bay. The Val d’Or-Noranda—Timmins ore field (1—3 in fig. 2), located in an Archean greenstone belt, contains about 1,500 occurrences of gold and about half that number of copper (Kutina and Fabbri, 1972). The aim of the plot of the Canadian ore fields is to initiate discussion of possible presence of important mineral concentrations in the Precambrian basement of the Eastern United States. If the HBP and its parallel structures with which some of the ore con— centrations both within the greenstone belts and out- side are spatially associated, extends southward into the basement of the United States, it may play a role in structural control of ore deposition. From this view- point there may be special interest in the swarm of north-south lineaments south of Lake Ontario and farther south, which may possibly reflect upward propagation of southern extension of the HBP from the Precambrian basement. The respective swarm of north-south lineaments intersects the east—northeast- trending “38th Parallel Lineament” in the area north of location No. 16 in figure 2. In the area of this inter- section, a coal basin has developed in western Vir- ginia. The shape of the basin as expressed in the “Geological Map of the United States” (King and Beikman, 1974) is symmetrical along the directions of the two intersecting sets of lineaments. Several areas with numerous gold occurrences and old small—scale mining operations for gold are known in the Eastern United States (Callahan, 1976). Though the primary source of this gold is in outcropping rocks which are considerably younger than the greenstone belts in Canada, the presence and distribution of gold in the Eastern United States is of special interest as it may represent remobilization of this metal from older rocks of the basement. LANDSAT CONTRIBUTIONS TO STUDIES OF PLATE TECTONICS 81 POSSIBLE EXTENSION OF MAJOR FRACTURE ZONES FROM THE INTERIOR OF THE SOUTH AMERICAN PLATE INTO THE ANDES OROGENIC BELT A discussion of this question, preliminarily presented at the Sioux Falls symposium, has been further ex- tended and included in another paper, “The Metallo- genic Role of East-West Fracture Zones in South America with Regard to the Motion of Lithospheric Plates (with an Example from Brazil),” to be published in Djalma Guimaraes Volume, Boletim, Universidade de Brasilia. ACKNOWLEDGMENTS The senior author thanks Bethlehem Steel Corpora- tion for the agreement to use some of his results from a consultant assignment. Thanks are given Dr. Stephen J. Gawarecki and Mr. W. S. Kowalik for references to some papers. SELECTED REFERENCES Callahan, J. E., 1976, Are there Witwatersrand types of gold deposits in the Triassic basins of the Southeast 2: Geol. Soc. America, Abs, v. 8, no. 2, p. 144—145, Callahan, W. H., 1970, Geology of the Friedensville mine, Lehigh County, Pennsylvania, in Ridge, J. D., ed, Ore deposits in the United States 1933— 1967: Am. Inst. Mining Engineers, v. I, p. 95—107. Carter, W. D., 1974, Tectolinear interpretation of an ERTS—l mosaic, La Paz area, southwest Bolivia, southeast Peru and northern Chile: Committee of Space Research Plenary Mtg, 17th, Sao Jose dos Campos, Brazil, preprint. Drake. C. L., and Woodward, H. P., 1963, Appala- chian curvature, wrench faulting, and offshore structure: New York Acad. Science, Trans, v. 26, p. 48—63. Favorskaya, M. A., Tomson, I. N., Baskina, V. A., Volchanskaya, I. K, and Polyakova, O. P., 1974, Global regularities in the distribution of big ore deposits (in Russian): Moscow, Nedra, 193 p. Guild, P. W., 1973, Massive sulfide deposits as indica- tors of former plate boundaries: US. Geol. Survey, Open file rept., 11 p. 1974, Application of global tectonics theory to metallogenic studies: Internat. Assoc. on the Genesis of Ore Deposits Symposium on Ore De- posits of the Tethys Region in the Context of Global Tectonics, Varna, Bulgaria, preprint. Heyl, A. V., 1968, The 38th Parallel Lineament and its relationship to ore deposits (abs): Econ. Geology, v. 63, p. 88. 1972, The 38th Parallel Lineament and its relationship to ore deposits: Econ. Geology, v. 67, p. 879—894. Isacks, B., Oliver, J., and Sykes, L. R., 1968, Seis— mology and the new global tectonics: Jour. Geophys. Research, v. 73, p. 5855—5898. Kahle, C. F., ed., 1974, Plate tectonics—assessments and reassessments: Am. Assoc. Petroleum Geol- ogists, Memoir 23, 514 p. King, P. B., 1970, Tectonics and geophysics of eastern North America, in Johnson, H., and Smith, B. L., The megatectonics of continents and oceans: New Brunswick, N.J., Rutgers Univ. Press, p. 74— 112. King, P. B., and Beikman, H. M., 1974, Geologic map of the United States: US. Geol. Survey, 12500,- 000. Kutina, Jan, 1969, Hydrothermal ore deposits in the western United States: a new concept of struc- tural control of distribution: Science, v. 165, p. 1113—1119. 1971, The Hudson Bay Paleolineament and anomalous concentration of metals along it: Econ. Geology, v. 66, p. 314—325. 1973, Structural control of volcanic ore de- posits in the context of global tectonics: Internat. Symposium on Volcanism and Associated Metal- logenesis, Bucharest, Romania, Sept. 1973, pre- print. 1976, Relationship between the distribution of big endogenic ore deposits and the base— ment fracture pattern~Examples from four con- tinents, in Internat. Conf. on the New Basement Tectonics, 1st, Salt Lake City, June 1974: Utah Geol. Assoc. Pub. 5, p. 565—593. 1976, Lithospheric plate motions—one of the factors controlling distribution of ore deposits in some mineral belts: Mineralium Deposita, in press. Kutina, Jan, and Fabbri, A., 1972, Relationship of structural lineaments and mineral occurrences in the Abitibi area of the Canadian Shield: Geol. Survey Canada Paper 71—9, 36 p. Kutina, Jan, Carter, W. D., and Lopez, F. X., n.d., The metallogenic role of east-west fracture zones in South America with regard to the motion of lithospheric plates (with an example from Bra- zil). Submitted to the Djalma Guimaraes Volume, Boletim, Universidade de Brasilia. 82 FIRST ANNUAL PECORA Lapham, D. M., 1970, Triassic magnetite and diabase of Cornwall, Pennsylvania, in Ridge, J. D., ed., Ore deposits of the United States 1933—1967: Am. Inst. Mining Engineers, v. I, p. 72-94. Le Pichon, X., 1968, Sea floor spreading and con- tinental drift: Jour. Geophys. Research, v. 73, p. 3661—3697. Morgan, W. J., 1968, Rises, trenches, great faults and crustal blocks: Jour. Geophys. Research, v. 73, p. 1959—1982. Regan, R. D., Cain, J. C., and Davis, W. M., 1975, A global magnetic anomaly map: Jour. Geophys. Research, v. 80, p. 794—802. Root, S. I., 1970, Structure of the northern terminus of the Blue Ridge in Pennsylvania: Geol. Soc. America Bull, v. 81, p. 815—830. Sclater, J. G., and McKenzie, D. P., 1973, Paleo- bathymetry of the South Atlantic: Geol. Soc. America Bull, v. 84, p. 3203—3216. MEMORIAL SYMPOSIUM Sheinmann, Yu. M., 1973, The new global tectonics and reality [in Russian]: Bull. Moscow Soc. Naturalists, Geol. Ser., v. 78, no. 5, p. 5-28. Sillitoe, R. H., 1972a, A plate tectonic model for the origin of porphyry copper deposits: Econ. Geology, v. 67, p. 184—197. 1972b, Relation of metal provinces in western America to subduction of oceanic lithosphere: Geol. Soc. America Bull, v. 83, p. 813-817. Sims, S. J., 1970, The Grace magnetite deposit, Berks County, Pennsylvania, in Ridge, J. D., ed., Ore deposits in the United States 1933—1967: Am. Inst. Mining Engineers, v. I, p. 108—124. Stewart, J. H., 1976, Late Precambrian evolution of North America: Plate tectonics implications: Geology, v. 4, no. 1, p. 11—15. US. Geological Survey, 1974, Mosaic of imagery from Earth Resources Technology Satellite-1 of the conterminous United States: in cooperation with the US. Department of Agriculture, Soil Con— servation Service, 125,000,000. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Landsat Data Contributions to Hydrocarbon Exploration in Foreign Regions By F. P. Bentz, Vice President, Santa Fe Minerals, Inc., Dallas, Texas 75219 and S. I. Gutman, Geophysicist, Santa Fe Minerals, Inc., Orange, California 92668 ABSTRACT Foreign exploration frequently requires that large areas of relatively unexplored territory be evaluated in the most expeditious and informative manner. The selection of areas with the greatest potential would normally start with the least expensive reconnais- sance tools like photogeology, field work, and aero- magnetics before proceeding to the more costly and possibly more definitive techniques such as gravity, seismic, and, eventually, drilling. In many areas, how- ever, this ideal sequence of evaluation can no longer be followed due to shortened exploration periods and other restrictive government regulations. Santa Fe Minerals‘ past experience with Landsat has proved that the ready accessibility of multispec- tral imagery provides for quick and inexpensive re- connaissance of foreign exploration areas. Firstly, it allows the construction of geographic base maps which are often more accurate than any existing maps. Secondly, Landsat imagery is an invaluable source of geologic information; if used in conjunc- tion with existing published data, it will, in most cases, improve the accuracy of geologic mapping and un- derstanding of an area. In addition, it has been found that the imagery complements and aids in the interpretation of aeromagnetic and gravity data; this relationship is reciprocal. Santa Fe Minerals has successfully experimented with Landsat multispectral imagery,‘ and, as a con- sequence, now routinely integrates Landsat data into its exploration efforts. Exploratory work in Egypt and Yemen serves to illustrate how Landsat imagery is used; some results are presented. INTRODUCTION The purpose of this paper is to present some of the major contributions of Landsat data to Santa Fe Minerals’ hydrocarbon exploration in foreign areas. Explorationists find themselves today in a highly competitive environment, complicated by shortening exploration periods and increasingly restrictive gov- ernmental regulations. For these reasons it is not always possible to follow a normal foreign explora- tion sequence, which is essentially a process of iso- lating and identifying targets with hydrocarbon po- tential from a large and relatively unexplored area. This process usually begins with reconnaissance tools such as aeromagnetometry, SLAR (side-looking air- borne radar), and photogeology and proceeds to the more costly and time consuming (but hopefully more definitive) methods such as seismic surveys and drill— ing. Landsat data, because of their ready accessibility, help significantly in shortening the early reconnais- sance stage of operations. In addition, Landsat imagery is, because of its low cost per unit area, one of the best data buys the explorationist can make. Landsat has proven itself to be a valuable source of geological information. It has invariably improved the contact and spatial relationships of outcrops in poorly or sparsely mapped areas. The mapping of linears helps to define a structural fabric or frame- work in the exploration area. This fabric is often in- distinct or invisible at the surface but, nonetheless, strongly influences hydrocarbon accumulations. With few exceptions, the pattern of linears is highly regular and can often be related to stress analysis. Subtle color and tonal anomalies have been observed which 83 84 FIRST ANNUAL PECORA are believed to be of significant exploration im- portance. Significant relationships between gravity and mag- netic data and alinements or trends have been ob- served in Landsat imagery. Initial conclusions suggest that the basement and intersedimentary structures, which are routinely interpreted from the potential field data, have greater (but subtle) surface expres- sions than we had previously anticipated. Because it has been successful, Landsat data are now routinely integrated into Santa Fe’s foreign ex- ploration programs. It aids in planning geophysical surveys and allows the creation of exploration base maps onto which the results of literature searchers and available data can be posted; one is now able to begin the data acquisition and interpretation process almost immediately. Exploratory work in Egypt and Yemen will serve to illustrate some of the contributions that Landsat data have made to our foreign exploration program. The choice of Egypt and Yemen was made because they essentially represent opposite ends of the informa- tion spectrum available to the explorationist. The permit area in Egypt has well-exposed surface geology, portrayed on relatively detailed geologic maps. In addition, there are subsurface data avail- able from nearby wildcat tests as well as Bouguer gravity and residual magnetic intensity maps. In con- trast, the onshore exploration area in Yemen is cov- ered with alluvium. There are no well data available from the five dry holes previously drilled there in the early 1960‘s nor is there an adequate or reliable geological map of the area. Despite the current in- terest in the southern Red Sea (as a consequence of the new global tectonics), there is little in the pub- lished literature concerning the onshore geophysics of the Yemen Arab Republic. LANDSAT DATA CONTRIBUTIONS TO HYDROCARBON EXPLORATION IN EGYPT The East Cairo concession straddles the Cario to Suez highway and encompasses about 1 million acres (fig. 1). It is bounded on the east by the Great Bitter Lake and on the west by the Damietta Branch of the Nile River where it comprises a small portion of the highly cultivated and densely populated Nile Delta. The rest of the concession area consists of gently rolling gravel desert bounded to the south by east- west trending limestone escarpments of Eocene age. These physiographic provinces are an expression of the geology, as shown on figure 2, a geologic sketch map reproduced with permission from Said’s The Geology of Egypt (Said, 1962). The 30th parallel MEMORIAL SYMPOSIUM (crossing the map in the center) forms the southern boundary of Santa Fe’s permit. The bulk of the escarpment-forming Eocene lime- stones lie south of the 30th parallel. Most of the con- cession is characterized by east-west trending out- crops of Oligocene to recent age, with Eocene rocks showing only in a few anticlines. Exposures of Cre- taceous sediments are reported from Gebel Shabra- wet overlooking Great Bitter Lake. The east-west trending outcrop pattern is indicative of the apparent folding axes; the main pattern of faulting, shown on the map, has a northwest-southeast direction. Figure 3 (p. VIII) is a Landsat mosaic which en- compasses a much larger area than the geologic map. One can orient oneself by Cairo at the neck of the delta in the west, the Suez Canal and Great Bitter Lake on the east, and the Mokkattam lime- stone escarpment forming the southern concession boundary to the south. Quite a number of geologic features shown on Said’s map can be readily identified on the Landsat image mosaic; for instance, the Eocene horst of Gebel Oweibid. Interestingly, there are many other geo- logic patterns which are more visible on the Landsat imagery than on the surface geologic map. As an ex— ample, note the almost circular syncline southwest of Gebel Shabrawet which is rather inadequately repre- sented on the geologic map. A great number of major and minor Iinears can readily be observed on the Landsat mosaic, as shown in figure 4. Close scrutiny reveals that these Iinears . rranea we " s e / 99 o Abandoned well a Gas well - Producing well 80 MILES O 20 4060 0 20 40 60 80 KILOMETERS FIGURE 1.—|ndex map of northern Africa showing Santa Fe MineraIs, |nc., exploration area. LANDSAT DATA CONTRIBUTIONS TO HYDROCARBON EXPLORATION 85 Abu Zo'ubov . 32-00“ 32' 30' Greot Bitter ~~~~~~ 32° 30‘ SP 30' 32°00‘ °A_ ,r._,?f5__ _ so Km - W“ Pleistocenea Recent + m no . Eocene d a Oligocene 05.55524 Pliocene E Nubia Sandstone m Middle Son 5 Gravels Miocene FIGURE 2.—-Geo|ogy of Cairo-Suez District (after Said, 1962). statistically represent a limited number of trends. A presentation in the form of a star diagram not only enhances their regularity but also shows that they are intersecting each other at angles of approximately 15°, 30°, and 60: (fig. 5). These are the expected angles for shear sets of a strain ellipse, and they can also be related to Moody and Hill’s wrench-fault tec- tonic scheme (fig. 6) first published about 20 years ago (Moody and Hill, 1956). This does not imply that the Cairo-Suez area is governed entirely by wrench faults. In an area that lies at the junction of seVeral important crustal plate boundaries which separate the African continent from the Mediterranean crust to the north and from the Sinai and Arabian blocks to the east, one can expect that divergent movements of these crustal blocks created a forceful stress system. This was likely re- lieved by a complex pattern of shear zones, compres- sional features, and tensional rifts. The major structural units are illustrated in a sim- plified manner in a regional tectonic sketch (fig. 7). The Gulf of Suez—Red Sea system is one of the most outstanding features in this area. It likely con- sists of a combination of tensional rifting and strike- slip movements (left lateral wrench faults). Initial block faulting occurred toward the end of Oligocene time and was associated with widespread volcanism. If sea floor spreading is indeed a factor in the later history of this area, it was restricted to the Red Sea proper and did not penetrate the Gulf of Suez. The Gulf of Aqaba—Dead Sea shear zone is equally impressive in its magnitude. It is probably along this left lateral wrench fault that any plate movement caused by Red Sea spreading occurred. A distinct N. 50° W. trend occurs at a 15° angle to the N. 35° W. Gulf of Suez—Red Sea trend. Said com» bines the two trends under the name “Erythrean or African faulting”; and, indeed, the shorelines 0f the Gulf of Suez and the Red Sea are controlled by 599‘ ments of both of these two trends (Said, 1962, p. 33). The N. 50° W. direction appears to be much older since it is expressed in shears within the Precambrian shield (Abdel—Gawad, 1969). In addition, the N. 50° W. direction was certainly rejuvenated in more recent times causing some of the more prominent right lateral shears in our area of interest. FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 86 a. #933 *0 2308 “33:3 to COEEEEBE EmEmchlé NEED: can ya ESEME OmAavU defi (qu0 a >% 33m “a u mm. g,» LANDSAT DATA CONTRIBUTIONS TO HYDROCARBON EXPLORATION 87 Primary Stress N. 20° W. East Atrican N. 10° w. N. 50° w. QGull ol Suez . ° w. N 35 Gull ol Aqaba IN' 1 N. 50° E. Primary Folds Drag Folds f N' 70° E' N. 65° w.\ .__‘ - 10 Linears all 223 Linear: Mapped FIGURE 5.—Star diagram Illustrating relationships between mapped lineaments. Primary-Stress Direction \ Complementary 1 st order wrench \\ \{If \ I Primary lst order wrench \ 2nd order wrench 2nd order drag fold ‘ Right lateral Len lateral 3'“ "‘9' 2nd order wrench 2nd order wrench L m t it???” e aera '9 3rd order Right lateral \g” 2nd order wrench drag Md \\ %/ Y Primary-Fold Direction / ‘ Primary-Stress Direction where a=0° hm, Q c=15° (‘7‘, FIGURE 6.—Wrench fault tectonics as related to interpretation. lineament The north-south or “East African" trend is described by Said as “the old grain of Egypt" (Said, 1962, p. 32), yet it still seems to have played an important role in more recent geologic events, being responsible for deflections of the Gulf of Suez shorelines and of the Nile River. An east-west trend related to the “Tethyan” direc- tion is prominent in Sinai and crosses the concession area in several places. At Gebel Oweibid it can be interpreted as a right lateral wrench fault based on the outcrop configuration shown on the geologic map. A major folding axis, described as the Gebel Maghara—Abu Roash line. In a N. 70“ E. direction, at right angles to the primary stress oriented at about N. 20° W. This primary fold axis is not obvious from the surface geologic map of crosses 0U!" area 29° 28° / fl/ / é 2,. ‘\ Xi 100 KILOMETERS ( l—g—i \\\\\/ ,_ (if, Center: at Cretaceous and Tertiary deformation / Major Antlclineel Trends / Malor Faults and Fault Zones V/////// Aruba-Nubian Mess" FIGURE 7.—-Regional tectonics of Northeast Africa. (Adapted from Said, 1962.) our concession area; however, the magnetic intensity (fig. 8, p. IX) and the Bouguer gravity map (fig. 9, p. X) do reflect this structural orientation. Considering the regional aspects of this main axis, one can discern a southwesterly plunge from the Jurassic outcrops at Gebel Maghara in the Sinai to the Middle Cretaceous outcrops at Gebel Shabrawet near the Great Bitter Lake and to the Lower Creta- ceous and Jurassic encountered 2,000 ft below the surface in the Abu Roash wells. At right angles to the primary stress lies also the direction of greatest tension which could, at least in part, be responsible for the rifting of the Gulf of Suez. And while a number of the mapped lineaments are probably related to strike-slip movements, others are associated with block faulting and tilting. It is be- lieved that the major movement along the Gulf of Suez trend was a vertical displacement along nwmal faults resulting in tilted fault b10CkS. This can be illustrated by the uplifted Middle Cretaceous sedi- mentary rocks at Gebel Shabrawet west of the Great Bitter Lake. A similar uplift along the same direction of faulting probably occurred at Abu Roash west of the Nile Valley near Cairo. This set of block faults 88 / FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM would result in a steepening of the southwestward axial plunge across the area. Evidence for horizontal displacements is illustrated by a set of drag folds that forms a distinctive pattern in the center of the concession area. In the course of our work, we have observed a surprising correspondence between gravity and mag- netic data and Landsat imagery. It usually is in the form of close associations of linears and lineaments with high gradients and magnetic accidents (fig. 8, p. IX). The magnetic anomalies which are essentially reflecting the configuration of the magnetic basement dramatically confirm the substance of the N. 70° E. regional trend that was first observed on the Landsat imagery. Please note the departure of the field from this trend in the south-central portion of the permit area. We have interpreted this as the reorientation of the basement along a major left lateral strike slip fault seen in Landsat data. This is an excellent ex- ample of how the potential field data and the Landsat imagery complement each other to arrive at a unified and consistent geological picture. The correspondence of the Bouguer gravity anom- alies (fig. 9, p. X) with the surface geology (fig. 2) is quite good. Note the roughly east-west alignments of gravity highs over the folded and faulted Oligocene sedimentary rocks in the north and observe that a broad, regional trend of N. 700 E. is disturbed by high frequency anomalies associated with contacts and structures, particularly in the center of the area. In the west, where there is a total absence of geologic outcrops due to the vegetative cover of the Nile Delta, we have a broad, elongated gravity low which trends north-south. The Landsat interpretation highlights the contact and fault relationship inherent in the gravity data as well as the regional N. 700 E. trend which we feel is quite important to hydrocarbon accumulations east of the Nile Delta. Please observe that jogs in the course of the Nile are associated with fiexures in the gravity anomalies and the extensions of lineaments mapped outside the delta. We conclude that not only is the present course of the Nile largely fault controlled but that these faults influenced the development of a gorge 0f Grand Canyon proportions which formed at the end of Miocene. This gigantic 10,000-1‘t deep ero- sion channel (discovered by Santa Fe’s seismic reflec- tion work) helps to explain the conspicuous north- south gravity low mentioned above. LANDSAT DATA CONTRIBUTIONS TO HYDROCARBON EXPLORATION IN YEMEN The Yemen Arab Republic lies on the southwestern tip of the Arabian Peninsula (fig. 10). The area in which Santa Fe Minerals was interested encompasses more than 5.7 million acres, of which close to 4 million acres are located onshore. A rather restricted explora- tion option of 8 months required us to survey this vast area in the most expeditious and cost effective man- ner. For this purpose, an abbreviated exploration program was undertaken consisting of an offshore seismic survey combined with marine gravity and magnetics. An aeromagnetic Survey and Landsat inter- pretations were conducted over the onshore portion. The Landsat mosaic (fig. 11, p. XI) gives an indica- tion of the nature and quality of the data as well as a feel for some of the interpretation problems that were encountered. Saudi i’ Arabia} j Ethiopia S. Y _ ’ emen /// l/Afars and lssas Terr. G"If of Ade“ FIGURE 10.—lndex map showing Santa Fe Minerals, Inc, ex- ploration area in Yemen. LANDSAT DATA CONTRIBUTIONS TO HYDROCARBON EXPLORATION 89 The Tihama Plain extends 400 km north to south along the Red Sea Coast. This coastal plain gently dips to the west from a mountain front which dra- matically rises to heights of more than 8,000 ft. Al- though faulting, fracturing, and jointing are clearly visible in the mountains, at first glance there appeared to be an almost total lack of these features in the plain due to the thick alluvial cover. On closer inspec- tion, however, it was observed that despite the lack of outcrops in the Tihama Plain, there are a profusion of subtle color or tonal streaks as well as circular anomalies visible from orbit which are totally indistinct either on the ground or in the aerial photos (fig. 12). Lineaments in the mountains define a structural fabric which is relatable to the stresses that this area has been subjected to as a result of regional upwarp or arching during the Oligocene and rifting during the early and middle Miocene. Linears in the coastal plain echo the structural fabric of the mountains lending support to the possibility that these color and tonal anomalies reflect in some way the configuration of the subsurface, specifically fault blocks and attendant structures. Subtle circular anomalies are observed in the Land- sat images and appear to be concentrated north of lat 14°30’N. Salt is being mined from outcrops of Miocene age which have surfaced along the Red Sea Coast, particularly around Salit (El Shazly, 1967). It is also understood that an oil company had drilled sev— eral of its wells on gravity lows, suggesting that salt domes were its exploration targets (privileged com— munication, 1975). The aeromagnetic survey‘firevealed that no magnetic anomalies are directly associated with these features. As a consequence, it is confident- ly felt that some of the circular anomalies refiect salt diapirs. Their concentration north of lat 14°30’N. sug- gests a thickening of the section in that area, strength- ening the results of the aeromagnetic interpretation. The subtle circular anomalies (such as those tar- geted on fig. 13) are best seen in band 7 images and in stretched false-color composites. While most of the anomalies appear as breaks in the tonal pattern of the plain, some are associated with vegetation occur— 'rences. Of the five wells drilled by an oil operator in the early 1960’s, three of them are located on circular anomalies. It is observed that all of the wells drilled on circular anomalies lie on the flanks rather than on top of structural basement highs as interpreted from the aeromagnetic data, further supporting the conten- tion that these circular anomalies reflect the presence of numerous diapirs. 44' // C \ 4‘ _//v i 16° 1 /\ 11 § - _/ //f W“ {l i 4, 4’ Lineaments in mountains and Seismically interpreted Faults offshore ‘3; Linear Tonal Anomaly Circular Anomaly Earthquake Epicenter FIGURE 12.——|nterpretation of lineaments, circular and tonal anomalies from Landsat data. The color or tonal anomalies to which we alluded are highlighted in a portion of Landsat—1 scene 1117— 06562 (fig. 14). The streaks, trending southwesterly appear to be related to the onshore extension of fault- ing as interpreted from our offshore seismic program. We feel that these represent a reorientation of the drainage patterns as a consequence of southwest tilt- ing of the section along these faults. The alinement of northwesterly trending streaks suggests that they may be the expression of faulting parallel to the Red Sea rift axis. Stream of‘fsets along one such tonal anomaly indicate approximately 3 km of right lateral offset. The association with earthquake FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 90 0935 zo—m-mu 20*“) O 00! IWTIDO .Nnow \m_‘ 33030 \wronoINwor 78%ch .BGCQEOU Lo_0u.m.m_fl vmucmsco >:qu¢me80;& m :0 UmcccwE >:mc_m:o 95:82 :m>> new mosz0cm _m_:u:UI.mF $50: .0 m-:§..N‘—n _§r4.m._mrz_;-m.vm:._fi..n$wa and .8 m m 8: Sawwflmwwwwz z 8-~$u\nm-m_z 0 «C82 I'M-MI lmN'“ 30%|“! “WIMI _8-mvom .8-§ 84g. LANDSAT DATA CONTRIBUTIONS TO HYDROCARBON EXPLORATION 91 FIGURE 14.—Detai| of streaks and tonal anomalies on Yemen coastal plain, Landsat—l image 1117—06562, bands 4, 5, and 7, November 17, 1972. epicenters (Fairhead and Girdler, 1970) further indi— cates that these streaks are the surface expressions of faulting (fig. 12). Compagnie General de Géophysique (COG) flew a 4,600-km aeromagnetic survey for this exploration effort in April 1975. The results of this survey were very gratifying because they not only confirmed the results of the offshore work but made many of the features seen in the Landsat imagery understandable in a structural context. A few of our observations may serve to illustrate this point. A north-northwest—trending fault system (inter- preted by CGG) parallels the mountain front and some of the linears observed within and at the foot of the mountains. The extensions of some lineaments from the moun- tains onto the Tihama Plain coincide with high mag- netic gradients not initially defined as faults. This aids in amplifying the interpreted fault pattern. Conversely, magnetic events interpreted as base- ment faults coincide with stream courses emerging from the mountains, suggesting structural control of some of the drainage patterns. The fianks and basinal axis of a 3-km deep mag- netic anomaly are reflected by surface linears. Numerous other interrelationships between mag- netic data and linears in the Tihama Plain assist in deciphering the structural development of the area. It is of particular interest to note that Landsat- observed linears are not solely related to fault and fracture patterns but often reflect deep-seated struc- tural trends and a variety of geologic phenomena. Considering its demonstrated versatility, low cost, and ready access, Landsat imagery deserves to be applied to all integrated exploration efforts, foreign or domestic. ACKNOWLEDGMENTS We wish to express our appreciation to Santa Fe International Corporation for encouraging us to pre- pare this paper. Our thanks to Dr. M. K. El Ayouty of EGPC for permitting us to present our findings in Egypt and to Elsevier Publishing Company for permis- sion to reproduce Said’s geologic map. We also thank Peter Coberly and Michael Wallace of Custom Color Labs in North Hollywood, California, who produced our Landsat false-color composites and Santa Fe’s Art Department which produced the illustrations for this paper. Special appreciation is due to Dr. William A. Fischer and Donald Orr of the USGS EROS Pro- gram for inv1ting us to publicize our work with Land- sat imagery. REFERENCES Abdel-Gawad, M., 1969, New evidence of transcurrent movements in Red Sea area and petroleum impli- cations: Am. Assoc. Petroleum Geologists Bull., v. 53, no. 1, pp. 1466—1479. El Shazly, E. M., 1967, Oil shows in Yemen and oil bubbles in rock salt: Arab Petroleum Congress, 6th, Baghdad 1967, Paper no. 40 (8—3). Fairhead, J. D., and Girdler, R. W., 1970, The seis- micity of the Red Sea, Gulf of Aden and Afar triangle: Royal Soc. London Philos. Trans, ser. A, . 267, no. 1181, pp. 49—74. Moody, J. D., and Hill, M. J., 1956, Wrench fault tec- tonics: Geol. Soc. America Bull, v. 67, pp. 1207— 1246. 92 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM Lowell, J. D., Genik, G. J., Nelson, T. H., and Tucker, 8., eds, Petroleum and global tectonics: Prince- P. M., 1975, Petroleum and plate tectonics of the ton Univ. Press, pp. 129—153. southern Red Sea, in Fischer, A. (3., and Judson, Said, R., 1962, The geology of Egypt: Amsterdam, Elsevier Publishing Co. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Measurement of Luminescence of Geochemically Stressed Trees and Other Materials By W. R. Hemphill,l R. D. Watson,2 R. C. Bigelow,“ and T. D. Hessen,” US. Geological Survey ABSTRACT The Fraunhofer line discriminator (FLD) is an air- borne electro-optical device which operates as a non— imaging radiometer and permits detection of solar- stimulated luminescence several orders of magnitude below the intensity detectable with the human eye. Procedures employing a laboratory fluorescence spec- trometer permit prediction of the FLD detectivity of a material prior to mounting an airborne test. Lumi- nescence spectra of the material are corrected for wavelength variation in spectrometer source and detector sensitivity and solar illumination. By compar- ing these results with the luminescence of a rhodamine WT dye standard, the luminescence of the material may be expressed in terms of rhodamine dye equival- ency at the wavelengths of several Fraunhofer lines. The FLD detectivity may be assessed at each Fraun- hofer line, and the optimum line for field observation of the material may be selected. Airborne tests of a redesigned FLD permitted measurement of differences in the luminescence of trees growing in soils containing geochemically anom- alous concentrations of copper (near Denver, Colo- rado) and molybdenum (near Gardnerville, Nevada) from trees growing in background areas nearby. In the tests near Denver, luminescence contrast between background and stressed trees tended to be higher during periods when cloud cover was less than 10 percent: in most cases, contrast was insignificant when cloud cover exceeded 10 percent. Although diurnal variations were also observed, these do not appear to be predictable or systematic. In other air- borne tests, the FLD distinguished luminescing phos- phate rock and gypsum from sandstone and siltstone ‘Reston, Virginia; ‘Flagstaff, Arizona; ‘Denver, Colorado. near Pine Mountain, California, and dispersal of oil in a natural seep from seawater in the Santa Barbara Channel. INTRODUCTION Laboratory and field study of the measurement and interpretation of luminescent materials began in 1964 using an active '* ultraviolet imaging system (Hemphill and others, 1965; Hemphill and Carnahan, 1965). This work led to the development of a prototype Fraun- hofer line discriminator (FLD), an airborne remote- sensing tool for measuring luminescence. The FLD was designed to operate on the Fraunhofer line depth principle (Kozyrev, 1956; Grainger and Ring, 1962) which uses the Sun as an excitation source and per- mits detection of luminescing materials under daylight conditions. Experiments with the instrument showed that selected materials, such as luminescing tracer dyes could be detected in very small quantities (Hemp- hill and others, 1969; Stoertz and others, 1969). For example, rhodamine WT dye was detected in the Pacific Ocean west of the Golden Gate in California in concentrations of less than 5.0 ppb (parts per bil- lion). Attempts to measure the luminescence of other materials were not successful because (1) the proto- type FLD operated at 589.0 nm (nanometres), the wavelength of the sodium D Fraunhofer line, and many materials luminesce at other wavelengths and (2) the sensitivity was an order of magnitude less than required for detection of some of these materials. In order to predict optimum wavelength and sensi- tivity requirements for detection of materials other than rhodamine WT, luminescence of a variety of "An “active” system uses an artificial source, in this case a cathode ray tube. to stimulate luminescence. 93 94 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM materials was measured with a laboratory fluores— cence spectrometer in terms of rhodamine WT, used as a laboratory standard. These results, coupled with those obtained with the prototype FLD. were used as a basis for redesigning an FLD with an order of mag- nitude increased sensitivity. This redesigned model, built by the Perkin-Elmer Corporation,“ operates at three discrete Fraunhofer lines, hydrogen [3 at 486.1 nm, sodium D at 589.0 nm, and hydrogen a at 656.3 nm and has the sensitivity to detect rhodamine WT dye in concentrations as low as 0.1 ppb in 0.5 m of water at 20° C. The objectives of this report are (1) to present the results of laboratory and field measurement of lumi- nescence of geochemically stressed trees, phosphate rock, and crude oil and (2) to demonstrate how the laboratory measurement of luminescence can be used to predict and interpret field measurements with the FLD. Since 1971, development of the redesigned FLD by the Perkin-Elmer Corporation, as well as associated laboratory and field studies, has been supported by the Advanced Applications Flight Experiments (AAFE) Program of NASA (NASA Order 1:58.514), the Earth Resources Observation Systems (EROS) Program of the Department of the Interior, and the Geologic Division of the US. Geological Survey. FRAUNHOFER LINE DEPTH METHOD Fraunhofer lines are dark lines in the solar spectrum caused by selective absorption of light by gases in the relatively cool upper part of the solar atmosphere. Line widths range from less than 0.01 nm to several tenths of a nanometre, and the central intensity of some lines is less than 10 percent of the adjacent continuum. The lines are sharpest, deepest, and most numerous in the near ultraviolet, visible, and near- infrared regions of the electromagnetic spectrum. The Fraunhofer line-depth method of measuring luminescence involves observing a selected Fraun- hofer line in the solar spectrum, and measuring the ratio of the central intensity of the line to a convenient point on the continuum a few tenths of a nanometre distant; this ratio is compared with a similar ratio of a conjugate spectrum reflected from a material that is suspected to luminesce. Both ratios normally are identical, but luminescence is indicated where the reflected ratio exceeds the solar ratio. Variation of "’ Use of trade name in this paper is for description purposes only and does not constitute an endorsement of the product by the U.S. Geological Survey. reflectivity with wavelengths is negligible for most materials over spectral ranges of only a few tenths of a nanometre; therefore, reflectivity differences be- tween the central intensity of the Fraunhofer line and the adjacent continuum can generally be ignored. Equations illustrating the relation of Fraunhofer line depth ratios to luminescence are presented in the Manual of Remote Sensing (Am. Soc. Photogram- metry, 1975, p. 116—118). LABORATORY PREDICTION OF THE FRAUNHOFER LINE DETECTIVITY OF LUMINESCENT MATERIALS In order to assess the sensitivity required to detect luminescing materials, Hemphill and Stoertz (1971) and Watson and others (1973, 1974) used a laboratory fluorescence spectrometer (fig. 1) to measure the lumi— nescence intensity of materials at six Fraunhofer lines (3968, 422.7, 486.1, 518.4, 589.0, and 656.3 nm) in terms of equivalent luminescence of specific concen— trations of rhodamine dye. The spectrometer was operated so as to produce excitation spectra; that is, the excitation monochromator was scanned, exciting the sample with monochromatic light in the near ultra- violet and visible spectrum, while the emission mono- chromator was stationary at the wavelength of a specific Fraunhofer line. This arrangement provided a system in the laboratory analogous to an FLD in the field, where the broadband excitation of the Sun pro- duces an emission in a luminescent material, the in- tensity of which is monitored at one or more Fraun- hofer lines. The spectrometer, equipped with a FIGURE 1.——Laboratory fluorescence spectrometer. Source (A), excitation monochromator (B), sample compartment (C), emis~ sion monochromator (D), and recorder (F). corrected spectra attachment (E), MEASUREMENT OF LUMINESCENCE 95 corrected spectra attachment, automatically adjusted for the wavelength dependence of the source and detector and produced a corrected distribution of excitation intensity. To relate all luminescence spectra to one set of conditions, rhodamine WT was used as the reference ”standard” prior to each measurement. When the area under the curve of the excitation spectra of the sample was compared to the area under the curve for rhoda— mine WT dye (at a specific dye concentration), a rela- tive equivalent rhodamine WT concentration was obtained. For example, a sample having an integrated excitation intensity of 50 was compared to a rhoda— mine WT concentration of 10 ppb, which also has an integrated excitation intensity of 50; the sample is said to have had an equivalent luminescence of 10 ppb rhodamine WT. This standardization makes it possible to correct for instrument response variations during measurement and to assess results achieved in the laboratory in terms of whether or not the same ma- terial in the field would be within the sensitivity range of an airborne FLD. To obtain the same wavelength and intensity de- pendence of luminescence in the laboratory as would be observed in an F’ILD measurement, the source- detector corrected spectra must be convoluted with the spectral intensity of total solar radiation (direct sunlight plus diffuse skylight). A further correction is required when adjusting the standard of rhodamine dye as measured in a centimetre quartz cell in the spectrometer to that measured in the field, where sunlight penetrates the material and stimulates lumi- nescence at a depth that is a function of the material’s absorption. Descriptions of the solar and depth cor- rections are presented by Watson and others (1974) and by the American Society of Photogrammetry (1975, p. 118—126)). DESCRIPTIQN OF THE FRAUNHOFER LINE DISCRIMINATOR The FLD consists of an optical head, electronic console, and light collector as shown in figure 2. The main components in the optical head are two tele- scopes, one Earth-looking and one sky-looking; a ro- tating optical chopper wheel; three replaceable optical filter sets; and a photomultiplier with its power supply. Sunlight and skylight falling on the diffuse surface of the light collector are reflected by a mirror into the telescope of the sky—looking channel. The Earth- looking telescope observes the target whose reflec- tivity and luminescence are to be measured. Light from the two telescopes is sequentially routed through two different paths by the rotating chopper wheel. In one path, light passes through a filter which is cen- tered at a specific Fraunhofer line but whose band- pass is several hundredths nanometres wider than the Fraunhofer line. This signal constitutes the light intensity measured on the solar continuum adjacent to and including the Fraunhofer line and is designated signal “a” in the sky-looking channel and signal “d" in the Earth-looking channel. In the other path, a Fabry—Perot interference filter, with half—width of less than 0.07 nm, passes light coincident with the intensity of the central part of the Fraunhofer line; this signal is designated signal “b” in the sky-looking channel and signal “c” in the Earth-looking channel. These signals are fed to a mini-computer that generates lumines- cence and reflectance by solving the following equa- tions: p= (d—C)/(a—b); 1‘: (d/a)~p, where p is reflec- tance and r' is the luminescence coefficient. Both p and r are displayed as four-digit numbers (from 0000 to 9999 counts) on the front panel and are fed to a digital printer for permanent record; hence, 100 counts for r implies a luminescence of 1 percent. Re- flectance values are displayed and recorded as per- cent directional reflectance. Inasmuch as displayed and recorded luminescence are only proportional to true luminescence for each substance measured, the luminescence count is referenced to the count from a standard (such as a dye sample or a photographer’s gray card), permitting relative luminescence measure- ments to be made. Reflectance and luminescence values are presented in this paper in terms of their mean, plus or minus the standard deviation of the mean (Hoel, 1965, p. 139). FIGURE 2.——Perkin-Elmer FLD, showing the optical head (A), electronic console (B), and light collector (C). 96 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM An automatic gain control is used to monitor and maintain the a—signal at a nearly constant level by varying the gain of the photomultiplier tube (PMT) detector. This permits measurements under a wide range of illumination and allows the PMT to operate at full gain under low illumination. Three sets of Fabry—Perot filters are available, permitting measure- ment at the 6563—, 5890—, and 486.1—nm Fraunhofer wavelengths. Additional filters can be made at other Fraunhofer wavelengths, including 396.8, 422.7, and 518.4 nm. Ground tests show that rhodamine WT dye can be detected at the 589.0-nm Fraunhofer line in 0.5 m of water under clear sky conditions in concentrations as low as 0.1 ppb; however, sensitivity under field con- ditions is generally considered to be 0.25 ppb. The relationship between dye concentration and lumines- cence measured by the instrument is approximately linear, with 1.32 ppb equating to 100 FLD counts. Luminescence measurements were about 25 percent higher under clear sky conditions than under overcast conditions. Figure 3 shows the FLD optical head and light col- lector installed in a basket atop the landing skid and adjacent to the fuselage of a helicopter. The system also includes (1) a television camera with a remotely controlled focus and zoom lens for monitoring the target measured with the FLD, (2) a second television camera which monitors a bank of light emitting diodes displaying reflectance and luminescence data that are continually updated by the digital output of the FLD, (3) a tape recording system for recording the data from both cameras and additional voice description, and (4) a television monitor on which is displayed in real time the target scene upon which is superimposed the reflectance and luminescence counts. Boresight- ing of the viewing system is achieved by positioning a luminescence panel on the ground and marking the position of the panel on the television screen by the luminescence indicated while hovering over the panel. The viewing system has proven dependable and per- mits discrimination of small luminescence targets which would otherwise be lost in data reduction. It also permits storage of both video and digital data in a form which can be replayed and analyzed in the laboratory. Current operational procedures for airborne meas- urements with the FLD are as follows: 1. Optical filter and electronic warmup—3O minutes. 2. Prefiight calibration using reflectant and lumines- cent chips. aawcr % V g. “afiofiéfii‘mfi‘éfi FIGURE 3.-—FLD optical head and light collector installed in a basket mid-ship of a Bell Jet Ranger helicopter. 3. Infiight systems check using a ground-deployed luminescent panel. 4. Airborne luminescence and reflectance measure- ments over target area. 5. Postfiight calibration using refiectant and lumines- cent chips. Air temperature, extent of cloud cover, and other meteorological parameters, as appropriate, are re- corded throughout the clay. GEOCHEMICALLY STRESSED VEGETATION The relationship between luminescence and photo- synthesis in plants is complex, but, in general, factors that both impede and enhance photosynthesis can modify luminescence (Hollaender, 1956). Several such factors exist in nature. Based on the air pollution work of O’Gara (1922), Thomas (1961, p. 236—237) ranked cultivated and native plants according to resistance to sulfur dioxide fumigation. Thomas noted substantial MEASUREMENT OF LUMINESCENCE 97 variability in sulfur dioxide susceptibility of some species, including conifers, but others were relatively constant. Alfalfa, barley, and cotton were most sensi- tive. Using alfalfa as unity, elm, birch, sumac, and poplar were about 2.5 times more resistant, pine 7—15 times more resistant, and live oak 12 times more re— sistant. Thomas (1961, p. 253) also noted that 3- to 4-month-old ponderosa needles were sensitive to hydrogen fluoride fumigation but that old needles were resistant. Another such factor is geochemical stress, where abnormally high concentrations of one or more metallic elements may subject plants to chemical and physiological changes. Copper and zinc are toxic if present in large quantities (Kramer and Kowlowski, 1960, p. 225-226) and are reported to produce symp- toms closely resembling iron chlorosis (Sauchelli, 1969, p. 65). J. E. McMurtrey (in Sauchelli, 1969, p. 156) notes that copper in excess of 0.1 ppm stunted tobacco plants. F. A. Gilbert (in Sauchelli, 1969, p. 158) states that growth of tomatoes in nutrient solu- tion was reduced by as little as 1‘ ppm copper. Sauchelli (1969, p. 112) cites zinc having a toxic effect in some cereal grains where levels exceed 150 ppm and in citrus where levels exceed 220 ppm. K. C. Beeson (in Sauchelli, 1969, p. 133 and 138) reports that concentrations of molybdenum in plants may vary from less than 0.1 ppm to more than 300 ppm without adverse effect on growth. Press (1974, p. 374—375) reports that bean plants grown in solutions containing varying concentrations of lead exhibited chlorosis and that spectral reflectance at 550 nm ranged from 15 percent for 1 ppm (part per million) lead to more than 30 percent for 500 ppm lead. Watson, Hemphill, and Hessin (1973) grew bean plants hydroponically and treated the nutrient solu- tions with varying metal concentrations, including molybdenum, copper, zinc, and lead, to demonstrate that luminescence is an indicator of geochemical stress produced by metal toxicity. Measurements were made both in the early stage of growth and at full maturity. Table 1 shows luminescence differences ob— served for 10 plants, 5 nonstressed and 5 stressed with 10 ppm sodium molybdate (NaMO.,). Both groups were grown in a Hoagland No. 2 solution. Differences TABLE 1.—Luminescence (source-detector, solar, in luminescence at 2 weeks are reversed at 5 weeks. The reversal not only occurs between stressed and nonstressed plants but also between large and small leaves. The difference in luminescence is reduced at 5 weeks for the large leaves. These results illustrate the complexity of luminescence during various stages of growth and tend to support the observations of Thomas (1961, p. 253) in his plant fumigation experi- ments noted above. To investigate the luminescence of needles of Pinus ponderosa, trees growing in a copper- and zinc-rich soil near Malachite Mine, JeffersOn County, Colorado (fig. 4), were measured in the field using the labora- tory fluorescence spectrometer and were compared to measurements of needles from background trees. Country rock where trees grow includes biotite schist and hornblende gneiss, and younger pegmatite dikes; all are of Precambrian age (Huff, 1963). Figure 5 shows the copper anomaly in the Mala- chite Mine area and the location of background and geochemically stressed tree groups. Measurements were made both diurnally (every 4 hours over a 24- hour period) and seasonally (every 4—5 weeks, July- November 1973 and April-May 1974). Data during the period from December 1973 through March 1974 could not be obtained because of heavy snow. Four twigs were collected from each of seven background and seven geochemically stressed trees making a total of 56 samples that were measured. The needles were removed from the twigs, placed on a black-matte non- luminescent surface, and measured in the front-surface mode in the spectrometer. The 4-hour spacing be- tween separate collection runs was determined by the time required for sample collection, measurements, and logistical considerations, but it was verified that the period of as much as 2 hours between time of collection and completion of measurement did not alter the luminescence of the needles beyond experi- mental error. Subsequent measurement of copper and zinc in the ash of the needles was by standard analytical tech- niques, and table 2 shows the results of chemical analyses of samples collected during late summer and autumn of 1973. Mean copper content of ash from stressed trees on four dates exceeds the mean copper 10 bean and depth corrected) of plants expressed in terms of rhodamine WT equivalency (ppb). The plants were grown hydroponica/ly in Hoagland #2 nutrient solution; five of these plants were geochemi— cal/y stressed with 10 ppm sodium molybdate (NaMoi). Stressed Nonstressed Description 2 weeks 5 weeks 2 weeks 5 weeks Top surface, large trifoliate leaves ____________ 0.62 1.05 0.45 1.19 Top surface, small trifoliate leaves ____________ .67 .64 .49 .98 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM [ 3 \ :E m ,/ / W... (1 COLORADO Lowereastponah ( Malachite mine .. ._/ N \. 3x: \2 .-/ \.. V is; r f ' 5 Idledale / 1,4.“ .f a. I, \ il .) GM : “ T 4 S I T 5 S ,3; /... FIGURE 4.—Location of Malachite Mine area, Jefferson County, Colorado. Stipple pattern shows area of figure 5. (From Huff, 1963.) EXPLANAUON Shall :‘Bock round _/ g /{ Portal 0! adit /_ 300\ Isopwihsindmonng sons of equal copper conionr In paris per minion Dashed where opproximomy located Dewy. COLORADO 0 100 ZOO 300 400 FEET lOO METRES CONTOUR INTERVAL 50 FEET TRUE wowru GNEr K ~0~7~ ”A Daclmallon, I963 FIGURE 5.—Distribution of copper in soil and location of sample trees growing in background and anomalous soils. (Modified from Huff, 1963.) MEASUREMENT OF LUMINESCENCE 99 TABLE 2,—Mean and standard deviation of copper and zinc content (ppm) of needle ash from 13 Pinus ponderosa growing in geochemical/y anomalous and background soils Seven stressed trees Six background trees Sample Date Copper Zinc Copper Zinc July 26, 1973 _________________________ 175:16 >2000 110:7 730:35 Aug. 30, 1973 ________________________ 180:8 2180:77 115:4 1130:30 Oct. 18, 1973 _________________________ 125:5 2330:36 100:7 700:84 Nov. 12, 1973 _________________________ 140:7 2080:35 100:4 880:30 content of the background tree ash by factors of 1.2 to 1.6. Mean zinc content of stressed tree ash exceeds the background trees by factors of 1.9 to 3.3. Distri- bution of copper in the soil in the vicinity of the mine is given by Huff (1963). Additional ash analyses for copper content of 16 trees in the area are given by Howard, Watson, and Hessin (1971). These analyses show that copper content ranges from 280 ppm to 660 ppm for eight stressed trees and from 120 ppm to 180 ppm for eight background trees. ‘ Figure 6 shows the spectral excitation curves ob- MPF—3 FLUORESCENCE SPECTROMETER Emission wavelength 656.3 Emission slit 1.0 tained in March 1973 for both groups of trees and also for rhodamine WT dye at a concentration of 1,200 ppb. The integrated excitation intensity is significantly greater for the background than for the stressed trees; solarcorrected rhodamine WT equivalents are 19 ppb and 5 ppb, respectively. Solar— and depth—corrected rhodamine WT equivalency is 0.4 and 0.2 ppb, re- spectively. In order to test the sensitivity of the FLD to chloro- phyll luminescence of stressed and nonstressed Pinus ponderosa, twigs were collected from seven stressed and seven background trees at the Malachite Mine site in early March 1974. Luminescence measurements were made within 2 hours outdoors at the Denver Federal Center and referenced to a nonluminescent gray-card standard. Mean and standard deviation of 100 _ Excitation wavelength Scan actuation slit 20-0 the mean of FLD luminescent counts was 107:0.5 an speed 3.0 Sensitivity 30.0 for background trees and 90:30 for stressed trees. 90 '- Remarks Water cooled: Front surface mode PINE NEEDLES Following this initial experiment, the FLD was used from a helicopter to measure the luminescence of E 80 i- _ m _ IEFZ 340 A ‘7 2° W stressed and nonstressed Pinus ponderosa trees at the 70- Em: 6563' :Ex in; 2::::::)nd) Malachite Mine, both on a diurnal and seasonal basis EX — for the period of September 1974 through July 1975. These results, combined with the results acquired 60— RHODAMINE WT: 1200;4ng iE :Eml A ,220 n," with the fluorescence spectrometer during summer x 440:1],748 and autumn 1973 and spring 1974, indicate that lumi- ' Em: 589.0 lEx RELATIVE QUANTA PER WAVELENGTH INTERVAL 8 nescence contrast between background and anomal- ous trees was independent of air temperature, rela- 40 .. Rhodamine WT 0‘ 72) tive humidity, and wind speed, but maximum contrast, 30 _ ranging from 1.4 to 2.2, tended to occur during parts Bac“"°“"° of 6 days when cloud cover was less than 10 percent 20— and the sun was not obscured by clouds (fig. 7)." On 6 days when cloud cover exceeded 10 percent, the 10 _ Stressed diurnal luminescence trace of both groups tended to be similar and contrast did not exceed 1.4 (fig. 8).6 o i i A Data collected on September 26, 1973, September 24, 300 400 500 600 EXCITATION WAVELENGTH (NANOMETRES) FIGURE 6,—Excitation spectra of Pinus ponderosa needles from geochemically stressed and background trees, Malachite Mine, Jefferson County, Colorado, as meas- ured with a laboratory fluorescence spectrometer (source-detector corrected only). integration of the area under each excitation spectrum permits comparison with the excitation spectra of a rhodamine WT standard. 1974, and April 22, 1975, are exceptions (fig. 9).6 A maximum contrast of 2.7 at 11:45 am. on September 26, 1973, was obtained during a period of loo—percent cloud cover. Contrast on April 22, 1975, does not 6Standard National Weather Service nomenclature: relative humidity in percent, wind speed in miles per hour, and cloud cover in percent. 100 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 907 90r JULY 26, I973 MPF—3 OCTOBER l8,l973 MPF-3 MAX CONTRAST IS AT I000 HRS. MAX CONTRAST 2I AT |300 HRS 80— Stressed trees 25 t 4 80- Stressed trees 30 t 4 Background trees I6 1 3 Background trees 6| 1 I5 70* / 70— 60— / 60» Cloud cover 50— 50— Background trees Relative hum1d1ty 40— \ i 40— a \ \\ Stresse trees \ 30— 30— Stressed trees . V ”— Relative numldlty 20 — / 20— \ / wind speed \" Clo d 10.. Background trees Wind speed 10> U W"; _ / —" \ __ _ __ — _ _/ / \ \ /- “a o 1 I Jr 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1/ 900 I000 IIOO I200 I300 I400 l500 IGOO I700 900 IOOO “00 I200 I300 I400 I500 I600 I700 90— 90 APRIL 30, l974 MPF—3 FEBRUARY 25, I975 FLD MAX CONTRAST 22 AT “45 HRS MAX. CONTRAST |.5 AT I445 HRS 80* Stressed trees 35 1 8 80” Stressed trees 47 2 6 Background trees I? 1 6 Background trees 73 1 4 7O —- 70— Bockground trees 60 — 60 — Relative humidity a \ g 50 — < 50: 1: . a1 Stressed trees 5 4o — \ 40— < Anomalous 3O _ 30_ trees *Background trees Stressed ""5 ‘ Relative humidity 20 — 20 — __ __ ',- x __ Background trees / '- ,0 I. _ |0_ _ __ W1nd speed , d \ . \ / __ medfl / \‘Cloud cover Cloud coverE: \ \ o 1 1 1 1 4 1 J o 1 1 1 \ 1 1 1 1 1 900 IOOO IIOO I200 I300 I400 I500 I600 I700 900 IOOO IIOO I200 I300 I400 I500 IGOO I700 90 I— MAY 24,1975 FLD 907 JUNE 23,1975 FLD / MAX CONTRA$T1.6 AT IOOO HRS. MAX CONTRAST I4 AT 0900 HRS 80— 80— Stressed trees 25 I 6 Background trees I8 1 3 Stressed trees 48 t 7 70 '— Bockground trees 7316 70 _ / Clo d on 60 _ Background trees 60 _ u ver\7 / 50»— Stressed trees 50F /Relolive humidity / / 40 — 4O — \ \\ REIOIIVQ humidity Stressed Z 30 ~ ’_ \ 30 ” trees / \ \ / Background trees 20 2 20 - Wlnd speed —— l0 — 10 — \ _/ // \ / ' -—’ — ——/ Cloud cover \ \ /_ -—— ; Wlnd speed 0 _; _1_ _c A _L _1 1 1 o I 1 / 1 1 1 ' 1 900 IOOO IIOO I200 I300 I400 I500 I600 I700 900 IOOO IIOO I200 I300 I400 I500 IGOO I700 MOUNTAIN STANDARD TIME FIGURE 7.—Tempora| luminescence of Pinus ponderosa and meteorological parameters. Maximum lumines- cence contrast between background and anomalous trees tends to occur during periods of minimum cloud cover. Standard deviation of the mean, computed for the time of maximum contrast, is at the 95—percent confidence level. MEASUREMENT OF LUMINESCENCE 101 ’— — —— ~ — Q IOO — — ‘ _ 90 ———., \ \— — — — Cloud cove, AUGUST 30. I973 MPF-3 F Cloud cover NOVEMBER 6, l973 mar-3 MAX CONTRAST |.2 AT “30 HRS \ 80— MAX CONTRAST II AT I200 HRS, 80— Stressed trees 30 13 Stressed trees 38 1 3 Background trees 25 t 2 Background trees 42 1 2 70 — 7o _ 5° — so P Relative humidity 50 _ 50 __ /\ \ / \ \ 40 \ RelatIve humidity 40 _ \ \ ‘ \ Stressed trees 30 / 30— Stressed trees Background "“5 \ __ __ _ 20 20 ~ Background trees . d d Wind speed IO _ Win spee IO _ __ __ fi __ __ __ _ -\ __ _ __ #— \ \ ’- I I I I I I I I I 0 I I I I I I l I\ ‘I 0900 I000 ”00 I200 I300 I400 |500 I600 I700 900 I000 ”00 I200 I300 I400 l500 I600 I700 I00 90_ / 90 \ MAY 24, I974 MPF-3 / l7 \ DECEMBER 3, I974 FLD MAX. CONTRAST I2 AT II00 HRS, MAX. CONTRAST l2 AT I430 HRS 80 — Stressed trees 30 t 2 80— Clo d r Stressed tress 22:8 Background trees 27 a 2 U °°Ve \Background trees 27:5 70 —- / 70 — \ \ 60 — X 60 — \ \ Cloud co \ w ver \ g 50 — / 50— : \ / _, I . . ESL 40 _ / 40 Relative humidity \ \ / 4 Relative humidity \, 3o _ Stressed trees 30— 20 — 20 _ \ / Background trees / \ ——~ / / \ _ Stressed trees IO _ / l0 \ _ ‘ __ ._— -——' _ _,—_ / 74 Wind speed '— —‘ Wind speed \ 0 I I | I I I I I I 0 I I I I l l I I 900 1000 “00 '200 I300 I400 [500 IGOO I700 900 IOOO IIOO l200 I300 I400 l500 IGOO I700 90 — —— — — 90 /— x l JANUARY I4, I975 FLD \ Cloud cover ’_ _/ ,__ —/ 8 MAX CONTRAST l.2 AT I430 HRS ( Temperature / / JULY- l8, I975 FLD 0— Stressed m,” 23 , 9 30— / MAXI CONTRAST I,I7 AT I300 HRS Background trees 301|2 Stressed trees 54 $3 70— / Background trees 46 t 4 7O 60 — so — /\ so ~ 50 4 —~ \ \ Cloud cover Stressed 40 _ 4O \ \7/ trees Background trees \ Background trees 30 _ 3O — X Relative humidity Stressed trees / \ zo- L / — ~ ‘ / 20 / Relative humidity \ \ l0 _ \ \ __.—— —-’ ‘ _ Wind speed lO \ \ Wind speed ' ‘ —— — L —— O l l I I | I I I I O / I I I I I I | J 900 I000 IIOO I200 I300 I400 I500 IGOO I700 900 I000 IIOO I200 |300 I400 I500 l600 I700 MOUNTAIN STANDARD TIME FIGURE 8.——Tempora| luminescence of Pinus ponderosa and meteorological parameters. Minimum luminescence contrast between background and anomalous trees tends to occur during periods of maximum-cloud cover. Standard deviation of the mean, computed for the time of maximum contrast, is at the 95-percent confi- dence level. 102 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 9.——Tempora| luminescence of Pinus ponderosa and meteorological parameters. Luminescence contrast of background and anomalous trees do not correlate with cloud cover conditions cited in figures 7 and 8. Standard deviation of the mean, computed for the time of maximum contrast, is at the 95-percent confidence level. AMPLITUDE IOO _ __ _ _ _ _— 90 — I I SEPTEMBER. 26.|973 MPF 80 _ MAXCONTRAST 2.8AT |045HRS. \ Stressed trees I5 2 8 70 ~ Background trees 42 1 II \ 60 \ Cloud Cover \A \ \ \ 5° ’ >~ Relative Humidity \ \ 40 ~ \ 3O _ Background Trees Stressed Trees Wind Speed 2° 2—, f ' q —_ l0 — I I I l 1 I I I J 900 I000 IIOO I200 I300 I400 I500 I600 I700 MOUNTAIN STANDARD TIME 90 — SEPTEMBER, 24,I974 FLD 80 - MAX. CONTRAST L5 AT 0930 HRS. Stressed trees 27 1 9 7O _ Background trees I9 * 5 60 - / /\ / \ 50 / \’/Cloud Cover Background Trees 30 _ \ Stressed Trees \ 20 ,_ \ _ I Relative Humidity \ IO _ Wind Speed ,y _—\- -—-— —\ / \ I I I 1 kl I \I i J 900 I000 “00 I200 I300 I400 I500 I600 I700 90 — 80 — Background Trees 70 P Stressed Trees / 60 — APRIL,22.I975 FLD / \ MAX. CONTRAST I.2 AT I430 HRS. 50 Stressed trees 66 t 5 Background trees 79 t I2 40 / \ / \ Cloud Cover~7 20 — / \ Relative Humidity>‘)R A / \ I0 ,_ Wind Speedy >4 / — — \ / / I 1 L I I l I 4 900 I000 IIOO IZOO I300 I400 I500 IGOO I700 MEASUREMENT OF LUMINESCENCE exceed 1.2 despite cloud cover of less than 20 percent prior to 2 PM. Contrast of 1.3 at 3:30 on September 24, 1974, is during a period of no cloud cover. Standard deviation of the mean for FLD ground measurements of twig samples from Pinus ponderosa at the Denver Federal Center in March 1974 (see above), as well as for the fluorescence spectrometer measurements of trees during 1973 and 1974 at the Malachite Mine site ranged from less than 1 percent to 3 percent of the mean. However, standard devia- tion of the mean of airborne FLD measurements dur- ing 1974 and 1975 were frequently somewhat higher. This was due to the relatively sparse foliage of the Pinus ponderosa (which permitted bare ground to be observed through the foliage at relatively high fre— quency), whereas, pine needles completely filled the field of view in both the spectrometer and ground based FLD measurements. This experience suggests that a smaller standard deviation may be assured in airborne measurements by avoiding species or individ— ual trees with sparse foliage. In addition to Pinus ponderosa, airborne FLD meas- urements were also made July 25, 1974, and May 6, 1975, on trees growing in soils containing background and anomalous concentrations of molybdenum 17 km southeast of Gardnerville, Nevada (fig. 10). Country rock includes limestone, monzonite, and volcanic rocks of Triassic to Jurassic age. Trees in the area are limited to pinon pine (Pinus monophylla) and juniper (Juniperus utahensis). In the 1974 measurements, the standard deviation of the mean of FLD luminescence counts for pinon pine commonly exceeded 50 percent of the mean; this is also attributed to viewing back- ground soil and grass through the low-density pine foliage similar to the experiment at the Malachite Mine described above. By restricting FLD observa- tions on both dates to denser junipers, standard devia- tion rarely exceeded 7 percent of the mean. Molybdenum content in plant ash of 32 juniper trees in figure 10 are shown in table 3. Figure 11 shows a linear regressive analysis correlating luminescence counts on both dates with molybdenum in concentra- tions of 20 to 300 ppm. Correlation coefficients of 0.73 for the July 1974 data (17 trees) and 0.80 for the May 1975 data (24 trees) are significant at confidence levels of 99.97. Luminescence measurements were not performed on all 32 trees because of small size and low foliage density of some individuals and because some grew in places difficult to approach with the helicopter. The 1975 data were measured at 1000 hours. Re- peated measurements on 12 stressed and background trees at 1200 and 1500 hours showed marked diurnal 103 TABLE 3.——Location, molybdenum content (ppm) in ash, and luminescence, expressed in rhodamine WT equivalence (ppb), of geochemical/y stressed and background juniper trees in the Alpine Mill area. Background trees are arbitrarily denoted as those containing 70 ppm molybdenum or less Molybdenum Tree No. content Luminescence 1 ______________________ 150 0.69 2 ______________________ 50 1.60 3 ______________________ 300 .73 4 ______________________ 500 .56 5 ______________________ 500 (Not measured) 6 ______________________ 300 .20 7 ______________________ 500 .74 8 ______________________ 20 (Not measured) 9 ______________________ 30 (Not measured) 10 ______________________ 70 1.02 11 ______________________ 50 .87 12 ______________________ 200 .33 ‘13 _______________________ 500 .53 14 ______________________ 500 .26 15 ______________________ 70 .92 16 ______________________ 50 (Not measured) 17 ______________________ 100 (Not measured) 18 ______________________ 500 0J8 19 ______________________ 150 .33 20 _______________________ 200 .38 21 ______________________ 500 .57 22 ______________________ 500 .58 23 ______________________ 100 .62 24 ______________________ 200 .51 25 ______________________ 70 (Not measured) 26 ______________________ 150 027 27 ______________________ 200 .88 28 ______________________ 30 1.09 29 ______________________ 70 .75 30 ______________________ 30 .87 31 ______________________ 20 (Not measured) 32 ______________________ 50 (Not measured) consistency. For example, mean FLD count for back— ground trees increased from 53 at 1000 hours to 58 at noon, and decreased to 49 at 1500 hours. Mean FLD count for stressed trees showed even less varia- tion, increasing from 37 at 1000 hours to 39 at noon and decreasing to 35 at 1500 hours. The 1974 data also appear to be diurnally consistent, although, be- cause of high winds and helicopter power problems, earlier data were not acquired sequentially and the time of each measurement was not recorded. Both the 1974 and 1975 data show a decrease in lumines- cence of juniper with an increase in molybdenum content up to 300 ppm. It is possible that luminescence may vary diurnally and seasonally with species; for example, although Pinus ponderosa repeatedly showed marked diurnal variability, Juniper utahensis was remarkedly consist- ent during the 2 days these trees were observed. A 104 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM exceeding 75 ppm Mo Cherokee Mine Area .- NEVADA .Gardnerville l °|0 EXPLANATION Area of trees ° 1 5 Sample tree Fault o 230 400 FEET o ICC 200 METRES Modified from Climax Molybdenum data FIGURE 10.~—Location of geochemically stressed and background trees in the Alpine MiII area, about 17 km southeast of Gardnerville, Douglas County, Nevada. Background trees are arbitrarily denoted as those containing 70 ppm molybdenum or less. ‘ MEASUREMENT OF 400 F 300 200 100 MOLYBDENUM CONTENT IN PLANT ASH (ppm) I I n 0.0 0.25 0.50 0.75 LUMINESCENCE 105 Correlation coefficientr=0.73 (7-74) 0 2080 (5—75) X I l 1.00 1.25 1.75 LUMINESCENCE (RHODAMINE WT EQUIVALENT ppb) FIGURE 11.—Linear regressive analysis correlating FLD luminescence counts for trees with molybdenum concentrates from 20 to 300 ppm. practical approach may be to restrict measurement of luminescence of geochemically stressed plants to single specie areas or to one or two species in mixed groups. Additional work is required to understand variation of diurnal and seasonal luminescence with species. PHOSPHATE ROCK Small hand-carried ultraviolet lamps have been used to stimulate luminescing minerals and rocks, but these methods of prospecting are limited because the lamps are low powered and effective range is limited to a metre or less. The work must be conducted at night because the low-intensity luminescence is obscured by bright sunlight. To assess the use of the FLD in prospecting for phosphate rocks, the luminescence of 10 samples of sedimentary phosphate rocks from several geographic locations, measured on the labora- tory fluorescence spectrometer, at the 486.1—nm Fraunhofer line, are shown in table 4. All samples luminesced within the sensitivity range of the FLD and pointed to the need for a field test. Several deposits of commercial-grade phosphate occur in southern California (Gower and Madsen, 1964). One of the highest grade deposits occurs near Pine Mountain, northeast of Santa Barbara, in the upper part of the Santa Margarita Formation of TABLE 4.—Luminescence of 10 phosphate samples measured with the laboratory fluorescence spectrometer at 486.1 nm in terms of rhodamine WT equivalence (source-detector, solar, and depth corrected) Sample Luminescence (ppb) Blue apatite from Brazil (low F, high rare earths) ______________________ 8.0 Cunday, Colombia, Cretaceous (25% P205, 2% F2, 100 ppm rare earths) ______ 7.9 Monterey, California, pelletal phosphate (25% P-_-O.~,, 2—3% F2, rare earths present) _______ 6.5 Tennessee brown phosphate (30"/o+P;O:., 3% F2, rare earths unknown) ______ 5.4 Diammonian phosphate-aqueous NHiOi + HJJPOI (N(15%), P(53%), K(0%), 0 F: ________________ 3.8 Cook Hollow, Tennessee (reprecipitable phosphate, low rare earths) ______ 2.9 Triple super phosphate CaxHyPOi (N(O°/o), P(48%), K(O°/o) some F2) ________________ 2.8 Homeland Mine, Florida (35% P30;, 3.5% F:, 200—300 ppm rare earth) ___ 1.6 Manatee County, Florida, recl phosphate (30% P305, 3% F2) ___________________________ .7 North Carolina heavy fraction (30% P205, 3.5% F:, 200 ppm rare earths) ______ .4 106 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM EXPLANATION 18°22'30" |l9°l18| so" QyOI 2 \ >— 25 8 an Er " A, 5-, Y II ' 5 1th 0 mm m > E , Tsm .210. g Landslide deposits < va //<: /" ._. Tm - 3 g / I '. Q' m * Older alluvium I I K '. m 1’ ‘ / \‘y 41 Tm ’3: g UNCONFORMITY ,/ °”4 m 0”) 20X Tsm l 5° TS'“ . 2 Santa Margarita Formation .. m T] .5 va 2 \0\ Monterey Shale, Rincon Fa,” A Mudstone. and Vaqueros ’ 5}; Formation, undifferentiated Tsm E g va (:3 Sespe Formation E 0| 2 Tcw E ° <1 8 Goldwater Sandstone L ; “J :2 Ted E 10 I I Tcw _°: 2 Cozy Del Sha e O m 0 Ted 000 O on L” Matilija Sandstone l I N Modified from Vedder. and others ([973) O l 2Miles Juncal Formation , l l J O l 2 3 Kilometres I I l J Contact Syncline, showing trouthina 34 Fault trace, dotted _;_ where concealed Strike and dip of beds Anticline, showing crestline FIGURE 12.—Geologic map of part of the Sespe Creek area showing location of a helicopter traverse (solid line east of Chorro Grande Canyon) and helicopter hovers (A through D). Numbers along traverse are FLD terms of rhodamine dye equivalency. Miocene age (fig. 12) where a phosphate zone more than 24 m thick occurs in siltstone and siliceous shale. Excitation spectra of gypsum and phosphate rock and soil samples from the Santa Margarita Formation measured on the laboratory fluorescence spectro- meter show that luminescence at the 486l1-nm Fraun- hofer line exceeds luminescence measured at 589.0- and 656‘3-nm lines. Table 5 shows that the luminescence of the phosphate and gypsum samples appear to be within the sensitivity limits of the FLD. Two experiments were conducted to measure these materials in the field. Through a cooperative arrange- ment with the Geological Survey’s Water Resources Remote Sensing group in Prescott, Arizona, the FLD was flown abroad their HUlB helicopter over the luminescence measurements in Sespe Creek area on November 26, 1974, at a hover altitude of 50 m. Reflectance and luminescence of gypsum, phosphate rock and soil, background soil, TABLE 5.—Luminescence of phosphate and gypsum samples, collected from the Santa Margarita Formation near Pine Mountain, California, and measured with the MPF—3 at 486.1 nm in terms of rhodamine WT equivalence (source- detector, solar, and depth corrected) Sample Luminescence (ppb) Phosphate pellets _____________________ 0.37—1.05 Phosphate soil ________________________ 0.33 Gypsum _____________________________ .65 Gypsum soil __________________________ .53 Assorted samples from non-phosphatic rocks (va and Ts, fig. 12) ___________ .22—0.34 MEASUREMENT OF LUMlNESCENCE 107 TABLE 6,—Reflectance, FLD counts, and luminescence in terms of rhodamine WT equival- ency (source-detector, solar, and depth ground reference materials. Field FLD corrected) of phosphate, gypsum, and back- measurements were performed in November 1974 Sample Reflectance FLD counts Luminescence (ppb) Gypsum _____________________ 0.2253 68 t 5 (66 i 2) 1.16 Phosphate rock and soil _______ .0471 52 i 3 (43 i 2) .95 Gray card _________________________ 0i3( 0:1) .26 Background soil and vegetation__ ______ 3 : 0.4 .30 Average of 10 readings from va and T5 (fig. 12) ______________ 16 i 4 .46 and vegetation along Chorro Grande Canyon are shown for the 486.1-nm Fraunhofer line in table 6. Although both gypsum and phosphate are lumines- cent, the two are significantly different in reflectance so that it is possible to distinguish them. Values in parentheses in table 6 were obtained by placing samples of the phosphate and gypsum under the FLD in a static ground measurement test. Agree- ment between the flight and static techniques is rea— sonably good, with a larger standard deviation oc- curring in the airborne helicopter measurements because of the nonhomogeneity of outcrops, soil, and TABLE 7.—Reflectance, FLD counts, and luminescence in terms vegetation in the FLD field of view. A calibration figure of 1.32 ppb/lOO counts was used to convert airborne counts to rhodamine WT equivalence. This experiment was repeated May 8, 1975, with a commercial rental helicopter at altitudes of 35 m for hover and 150 m for traverse. FLD counts for a single traverse are shown along the bold north-south line in figure 12. The counts are substantially higher in the Santa Margarita Formation than in the less phosphatic older rocks. In addition, luminescence and reflectance of gypsum, phosphate rock, phosphate soil, background sandstone, and background grass of rhodamine WT equivalency of phosphate, gypsum, and back- ground materials. Airborne FLD measurements were performed May 8, 1975 Run 1 Run 2 Luminescence Luminescence Sample Reflectance FLD counts rho. WT (ppb) Reflectance FLD counts rho. WT (ppb) Gypsum (hover A) ______________________ 0.111 97 t 5 1.54 0.0948 92 i 15 1.47 Phosphate rock and soil (hover B) _________ .06 77 : 7 1.28 .051 72 i 8 1.21 Phosphate soil (hover C) _________________ .042 56 : 6 1.0 .045 64 i- 8 1.10 Sandstone (hover D) _____________________ .082 41 i 4 0.80 .147 50 i 11 0.90 Background grass and soil ________________ .041 35 t 2 0.72 .052 21 i 3 0.52 Gray card ___________________________________ 0 1- 5 0.26 _____ 0 -+_- 4 0.26 and soil were determined by hovering over each ma- terial along Chorro Grande Canyon (points A, B, C, and D, fig. 12) and collecting approximately 50 data points for each material. Reflectance, FLD counts, and rhodamine WT equivalent luminescence for two hover runs over Chorro Grande Canyon are shown in table 7. , Comparison of Tables 6 and 7 shows that relative values of reflectance and luminescence for gypsum and phosphate correlate well for both dates, although the 1975 data are higher; background grass and soil are substantially higher. These increased levels of luminescence could be due, in part, to greater insola- tion in May than in November. The differences could also be attributed to the fact that 1974 and 1975 hovers were in the same area but over slightly dif- ferent outcrops. Comparison between values of rho- damine WT equivalents calculated from FLD counts (tables 6 and 7) and those values based on measure- ments with a laboratory fluorescence spectrometer show that higher values were obtained for lumines- cence of phosphate from the spectrometer than with the FLD. This is probably caused by phosphate con- tent of the pellet sample measured in the spectro- meter in the laboratory being higher than the bulk phosphate rock measured with the FLD in the field. It is not understood why the luminescence level of gypsum measured with the FLD exceeds the level measured with the spectrometer by nearly a factor of 2. In addition to the measurements noted above, phos- phate rock was measured with the FLD during a 108 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM TABLE 8.—FLD counts and luminescence in terms of rhodamine WT equivalency of asso- ciated phosphate and gypsum materials in the Lake/and, Florida, area Luminescence Source FLD counts rho. WT (ppb) Gray card ________________________________ 0 j: 2 (overcast) 0.26 2 i 3 (clear sky) .29 Phosphate venier in dredged pit _____________ 65 i 12 (overcast) 1.12 Phosphate rock pile (Royster plant) ___________ 116 i 4 (overcast) 1.79 143 i 4 (clear sky) 2.15 Gypsum pile _______________________________ 137 i 11 (overcast) 2.04 189 i 15 (clear sky) 2.75 series of airborne experiments in the Lakeland, Florida, area. The materials measured had been mined and are, therefore, not representative of weathered surfaces on natural outcrops as was the case in the California experiment. Luminescence of these materials is shown in table 8. Comparison of the gray card, phosphate rock pile, and gypsum meas- urements shows that the FLD performs adequately under overcast conditions, although rhodamine WT equivalency for these three materials is reduced 10, 17, and 26 percent, respectively. OIL SEEPS Riecker (1962, p. 60—75) analyzed 103 crude oil samples from California, Colorado, Wyoming, and Alberta with a laboratory spectrometer; he found that all of them exhibited luminescence peaks in the vis- ible region between 444 and 630 nm. Twelve per- cent of the samples peaked in the red at 630 nm, 54 percent peaked in the yellow at 578 nm, and 34 percent peaked in the blue at between 482 nm and 442 nm. Fantasia, Hard, and Ingrao (1971) used a pulsed ultraviolet laser installed at dockside to dis- criminate several oils on seawater by measuring the ratio of their luminescence through narrow band filters centered at 433 and 533 nm. In order to quantify the luminescence of selected oils in terms of rhodamine WT equivalence, the excita- tion spectra of 29 crude oils were measured with the laboratory fluorescence spectrometer by Watson and others (1974). The integrated excitation intensi- ties, corrected for source—detector, solar, and depth effects are shown in table 9. The crude oils exhibit a peak rhodamine WT equivalency at Fraunhofer wavelengths of 518.4 nm or shorter, with 19 samples (65 percent) having a maximum at 518.4 nm. Lum— inescence of 28 of the 29 samples exceeds 0.25 ppb rhodamine WT equivalency and appears to be within the sensitivity range of the FLD. Specific gravity of the 29 crude oils ranges from 0.7317 to 0.9677. The dependence of bulk oil lumines- cence on specific gravity is shown in figure 13. A regressive analysis was performed on the 29 crude oil samples to obtain the best least-squares fit to the data points, with both the correlation coefficient and the standard error determined for six selected Fraun- hofer lines throughout the visible spectrum. Correla- tion coefficients range from 0.34 at 656.3 nm to 0.77 at 396.8 nm and indicate that the best correlation between luminescence and specific gravity occurs at 396.8 nm. Natural fractures in the ocean floor of the Santa Barbara Channel, California, permit oil of varying densities to seep to the surface. Throughout the channel, patches of oil films at various times can be seen with the naked eye. At times heavy crude seeps from narrow filaments near Coal Oil Point that ex- tend for several miles parallel to the shoreline To measure the luminescence of crude oil with the prototype FLD, a sample of oil from the Santa Barbara Channel was poured into a tank of water; luminescence of the resulting film was not detected (Hemphill and others, 1969) because (1) the emission peak of the Santa Barbara oil is at a shorter wave- length than 589.0 nm where the prototype FLD operated and (2) sensitivity of the prototype FLD was inadequate. Luminescence of oil seeps off Coal Oil Point were measured during the week of November 24, 1974, with the redesigned FLD operating from a helicopter at the 486.1—nm Fraunhofer wavelength. Data obtained on November 30 are shown in figure 14. Lumines- cence intensities of small lakes several miles inland from the coast average 20—40 counts less than the counts of so—called “clear” Santa Barbara Channel water, suggesting that small amounts of oil or other luminescing material is present in the channel. MEASUREMENT OF LUMINESCENCE 109 TABLE 9.—lntegrated excitation intensity of crude oils at specific Fraunhofer wavelengths in terms of equivalency with rhoda- mine WT. Samples are ordered by decreasing intensity at the 486.1—nm Fraunhofer line and corrected for source-detector, solar, and depth effects Wavelength (nanometres) Description Specific 396.8 422.7 486.1 518.4 598.0 656.3 Sample No. Location gravity (ppb)* (ppb)* (ppb)* (ppb)* (ppb)* . - _ t — .- m - - O 3 g 10 5 103 = 2 : F. 3 - _ _l I- ,, E h h 2 ,: 10 5 102 z < : 0 E 0 _l - m: —0.146538157704ex 02 - m: —0.127042755296ex 02 “J : b=0.l971537l4083ex 02 : b=0.l792361054382x 02 I _ Sy.x=0.88400497053lex 00 Sy,x=0,883699253917ex 00 Sm =0.332782229794ex 01 — Sm=0.332667143273ex 01 1 r: -0.633012244860ex 00 C 1 r =0.578463406748ex 00 D 10 I l 10 I l 0.7 0.80 0.90 1.00 0.70 0.80 0.90 1.00 10" 10‘ = o E 3 10 103 = '- O " o 2 10 10 2 E ‘ m: -0.9203134654306x 01 O m: “0.668334528749lex 01 . b:0.l42316031557ex 02 0:0 11119363817lex 02 Sy.x=0.840909979804ex 00 Sy.x=0.909l88646027ex 00 Sm =0.316559190801ex 01 E Sm=0.342262583372ex 01 1 I: -0.4750561242286X 00 1 r: —0.340892588926ex —00 O F 10 I I r 10 I t ' 0.70 0.80 0.90 1.00 0.70 0.80 0.90 1.00 SPECIFIC GRAVITY FIGURE 13.—Statistical plot showing correlation between relative luminescence and specific gravity of 29 crude oils at the following Fraunhofer lines: A, 396.8 nm; 8, 422.7 nm; C, 486.1 nm; D, 518.4 nm; E, 589.0 nm; and F, 656.3 nm. ing those periods when cloud cover was less than 10 percent. 6. FLD measurements of juniper showed significant correlation between molybdenum intake and luminescence for values up to 300 ppm molyb- denum in plant ash. 7. Airborne FLD measurements in the Sespe Creek area, California, show that contrast between luminescing gypsum and phosphate rock out— crops, and older non-phosphatic sedimentary rocks, ranges from 1.5>< to 3X. 8. Of 29 crude oils measured in the laboratory, 28 luminesce within the sensitivity range of an air- borne FLD; laboratory work also shows cor- relation between thickness of an oil film (such as would occur in a marine oil spill) and luminescence for oils of specific gravity of 0.8327 or less. 9. Airborne FLD measurements in the Santa Barbara Channel show that luminescence of thicker, nearshore layers of oil from the marine seep exceeds “clear water” by a factor of 10; luminescence of thinner layers which have dis- persed seaward are marginally detectable. MEASUREMENT OF LUMINESCENCE 111 OIL'IN SANTA BARBARA CHANNEL. CALIFORNIA NATURAL SEEP—HEAVY CRUDE 2.02 :0.1 ppb COAL OIL PT. I SANTA BARBARA ‘~ --I ~‘ ~ : I \y I t‘ ,’ THIN OIL LAYER \\ I 0.30:0.03 ppb x \ \\ \ \ \ I \\ \‘ \\‘ I ‘_ x “““““ x, “CLEAR WATER" 0.21 :0.01 ppb N ;—____1 O 16 KM SANTA CRUZ FIGURE 14.—Map showing: (1) Dispersal of oil from natural seep off Coal Oil Point, Santa Barbara Channel, and (2) FLD luminescence responses from clear water, thin oil film, and heavy crude filament. REFERENCES American Society of Photogrammetry, 1975, Manual of remote sensing: Falls Church, Virginia, 2,144 p. Fantasia, J. F., Hard, T. M., and Ingrao, H. C., 1971, An investigation of oil Fluorescence as a tech- nique for the remote sensing of oil spills: Natl. Tech. Inf. Service PB 203585. Gower, H. D., and Madsen, B. M, 1964, The occur— rence of phosphate rock in California: US. Geol. Survey Prof. Paper 501—D, p. 79—85. Grainger, J. F., and Ring, J., 1962, Physics and astronomy of the moon, in Kopal, Zdenek, ed., The luminescence of the lunar surface: New York, Academic Press, p. 385—405. Hemphill, W. R., Fischer, W. A., and Dornbach, John, 1965, Ultraviolet investigations for lunar mis- sions: Advances in the Astronautical Science, v. 20, pt. 1, p. 379—415. Hemphill, W. R., and Carnahan, S. U., 1965, Ultra- violet absorption and luminescence investiga- tions: US. Geol. Survey Tech. Letter NASA—6, 27 p., 53 figs. Hemphill, W. R., Stoertz, G. E., and Markle, D. A., 1969, Remote sensing of luminescent materials, in Internat. Symposium on Remote Sensing of Environment, 6th, Ann Arbor, Michigan 1969, Proc., p. 565—583. Hemphill, W. R., and Stoertz, G. E., 1971, Fraunhofer line discriminator progress report: Principal In- vestigator’s Review, Advanced Applications Flight Experiments, NASA Langley Research Center, Proc., p. 171—185. 112 Hoel, P. G., 1965, Introduction to mathematical statis- tics: New York, John Wiley and Sons, 427 p. Hollaender, A, 1956, Radiation biology: New York, McGraw-Hill, p. 765. Howard, J. A, Watson, R. D., and Hessin, T. D., 1971, Spectral reflectance properties of Pinus pon- derosa in relation to copper content of the soil— Malachite Mine, Jefferson County, Colorado, in Internat. Symposium on Remote Sensing of En- vironment, 7th Ann Arbor, Michigan 1971, Proc., p. 285—297. Huff, L. C., 1963, Comparison of geological, geophysi- cal, and geochemical prospecting methods at the Malachite Mine, Jefferson County, Colorado: US. Geol. Survey Bull. 1098—C, p. 161—179. Kozyrev, N. A, 1956, The luminescence of the lunar surface and intensity of the solar corpuscular radiation: Izvestia Krymskoi Astroizitcheskoy Observatorye, v. 16, p. 148—161. Kramer, P. J., and Kowlowski, T. T., 1960, Physiology of trees: New York, McGraw-Hill, Botanical Science Series. O’Gara, P. J., 1922, Sulfur dioxide and fume prob- lems and their solutions: Industrial Engineering Chem, v. 14, p. 744. Press, N. P., 1974, Detecting the toxic effects of metals in vegetation from Earth observation - satellites: British Interplanetary Soc. Jour., v. 27, p. 373—384. FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM Riecker, R. E., 1962, Hydrocarbon fluorescence and migration of petroleum: Am. Assoc. Petroleum Geologists Bull, v. 46, no. 1, p. 60—75. Sauchelli, Vincent, 1969, Trace elements in agricul- ture: D. Van Nostrand Reinhold, Co., 248 p. Stoertz, G. E., Hemphill, W. R., and Markle, D. A, 1969, Airborne fiuorometer applicable to marine and estuarine studies: Marine Technology Soc. Jour., v. 3, no. 6, p. 11—26. Thomas, M. D., 1961, Effects of air pollution on plants: New York, Columbia Univ. Press, World Health Organization Monograph, no. 46. Vedder, J. G., Dibblee, T. W., and Brown, R.‘ D., 1973, Geologic map of the upper Mono Creek- Pine Mountain area, California: US. Geol. Sur- vey Misc. Geol. Inv. Map I—752. Watson, R. D., Hemphill, W. R., and Hessin, T. D., 1973, Quantification of the luminscence intensity of natural materials, in Am. Soc. Photogram- metry, Symposium for the Management and Utilization of Remote Sensing Data, Sioux Falls, South Dakota, Proc., p. 364—376. Watson, R. D., Hemphill, W. R., Hessin, T. D., and Bigelow, R. C., 1974, Prediction of the Fraun— hofer line detectivity of luminescent materials, in Internat. Symposium on Remote Sensing of Environment, 9th, Ann Arbor, Michigan 1974, Proc., p. 1959—1980. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Use of ERTS—l (Landsat—1) Images for Engineering Geologic Applications in North—Central Iran By Daniel B. Krinsley, US. Geological Survey, Reston, Virginia 22092 INTRODUCTION Most of the Iranian population lives adjacent to the margins of the interior desert basins because of the availability of flat land and moderate supplies of ground water at relatively shallow depths. The basins are potentially valuable sources of water and chemi- cals and could be utilized for the emplacement of roads and inexpensive airfields. The basins have not been fully utilized to date because of the lack of adequate knowledge concerning the seasonal changes in their surface and ground water hydrologies and in the physical properties of their sediments. Most desert basin investigations have been conducted during the summer when the surficial sediments are dry and have sufficiently high bearing strengths to support men and vehicles. The repetitive coverage of ERTS—l (all images used are ERTS—l and to avoid confusion that designation will be used in this paper) is ideally suited to provide seasonal images of the Iranian basins from which changes in the areal extent and morphology of the surficial materials may be recorded in conjunction with contemporaneous or previous ground truth studies of actual surficial conditions. Data derived from the analyses of ERTS—l images can provide a rational basis for planning the economic utilization (salts or water extraction and agriculture) and engi- neering development (roads and airfields) of these desert basins. ERTS—l images used in this study have been ex- amined in single bands and in false-color composites of several band combinations. PHYSICAL SETTING OF THE IRANIAN DESERT BASINS The Elburz and Zagros Mountains (fig. 1) were com- pressed against the relatively oIder inflexible block of the Iranian Plateau to the east during the Pliocene- Pleistocene phase of the Alpine orogeny. Lower but nevertheless rugged mountains that formed along Iran’s eastern border divide the plateau into a major eastern basin in Afghanistan and a major western basin in Iran. Within the western basin are several lesser mountain chains and isolated mountains as well as extensive desert plains. The lowest depressions in these deserts are occupied by smaller desert basins (fig. 1). Approximately half of Iran consists of basins from which there are no outlets and from which the col- Iected drainage is removed by evaporation. Moist winds bring considerable precipitation during the winter half of the year to the northern slopes of the Elburz Mountains which effectively bar movement of moisture to the south. Similarly, the Zagros Mountains act as a barrier to the moisture-laden westerlies. Rain- fall is quickly reduced in amount southward and east- ward so that the interior lies within a vast rain shadow and becomes increasingly arid from west to east and from the north to south. There is exterior drainage only along the north, west, and south margins of the country; the remainder has interior drainage. There is a considerable water deficit in the interior during all seasons, and streams that do reach the basin sumps are generally fed by ground water. The upper aquifers in the alluvial fans are phreatic, but towards the center of the basin artesian conditions develop as a result of confining clay layers (Issar, 1969, p. 94). The uppermost aquifer does remain phreatic to the margin of the basin sump (playa), where the water from the toe of the fan spills out over the playa surface creating a “wet zone.” Evaporation through the capillary zone from the phreatic aquifer results in salinization of the water. Salinization is ac- centuated in the north-central basins that are under- lain by Miocene evaporites. 113 114 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM \4’4 TURKE NEUTRAL ZONE Yerevan . \ \- SAUDI.‘~/'ARABIA \aa U.S.S.R. N Bas 3' UP.. . / KUW 9 fix 0 ‘ UWAIT \ o E 0 50 0 50 I00 150 IRAN Altitude: above 2,000 metres Playas and asso- ciated features 100 Iso 200 Miles 200 Kilometres 9 Bushire 52 =aku CASPIAN SEA V I BOUNDAav REPRESENTATION IS NOT NECESSARILV AUTHORIYATIVE Krasnovodsk Bukhara o U. S. S. R. . Kizyl-Arvat 5;, . ~ 45. \,_ 0’..— 23 edit" “ 1 ) ‘ \ Nok Kuqdi 4’ ‘ KISTAN ’ /\ \..\P1: , x - 28— f}: Q Bandar ‘Abb‘as %' Bandar‘e Leng' . up \ 2.. “o— . h 1 ‘ Iransha f 9 MAKRAN MO NTAW‘S f. < I A Jask Chéh TRU AL . Bahbr G UL P 0' ST TES 0114 N 59 /" Although man has severely altered the natural vege- tation by extensive cutting, it is doubtful that vegeta- tion within the inner plateau seriously affected the FIGURE 1.—The relationship of the desert basin sumps (playas) and associated features to relief in Iran. 1972b, p. 116). amount and intensity of overland flow during the Pleistocene. It seems more likely that increased cold during the glacial periods more than counteracted any relative increase in humidity that might have been favorable to growth. Significantly, no evidence has been found in any of the upper Pleistocene sediments, GREAT KAVI R WATERSHED The Great Kavir watershed (fig. 2) occupies an area of 200,747 km2 in the northeastern oiviadrant of Iran. The northwestern divide; in the Elburz Mountains, has along the inner mountain flanks or in the playas, of buried trees or other vegetative debris (Krinsley, USE OF ERTS—1 FOR ENGINEERING GEOLOGIC APPLICATIONS 115 BOUNDARY REPRESENTATION IS 50 N07 NEcEssAva AurHomTAYIVE l Krasnovodsk I aukhara. TURKE ; U S. S.I.R .0 CASPIAN IKIZYI‘AWM l SEA I ‘ .AshkhabadIl ‘ ‘Vfl Mary. r "i. _ l GREAT KAVIR .MIeshed :6— \i% \‘\ 'Daman-nWATERSHED ‘ ,/ (“.41 - l .- TEHERAN 3 . Semi—.5". .__)~_/ 0 Sananda k ‘ \ ImTERSHED . Kush Hamadan ‘ KermanShah Qom' ‘ l . l : - ”in” l \ AFGHANISTAN Biriand. I " ISfahan . I \ Farah. Shahr Ker ~ “ I .Yazd 1 f wad“ ’1. 090 -andar-e Shéhou S .Gach Sa' ("@2519 ‘9 ‘ 2:4me , Kerman I '/ fish“ n. \ I . Nok KquI l . FAKISTAN .\ IRAN Ix- -—~~ "'4‘“ —23 I . «o I -' —---— International boundary @ ”‘- ® Nationalcapital Q $0 I .i'an ahr ‘ Railroad I ‘/.- 0 Road Ad Dammém I Bandar-e Lens r BAHRAIN ' CO C so 100 150 200 Miles I) ‘ A . V o 50 too 150 200 Kilometres 'A'Huf QA R TRU AL GULF 0F 52 ST TES 0",. N so FIGURE 2.—|nterior watersheds of Iran: the Great Kavir and the superposed boundary of the ERTS—1 (Landsat—1) images summit altitudes which decrease from 4,002 m in the area of 52,800 km '-’ (fig. 3). In Iran, the term “kavir” west to about 3,000 m in the east. The southwestern is a generic name for playa. The Great Kavir contains and southeastern divides have summit altitudes which numerous depressions occupied by true kavirs which generally range from 1,500 to 2,000 m with a few have imparted their character to the entire desert. peaks just under 2,400 m along the southeastern di- The annual precipitation decreases from a max- vide. Within the watershed the lowest altitudes are imum of 300 mm near the divide, but within the rain found at the surface of the salt crust basins, and the shadow of the Elburz Mountains, to less than 100 mm lowest of these lies at an altitude of 650 m. in the Great Kavir. Mountains along the northeast ’ divide benefit from moisture-laden air skirting the GREAT KAVIR southern shore of the Caspian and penetrating ‘the The central and lowest area of the Great Kavir valleys and lower passes northeast of Damghan (fig. watershed is occupied by the Great Kavir, a desert 1). Precipitation in this eastern area ranges from 100 116 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM CASPIAN' sEA ' 36° _ 4002 ELBURZ MOUNTAINS TEHRAN o O _ \ SURFACE TYPES FAN DELTA WET I‘Z‘ONE CLAY FLAT 34° — SALT CRUST \2 m MIOCENE BEDROCK (UPPER RED FORMATION) " \ \ \ \ SALT DOMES SAND DUNES AND PREVAILING WIND I p 0 50 320 — MOALLEMA \\ PROPOSED GREAT KAVIR ROAD 1 00 KILOMETRES A ALTITUDE IN METRES “Mg FAULT 09$” 27"} ’1’ 52° 54° 56° 58° , FIGURE 3.—E|ements of the Great Kavir. Map based on author’s fieldwork and photointerpretation and on the Geological Map of Iran (National Iranian Oil Company, 1959). to 150 mm in the mountains, but there is considerably less in the adjacent lowlands. In spite of the rapid decrease in precipitation south- ward, broad valleys with underfit and intermittent streams, below the mountain front, attest to the past erosional effectiveness of the southward flowing streams. Coarse alluvium including gravel beds, which are interbedded with clay, extends farther into the basin with increasing depths in the section. This strati- graphic relationship suggests previously greater streams. Coarse alluvium, including gravel beds, which materials farther from their source. The surface of the Great Kavir ranges in altitude from 850 m in the northwestern part to 650 m near the northeastern boundary. It is underlain by Miocene rocks of the Upper Red Formation (Gansser, 1955, p. 291). These intricately folded siltstones, marls, an- hydrites, and salt beds have been eroded to a pene- plain where they are exposed over an area of 19,300 km 2. The Great Kavir may be subdivided into eastern and western basins separated by the uplifted pene- plain cut into‘ the Miocene rocks. Three central areas of salt crust in a north-south alinement (fig. 3) approxi- mate the peneplain divide. The location of this divide near the western boundary of the peneplain and the development of two east- to northeast-trending troughs (occupied by linear “wet zones) in the eastern half of the peneplain indicate that the regional slope has long been toward the east. It is likely that currently active streams flowing westward into the western basin will eventually breach the divide by headward cutting and shift the divide eastward. The present stream cutting is due to increased gradients resulting USE OF ERTS—1 FOR ENGINEERING GEOLOGIC APPLICATIONS from recent tectonism. The Miocene rocks and the younger sediments are cut by currently active north- east—trending faults, the principal one of which is the Great Kavir fault (fig. 3). SURFACE TYPES WITHIN THE GREAT KAVIR The surface types within the Great Kavir are the Miocene rocks, locally pierced by salt domes, dune fields, fan deltas, wet zones, clay flats, and salt crusts (Krinsley, 1970, p. 107). The areas occupied by these surface types have been generalized at a scale of 12,500,000 and then reduced in figure 3. Small areas of these types which appear in ERTS—l images (scale 121,000,000) may not be shown in figure 3. Miocene rocks consisting of evaporites and other generally saliferous sediments occupy 35 percent of the area of the Great Kavir. At the peneplained sur- face of these rocks is a thin regolith of clay, silt, and sand with up to 47 percent halite. As a consequence of repeated dissolution of the salt in winter rains and crystallization upon the evaporation of surface moist- ure, or the summer evaporation of capillary water, the regolith has been churned into a rough surface resembling a plowed field. Across this vast almost flat surface, narrow drainageways terminate in the wet zones or salt crusts. Salt domes, which have locally pierced the Miocene rocks, are similarly cut by the peneplain. The dune fields occupy 13 percent of the Great Kavir and are situated along its southern and south- eastern borders. Fan deltas are large alluvial fans which transgress other surface types. The toes may be completely inundated during the principal runoff period, with the result that peripheral deposition, occurs in water, as at a delta. The two large fan deltas in the Great Kavir occupy 5 percent of its area. The wet zone is a transitional zone which is periodi- cally inundated and always wet. It commonly borders the toes of alluvial fans and fan deltas, but it also occupies linear basins or narrow troughs within the area immediately underlain by Miocene rocks. The linear wet zones have surface gradients of less than 1 degree, sloping toward the salt crust areas. Wet zones constitute 4 percent of the kavir surface. The clay flat, which occupies only 2 percent of the area of the kavir, is a generally firm surface underlain by dry clay and silt, with variable amounts of salt. It is dis- tinguished from the wet zone, with which it is pedo- logically identical, on the basis of its higher position above the dry-season water table. The salt crust oc— cupies 41 percent of the kavir surface in basins in the east, center, and west. The eastern basins are gen- 117 erally continuous but are locally interrupted by Miocene outcrops in the form of northeast—trending ridges. The surface of the southern basin in the east- ern area is 50 m higher than the northern salt surface adjacent to the fan delta. There is thus significant gradient northward toward the fan delta. Three salt- crust basins in the central area of the kavir occupy downwarps in the Miocene rocks. The salt crusts of these basins lie at different levels, and they are prob- ably hydrologically independent. The larger middle basin surface is estimated to be at an altitude of ap- proximately 850 m. In areas of thick, white salt crusts and seasonally high water tables, the wet briny muds beneath the crusts may be more thermally expanded than the crusts. This process is facilitated by the transparency of pure salt to infrared rays and by the heat absorp- tion of the black mud. Occasionally, the plastic muds are forced from beneath the salt plates outward and upward through the peripheral cracks. Rapid evapora- tion of the mud brines results in the formation of black salt dikes along the polygonal cracks. Grooves and scratches preserved in the solidified dike that were cut by preexisting crystallized salt during extrusion of the mud attest to the rapidity of the solidification. The uneven extrusion of the briny mud frequently results in the tilting of the polygonal salt plates. Even- tually, the area becomes a rough, jumbled mass of sharp and angular black salt blocks with ridges and pinnacles (fig. 4). Wind-driven rain and silt tend to modify and sharpen the rough surface features of these salt fields. Aeolian or alluvial silt added to the old salt, already darkened by the admixed black mud from beneath the crust, imparts a dark matte appear— ance to this grotesque surface. Relief among the salt pinnacles and ridges ranges from 30 to 50 cm. Salt crusts and related surface features are pri- marily a reflection of the hydrologic regime within a basin. The crusts tend to buckle when they are thin, darkened by silt and underlain by light-colored sedi- ments and a seasonally depressed water table. Saline muds are more likely to be extruded when they are black, and water saturated, beneath a relatively thick, white salt crust (Krinsley, 1972a, p. 173). POTENTIAL FOR ENGINEERING DEVELOPMENT Almost all materials produced in northern and cen- tral Iran are exchanged by means of trucks moving along the east-west and northwest-southeast road that passes through Tehran (fig. 5). Goods from the north destined for the region immediately south of the Great Kavir must travel more than 900 km rather than the 118 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 4—Rough black salt ridges and pinnacles adjacent to the dry- season road across the Great Kavir, 40 km south of Moalle- man. Local relief ranges from about 30 to 50 cm, 200 ktn that would be required if an all-weather road existed across the desert. Less dramatic but never- theless significant savings in time and in transportation costs could be achieved with such an all—weather road terminating at Nain farther to the southeast (fig. 5). Trucks traveling from Damghan to Nain cover the 750 km in ‘9 hours; the route across the Great Kavir would be 45 km and require 6‘ hours. Rough dry—season roads of gravel and dirt exist for caravd‘ns or 4—wheel drive vehicles along the route indicated in figure 5, between Damghan and Nain. Engineering procedures associated with the construc- tion of an all-weather road across the Great Kavir view northwest, date September 2,1966. would be concerned with stabilizing, strengthening, or constructing a subgrade over the saliferous surficial materials and installing culverts and bridges to permit normal drainage across the route of the eventual road alinement. This section of the report is concerned with the use of ERTS—l scenes in the selection of a pre- liminary road alinement through the Great Kavir. ANALYSIS OF THE ERTS—l IMAGES OF THE GREAT KAVIR The basis for the analyses of the ERTS—l images of the Great Kavir is familiarity with its actual ground USE OF ERTS—1 FOR ENGINEERING GEOLOGIC APPLICATIONS 119 '05 5'2 V IOUNDARV ntrnntnnirvou IS 60 NOT NECEII‘IILV AUTfloni'AYH/E : 'k” Krasnovodsk Bukhan . U. S..S. R- CA S PIA N d anyIAArvat SEA Ashkhabad Sananda, Hamadan Qom Isfahan Sham Kord. ain ,badan = andar-e Shéhpu .Gach Sa’ an TEHERMK/ Binand Yazd Kerman VEUTRAL ~ 4-. ZON » ‘ \_ u’..— ~W$..1 Nok Kundi SAUDI " ARABIA V. I I ‘ I \. pAKlSTAN I .\- IRAN l . - 23 ~— 28 '_ _ _,__ —--— International boundary ‘0 l Lu». (9 NahOnaI capital '§ 419’ andar 'Abbias iranshah' 9 1¢ O I" a ‘ 0' R030 Ad Dammam Banday.e Len. _ . &'7 r BAHRAIN o _ I 0‘ .' o 50 100 150 200 Miles 4“ Jask Chéh 5 ' V aha! o so 100 150 200 Kilometres AT R TRU AL 8 AI Hu ST ES 0 UL r or 52 ”AN 60 FIGURE 5.—Existing Iranian road net and the proposed Great Kavir road, which is shown by a bold dashed line. The north and south extensions of the proposed road are shown by solid bold lines, conditions and sediments during the summer and autumn of 1965, 1966, and 1974. In addition to ground control, this area was viewed from low—flying aircraft, and many of its details were compiled from aerial photos at scales of 1:30,000 and 160,000. Initially, false-color diazo transparencies were pre- pared from positives of ERTS—l images. Yellow, red, and blue were used for MSS bands 4, 5, and 7 re- spectively. These three colors were then composited to form a false-color transparency. The composite was then compared to actual ground and aerial photos and interpreted in the light of the investigator’s ex- perience. The false-color illustrations used in this report are products of a subtractive color system. The dyes used in this system are magenta, yellow, and cyan, but the resultant pure colors are red, green, and blue; In the false-color composites, water is blue, salt is white and the saliferous silt and clay regolith of the Miocene sediments is buff to light brown. The purity 120) FIRST ANNUAL PECORA or il‘ntensity of these colors reflect differences in the composition, hydrology, or morphology of the surface materials. With increasing depth, water ranges in color from aquamarine to dark blue to black. With increas- ing saturation, the salt crust ranges in color from pure white to aquamarine. With increasing roughness or micfiorelief, the salt crust ranges in color from pure white (smooth crust) to black (rough salt ridges; figs. 4 and 6, p. XII). Combinations of wet, silty and rough salt crust result in mixed false colors which require special considerations in their correct interpretation. In the ERTS—l scene of September 20, 1972 (fig. 6, p. XII), white areas are smooth salt crust and black areas, adjacent to the dry—season road, are rough blaclt salt ridges and pinnacles (fig. 4). There is no standing water, in that area, at that time of year. In the ERTS—l scene of May 12, 1973 (fig. 7, p. XIII), the black) areas adjacent to kilometres 75 and 90 are white salt crusts covered by water (compare figs. 6, p. XII, and 7, p. XIII). MEMORIAL SYMPOSIUM SELECTION OF A PRELIMINARY ROAD ALINEMENT THROUGH THE GREAT KAVIR The area of the Great Kavir most suitable for the construction of an all-weather road lies entirely within one ERTS-l scene (figs. 2 and 6, p. XII). The center point of the September 20, 1972, image of that scene is 34°35’N.; 55°12’E. False—color diazo composites were prepared from positives of ERTS—l images .taken of the Great Kavir on September 2 and 20, 1972, December 19, 1972, February 11, 1973, March 1, 1973, and May 12, 1973. The route of the existing dry-season road was overlain on each diazo com- posite, and the seasonal hydrologic conditions along four critical segments of that route south of Moalle- man were recorded (table 1). The four critical road segments represent areas that are subject to soil saturation by water or even local flooding and thus require special engineering consider- ations. The road segment from kilometres 25 to 50 TABLE‘1.——Hydrologic conditions along critical segments of the dry—season road across the Great Kavir as inferred from ERTS—1 ‘ images I Identification of Distances in kilometres south from Moalleman ERTS—I—1 image" 25—50 60—63 90-100 110—116 I Sept. 2) 1972, Moist surface at km 40 _ Moist surface at km 62 _ Dry salt crusts ________ Moist surface at km 34°35'N., 55°09’E., 113. 8104106265. I Sept. 20, 1972, Moist surface at km 40 _ ____do ______________ Dry salt crusts ________ Do. 34°35'N., 55°12’E., 8105906265 Figure 6. I Dec. 19,, 1972, Wet surface at km 40; ____do ______________ Moist to wet salt crusts; Moister surface at km 34°35’Nt, 55°09’E., dissolution of salt dissolution of salt 113. 8114906273. crusts. crusts. Feb. 11,“ 1973, Wet surface at km 40; Obscured by cloud ___ Wet soil and salt crusts; Obscured by cloud. 34°35’N.i, 55°05’E., dissolution of salt dissolution of salt 8120306274. crusts. crusts. Mar. 1, 1973, Wet surfaces at km 32 Partly obscured by Wet soil and salt crusts; Wet surface at km 113. 34°36'N.j 55°01’E., and km 40; dissolu- cloud, probably wet. dissolution of salt 8122106275. tion of salt crusts at crusts; standing ‘ km 40. water. May 12, 1973, Wet surfaces at km 32 Wet surface at km 62 _ Wet soil and salt crusts; Wet surface at km 113. 34°43’N.,“ 54°56’E., and km 40; dissolu- dissolution of salt 8129306274 tion of salt crusts at crusts; standing Figure 7.) km 40. water. Summary Alinement to west with culverts probably re- quired over wet areas. Bridge or raised road- bed required. Raised roadbed and bridge required. Alinement to west around wet area. *Date, location, identifier (EROS Data Center identification number), and figure number in this report. USE OF ERTS—1 FOR ENGINEERING GEOLOGIC APPLICATIONS lies in the area underlain by Miocene bedrock (fig. 3). It is always moist along a narrow drainageway at kilometre 40 during the dry season, but this area be- comes wet by mid-December. Dissolution of the local salt crust parallel to the drainageway suggests satura- tion of the surficial materials. By March, the surface in a slight depression is also wet at kilometre 32. The road from kilometres 25 to 50 can be alined inexpen— sively to the west on good natural subgrade to avoid the two wet areas (fig. 7, p. XIII); culverts will be re- quired along drainageways. The segment from kilometre 60 to 63 lies in a de— pression within the area underlain by Miocene bed- rock (fig. 3). It is always moist at kilometre 62 during the dry season and becomes moister by mid- December. This segment is wet by mid-May and a bridge or raised roadbed would be required at kilo- metre 62 (fig. 7, p. XIII). The segment from kilometres 90 to 100 lies in an area that is occupied principally by salt crusts (fig. 3) which undergo significant changes in their surface conditions during the seasons. Salt crusts are dry dur- ing September but become moist by mid-December with some dissolution. By early February, both soil and salt crusts are wet with continuing dissolution of the salt. Standing water is present by March 1, and its areal extent has increased by mid-May. This seg- ment is the most difficult and would require both a raised roadbed and a bridge. Although an alinement to the east would avoid the wettest areas, the costs of engineering for the poor subgrade over the in~ creased length of the road (up to 30 km) would be prohibitive (fig. 7, p. XIII). The segment from kilometres 110 to 116, which lies within the area underlain by Miocene bedrock, is always moist at kilometre 113 during the dry season, and becomes moister by mid-December. By March 1, the surface is wet at kilometre 113, and this condition continues into May. This area can be avoided by a short alinement to the west (fig. 7, p. XIII). The use of ERTS—l images provides a method for examining areas that are seasonally inaccessible in order to determine hydrologic changes that affect soil conditions and thus their engineering properties. There must be some knowledge of actual ground conditions in order to correctly interpret the ERTS—l images, and the eventual determination of the location of any engineering project such as a road alinement should be based on a longer record of observation and on- site investigation. 121 POSTSCRIPT The author met with His Excellency Javad Shahre- stani, Minister of Roads and Transport of Iran, on September 24, 1974, at the latter‘s invitation for the purpose of discussing the construction of an all- weather road across the Great Kavir. Further infor- mation was provided to Minister Shahrestani by the author in November 1974 and during January and March 1975. In August 1975, the Ministry of Roads and Trans- port of Iran awarded a contract to a joint venture company, Peyma—Harland (Harland Bartholomew and Associates International, Inc., of Washington, DC, and Peyma Consulting Engineers of Tehran, Iran) to provide consulting engineering services for a feasi- bility and location study for a highway that will begin in the vicinity of Damghan (fig. 5) and terminate in the vicinity of Nain. The first phase of this project will be to delineate the exact route of the highway. This phase will be completed in approximately 7 months, at the end of February 1976. The second phase, which will require 18 to 24 months, will involve the final design and contract documents for the highway con- struction. The final construction phase will require approximately 2V2 to 3 years and should be com- pleted no later than February 1981. REFERENCES Gansser, A., 1955, New aspects of the geology in central Iran, in World Petroleum Congress, 4th, Rome 1955, Proc., sec. I, Geology Geophysics, p. 279—300, lssar, A., 1969, The ground water provinces of Iran: Internat. Assoc. Sci. Hydro]. Bull, v. 14, no. 1, p. 87—99. Krinsley, D. B., 1970, A geomorphological and paleo- climatological study of the playas of Iran: U.S. Geol. Survey Interagency Rept. Military—1, 329 p., 4 plates, 155 figs, 17 tables. 1972a, Dynamic processes in the morpho- genesis of salt crusts within the Great Kavir; north-central Iran, in Internat. Geol. Cong, 24th, Montreal, Proc., sec. 12, p. 167—174. 1972b, The paleoclimatic significance of the Iranian Playas, in Zinderen Bakker, E. M. Van, ed, Palaeoecology of Africa: Cape Town, A. A. Balkema, v. 6, p. 114-120. National Iranian Oil Company, 1959, Geological map of Iran (12,500,000) with explanatory notes: Zurich, Orell Fussli. I. ‘ I \fl «fitlxsl PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Applications of Remote-Sensing Technology for Powerplant Siting By Leo Eichen and Richard F. Pascucci, Dames & Moore, Cranford, New Jersey 07016 INTRODUCTION Remote sensing as a practical service consists of acquisition of various data acquired remotely, e.g., satellite observations and aerial overfiights; the inter- pretation of the source data in the context of the problems involved, e.g., geotechnical evaluations and environmental assessment; and the presentation of the results in suitable format to facilitate later phases of an investigation, e.g., planning detailed field sur- veys and the layout of bore-hole drilling programs. It should be noted that remote sensing is never a solution in itself but rathera significant tool to assist the investigator in his basic skills. For example, the analysis of imagery acquired from satellites or aircraft from which specific information is sought (geologic mapping or environmental assessment) is best per- formed by scientists trained in these disciplines. Assistance, where required, in the techniques of image ‘ enhancement or computer-assisted analysis can be provided by remote-sensing specialists. DATA ACQUISITION Remote sensors generally provide the following advantages: 0 When carried aboard aircraft or spacecraft, they offer a synoptic overview not achieved by ground survey methods. Observations of the total scene recorded as an image present a visual set of data patterns not merely a group of data points as would have been provided by ground methods. 0 The sampling technique is an unobstrusive way of gathering data. The mere presence of a ground survey team, for example, investigating potential sites for development may result in the spread of unfounded rumors and cause unwarranted ad- verse reactions that may hinder further evalua- tion of the site. 0 Images have a very high information density com- pared with graphic, textual, or electronic storage media. 0 Remote-sensor observations provide a more com- prehensive picture of the terrain than do field methods. Although the level of detail recorded by a sensor may not be so detailed for a small area as ground observations, the sensor record affords a valuable overview in a manageable form. The cost/benefit ratio between overhead coverage and ground traverses for a given area greatly favor the remote sensing approach ex- cept in cases where investigation of only a very small area is required. TYPES OF REMOTE-SENSING DATA Satellite-Acquired Data—Landsat and Skylab For broad, regional coverage requirements (e.g., site selection and regional tectonics) the use of satel- lite imagery from NASA’s Landsat and Skylab pro- grams has proven to be an invaluable aid. In July 1972, NASA launched the first unmanned satellite specifically designed to acquire Earth re- sources information. Since then a second vehicle was launched in January 1975, with a third scheduled for 1977. A fourth vehicle is planned for 1978—79 with insertion into Earth orbit from NASA’s Space Shuttle. The capabilities of the Landsat systems currently in operation provide for imaging the entire globe in ' 100X100-nautical-mile increments from an orbital altitude of approximately 500 nautical miles (fig. 1). This orbit was designed so that sensing is always con- ducted at a constant sun angle (approximately 9:30 123 124 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE '1.—Landsat satellite. am. local time). The only deviation in illumination occurs as a result of the seasonal variations in sun elevation. Each satellite repeats coverage of any given swath of the United States every 18 days, and, with both satellites operating, each swath is covered every 9 days. This allows changes on the surface to be moni— tored regularly. Overseas coverage is obtained on a selected area basis. The sensor system aboard the satellites consists of a multispectral line-scanner. This system generates individual scenes, with each scene scanned in four different wavelengths to produce four similar images simultaneously, as follows: Green Band . . .records in the 0.5 to 0.6 ,Lm range Red Band ..... records in the 0.6 to 0.7 mm range IR Band ...... records in the 0.7 to 0.8 [H.m range IR Band ...... records in the 0.8 to 1.1 ,im range As the vehicle travels along its orbit, each frame overlaps the previous one by about 10 percent so as to provide continuous coverage. Adjacent parallel orbits are spaced so that side-lap coverage is about 10 percent at the Equator with greater side-lap nearer the poles. The observations of the Earth are transmitted to ground receiver antennae as electrical signals and are subsequently converted to facsimile images in the laboratory. Figure 2 is an example of such imagery. The three Skylab missions were provisioned with two film camera systems. The astronauts photo- graphed swaths of the Earth’s surface with different film-filter combinations (black and white, color, color- infrared photography) in two photo scales. Exposure intervals were cycled to accomplish stereoscopic cov- erage. Under good photographic conditions, it has been reported that ground resolutions approximating 30 ft (10 m) have been obtained. Airborne-Acquired Data A vast amount of remotely sensed data is acquired by aircraft, including inexpensive light aircraft, using hand-held, 35-mm cameras or mapping cameras over open ports in the floor of the plane. For broader re- gional coverage, more sophisticated imaging systems are carried aboard multiengined prop and jet aircraft (including the RB—57 and the U—2) for moderate- and high—altitude missions. Aerial cameras with different capabilities are de- signed for different photographic missions utilizing a wide range of film and filter combinations. The most common cameras are those employed by aerial survey firms where metric data are required for high pre— cision applications such as in contour mapping or APPLICATIONS OF TECHNOLOGY FOR POWERPLANT SITINC “0751-88! 14873-30! 311873-8123! 7 ‘7 #1074881 N874-68l ”1039*30 ”463-38 ”8.34! BSY‘IHY?“ C 5140-28/14873—43 N NAG-IS/N673-33 ”SS 5 D SUN ELSE H2126 188's}! lB-N-leVjD'ZLEISH ERTS E~L654-IS@I3- FIGURE 2,—Landsat image of New York City and vicinity. 125 0! 562 IQQJ ‘ M‘JSE 163(4) V 8 [>632 I®®~®D®Z ISLO I CDCOGJZ 126 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 3.—Therma| infrared image recording of surface heat emissions. (Daedalus Enterprises, Inc.) compiling volumetric quantities, etc. These carto- graphic cameras are designed to provide low image distortion with high illumination tolerances to main- tain photometric quality. The most common carto- graphic cameras provide 9><9-inch image frames on rolls of aerial fllm. Image scale factors can be altered by varying (1) the flight altitude or (2) the camera’s focus through the use of interchangeable lens cones of varying focal lengths. Stereoscopic photography is accomplished by selecting exposure intervals so that successive frames overlap each other by 50 to 60 percent. Thermal Infrared Scanners An improved remote-sensing device which detects heat emissions from the surface of terrain or water bodies employs an optical scanning technique to re- cord observations beyond the visual range (thermal infrared, see fig. 3). These observations are accom- plished by virtue of surface phenomena which illu- strate a “thermal contrast” with respect to its back- ground, e.g., warmer outfall effluent into a colder stream. Scanners are configured with single or arrayed de— tectors which sense the incoming signal from a col- lector optic (scanning mirror). Scanners look at a single “spot” of ground at any given instant in time. The “spot” is scanned laterally to produce a line of imagery. The forward motion of the aircraft “collects” successive lines to produce a swath of the scene. The incoming optical signal image is converted to a modu- lated electrical signal which can be either recorded on tape for later reproduction or used inflight to vary a point source of illumination to photographically re- cord the image on film. Side-Looking Airborne Radar (SLAR) SLAR is an active sensor which can obtain images either during the day or at night and through most cloud cover. The system carries its own “illumination" source that transmits a radar beam off to the side of the aircraft, normal to the flight path (hence the term “side-looking") and detects the back—scattered radar from ground objects. These signals modulate a light source (cathode ray tube sweep), through focusing APPLICATIONS OF TECHNOLOGY FOR POWERPLANT SITINC 127 FIGURE 4.—Sketch diagram, typical side-looking airborne radar system. (Modified from Westinghouse Elec- tric Corp., 1967, Side Look Radar.) optics, and onto a film recorder (fig. 4). Imagery ob- tained from this system depicts the terrain in a way unlike conventional photographic records. Gray shades in SLAR images are a function of several terrain characteristics, such as object roughness and the basic electrical properties of the materials. The moisture content of soils also affects the film record. Due to the low gazing angles of the system’s “illumination” sources, especially toward the horizon, SLAR images highlight terrain relief features so as to create the impression of a “three-dimensional” display. This rendition often emphasizes subtle relief features, as in a scene illuminated by the sun at a low angle. These enhancement characteristics are often most useful in detecting fractures, faults, stream patterns, and other terrain relief features of significance in geologic studies (fig. 5). Aeromagnetics Airborne aerial magnetic surveys measure the dif- ferences in magnetic susceptibilities of rock types over a study area. A series of parallel and perpen— dicular flights is designed to provide a complete set of “gamma” profiles recorded by the magnetometer and referenced to a common magnetic datum. The data are processed to produce magnetic contour plots from which geophysicists interpret anomalous zones associated with geologic structures (fig. 6). The interpreted magnetic map is often used in con- junction with overlays generated by image analysis showing lineaments. These overlays are superimposed over the aeromagnetic map and provide corroborative evidence of significant geologic structures, such as fault and fracture zones. Ground-Acquired Data To supplement information derived from overhead coverage, geological and geophysical ground surveys are performed to provide details over specific site locations (e.g., suspect areas which may impact in the location and design of nuclear power generating facilities). 128 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 5,—Side-Iooking airborne radar image by International Aeroservices Corp. and Goodyear Aerospace Corp. For these purposes, portable ground magnetometer, gravity meter, and seismic surveys are conducted. DATA ANALYSIS AND SYNTHESIS In most site investigations, the analysis of remotely sensed data is limited to Landsat imagery and off-the— shelf aerial photos for reasons of cost effectiveness. The cost of these data sources, where purchased from the Federal or State government, ranges from a few tens to a few hundreds of dollars. Interpretation and analysis costs rarely exceed $10,000 to $15,000, and a considerable quantity of structural information usually results. Other sources of remotely sensed data are not often used if they have not previously been acquired. However, in order to illustrate the methods and procedure of remote-sensor data analysis, it will be assumed that data from all of the previously de- scribed sensors exist (i.e., Landsat, Skylab, SLAR, aeromagnetics, gravimetrics, aerial photos, and thermal infrared). In general, the procedure of analysis begins with small-scale formats and progresses to larger scales. The Landsat imagery, typically at a scale of 12500000, is usually analyzed first; followed by SLAR, typically at a scale of about 11250000; and proceeding on to aerial photos at about 125,000 and a thermal infrared scan at a scale of 1210000 or larger. The aeromag- netic and gravimetric data are interpreted separately at about the same time as Landsat and SLAR. The following describes the methods and criteria em- ployed in the analysis and interpretation of each remote-sensor data product: 0 Satellite Analysis—The interpretation of Landsat imagery is usually performed on spectral bands 5, 7, and the color-infrared composite. Supple- mentary coverage from Skylab where available is included. The principal effort is directed toward the detection and delineation of linear features that are neither manmade nor caused by bedding or foliation (fig. 7). It is evident, by the process of elimination, that such features are very likely to be either faults or fractures at the surface of the Earth or surface expressions of deep-seated dislocation. Each lineament is interpreted as be- ing either a “probable fault” or a “possible fault.” APPLICATIONS OF TECHNOLOGY FOR POWERPLANT SITING 129 78°00'00” 77°52’30" 77°45'00" 77°37'30” 38°07'30" . // nunéaqsnm ‘ / k// i 38 °00'OO” ~ ’ />//4 = ego W/ /// ,7 / , i l. i, " (757.4» , //,/x\V‘{ M N/ ,. / 7/” Kr: ////////' ./ ,(’, //y I k / w // d/“H/ ///% / , . r W . . ”W 3795230” v ,' 2: 7 [\lfl) '/\ / M)“, / l V) / NEUSCHEL'S LINEATION 1 0 1 2 3 4 TOTAL INTENSITY IN GAMMAS CONTOUR INTERVAL 25 GAMMAS 5 MILES m BASED ON HALF MILE SPACED, EAST-WEST FLIGHT LINES 5N FEET ABOVE GROUND FIGURE 6.—Typical aeromagnetic map, Mineral, Virginia. (Modified from Neuschel, S. K., 1970, Correlation of aeromagnetics and aeroradioactivity with lithology in the Spotsylvania area, Virginia: Geol. Soc. America Bu||., v. 81, p. 3575—3582, fig. 2.) An example of a “probable fault” is offset bedding or an offset ridgeline. An example of a “possible fault” is a very straight stream seg- ment, which may have developed by headward erosion along a fault or which may simply be a consequent stream following the original slope of the surface upon which it developed. All of the lineaments are selected on the basis of satisfying one or more of the following criteria: 1. Alignment of drainage. This refers to a recti- linear segment of a stream, river or lake. If a valley is observed, but not the stream itself, criterion number 3—alignme‘nt of geomorphic features—is used. 2. Alignment of lithology. A rectilinear inter- face between two lithologic types as- sumed to be evidence of a “possible fault.” 3. Alignment of geomorphic features. This in- cludes straight ridgelines and valleys that do not appear to be caused by bedding. 4. Alignment of vegetation. This criterion in- cludes rectilinear interfaces between veg- etation types as well as long, straight 130 FIGURE 7.—Lineament analysis of Landsat image (1072—15190). (From Withington, C.F., 1973, Lineaments in coastal plain sediments as seen in ERTS imagery, in Natl. Aeronautics and Space Admin. Goddard Space Flight Center, Symposium on Significant Results obtained from Earth Resources Tech- nology Satellite—1, 2d, New Carrollton, Md., March 1973, Tech. Presentations, v. 1, p. 517—521.) narrow bands of vegetation, such as those bordering a watercourse. 5. Alignment of optical density (tone). This criterion is employed when the cause of an observed alignment could not be iden- tified further; that is, when only a recti- linear tonal difference could be seen. 6. Offset of drainage. This refers to a recti- linear, lateral displacement of a stream. 7. Offset of lithology. A rectilinear discontinu- my in the strike of a contact is regarded as evidence of a “probable fault.” In most instances, lineaments that are probably caused by an offset in lithology were identified as offset or geomorphic features. This is due to the fact that the interpre- tation of geomorphic features is virtually certain while the interpretation of litho- logic types is conjectural. 8. Offset of geomorphic features. 9. Offset of vegetation. (When interpreted as not being the work of man.) 10. Offset of optical density (tone). In general, but by no means always, criteria 6 through 10—offset—are regarded as “probable FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM faults," while criteria 1 through 5—alignments— are regarded as ”possible faults" unless the align- ment is very rectilinear and well-defined. The suspected faults are delineated on a transparent overlay registered to the Landsat image. 0 Analysis and Interpretation of SLAR—The same procedure is followed, and the same criteria are employed as in the analysis and interpretation of Landsat imagery; that is, the analyst searches for evidence of alignment and offset. The SLAR is first examined for evidence corroborating the lineaments found in the Landsat imagery and is then searched for evidence of additional linea- ments. As with the results of the Landsat analy- sis, the suspected faults found in the SLAR imagery are delineated on a transparent overlay registered to the SLAR. Figure 8 is an example of SLAR analysis. 0 Analysis and Interpretation of Aeromagnetic Data ——The aeromagnetic contour maps are inter- preted without consulting the Landsat or SLAR data. The criteria used for recognizing aeromag— netic “lineaments" include: Termination of highs. Termination of lows. Changes in gradient. Linear contour patterns. Alignments of highs and lows. Long, linear “alteration” lows. A combination of the above criteria. NerbWN!‘ Lineaments that conform to these criteria are delineated on a transparent overlay registered to the aeromagnetic contour map. 0 Analysis and Interpretation of Gravimetric Data— The gravity contour map is interpreted in con- junction with the magnetic map. Differences in map construction introduce small errors in the alignment of the interpreted gravity lineaments and magnetic lineaments, but these displacements do not invalidate the existence of anomalies. Gravity lineaments are interpreted according to the same criteria used in interpreting the mag- netic lineaments described above. The results are delineated on an overlay registered to the gravity contour map. 0 Analysis and Interpretation of Aerial Photos—The principal difference between analyzing aerial photos and satellite (or SLAR) images is that the synoptic overview of the latter is exchanged for the stereoscopic view of the former. In other words, the integrating effect of the large-area, small-scale data set is supplemented by the great APPLICATIONS OF TECHNOLOGY FOR POWERPLANT SITING 131 density of detail provided by the large-scale, three-dimensional format of the photo. The analy- tical criteria remain the same, however, in that the analyst is searching for evidence of alignment and offset. The principal function of the photo— interpretation phase is to detect corroborative evidence of lineaments found on Landsat and SLAR and to locate smaller lineaments that Landsat and SLAR failed to detect. 0 Analysis and Interpretation of Thermal Infrared Imagery—Thermal infrared imagery is seldom used in site investigations unless it has been pre- viously acquired for some other purpose. This is because it is generally believed to be noncost effective in investigations of this kind, where it is unlikely to detect structures that have not already been detected less expensively through the use of Landsat and aerial photos. However, if thermal imagery exists and can be purchased at low cost, or, if the site is underlain by a lime- stone terrain in which near-surface cavities can be expected, it is well worth the additional cost. Unlike thermal pollution studies, site investiga- tions do not require that the thermal imagery be calibrated so as to give accurate temperature readings. The only requirement is that tempera- ture differences and anomalies be observable FIGURE 8,—Lineament analysis of SLAR imagery. and caused, for the most part, by differences and anomalies in soil moisture or vegetation, which in turn may be caused by faults and near-surface cavities. Figure 9 (p. XIV) shows a machine-aided analy- sis of thermal imagery in which the colors repre- sent preselected temperature ranges. For example, faults serving as ground-water conduits may cause the superjacent soil to be wetter (cooler) than the surrounding soil, and cavities may drain the overlying soil causing it to be dryer (warmer). 0 Synthesis of Remotely Sensed Data—When the in- terpretation and analysis of all of the remote sensor records have been completed, a transfer scope is used to plot all of the delineated fea- tures of structure and lithology at a common scale and format on separate transparent over- lays. These are registered to a common base map on which known faults have been delineated. Superimposition of the several transparent over- lays results in an effective synthesis of the separate data sets contributed by each of the remote sensors that were employed. A syn- thesized overlay is then prepared which combines all of the geological features from the separate overlays. The separate structural features are 132 indicated by different line weights and patterns— one weight or pattern indicating that the feature was observed on all sensors and on the “ground truth” map, another weight indicating that the feature was seen on all of the sensors but one, and so on down to those features that were ob— served on one sensor only. Figure 10 (p. XIV) shows a synthesis of Landsat, aeromagnetic and and gravimetric data. APPLICATIONS Nuclear Powerplant Siting Studies Dames & Moore routinely utilizes satellite and aerial imagery to support investigations for the loca- tion of potential plant sites and as an integral part of studies for the safety analysis of selected sites. The interpretation of these images is performed to extract information relating to the regional geology and tectonics. Utilizing imagery from the Landsat and Skylab Programs, analyses are conducted to delineate re- gional lineaments from which indications of fault- ing and major fracture patterns can be assessed. This information is presented on overlays to standard 12250000 maps of the project area. Correlations are made between the linears interpreted from the over- head coverage and known faults as shown on existing geologic maps. Linears which may have been erron- eously compiled as manmade features (i.e., roads or railroads) are corrected (eliminated); existing fault traces which appears on geologic maps may be ex— tended. By virtue of the broad (synoptic) view afford— ed from satellite altitudes, new lineations, heretofore undetected, add to the store of geologic information about the region. In addition to the linear compilation, annotations showing geologic contacts in greater detail can be compiled and correlated to formational units from existing geologic mapping. By this means the final overlays will have provided investigators with litho- logic and soils maps of significantly more compre- hensive information. As a complement to the satellite data, Dames & Moore has acquired available SLAR coverage for a number of projects. SLAR imagery has been used as correlative information and has been found to be extremely useful in that subtle surface linears are en- hanced by SLAR, thereby facilitating detection. Another data input applied to the geologic analysis is aeromagnetic mapping. Often these data are not available; however, where costs are justified, new surveys have been executed and the results indicative FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM of subsurface structure are applied to those lineament features obtained by visual analysis. Regional mag- netic anomalies, where found, can be matched to the scale of the linear compilation and often highlight zones for priority field checking at later stages of a siting program. The procedures described above constitute an initial reconnaissance phase from which a “first-cut” evaluation of desirable sites can be made. Once this decision is made, a followup analysis utilizing high— and low-altitude aerial coverage is conducted. The higher resolution aerial photography provides the greater ground resolution from which detailed geo- logic mapping of localized areas can be performed. The resulting products (annotated maps, photo- graphs) are useful in orientating field crews and di- recting them to outcrop locations, suspected faulting, et cetera. These are supplied to the project field of fices and utilized in the field for planning detailed surveys. In addition to remote—sensing analysis for geologic/ tectonic information. Dames & Moore has performed visual image interpretation and computer processing for land use mapping and ecological assessment. Digital magnetic tape data have been used to create land—use maps for several siting studies using com- puter-interactive techniques. Figure 11 shows such a map, in which four categories of use have been iden- tified. Parcels of land down to about 1 acre in size are shown. Where secondary or tertiary levels of de- tail or higher ground resolution are required, aerial photography and visual interpretation methods are employed along with ground truth surveys to delineate these subgroup requirements. The land use thematic maps output, from the multi— spectral processing, may be compiled in a variety of formats. Different themes or categories may be dis— played as alphanumeric characters, colors, or shades of gray. Projects where similar requirements are applicable include: environmental impact reports, wetlands mapping, agricultural and forest stand inventories. The same techniques employing electronic process- ing for land use data are also extremely useful in sur— face water mapping. Using color-infrared data, accurate delineation of land/water interfaces can be made down to the highest resolution of the data. In some cases, physical and chemical pollutants can be identified and traced back to their sources both in estuarine and marine environments. Using thermal infrared data, surface temperature differences of less than 1°F can be mapped, providing an excellent tool for mapping differential thermal effluents, spotting APPLICATIONS OF TECHNOLOGY FOR POVVERPLANT SITINC 133 a. t W.. N . .t x‘ g, {1:14 {’4 FIGURE 11.-—Digital|y processed land use data from Landsat observation. water seepage, or freshwater sources on the terrain surface. EXAMPLES A brief description and illustrations of three nuclear powerplant studies involving typical tasks performed by Dames & Moore employing remotely sensed data and techniques are as follows: Site Selection, South Carolina—A preliminary in- vestigation utilizing Landsat imagery (bands 5 and 7) was conducted to ascertain major regional lineaments over the State of South Carolina. This “front-end” survey was performed to delineate indicators of subsurface structure which provided initial information as to: 1. Areas that might be eliminated on the basis of potential fault—related zones. 2. Where other suitability factors favored plant siting near these zones, further studies could be planned to direct subsequent ground investigations to these areas. 0 Site Assessment/Regional Tectonics, Ameri and Bandar Bushehr, Iran—The area circumscribed by a ZOO-mile radius which originates at each of these two sites was evaluated and supported by Landsat image analysis. The analysis provided overlays upon which observed lineaments and existing fault traces were compiled. These prod- ucts were correlated with a plot of epicenter locations to determine subsequent requirements for geologic field investigations. At a later stage, critical areas were analyzed in detail using 1:50,000-scale aerial photos, in stereo, and uti- lized in field surveys to verify faulting. a Site Specific Investigation, Fulton, Pa.—The result of a regional tectonic study via Landsat visual interpretation detected a number of lineaments within a 10- to 20-mile radius of the site area. Corroborative information was achieved from SLAR, high—altitude U—2 overflights and detailed aerial photos at a relatively large scale. In addi- tion, an aeromagnetic survey was performed so that the synthesis of all source outputs provided FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 134 .QNOFIMFOF A. .m.. .> ruck. £5: 9:: :xm... €0.30... .E:_man>m >w>5m 3330mm”. stmm. .:_EU< macaw Ucm mu..:m:o._m< .52 E $9.353“: w:_mmmuo.q-wme. .mzwfi Ucm Emu .‘Ifimncm. main mam; wEImonéuEzw .0. 5.8m .mnmr 5&3. 5.92:3. Ucm :m 28.0 :0 .y. aEEsum E0: “Imam—3... .8... :23 E :5me m. cocmmzfimEE 02:3 tam c0390.? .mE.v£o‘_.u>; 9:8 .9: U039? 8.906 3.0: oucmmmficcogm :fimESmEE: mm .uwmmflu manta w>c wESOLm 63$me. \__.mD._-..0v. .o .83 «Em :m .0 QmE cozmucfimflu Uvmmmuoa .83a80u .m..umn.m_._z§ll.mw $50.. II IIIII. IIIIIIIII III VIII.I.I .I x 1.- I_xI_.- xx -x II II III..I ..xrx .2: .I.. .1 II xILIII ....... ...I II II. IIIIIII III-I IIIII II II... _ II~ I . xx: xx xx _ II IIIII x t;-: 2:1. I xx ._.__ I II II _ I IIIIIII III I n. x I I I1 I . _I I ”I I 1 .I- .III. .1 II _ _ IIIx III I I .1 1- x :x:: I I.xx x III-III. .x.II . 112.... . .xIx III: .2: I z. 4— HI I I.. x xx Ix. II: I z I II I. - -:_xx;:xIII:.z;IIIn . IIxII I .IIx.IIx:I II I». II xI 11.x:xx 1:.z :aIII I :IIII..I. . In: .I :. IIxI..x 1:1 ...rx‘: II In_I . I I.xxxxII I. I 1- I I I. IIqusz II IIIIIII III .. x:- u. I: z:::_ . . _—. I ImI II .. II IIII II. 1 I : _I ._ II III .._ _:Ix IIIIJIEXIZ? IIII III I III-IIIIIIIIII... Iztxxzx.. IIIIIIIIIIII .IIIIIIIII IIII III. I IIII xIx :.Inxx I I. .x.-#_. IIIIIII I r. IIII. I II... x: .21”: I. w .II I II II II I:_ :. .IIIuII IIII. III" IIIII II IIIII I I.1 I III I I III I I r:---:1-.rz.. IIIIIII III I IIII III I IIII IIIIII II III” I I I IIIIIII . .. I I I IIII I . I IIII I IIIIIIIIII . I .IIIHH IIIIIIII II III IIIIIIII III I I IIIIIIIIII I IIII II I I IIII IIIIII :xI .1 2: I x. .1 .lJHIII IIIIIITI IIII IIIII II IIxI IIIII IIIIII IIIIIIIIII x I_. IIII II.II..I IIIIIII : II .IIIIIIIIII x t: .:I ..I u 1.2:: _ . ...: I ..III .IIr .IIIIIIIIIII I1.:12II . II I I III II: IIIII I I IIquIIIIIII .22.... I. II III. II.IxxxxII II. I..xI1 :1 :IIxy__x IIIIII .3523. 17:17 I .. 31:11.: I I ..1II .m .x-IIzI III.2~1zI I I I —Qx IXZ IX Inlx. ._ . I I 2.qu I I IIII II. I .xx . z r I. . — IIII _ I 1..II:I II: II. I IIII II III 3 I II Ixx I II . X X ‘3‘ XX. 0 I X! — _- . XXX-u: II I: _ .I III-III: I I; :2... xx I. ._xI. IIIII. I .I: .I. I .x x. Ix..- APPLICATIONS OF TECHNOLOGY FOR POWERPLANT SITING 135 isolated areas to be used in planning ground mag- netic surveys, seismic investigations and ultimate verifications by bore-hole drilling. Figure 12 (p. XV) shows lineament data from three sources: Landsat, SLAR, and U—2 photos. 0 Uranium Exploration—We are witnessing a new surge in exploration activities for increased sources of uranium to support the projected needs of the nuclear power industry. Remote— sensing techniques have been developed which now afford cost-effective means for surveying vast areas and mapping zones which may contain potential new deposits. Two methods which are employed utilize satel- lite and aerial photography along with geophysi- cal survey data, as follows: 1. Visual interpretation for lineaments coupled which may be associated with faults and fractures. Figure 10 (p. XIV) shows areas in which lineaments derived from Landsat data, aeromagnetic data, and gravimetric data are coincident (dark swaths), and areas in which Landsat lineaments are coincident with either aeromagnetic or gravimetric lineaments (light swaths). Areas of known mines (shown as light dots) show a high de- gree of correlation with these areas of coincident lineaments. 2. Multispectral computer processing to de- lineate potential mineralized zones which may be associated with surficial features (such as halos) and which show a unique spectral response (signatures) on Landsat multichannel scanner data (fig. 13). with aeromagnetic and gravity data reduc- By these procedures an area survey may be accom- tions and correlated with geologic maps plished to delineate potential mineralized zones whose and known deposits provide an overview to occurrences may be related to deposition in faults/ delineate areas of potential mineralization fractures or in sedimentary structures. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Landsat—1 Image Studies as Applied to Petroleum Exploration in Kenya By John B. Miller, Chevron Overseas Petroleum, Inc., San Francisco, California 94105 ABSTRACT The Chevron-Kenya oil license, acquired in 1972, covers an area at the north end of the Lamu embay- ‘ ment. Immediately after acquisition, a photogeologic study of the area was made followed by a short field inspection. As geophysical work got underway, an interpretation of Landsat—1 (ERTS—l) images was completed as a separate attempt to improve geo- logical knowledge. This paper describes the method used in the image studies, the multispectral character— istics of rock units and terrain, and the observed anomalous features seen in the Landsat imagery. The observed lineaments seem to be part of the regional fracture system associated with East Africa rifting and crustal expansion. A trend intersection at the north end of the Lamu embayment might be inter— preted as a failed, immature triple junction, with one postulated arm opening into the Rudolf trough and a second arm into the Ogaden basin. The Lamu em- bayment is viewed as a possible aulacogen. A feature of some significance to the geologic inter- pretation is an early Quaternary depositional summit surface formed from merging alluvial fans. The surface is slightly cut by shallow modern valleys eroded into the top part of Pliocene sedimentary deposits just beneath. Absence of a fan near the northwest license corner and a swampy area along the Ewaso N’Giro drainage suggest subsidence along what may be a structural trough or downwarp. The postulated local trough along the Ewaso N’Giro is confirmed by the results of magnetometer and gravity surveys over the area and some early results of reflection seismic studies. The geophysical results also support certain predictions made from the photo- geology and image studies relative to the basin boundaries and larger lineaments. For example, anom- alies are found along the Lagh Choichuff, Lagh Bogal, Ewaso N'Giro, and Hagadera-Liboi lineaments. However, convincing geophysical evidence of the Habaswein-Wajir Bor lineament has failed to material- ize. Postulated fault relief along the trend, down- dropped to the south, seems to be present relative to a structural block of basement rock at and west of Wajir but is not apparent relative to the sedimentary units and underlying basement farther east, toward the Somali border. Something of a puzzle in inter- preting the geology of this northeast corner of the license area is the fact that the sedimentary section north of the lineament as indicated from the pre- liminary geophysics is very much thicker than the section measured in surface exposures. Anomaly A, a large circular feature seen in the Landsat imagery, partly coincides with the apparent “basinal” (thick sedimentary) area east of Wajir. The real nature of this anomaly is not yet apparent. The Landsat imagery yielded a surprising amount of information in this flat, relatively featureless area. The success obtained strongly supports Landsat studies as a valid exploration tool. Such image studies and photogeologic work nicely complement one an- other. The study has helped to define the relationship of the Lamu embayment and its internal structure with surrounding regional features, such as the East Africa rifting, the Rudolf trough, the Bur-Acaba structural ridge, and the Ogaden basin. INTRODUCTION When Landsat—1 (ERTS—l) was launched by the Na- tional Aeronautics and Space Administration in July 1972, a door was opened toward a new type of infor— mation, quick and inexpensive to use. Among many 137 138 FIRST ANNUAL PECORA uses, the information can assist in exploring for oil, gas, and minerals. The task of Landsat—1, also Landsat—2 and future units, is to provide an orderly array of pictorial images over what eventually may be most of the Earth's sur- face. Already the coverage is large and in many in- stances repetitive through time and changing seasons, snow, flood, and drought. Many geologists who have made serious and well- informed efforts to use Landsat imagery have ob— tained good results. Also, they have recognized some of the limitations of such work. These things have been reported principally through the NASA ERTS—l symposia, especially those of March and December 1973. Several things contribute to a rather unique value of the Landsat imagery as it applies to exploration. One is the coverage that is already great and promises to be all but worldwide. Also, the multispectral capa- bility, supplementing what can be seen on conventional airphotos, helps the geologist to distinguish different types and combinations of rock, soil, and vegetation and to find anomalies. The orthographic exactness and scale uniformity of Landsat imagery is of special value. An important application of the data is to pro- vide quickly useful planimetric and geologic maps in areas where information is sparse. Inaccuracies in ex- isting maps can be corrected while posting the maps with additional new information visible on Landsat images. The Landsat—1 imagery is of two types: MSS (Multi- spectral Scanner) data and RBV (Return Beam Vidi- con) data. However, the bulk of Landsat—1 data and all the Kenya imagery is M58. The Landsat imagery of Kenya was not obtained in time to be helpful in selecting the initial Chevron acreage. It was obtained soon afterward, however, and was used in making an early appraisal of the area and in planning a work program. Also, it was part of the basis for filing later on a smaller additional op- tional block which adjoins the initial block at its northwest boundary and corner. At each stage after its original presentation in Cairo, Egypt, additions have been made to this paper1 to keep abreast of the progress in acquisition and ex- ploration. This revision adds geophysical information not previously given. ‘Updated and slightly edited from a paper which was pre- sented November 1974 at the Fourth Exploration Seminar, Egyptian General Petroleum Corporation, Cairo; in February 1975 at the Research Conference on Remote Sensing, Univer- sity of Kansas, Lawrence; and in June 1975 at the NASA Earth Resources Symposium, Houston. MEMORIAL SYMPOSIUM GEOLOGIC BACKGROUND The Chevron oil exploration license covers the northern part of the basinal feature of the Kenya coast known as the Lamu embayment (fig. 1). That much was known in 1972 when the exploration right was acquired. But not much was known about the exact basin shape, its limits, and structural nature. FIGURE 1.——|ndex map showing regional features and areas of the Chevron exploration license and option (shaded line). Lineaments are numbered: (1) Lagh Bisika, (2) Lagh Bogal, (3) Lagh Katulo, (4) Habaswein-Wajir Bor, and (5) Hagadera— Liboi. Area of Landsat image mosaics, figures 2 and 3, shown as dashed line. Previously, British Petroleum—Shell had held an oil lease over a large part of the Lamu embayment, in- cluding some of the ground taken by Chevron. Most of the BP—Shell rights had been relinquished, leaving only a relatively small area along the Kenya coast. BP—Shell had explored extensively and drilled 16 “dry” wildcat wells in the Lamu embayment. Only three of these—Meri—l, Waligero—l and Wal Merer—l ——are on the Chevron tract. These were enough to demonstrate the presence of more than 12,000 ft (3,660 m, approximately) of sedimentary rocks within the land acquired by Chevron. Several other wells of substantial depth were known to exist some distance to the east, in neighboring Somalia. But while information on the embayment itself was sketchy, certain information on surrounding areas was well documented by the Geological Survey of Kenya. In the area west and north of Mombasa, Caswell LANDSAT—1 STUDIES APPLIED TO PETROLEUM EXPLORATION IN KENYA (1953, 1956), Sanders (1959), Thompson (1956), and Williams (1962) reported a thick section of sedi- mentary rocks. This includes Karroo-type red beds (mainly continental sandstones, Carboniferous and Permian through Triassic), a section of marine shales and local shelf—type limestone (Jurassic and Creta- ceous), and locally along the coast onlapping marine deposits of Tertiary age. The aggregate section of Karroo-type beds totals about 22,000 ft (6,700 m, ap- proximately). The Jurassic section totals 6,600 ft (2,010 m). The Cretaceous is incompletely represented in exposures by only about 300 ft (90 m) of section. In the area north of the license area, extending to the Ethiopia border, the stratigraphic column begins with Triassic strata and ranges through the Lower Cretaceous. These rocks have also been described by geologists of the Kenya Geological Survey, par- ticularly in reports by Thompson and Dodson (1958), Baker and Saggerson (1958), Saggerson and Miller (1957), and Matheson (1971). As compared with sedi— mentary rocks at Mombasa, the thickness is much less and the lithology is somewhat changed. The Triassic again is a Karroo-type facies, but the exposed thickness is only 2,000 ft (610 m). The Jurassic strata, roughly 5,500 ft (1,680 m) thick, are mainly dense, gray shelf—type limestones. About 2,500 ft (760 m) of Cretaceous section consists mostly of shallow-water sandstone and some interbedded marine shale. Contrasts both in thickness and lithology are found when the stratigraphic units north of the license area are compared with those observed near Mombasa. However, it is not clear where and why the changes occur nor what may be their significance to the pos- sible presence of commercial hydrocarbons. Available maps show most of the surface in the license area as Quaternary alluvium. Jurassic lime— stones abut and enter the area east of Wajir, some late Tertiary volcanic rocks occur in the northwest corner, and Precambrian metamorphic rocks crop out in the southwest. Several small Jurassic exposures are reported just outside the area, north of Mado Gashi. Preliminary investigations have shown that, over much of the area, erosion has cut very shallow valleys into underlying clays considered of Pliocene age. While the exact age of these is subject to ques- tion, the clays are mainly older than sands and soils found on the undissected summits, and older by yet another stage than the alluvium in the modern stream- fiats. They appear to be the top part of a fairly thick Miocene-Pliocene section known in much of the Lamu embayment. A complex, sloping plateau rises from an altitude of about 1,200 ft (370 m) above the sea at the west 139 margin of the Lamu embayment north of Garissa to a structural swell reaching a height of more than 6,000 ft (1,830 m) along the Kenya (Gregory) Rift Valley. This is mainly an area of Precambrian rocks, includ- ing a variety of metamorphic and intrusive rocks, but there are also great areas of volcanic rocks and several large volcanic peaks. The more resistant crystalline masses stand as prominent inselbergs. Many belts of hills representing old topography occur within and between the valleys. In this terrain are well-preserved parts of three planation surfaces, pro- duced by denudation after principal stages of struc- tural uplift associated with the African rift system. The plateau area was tilted seaward in Late Creta- ceous, middle Tertiary, and late Tertiary episodes of domal uplift (Saggerson and Baker, 1965; Matheson, 1971; Baker and Wohlenberg, 1971). Two geologic papers containing new, important in- formation relative to the Lamu embayment were published in 1973. One of these, by Walters and Lin- ton of BP—Shell (1973) provides subsurface informa- tion on the Lamu embayment. The other, by Gulf Oil geologists Beltrandi and Pyre (1973), described the geology of adjoining basinal areas in Somalia. Al- though these add substantially to available knowledge, they provide incomplete answers about basinal struc— ture, bounding limits, and stratigraphy. GEOLOGICAL INTERPRETATION METHODS To conduct this study, Chevron purchased 70-mm positive transparencies of all four spectral bands of the required Landsat—1 imagery. These were com- posited into color, using an early model International Imaging System (128) additive-color viewer then avail- able at the Chevron Oilfield Research Laboratories at La Habra, California. Several color arrays were recorded as 35-mm color slides on Ektachrome high- speed film. For interpretation of the imagery, the 35-mm slides and also the 70—mm black and white images were projected on the wall of a darkened room. A sheet of white drafting paper fastened to the wall served as a projection screen and as a base for annotating the geology and other photographic detail. Since completion of this work, Chevron Overseas has obtained, at its home office, an improved IZS viewer. This eliminates dependence on color slides, and makes it possible to annotate on a transparent medium directly over the screen of the compositor. This is done at a favorable scale of 1:500,000. During annotation, the image is examined with a binocular 140 FIRST ANNUAL PECORA headpiece which provides 11/2 X and 21/2 magnifica- tion. Just prior to the Landsat study, even predating the availability of the Landsat imagery, I conducted a photogeologic study of our area (airphoto scales partly 1240,000; partly 1:80,000) and made a 6-day trip into the field. This work extended from Septem- ber into early November 1972. This experience in Kenya was most fortunate as a prelude to the work on the Landsat images. Of course, such experience is never amiss. As background toward evaluating a new and unfamiliar type of imagery, it proves to be espe— cially helpful It is a framework for judging the physio— graphic and geologic phenomena as registered by the Landsat sensors and for evaluating the usefulness of the data. Two Landsat image mosaics are used to illustrate this paper. These copy fourth—generation products from the 70-mm positive transparencies acquired from the EROS Data Center at Sioux Falls, South Dakota. The first step was to photographically enlarge the images into a set of negatives at 1:1,000,000 scale, produced on 9X9-inch film. Contact prints from these formed the original mosaic, which then was repro~ duced into a single mosaic negative and into photo- graphic prints. The remainder of this paper is an appraisal of in- formation recovered from the interpretation of the Landsat images. Comparisons are made, from time to time, with detail as seen on airphotos and occasionally as seen on the ground. A comparison is made also with geophysical data obtained since the Landsat analysis was completed. PHYSIOGRAPHY Much of the drainage pattern is seen on the Landsat images with remarkable clarity, even into the head- waters of relatively minor drainage courses. In view- ing the Landsat images, this is almost the only evidence of relief. An impression of relief is better when there are ground shadows, but very few slopes in the area are steep enough to produce shadows. The exceptions are the cuestas of Triassic sandstones roughly 80 km north of Wajir, several lava features west of Wajir, the Bur Wein anticline south of El Wak, and inselbergs of basement. All of these hills are small, even at the scale used during the interpretation. At the reduced size of the present mosaics and maps (figs. 2—5) they become all but invisible. The drainages are gullies and wide, flood-washed channels in a semi-arid terrain. These are seen clearly in Landsat images and can be drawn accurately. The drainages that originate on the plateau and enter the MEMORIAL SYMPOSIUM region from the west have relatively wide streamfiats. The Tana River carries a good How of water. Through the use of the infrared image particularly, the dark meandering line of this can be followed readily (fig. 2). The other streams are seasonal. The Ewaso N’Giro spreads and much of its flow is arrested at the Malka Galla swampy area north of Mado Gashi. Lush vegeta- tion at the “swamp” produces high response in the infrared as compared to the green and red band re- sponse, so that a false-color image registers flaming hues of red. Below this spot, the Ewaso N’Giro drain— age channels become narrow and dispersed, and, while borders of the streamfiat are sharply defined, the channels within it are indistinct. The same is true for other similar streams. For example, a 15-km seg- ment of Lagh Katulo as sketched differs from the main stream course marked from airphotos, but the difference is inconsequential in small-scale mapping amid shifting channels. The relief, indeed, is low. Except for an escarpment along the Yamicha area lava bench, the only hills with substantial sloping profiles are either west or well to the north of the Chevron license. Although there is about 180 m of slope in 220 km from the west edge of the license to the southeast corner, the horizon, as seen on the ground, is commonly flat. The most. rugged ground generally consists of a gentle, faintly rilled slope into a shallow valley, a low-cut bank, and then an equally gentle rise to a low, broad summit. In perhaps half of the license, these slopes are so slight as to be imperceptible during ordinary stereoscopic study of airphotos. They are defined mainly by direc- tions of rainwash and drainage rather than visible relief. Remarkably, much of the drainage of these in- substantial slopes is seen almost as clearly on Landsat images as on the airphotos. ROCK SIGNATURES Quite a number of boundaries between rock types can be traced on the Landsat images even though the nature of the lithology is not clear. To some degree, even the approximate nature of the lithology can be surmised. In this text, the visible characteristics of units will be reviewed in successive order of decreas- ing age. Many of the characteristics are not seen in figures 2 and 3 because these are much smaller than the images employed for interpretation. The migmatitic character of the basement rocks seems to be subtly evident. A peculiarity of folia is that they show on many scales, which range from microscopic size into the outcrop and then to a textural grain seen on airphotos. In the Kenya LAN DSAT—1 STUDIES APPLIED TO PETROLEUM EXPLORATION IN KENYA 141 FIGURE 2,—Landsat image mosaic Carissa-Wajir-EI Wak area, Kenya, band 7 scenes. The lines where the images are matched show as sharp, straight boundaries parallel and normal with the slant margins. The dark Jurassic lime— stones northeast of Wajir, transected by the Lagh Katulo alluviated valley and fault line, are clearly visible. The Hagadera Liboi lineament is also prominent. The white flecks in the lower right hand scene and at extreme lower left are clouds. Landsat images 1190—07052 (upper left), 1189—06593 (upper right), 1190—07054, 1189—07000, 1190—07061, and 1153—07003. migmatites, some of the foliated grain is reproduced even in the Landsat images. It appears as faint, shadowy bands, probably representing rock and topography and interwoven layers of differing litho- logy. In places, there are inselbergs indicating resist- ant rock. A wide color range, from eggshell and flesh tones into green and gray very likely represent compo- sition ranges from light—reflective bands of quartzose rocks and feldspar through rocks of intermediate composition, mantling residual material rich in iron oxides, into belts of relatively fresh rock containing unoxidized femic minerals, gray slate, and, locally, marble The Triassic conglomerate and sandstones north of Wajir stand out from the metamorphic complex mainly because of their sharp, arcuate- and irregular-shaped cuestas. The well-stratified and resistant character of the rocks is revealed by the erosional form of the 142 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 3.—Landsat image mosaic, Carissa-Wajir-El Wak area, Kenya, band 5. The Lagh Bisika, Lagh Bogal, Lagh Bor, and at least part of the Wajir Bor Iineaments are quite apparent. The Pliocene-Pleistocene dep- ositional surface is better defined by band 5 than band 7 (fig. 2). The expanding wedge of ”upper” Merti can be found by comparing band 5 with band 7 imagery; it is seen in mainly the Tana River valley wall north of Carissa. With apex near the metamorphic rocks and expanding as a narrow wedge eastward, the unit contrasts sharply with the rim of darker duricrust, just north. Its base, seen as a contrast with the lighter colored "lower” Merti, shows better on the band 7 imagery of figure 2. ridges. The low dip present in the strata can be con- strued from down—dip expanse of the dip slopes. A basal boundary, of the Jurassic limestone can be followed only inexactly. The limestones appear quite dark in bands 6 and 7, somewhat less dark in bands 4 and 5, and show as dark gray partly flushed with red in the composite image. The red is a response from a combination of thin soils and sparse vegeta- tion. Cretaceous sedimentary units show in the extreme northeast as sharply defined ridges and cuestas. The direction of dip is generally discernible. Pliocene and Quaternary volcanic rocks, mainly basalts, are present in the northwest part of the area. In fresh outcrop, these register darkest in all color bands, and, in the color composites, they show as nearly black. There is a weathered surface and some soil across the basaltic mass at Yamicha, but, in its LANDSAT—1 STUDIES APPLIED TO PETROLEUM EXPLORATION IN KENYA 143 ’ . yoga—mm" <1_,:> ROCK UNIT BOUNDARIES Km ARCUATE ANOMAUES- Q INTERPRETED AS ____._ LINEAMENTS LA)“. POSSIBLE BEDFORMS. , LINEAR UNITS a ’ ”M’ STRATI FICATION — —————— Egggoanoumanv FIGURE 4.—Lineaments, rock unit boundaries, and other features as interpreted from the Landsat imagery. The Qs symbol designates part of the Pliocene-Pleistocene summit depositional surface, regarded primarily as Qua- ternary material. Centered at A and near its margins by a circular pattern of drainage is Anomaly A. Scale 122,500,000. southern escarpment, which probably marks an Of somewhat the same age are the Pliocene- original flow edge, the volcanic rocks are very fresh. Pleistocene clastic deposits. These show the most Farther north, a virtually unweathered volcanic fiow varied color signatures of all—a greater range than is defined both by its lobate form and nearly black are accounted for by known differences in lithology. color. The Pliocene part can be locally divided into two 144 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM " "WW II uni ”WW ”(Hi/UH 7 §EDIMENTABI W JURASSIC (J) E] RECENT ALLUVIUMa OUTWASH - TRIASSIC PLIO- PLEISTOCENE DEPOSITIONAL SURFACE METAMORPHIC g IGNEQJ§ m CALCRETE w: TERTIARY a OUATER- NARYVOLCANICS. PLIOCENE PRE-CAMBRIAN [MW] CRETACEOUS m METAMORPHIC (B) alsusous FIGURE 5.—R0ck unit distribution, broadly defined from Landsat imagery and other sources. LANDSAT—1 STUDIES APPLIED TO PETROLEUM EXPLORATION IN KENYA parts. At the top and probably Quaternary in age is a depositional surface and calcrete (fig. 5). The main unit, the Merti sands and clays named in Kenya geological literature, is regarded as late Plio- cene. These sediments fill a late structural sag along the Lamu embayment. The observed relationships strongly suggest faulting along the western basin margin, because the “End Tertiary” erosion surface extends across basement rocks at elevations of 600 ft (182 m) or more west of Mado Gashi but drops be- neath the Merti in the basin. As shown by the thick— ness of Merti beds in the Dera borehole, near the basement rock boundary—240 ft (73 m) to top of basalt; 408 ft (124 m) to base of volcanic debris: Matheson, 1971, p. 19—the “End Tertiary” surface seems to drop structurally at least 250 ft (76 m) in crossing the basement boundary. The boundary of the Merti deposits with basement rocks is plainly seen on Landsat—1 imagery at the Tana River. The basement appears banded. The adjoining Merti is finely rilled and, especially in band 7, has a lighter tone (fig. 2). Extrapolating from topographic maps, the thickness of Merti beds exposed in the valley slope near the basement boundary is about 600 ft (180 ml. Above this, and within my knowledge recognized only on the Landsat—1 imagery, is a second unit. It begins at the basement boundary and thickens wedge- like into the basin. It is slightly darker in bands 5, 6, and 7 than the sediments beneath. I distinguish this wedge-shaped unit tentatively with the name “upper Merti,” but suppose it may not be readily separated from the underlying part of the Merti except in the Tana Valley. Saggerson and Baker (1965, p. 57) men- tion 300 ft (90 m) of Merti sediments at Garissa. On the basis of the Landsat—1 image, at least 200 ft (61 m) of these seem to belong to the “upper Merti.” Seemingly they will entirely overlie the estimated 600 ft (183 m) of Merti beds exposed near the basement boundary. The distinguishing tonal difference of the “upper” Merti from underlying beds is clearly seen in the color image and in the longer wavelengths of bands 5, 6, and 7. It is seen hardly at all in band 4. The smaller sensitivity of shorter wavelengths to these lithic divi- sions may explain why the divisions were not recorded during work on conventional airphotos. Coupled with this is another reason: The sediments are uncon— solidated. They are intricately scored by rilled drain- ‘age, and small masses of the material have moved downslope through the action of crumbling and mud- fiow. The boundary, as seen on the airphoto, has probably been softened and blurred by this action. On 145 the other hand, as seen in the overview of the Landsat image, the boundary is relatively sharp and con- spicuous. Above the Merti is a thin unit of sand, marl, and calcrete deposits or “caliche.” This tends to form a duricrust with thickness commonly ranging from 1 ft to perhaps 8 ft (about 2 or 3 m). The top surface, probably Pleistocene, is a thin layer of soil thinly mantled with sand. Near the town of Wajir, the Quaternary deposits—about half sand and shale, and the top half mainly calcrete—thicken to as much as 100 ft (30 m) and contain a substantial amount of ground water. Calcrete outcroppings in this area appear on the Landsat images as white patches. The depositional surface of the Quaternary duri- crust (mapped Qs) has been cut and partly destroyed by the modern system of streams. Nevertheless, there are numerous remaining patches. The distribution of these bears on several aspects of the geology. includ- ing the reconstruction of previous drainage and a definition of lineaments. Remnants of the surface, as seen on airphotos, have a relatively dark gray cast and slight, benchland-type rims. Commonly there is a pattern of widely spaced “pits" that appear to be subtle, periodically wet de- pressions. Where the surface has been destroyed, there are rilled slopes or an expansive surface of clay which is broken, probably along jointing, into polygons or prisms. Each polygon has an almost imperceptible hummock and central cluster of brush (as confirmed on the surface) surrounded by bare clay and a wash of white sand. The depositional surface stands out quite well on the Landsat imagery, although a color composite is needed for most effective mapping. It supports grass and significantly more trees than are found over most adjoining rock units. Consequently, in addition to a darker cast than most surrounding rock units except limestone and basalts, the false-color images rendered by the surface are distinctly reddish. A response that comes from grass and trees, the reddish tint is prob— ably somewhat seasonal, dependent on rains. Examination of the surface suggests two main parts, a sloping plain from the north-northwest, and what may be called the Tana fan from the southwest. Judged from valley entrenchment and airphoto stere- oscopy, there is a definite easterly slope to the Tana fan. Also, the plain beyond the Ewaso N’Giro tends to undulate slightly across axes parallel to the present drainage. As judged from the distribution, destruction of some of the surface, e.g., the northwesterly part and a belt along the Ewaso N’Giro, seems to have 146 FIRST ANNUAL PECORA started before development of the southern part of the surface was complete. Examination of the Landsat images reveals several elements within the plain that were not noticed on airphotos. Located between the Ewaso N’Giro and Lagh Bogal (fig. 4), these can be effectively defined only on the color image (fig. 6, p. XVI). They show as four or possibly five differing arcuate bands of color. Their shapes suggest giant bedding structures. Preserved parts of the Quaternary depositional sum- mit coincide with the associated arcuate forms, indi- cating these might be anchored by resistant layers among somewhat complex stratigraphic forms just beneath the surface. For example, they could mark the upper edges of several shallow, trough-shaped features related to erosion and depositional patterns. Although probably not sensitive to structural trends, these might be significant to the distribution of pos- sible shallow ground water. LINEAMENTS As used in this paper, the term lineament refers to apparently natural lines, bands, and anomalous zones generally more than 10 km (6 mi) long. These lines and linear bands are associated variously with straight or slightly arcuate drainage courses, tonal lines, aligned erosional breaks in the Quaternary depositional sur- face, and occasionally other phenomena. A set of linears aligned about N. 50°—60° E. prob- ably is paired with another set aligned N. 45°—60° W. Representatives of the northeasterly set are the Habaswein-Wajir Bor and Hagadera—Liboi lineaments. Trends in the northwesterly set are along the Lagh Katulo, Lagh Bogal and a segment of the Ewaso N'Giro. Considered as conjugate directions, the in- tersection angles generally fall between 60° and 80°. Another set of lineaments (e.g., Lagh Choichuff, Lagh Bor) commonly bears about N. 10° W.—N. 20° W. Probably conjugate with this is the Lagh Bisika and several smaller trends bearing N. 70° W. The inter- section angle ranges between 50° and 65°. A number of lineaments, accentuated in figure 4, are discussed below. Among the many linear features present, these seem to relate, as seen in figure 1, to geologic phenomena well beyond the figure 4 map boundary. HABASWElN-WAJIR BOR TREND This wide, subtly defined linear anomaly (lineament) crosses the license area at its middle west boundary and, east of Wajir, continues into Somalia. The anom- aly forms a band distinguished in part by sparsity (a result of erosion) of the Quaternary duricrust. MEMORIAL SYMPOSIUM The trend is defined by two bounding' linear fea- tures comprised of separate, aligned, but not quite continuous, lineaments. Parts of these are formed by straight stream courses and other parts by tonal changes occurring, for example, along straight edges of remaining areas of the duricrust where fractures are probably a controlling factor in the erosion. The ending of ridges where Jurassic limestone is exposed northeast of Wajir seems to be part of the tonal change along part of the lineament. This sug- gests that the trend, in this sector, may be an ex- pression of a fault that steps the limestones downward into the basin where they may then be overlain by younger sediments. An easterly extension of the Habaswein—Wajir Bor trend may play a part in an offset in basement rocks along the north flank of the Bur-Acaba structural ridge in Somalia. This is mentioned further in later paragraphs. In the southwesterly direction, the Habaswein- Wajir trend extends toward distant Mt. Kenya, a major volcano. Some 60 km south of the Wajir Bor trend is a parallel Landsat lineament, mainly tonal. It is not named, but on the map passes nearly through the center of the feature called Anomaly A (fig. 4). The lineament may play a part in the tectonic scheme and therefore will be mentioned again. HAGADERA-LIBOI LINEAMENT About parallel to the Habaswein—Wajir Bor fea— ture, the Hagadera—Liboi lineament crosses the south- ern part of the license and then extends into Somalia. Much of it is defined by drainage along a southerly leg of the Ewaso N’Giro (locally called Lagh Dera) past the Liboi military post. Toward the west, it runs along lesser drainage courses and finally extends along the south margin of the Tana River valley. In this segment, it partly outlines the edge of the pre’ served Quaternary surface. As it is traced eastward on Landsat—1 imagery, the Hagadera—Liboi lineament aligns directly along the southerly edge of the Bur-Acaba structural ridge. This ridge, in Somalia, is a major uplift exposing an area of granite and metamorphic basement. The local geology indicates that at least its southern flank is faulted. The Habaswein-Wajir Bor and Hagadera- Liboi linear features were first defined through the Kenya photogeology. Their exact coincidence, when projected by means of Landsat imagery into Somalia, with the boundaries of the Bur-Acaba structural ridge is considered geologically significant. LANDSAT—1 STUDIES APPLIED TO PETROLEUM EXPLORATION IN KENYA CHOlCHUFF, BOR, AND EWASO N’CIRO LINEAMENTS The Lagh Choichuff and its companion trend, Lagh Bor, trend southward across the license boundary along drainage courses bearing these names. Just within the license, each is crossed by south— eastward features—the Lagh Bogal and Lagh Bisika lineaments. The Lagh Bor feature is not seen south- ward of its juncture with these features. The Choichuff trend meets the valley of the Ewaso N'Giro and—if structural aspects are continuous—changes in trend from S. 20° E. to a new direction of S. 25°—30° E. The apparent bend along the Ewaso N’Giro appar- ently is caused by an interaction between trends and by structural slivering. When viewed with the aid of a regional map, the Choichuff lineament is found to align with a segment of seacoast north of Malindi and the lower stream course of the Tana River. This sug- gests a deeply rooted fracture nearly parallel to the fault which bounds the west margin of the basin. LACH BOCAL AND LACH BISIKA LINEAMENTS These two trends cross the northern part of the license area and intersect at a narrow angle. Some structural interference may occur where the two trends meet. There is a hint of left-lateral displace- ment of the Lagh Bogal feature, but exact placement of the lineaments and definition of offset are uncer- tain. The Lagh Bogal lineament is about axial relative to a structural-morphologic feature called the Rudolf trough by Brock (1965, p. 101—102). From a regional standpoint, it might be appropriate to name the linea- ment from North Island in Lake Rudolf because it can be traced this far to the northwest (fig. 1) and prob- ably is responsible for the North Island volcanic center. It is also associated with a concentration of volcanic craters northeast of Mount Marsabit. The Lagh Bisika feature also passes northwestward, but lies slightly askew to the trough. This trend flanks the south side of Mount Marsabit and merges with a bounding linear of the trough about at the south end of Lake Rudolf. Along the Kenya-Ethiopia border, linears Flanking the north side of the Rudolf trough are seen on the Landsat imagery. LACH KATULO FAULT A fault displacement of Jurassic sediment along the Lagh Katulo can be detected both on Landsat imagery and on the Kenya airphotos. The displacement can be seen in the structural attitudes, also in the distribution and expanse of Mesozoic rock units. The faulted lineament can be traced northwestward into Ethiopia, also southeastward into Somalia. 147 OTHER ANOMALIES ANOMALY A This anomaly is partly within the license area. It centers east of the international border, in Somalia, and is a large anomaly of relatively pale tone and color, outlined by a circular drainage pattern. Its eastern edge is marked by a prominent deviation of the Lagh Katulo stream course. A northern boundary is the Wajir Bor lineament; a southern boundary is the Liboi lineament. The feature therefore may be related structurally to the Bur-Acaba structural ridge which apparently develops astride this same trend and structural block, farther east. In a fold regime, such an anomaly is suggestive of a domal uplift. Within this extension regime amid a strongly developed fracture system the domal implica- tion becomes much less forceful. The stream systems seemingly are deflected across major fracture lines, then follow along polygons of fracturing. The anomaly as a whole is outlined by a rimlike margin of the preserved Quaternary duricrust, reduced by sapping action to a circular outline. Whatever the difficulties in exact interpretation, the anomaly probably is related in one way or another to the conditions of subsurface structure. MALKA CALLA AREA The periodic collection of water over the Ewaso N'Giro streamfiat is not strinkingly abnormal at sites such as this where extensive drainage from highlands empty into arid plains. Yet it does suggest a break in the stream gradient and a possible structural sag. Possibly more significant than the “swamps” is the apparent absence, in this sector, of any large alluvial fan coincident with the Ewaso N’Giro. The Quaternary duricrust developed mainly over fans entering from the southwest (Tana River sector) and north (Lagh Choichuff and Lagh Bor, and similar drainage). The absence of a substantial fan on the Ewaso N’Giro may be construed as evidence of structural subsidence. Acting as a sediment trap, substantial subsidence would arrest the movement of sediment and minimize the growth of fans. Construed in this way as a discrete structural sag filled with young clastic sediments, the area may be ideal for large, relatively shallow supplies of ground water. BUR WEIN The Bur Wein anticline (Baker and Saggerson, 1958) lies within Cretaceous outcrops about 40 km south of El Wak. This feature, a sharply defined domal up- 148 lift, shows in the Landsat image as a unique feature. Its steep flanks, clearly visible, set it apart from all other recognized features. The characteristics of the Bur Wein anticline, as contrasted with the surrounding geology, suggest it as a diapiric feature possibly de- veloped over an igneous plug. ANALYTICAL REVIEW Four basic directions have been described in the pattern of lineaments. Three of the directions are coincident with major rift directions mapped in East Africa. The examples, classified roughly in terms of regional direction groups, are northward, parallel to the Gregory Rift Valley (Choichuff), northwestward (Lagh Bogal—Rudolf), and northeastward (Habaswein- Wajir Bor)——the northwestward and northeastward systems parallel regionally to the Rukoban and Ruaha fracture sets in Tanzania. Except for local minor deviations, the directions are remarkably consistent. Not only the younger Cretaceous and post—Cretaceous rifts but also the older basins filled with Karroo—phase sandstones as old as Permian conform with this pat- tern. The rift and fracture system tends to divide the continental block into entire and partial hexagonal prisms. The regional tectonics apparently are dominated by vertical movements and tensional stress. The Lamu embayment, which contains a succession of sedi- mentary rocks beginning with Karroo sandstones, is postulated as an extension basin of possible failed- arm type. Any compression—type structures, within the embayment are probably of secondary importance. Evidence of downfaulting of Pliocene sediments at the basement boundary west of Garissa is construed from the geometry of bedding and sedimentary fill (observed in the Landsat image) as these combine with known relief. The faulting suggests this line as a main, if not necessarily a unique western boundary to the basin, rather than as a less important line of sedimentary onlap and tilting. The anticipated northern limit of the Lamu embay— ment, east of Wajir, is the line of postulated fault relief across the Habaswein-Wajir Bor lineament. North of this, Triassic sandstones and Jurassic car- bonates are found in outcrop, resting on metamorphic basement. South of it, one might expect a much thicker section, ranging to Tertiary age. Westward from Wajir, the situation changes. It is possible to relate three separate phenomena and conclude from them that an apex of the basin may extend into the northwest corner of the license area— its margin staggered, perhaps, between several inter- secting faults: (1) the semi-swamps of the Ewaso FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM N’Giro, (2) an absence of a clearly defined “Quater- nary surface" fan, as previously discussed, combine to indicate an active structural sag, and (3) Jurassic outcrops reported in this area, stepped far to the west of the Jurassic boundary northeast of Wajir. These outcrops must be preserved in a downwarped or down-dropped block and suggest a major offset or displacement of the basin margin. Construed from this evidence is the idea that a somewhat discrete trough or graben may extend to the northwest corner of the Chevron holding. A number of lineaments converge within the north- ern part of the tract. These may relate partly to the termination of the Lamu embayment if the embay- ment is, in fact, an aulacogen or failed arm. The northerly trend of the Lamu embayment combined with two lesser branches may form a triple junction. One of these, behaving as a carbonate shelf northwest and landward from the Bur-Acaba ridge, goes into the Ogaden basin of Ethiopia and Somalia. The other, the narrow grabenlike trough toward Lake Rudolf, seemingly could have been a Mesozoic valley which directed Mesozoic river drainage into the Lamu embayment. This could bring about major changes in sedimentary thickness and facies—changes not yet known, but possibly important to oil occur- rencefiwithin the Chevron holding. Possibly, also, the trough could have held a short—term arm of the sea. GEOPHYSICAL PROGRAM During 1973, an airborne magnetometer survey and helicopter-mounted gravity survey were made. These surveys were planned with the benefit of photogeo- logic knowledge. The work was already in progress during the Landsat study. A seismic program, which benefits from all previous work, is going forward in 1974 and 1975. There is a limited correlation between the geo— physical results and the analysis made from Landsat images and the airphotos (fig. 6, p. XVI; figs. 7, 8). At this preliminary stage, neither the coincidence of anomalies where coincidence occurs, nor a lack of coincidence is very well understood. The best cor- relations are these: 1. A good magnetic definition, and a modest gravity definition closely matching the construed fault along the western basin boundary. The data do not suggest a single profound structural drop-off, but instead one that is stepped or, most likely, involves an element of tilt as well as faulting. 2. Defined in all geophysical results is a troughlike feature in the northwest part of the license area LANDSAT—l STUDIES APPLIED TO PETROLEUM EXPLORATION IN KENYA 149 g ‘s >/ i ‘1'... ELWIIIK R I; z (09 R xx as 4 z “ x‘ m. x Ex 2 § *9. ‘s 5 ‘~ x .3! ‘5. 1‘ (42 "x GlRfllFTU 5‘ I: X In...“ "'-. "x \ ‘5. ‘1‘" A ”M‘- ..-o-""' “s \ ~O~ Lin. M Qfir’r lI'IGARISSA % 8 L E G E N D “moon-0000 LINEAMENTS ANOMALIES AND AXES , LOW OR BASIN (MAGNETIC) (GRAVITY) (SEISMIC) T:::::::::::::z:z: I-‘ZLT‘ISIVE HIGH Cl} + "IIII" HIGH AS INTERPRETED 1- 1- (GRAVITY 0R MAGNETIC) :1 _ m LOW {STRUCTURALLY CHEVRON LEASE BOUNDARY FIGURE 7.——Comparison of Landsat imagery structural features with crudely represented geophysical data. 150 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM GIRIFTU BLOCK STRUCTURALLY / EXPLANATION fix/ 53:133.} 3355:1735?“ l MAJOR STRUCTURAL HIGHS (Moon-tic futuru. support“ by ottm data) SUBSURFACE NG / / ll \ / I I If \ x’)\J° FIGURE 8.——Structural pattern derived from geophysical data over the northern part of the initial Chevron license. In Somalia, a less detailed pattern is located as well as possible from the small-scale published maps of Beltrandi and Pyre (1973). The shaded outlines define the position of the colored Landsat images reproduced in figure 6. These permit a comparison with imagery, maps, and descriptions given in this paper. The position of Anomaly A, in right part of map, is indicated. Magnetometer and gravity data spread rather uniformly over the area of the Chevron license. There is also a grid of seismic lines over the whole area, but the principal density of seismic data is in the west. Very apparent is the triangular anomaly formed around the magnetic high in the eastern half of the Chevron holdings, caused evidently by a rela- tively shallow body of igneous rock, probably mainly intrusive. This anomaly appears as a central point within a distinct three-way branching of structure. The details shown within Chevron holdings are adapted from interpretations of geo- physical data by Chevron geophysicists R. F. Flege and R. M. Wright. There is insufficient information on structural control in Somalia to properly evaluate this part of the map. The differences in geology from the Kenya to the Somalia side of the border may well be caused by the relative spacing and position of data points. For example, inspection of the Landsat imagery, not all shown here, suggests the southeastward-trending fault which passes the Bur Amber-Bur Acaba ridge in Somalia as a probable extension of the Lagh Katulo fault. It is simply a little askew, as shown, and poorly located in the sector near the Kenya border. more or less coincident with the inferred struc- tural sag along the Ewaso N’Giro. A northeast and east boundary of this feature is evidently a fault. For a considerable distance, the fault lies along the Lagh Bogal lineament. About at the juncture of the Lagh Bor lineament it turns south through an area where there is no dis- tinct, matching linear. It tends to die away near the center of the Chevron holding. 3. Rocks having high seismic velocities and the mag- netic properties of basement project slightly into the license area in the Giriftu structural block, just east of the Lagh Bor lineament. 4. Shallow magnetic basement, probably igneous, lies along the Lagh Bogal and Lagh Bisika linea- ments southeast of their juncture and about 80 km southeast of Wajir. A tonal Landsat linea- ment parallel to the Wajir Bor lineament extends northeastward from this point and bisects the feature called Anomaly A. As noted by geo- physicist R. F. Flege, the magnetic and gravity anomalies which signify the presence of the igneous body are roughly triangular. Their shapes correspond with magnetic linearity in three directions about parallel to the Wajir Bor, Rudolf trough and Lamu embayment directions of structure. Conceivably, the magnetic anomaly may represent an igneous mass related to triple junction spreading. 5. The area north of this basement high, over part of Anomaly A and the entire northeast corner of the license area, has a thick section of sedi- ments. This relatively basinal feature is con- firmed by early results of seismic work. 6. A general gravity high trend crosses the area along a line just north of the Hagadera—Liboi LANDSAT—1 STUDIES APPLIED TO PETROLEUM EXPLORATION IN KENYA lineament. A magnetic high trend follows this in a general way. Basement along the crest of this feature, as compared to the flanks of the feature, appears relatively shallow. Some differences and things yet unexplained are apparent also: 1. There appears to be no distinct downstepping of the basin from the carbonate shelf northeast of Wajir toward Anomaly A. Also, the apparent depth to basement within and immediately ad- jacent to the Jurassic outcrops is at least twice what it should be as judged from the strati- graphic measurements of the surface sections reported by Thompson and Dodson (1958), Joubert (1963), and others. Why this is so is not yet clear. 2. No substantial geophysical anomaly has been found along the Habaswein—Wajir Bor lineament. The Giriftu basement block seems to end near the lineament probably through downfaulting. West of the Lagh Bor lineament, no geophysical anomaly at all seems to be associated with the Habaswein-Wajir lineament. East of the town of Wajir, gravity data fail to detect the linea- ment. Magnetic and seismic data show a local, minor feature which seems to have neither great displacement nor continuity. 3. The transverse basement high aligned parallel to the Hagadera-Liboi lineament seems to termi- nate the structural trough which extends along a part of the Ewaso N’Giro. This trough, which lies in the northwest part of the license area, tends to end against the northwest flank of the high. 4. From geophysical evidence (fig. 8), the central structural depression along the Ewaso N‘Giro is not directly along the valley of this stream as my foregoing analysis implies. Instead, it lies farther to the north and east, about at the Ewaso N’Giro—Lagh Bogal drainage divide. Possibly the present stream valley is eroded in softer ground where the duricrust was thin or along a line where faulting has weakened it. 5. Low velocity beds that seemingly would be Terti- ary do not commence just southward of the Wajir Bor lineament, as predicted. Their first appearance is approximately at Lagh Bisika, some 70 or 80 km farther south. 6. Although phenomena coinciding with many of the major imagery lineaments are seen in the geo- physical data, the features generated along a single lineament are often highly varied. A sub- 151 stantial number of the individual elements among these are elongated in direction of the linea- ment identified on airphotos. Yet from the geo- physical evidence alone, one generally would not integrate them as a structural trend. The idea that the Lamu embayment may have originated as an aulacogen, even though unproven, is partly supported by gravity data. Readings over the embayment are consistently higher than strongly nega- tive readings generally obtained over the outcropping Precambrian rocks to the west. The readings are not anomalous to what is commonly regarded as a normal crustal thickness for continents. Significant, however, is a three-way alignment seen among the structural components of the basin. This is quite apparent in figure 7 as well as in detail added by figure 8. CONCLUSIONS The results of the Landsat image study are analyzed here as they fit within an evolving program of explor- ing for oil in Kenya. They start with a review of geological knowledge initially available, progress through a comparison of Landsat image analysis with photogeology, and finally compare the results with what has been learned during the beginning and inter- mediate stages of geophysical exploration. The Landsat information assists the geological pro- gram in two ways: 1. By helping to define an oil license area and the end of the Lamu embayment relative to a re- gional geological setting, and 2. By improving the definition of anomalies and lineaments significant to interpreting structure and stratigraphy. The area provides a severe test of the use of Landsat data, because surface geological differences within the license area are few and the relief very low. Almost all the important basin geology is concealed by young, fiat-lying Pliocene and Quaternary fill. Useful ideas have sprung from the information ex- tracted, and this is counted as progress. In addition, the Landsat studies contributed evi- dence for suspecting subsidence along the Ewaso N’Giro near Malka Galla and suggested a branching arm of the basin toward the Rudolf trough. When supported by geophysical evidence, this led to filing on an option area adjoining the original Chevron license, thus extending the Chevron holdings. Several of the anomalies are defined better on the Landsat images than on the airphotos. The Lagh Bisika lineament, as a distinct fracture direction, was 152 defined only from Landsat data. On the airphotos it might be thought of simply as a sub-fracture parallel to the Lagh Bogal lineament. The Hagadera-Liboi lineament and the Lagh Katulo fault are defined more completely on Landsat images than on airphotos. Anomaly A was not covered completely in airphoto work and did not become apparent until the Landsat images were examined. On the other hand, substantial evidence in support of the Habaswein—Wajir Bor trends comes from the pattern of smaller fractures observed only on air- photos. In the Landsat imagery, this lineament-— mainly seen as a tonal band—has its best expression northeast of Wajir extending into Somalia. Its coinci- dence with the north flank of the Bur-Acaba structural ridge appears more than coincidental. I believe the lineament is an important, valid structural feature and do not understand why the geophysical data in Kenya do not show it clearly. The similarities of trends mapped by means of geo— physics with many features of the Landsat and air- photo interpretation suggests that a great number of the chosen anomalies and lineaments are structurally valid. The directions are generally alike, even when the positions differ. Many Landsat lineaments are wide bands—10 km and more—where the position of the annotation or drawn line is rather arbitrary. There is also the chance of fault-line anomalies: surface traces shifted laterally from the phenomena that cause them. The complex overall associations point to complex‘ rather than simple geology. They are not conducive to an early and complete understanding. During the Landsat image study, evidence was found of a previously unrecognized lithologic division within Pliocene deposits west of Garissa and also an internal depositional structure within Pliocene-Pleisto— cene deposits along a summit east of the Ewaso N’Giro. These proved helpful in understanding struc- tural and basinal depositional patterns and in recon- structing basin history. And significantly, they show how Landsat images can function as a separate source of information. In the Landsat interpretation, I have used a modified form of the usual techniques and skills of photogeo- logy. The Landsat studies give an overview predis- posed to finding regional anomalies, while airphotos more effectively look at local detail. The Landsat and airphoto work logically should go hand—in-hand, neither of them separately and alone, because each of them may lose the things most easily seen by the other. Landsat imagery adds multispectral data that are geo- logically useful. Most significant of all, Landsat studies FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM have the advantage of very quick interpretation over new areas and regions. Such work saves time and money and gives a substantial basis for new ideas. REFERENCES Baker, B. H., and Saggerson, E. P., 1958, Geology of the El Wak-Aus Mandula area: Nairobi, Geol. Sur- vey of Kenya, Report no. 44, 48 p. Baker, B. H,, and Wohlenberg, J., 1971, Structure and evolution of the Kenya Rift Valley: Nature, v. 229, p. 538—542. Beltrandi, M. D., and Pyre, A., 1973, Geological evolu- tion of Southwest Somali, in Blant, G., ed., Sedi- mentary basins of the African coasts: Paris, Assoc. of African Geol. Surveys, p. 159—178. Brock, B. B., 1965, The Rift Valley Craton, in The World Rift System, International Upper Mantle Project: Geol. Survey of Canada Paper 6614, p. 99—123. Caswell, P. V., 1953, Geology of the Mombasa-Kwale area: Nairobi, Geol. Survey of Kenya, Report no. 24, 69 p. Caswell, P. V., 1956, Geology of the Kilifi-Mazeras area: Nairobi, Geol. Survey of Kenya, Report no. 34, 54 p. Joubert, P., 1963, Geology of the Wajir-Bor area: Nairobi, Geol. Survey of Kenya, Report no. 57, 34 p. Matheson, F. J., 1971, Geology of the Garba Tula area: Nairobi, Geol. Survey of Kenya, Report no. 88, 30 p. National Aeronautics and Space Administration, God- dard Space Flight Center Earth Resources Tech- nology Satellite Symposium, 2d, New Carrollton, Md, 1973, Abs. —— Earth Resources Technology Satellite Sym- posium, 3d, Washington, DC, 1973, Summary of Results. Saggerson, E. P., and Baker, B. H, 1965, Post-Juras- sic erosion surfaces in Eastern Kenya and their deformation in relation to rift structure: Geol. Soc. London Quart. Jour., v. 121, p. 51—72. Saggerson, E. P., and Miller, J. M., 1957, Geology of the Takabba-Wergudud area: Nairobi, Geol. Sur- vey of Kenya, Report no. 40, 42 p. Sanders, D. L., 1959, Geology of the Mid-Galana area: Nairobi, Geol. Survey of Kenya, Report no. 46, 50 p. Thompson, A. O., 1956, Geology of the Malindi area: Nairobi, Geol. Survey of Kenya, Report no. 36, 63 p. LANDSAT—1 STUDIES APPLIED TO PETROLEUM EXPLORATION IN KENYA 153 mentary basins of the African coasts: Paris, Thompson, A. O., and Dodson, R. G., 1958, Geology Assoc. of African Geol. Surveys, p. 133—158. of the Kerkali—Melka Murri area: Nairobi, Geol. Survey of Kenya, Report no. 43, 35 p. Williams, L. A. J., 1962, Geology of the Hadu Fundi- Walters, R., and Linton, R. E., 1973, The sedimentary Isa area: Nairobi, Geol. Survey of Kenya, Report basin of coastal Kenya, in Blant, G., ed., Sedi- no. 52, 62 p. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Regional Linear Analysis as a Guide to Mineral~ Resource Exploration Using Landsat (ERTS) Data By R. A. Hodgson, Gulf Research and Development Company, Pittsburgh, Pennsylvania ABSTRACT The direct association of commercial mineral de- posits with geologic features having a linear surface expression, such as faults and fracture zones, has long been known. Landsat data, displayed in appropriate formats, comprise a remarkably effective tool for recognizing and mapping several orders of such fea- tures in considerable detail. The data also allowed the definition of complex fracture—lineament systems over significantly large regions. Regional fracture-lineament systems mapped to date show patterns that appear to be reasonably con- sistent with those predicted by theories which postu- late such fracture patterns result from stresses gen- erated by systematic and nonsystematic changes in the Earth’s shape and/ or shifts of the lithosphere with respect to the geographic pole. As sufficient detailed data become available for large regions with respect to the areal distribution and nature of the various orders of linear features, it will become possible to evaluate better present frac- ture-tectonic theories as well as to establish empirical- ly significant spatial relations between various types of mineral deposits and elements of the fracture net- work. Perhaps at that point it may become possible to develop a valid genetic theory for predicting in advance the location of environments specifically favorable for mineral deposits. INTRODUCTION The direct association of commercial mineral de- posits with several orders and types of linear struc— tural features such as faults, fracture zones, and joint systems has long been established. The fact that such structures commonly have a systematic spatial dispo- sition has been well documented for over a century. Many mineral deposits occur along or at the intersec- tion of linear structures where geologic conditions are favorable. Thus, any new technique should be investi- gated which promises to aid in the delineation of the spatial relations and structural characteristics of the various orders of linear features. The Landsat space- craft provides a new platform for observation of such features and a very effective digital—analog system for the recording, display, and analysis of data. The use of Landsat data to delineate and analyze regional linear features is, of course, quite recent. The Landsat system was not designed specifically for such a job but, as it turns out, has several features which make it particularly well suited for the purpose. The most important of these features from the standpoint of analysis are: 1. Synoptic nature of the imagery. 2. Temporal character (capability of recording the same scene every 18 days or every 9 days where both Landsat—1 and —2 are operational). 3. Very extensive areal coverage. 4. Capability for both digital and analog enhancement of images. 5. Unrestricted availability and low cost of raw data. 6. Variety of useful formats for analysis and display of data. The Landsat system operates within a well-defined digital-analog framework and, therefore, imposes a specific approach in the analysis of the data by the investigator. Figure l is a flow chart showing the various steps and options in the use of Landsat data for minerals exploration. 155 156 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM PROGRAMMING DATA OF DATA DATA RECEPTION DATA --D ACQUISITION —D TRANSMISSION —'D TRANSLATION ACQUISITION (SATELLITE) (SATELLITE) OF SIGNALS I /////////// gag/{m W LA NDSAT/ /‘ $AND FALSE §+_%GENERATION \\\ ‘— COMPUTER ///////// jMOSAIcs / OR PAPER / TAPE (CCT) ////////////// /////// //// A I ANALOG \ \ \ ENHANCEMENT USER USER PROCEDURES ¢ \SQS‘SSSS _ OBJECTIVE BJECTIVE §color additive¥§ DETERMINE DETERMINE Viewers-mosaic PROCESSING PBQQESSING DIRECT SIGHT DIRECT SIGHT IMAGE \ ////////////// IMAGE ANALYSIS IMAGE ANALYSIS ATlON §—./E ENHANCEMENT; AND AND 0 INTERPRETATION INTERPRETATION (Level 1I _, (Level II I I SYSTEMATIC AND , , NONSYSTEMATIC / SYSTEMATIC ; ANALYSIS AND ‘____IMAGE ANALYSIS ¢ANALYTIC / INTERPRETATION AND / PROCEDURES? OF LEVEL 1 DATA INTERPRETATION , (OPTIONAL) ' APPLICATION TERPRETAT TIO/N “ /'///////////; FIGURE 1.—Chart showing sequence Of Landsat data flow from acquisition to application in minerals exploration. PRIMARY FACTORS IN ANALYSIS Before undertaking the analysis and interpretation of linear geologic features from Landsat images a number of factors should be considered which in- dividually and together determine the effectiveness and validity of the results. These factors are outlined in figure 2. The hardware and software Characteristics of the Landsat system are most important in that they de- termine the ultimate resolution of any data acquired REGIONAL LINEAR ANALYSIS, GUIDE TO MINERAL EXPLORATION 1. CHARACTERISTICS OF DATA ACQUISITION SYSTEM 2. FORMATS OF DATA PRESENTATION 3. METHODS OF DATA ANALYSIS (METHODS MUST BE APPROPRIATE TO DATA) 4. METHODS OF DATA INTERPRETATION LEVEL l—DIRECT SIGHT DELINEATION FROM PHOTOGRAPH LEVEL ll—APPLICATION OF THEORETICAL CONCEPTS 5. APPLICATION (EXPLORATION OBJECTIVES) EXPERIENCE THEORY FIGURE 2.—Primary factors considered in use of Landsat data for analysis and interpretation of regional linear geologic features. as well as the various data formats available for analysis and interpretation. Analytic data handling procedures differ with digital and analog formats even though the objectives may be entirely similar. More detailed information resides in the computer compatible tape (CCT), for example, than in the photographic images for the same scene. With respect to regional linear analysis more informa- tion often can be gained from mosaics composed of the images of several scenes than from a single scene even thOugh the scale and definition of the two photo- graphic representations are the same. It is important that the methods used to analyze Landsat data be appropriate to the nature of the data. This implies a working knowledge of the data systems as well as a knowledge of the geological parameters involved. For example, because of the synoptic nature of the imagery, the areal pattern of linear features can be established more effectively by direct inspection of the image than by a statistical treatment of data from a series of observations at points over the image. The various options on digital enhancement opera- tions should be considered carefully as some are more useful than others for enhancement of linear features. Inspection of images shows that the linear features are represented by a series of tonal or morphological events which can be visually integrated into a single large-order feature. Mathematical analytic procedures such as linear stretching and edge enhancing of the data tend to magnify or emphasize the changes in reflectance which identify linear features. Procedures such as clustering may help establish significant but not obvious areal differences in reflectance across a linear feature suggesting it may mark a fault zone or a geologic boundary, etc. There are two distinct levels of interpretation in- volved in using the imagery to investigate regional 157 linear features. The first level involves the identifica- tion and delineation of the various orders of linear features from the raw data. This may be accomplished by direct sight interpretation of photographic images or may involve a complex technology of data enhance- ment and viewing techniques. The end result is usually a map showing directly the structural character and spatial relations of the features sought. At the second level of interpretation, theoretical concepts can be applied to the “observed” data and various established analytic techniques used to study some portion of the data content of specific interest. For example, one might consider that areal changes in angles of intersection of a certain class of linears is important or that a certain spacing or azimuth is critical. The results of such systematic analyses can be presented using contour maps, graphs, etc. Where analysis and interpretation are being done for a specific exploration purpose, the exploration ob- jectives must be kept in mind and the methods of investigation designed to abstract as much as possible of the desired information from the imagery. How the results are applied to the problem at hand depends, of course, on the experience and theoretical views of the user. LANDSAT DATA ACQUISITION SYSTEM Of particular interest to the interpreter are the main elements of the Landsat data acquisition system, inasmuch as the primary limitations on the use of Landsat data for regional linear analysis appear to reside primarily in the Landsat system itself and have to do with parameters such as spatial resolution, spec- tral bands recorded, the Earth’s atmosphere, and data handling formats. Landsat is a remote-sensing system that operates in limited and specific wavebands. It has a lower limit of spatial resolution which is significant from the standpoint of the interpretation of linear features. The manner in which the system records and transmits data determines to a great extent the digital and analog formats in which the data can be presented for analysis. The basis for these formats should be understood at least in outline so that unwanted bias is not introduced into the interpretation of linear fea- tures from the imagery. The Landsat—1 spacecraft, from which the imagery discussed in this paper was derived, has two image acquisition subsystems. These are the Return Beam Vidicon, or RBV, and the Multispectral Scanner, or MSS. All the data used here for analysis are from the M88. Figure 3 shows the main features of the M88 subsystem. 158 SCANNER 6 Detectors Scan mirror per band: (OsciIlates 24 Total nominaIIyt2.89°) Field of View = 11.560 Spacecraft Travel FIGURE 3.—Main features of the Landsat—1 MSS subsystem. The M88 is a four-band scanner which operates in the solar-reflected spectral region from 0.5 to 1.1 “m. The four spectral bands are 0.5 to 0.6 ,tm, 0.6 to DATA DS FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 0.7 ,Lm, 0.7 to 0.8 pm, and 0.8 to 1.1 ‘ILm. There are six detectors for each of the four bands, and the spectral bands do not overlap. Scanning is accom- plished by means of an oscillating fiat mirror placed between the ground scene and a double reflector telescopic optical chain. The video outputs from each of the detectors are sampled, commutated, and multiplexed into a modu- lated stream so that the data can be encoded and transmitted to a ground receiving station. At the re- ceiving station, the raw data are compiled on video tapes which are transmitted to the NASA Data Pro- cessing Facility at the Goddard Space Flight Center. At this point the data are converted into computer compatible tapes and 70-mm film images through the Initial Image Generating Subsystem, shown schemati- cally in figure 4. These two products are the basic material which is used in one form or another for the mapping, analysis, and interpretation of geologic fea- tures. A single Landsat scene comprises about 34,000 km 3 on the ground and the ground resolution of the system is about 80 m from the operational altitude of 918 km. As shown schematically in figure 5, any given -—> CONTROL SYSTEMS I SUBSYSTEM DIGITAL A INITIAL IMAGE GENERATING SUBSYSTEM T MSS VIDEO ----—-—-—.’ DATA DATA I'GS SYSTEMS _" CONTROL TA PE FIGURE 4.—Diagram showing configuration of the Initial Image Generating Subsystem. REGIONAL LINEAR ANALYSIS, GUIDE TO MINERAL EXPLORATION image is composed of parallel scan lines each contain- ing a large number of video data points. A film image of a scene can be generated from the M88 video data where the radiance levels recorded on the tape can be projected on 70-mm film at successive points along a scan line to produce pixels, or picture elements, in specified intensities or levels of gray. The resulting image looks very much like an aerial photograph from which, however, it differs significantly in several ob- vious respects. DIRECTION 2340 OF SCAN LINES SCAN SPACECRAFT LINES FIGURE 5.—Diagram showing the main elements comprising a Landsat image. 159 DEFINITIONS The term “linear" as used by photogeologic inter— preters has acquired some specific connotations with respect to length which are not implied where the term is used in this discussion. The term is used here in the general descriptive sense as defined by Webster; that is, the term pertains to a line or lines, consisting of lines, in a straight direction, resembling a line, narrow, long, and uniform in width. In general, the regional linear features of the Earth’s surface as seen on Landsat imagery can best be described by the technical term “lineament” as originally defined by William Herbert Hobbs (1904, 1911). Lineaments are generally found to be composite geomorphic and structural features when examined in detail and as shown schematically in figure 6. The term “fracture zone” is used here to describe linear structural features (other than known faults) which have a direct expression on the Landsat imagery and can be reasonably inferred to be a direct expres— sion of fracturing. The term “planetary” is used to describe those re- gional or super-regional lineaments which are of a length, width, and linearity which would indicate they extend to great depths in the lithosphere. They are azimuthally distributed in systems that reasonably can be inferred to reflect their generation by the body forces of the Earth. In addition, they show morpho- logic similarities to equavalent features on the other SCHEMATIC DIAGRAM INDICATING THE COMPOSITE EXPRESSION OF A LINEAMENT after HOBBS (I9II) SCHEMATIC DIAGRAM INDICATING THE COMPOSITE NATURE OF DISLOCATION LINES after HOBBS (l904) FIGURE 6.—Schematic diagrams showing the composite structural and physiographic aspects of lineaments. 160 FIRST ANNUAL PECORA terrestrial planets and so may be features which are common to all such planets and which are generated initially by similar forces. BASIC TYPES OF GEOLOGIC STRUCTURES The study of the nature of geologic structures sug- gests that there are two basic types which can be classed as continuous or discontinuous and distinguish- ed on the basis of their distribution in time and space. The types of structures and the main characteristics which distinguish them are outlined in figure 7. The MEMORIAL SYMPOSIUM major linear features viewed on Landsat imagery clearly fall into the category of continuous structures. This is of great importance to minerals exploration as the universal and remarkably uniform distribution of these features as well as their persistence across all other structures of whatever age suggests some level of constant or intermittent structural activity along their length and at their intersections. As a re— sult we may expect to find structures favorable for mineralization distributed more or less uniformly around the Earth in rocks of all types and ages. BASIC TYPES OF STRUCTURES CONTINUOUS DISCONTINUOUS SYSTEMATIC JOINTS MASTER JOINTS JOINT ZONES LINEAMENTS BELTS OF LINEAMENTS FAULTS FOLDS CLEAVAGE VOLCANIC CRYPTOVOLCANIC IMPACT MAIN CHARACTERISTICS MAIN CHARACTERISTICS 1. UBIQUITOUS OCCURRENCE IN ROCKS OF ALL TYPES AND AGES—NON- REVERSIBLE 2. MAINTAIN UNIFORM AREAL PATTERNS OVER VAST AREAS 3. UNIFORM SPACING MAINTAINED BE- TWEEN ELEMENTS OF EACH ORDER 4. REGIONAL AND PLANETARY ORDERS CROSS ALL OTHER STRUCTURAL FEATURES WITHOUT DEVIATION OR OFFSET 5. ALL SYSTEMATIC LINEAR FEATURES CROSS EACH OTHER WITHOUT DIS’ CERNIBLE OFFSET 6. LARGER ORDERS—COMPLEX STRUC- TURES ‘I. ISOLATED STRUCTURAL EVENTS IN TIME AND SPACE 2. NONUNIFORM AREAL DISTRIBUTION 3. CLEAVAGE AND FLEXURES—NON- REVERSIBLE—EPISODIC 4. FAULTS—REVERSIBLE (?)—EPISODIC S. VOLCANIC—EPISODIC 6. CRYPTOVOLCANIC AND IMPACT—— UNIQUE—NONREPEATED FIGURE 7.—Basic types of geologic structures. ORDERS OF SYSTEMATIC LINEAR FEATURES Linear features seen at the Earth’s surface occur in several distinct orders and, individually, range from a few metres to hundreds of kilometres in length. Within each order, they tend to occur in well-defined sets or systems which extend quite uniformly over very large areas. The main orders of systematic linear features are listed in figure 8. The two largest classes of linear features are the subject of discussion here, but it is useful to consider that they may be end members in a continuous or intermittent series of genetically related structures at the other end of which is the systematic joint. A field example of regional linear feature is the Florence Pass lineament (Hodgson, 1965). This fea- ture is some 100 km in length and crosses the center of the Big Horn Mountain Range in Wyoming. The lineament is marked along its length by a series of deep valleys in Precambrian terrane where it crosses the highest part of the mountain. It is expressed as a monoclinal fiexure in the sedimentary rocks on the gentle west flank of the range. These characteristics are shown in figure 9. Some normal faulting is present along the line of the lineament at the very crest of the range. A lineament of this magnitude is usually a com- plex structure of some width as was recognized long ago by Hobbs. Field examples of smaller orders of linear structures are shown in figure 10. REGIONAL LINEAR ANALYSIS, GUIDE TO MINERAL EXPLORATION ORDERS OF SYSTEMATIC LINEAR FEATURES DISTINCTION MAY BE BASED ON LINEAR 1. PLANETARY EXTENT AS WELL AS CHARACTER 2. REGIONAL 3. LOCAL FRACTURE ZONES MASTER JOINTS SYSTEMATIC JOINTS FIGURE 8.—-Orders of systematic linear features. ‘ A 25m A (in; hip,“ \ 1 ‘A FIGURE 9a.—Aerial view west along Florence Pass lineament showing physiographic expression of the structure. (Sketch after photo by R. A. Heimlich in Hodgson, 1965). See figure 9b, p. 162. Figure 11 shows graphically the order of linear features which we are likely to see on Landsat images considering the ground resolution of the system as 80 m. As indicated in the figures, we can see only the widest of fracture zones directly, and then only under favorable lighting and environmental conditions. It is possible, however, to infer the presence and even the spatial relations of smaller orders of linear features indirectly. This is because of the distinctive reflectance signatures of vegetation and soils above fractures 161 where the environmental conditions are slightly differ— ent from those in the immediately adjacent areas. Topographic features which develop preferentially along such linear fractures also reveal their presence. Obviously, there is no difficulty in seeing regional linear structures in Landsat imagery. INTERPRETATION Regional lineaments are not identified by a unique spectral signature but rather by a series of signatures. It is the difference in spectral signatures between any point or area along or within the lineament and that of areas immediately adjacent that makes the linea- ment visible. The prominence of a given lineament may vary according to the spectral band in which it is viewed, but the larger lineaments are usually visible in all bands primarily because they are identified by differ- ences rather than uniformity of spectral signatures and so may be marked by a wide spectral band of refiectances. Digital data enhancement procedures, where used, should be designed to emphasize local spectral differ- ences. This can be done by a number of mathematical operations such as linear stretching, edge enhancing, or by developing signatures and themes for areas along a lineament which, by inspection, appear favor- , able for defining the total extent of the feature. Ana- log enhancement procedures involve the use of color additive viewers, changing of scales and total area viewed, and the selective use of printing screens in making black and white photographic copy. The screen technique is particularly useful in that detail of the image is eliminated or de—emphasized without sig- nificant loss in distinction among gray levels. The end result is the enhancement of the larger linear features by increasing the apparent contrast between them and their surroundings. The order, or size, of linear features recognized in direct viewing of images and mosaics is a function of the maximum ground resolution of the imagery which, of course, is the limiting factor in the ultimate amount of detail which can be seen and so determines the smallest order of linear structure which can be mapped. The two parameters of the imagery which can be varied within wide limits are scale and total area of the Earth’s surface presented in a single view. Features not at all apparent, or seemingly poorly de- fined in a single scene, may show clearly on a mosaic at the same scale and in the same spectral band but depicting a much larger area. Decreasing scale tends to enhance the appearance of increasingly larger linear features as scale is reduced. 162 I08°00' FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 45°00' l I r NORTHERN BIGHORN MOUNTAINS WYOMING Structure contours drawn at 500' interval on top of Cloverly form. ‘ 3, Boundary of Precambrian Outcrop 44°oo' 1 08°00‘ 107°oo’ FIGURE 9b.——-Structure contour map of the northern Big Horn Mountains, Wyoming, showing location and structural geometries of the Tensleep and Florence Pass lineaments. (After Hodgson, 1965.) These observations and principles can be illustrated by showing examples of regional and planetary linea- ments. Figure 12 (p. XVII) shows a color composite image of a Landsat scene in the Great Basin and includes the area around Twin Falls, Idaho, and the region to the south. Looking at the image itself one can see a nar- row but very distinct linear feature extending diagonal- ly across the top of the image from left to right just to the south of the predominantly red area. From the changes in the colors of the surficial material and the fabric of the topography, it is evident that the linea- ment crosses a variety of physiographic and geologic features. It extends unbroken for a distance of nearly REGIONAL LINEAR ANALYSIS, GUIDE TO MINERAL EXPLORATION 163 K_,5mI___/‘ MASTER JOINTS Joint zone deforming graywakes of the 0 Lower Carboniferous strata dips about 25 - a quarry NE of Olomouc.Czechoslovakia. (_ Photo by Miroslav Pl iéka) FRACTURE ZONES FIGURE 10.—Smaller orders of linear geologic structures. 164 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM RESOLUTION OF LANDSAT IMAGERY & INDIRECT IDENTIFICATION OF LINEAR. FEATURES (INFERENCE) 80M LAN DSAT LANDSAT SCENES MOSAICS I I V/////// I I SYSTEMATIC FRACTURE REGIONAL PLANETARY JOINTS ZONES MASTER JOINTS RESOLUTION OF LANDSAT IMAGERY & DIRECT IDENTIFICATION OF LINEAR FEATURES (OBSERVATION) 80M IM 3M I I I I I I I’////////fl SYSTEMATIC T FRACTURE JOINTS ZONES MASTER JOINTS IOOM LANDSAT SCENES LANDSAT MOSAICS REGIONAL PLANETARY FIGURE II.—Smallest order of linear geologic structure seen on Landsat imagery by direct observation or by inference. a hundred miles in this scene To determine how much farther it may extend in either direction would require enlarging the area viewed. The feature itself appears to be a wide fracture zone and may well include a number of normal faults. It maintains its remarkably straight course through mountainous terrain as well as flat lowlands and so must be essentially vertical in nature and extend to great depth. Figure 13 (p. XVIII) shows the same lineament mapped to scale on the “Tectonic Map of the United States” (USGS and AAPG, 1961). The feature follows the trend direction of mapped faults and, in places, is coincident with them. It appears from the lineament that the structural trend of the west limb of the Snake River downwarp continues far to the southeast of Twin Falls. The fact that the Landsat feature shows so plainly is a good indication of recent tectonic move- ments along it. This also is information which can be readily checked in the field. By increasing the size of the area viewed without losing essential detail, it becomes possible to map similar features to their full extent and to delineate larger features which cannot be defined initially on a single image. Figure 14 shows the western United States as depicted by the band 7 US Department of Agriculture Landsat mosaic at the scale of 125,000,000. Less than half the visible lineaments have been mapped here, but enough are marked to illustrate the remarkable parallelism among sets of lineaments over very great distances. The light lines show individual linear features which appear to belong to a "given set but are generally more irregular in their course than their neighbors. Note also the remarkable uniformity of spacing in places between members of the same REGIONAL LINEAR ANALYSIS, GUIDE TO MINERAL EXPLORATION 165 FIGURE 14.—Sets of major Iineaments identified on the 1974 band 7 USDA Landsat mosaic of the United States at scale 125,000,000. 166 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 15.—First-order lineaments of the United States taken from the band 3 Mosaic of the United States as published in the July 1974 issue of Ceotimes. set and the orthogonal spatial relations between at least two of the sets. Of great tectonic significance is the apparent lack of offset of one great lineament by another where they cross each other and the lack of significant deviation where they cross major fold belts. The broad band delineates a wide zone of several closely spaced parallel lineaments which, taken to- gether, may comprise the largest order of lineaments. One can continue to reduce scale and increase the area viewed to define larger and larger linear fea- tures, if they exist. As noted previously, the printing screen can be used to enhance the contrast of the larger features. Figure 15 shows the results of mapping lineaments which have a transcontinental extent (or at least a transmosaic extent) on the band 5 mosaic of the United States published as the center fold in July 1974 issue of Geotimes. (This is an example where the dot screen used in making the plate for printing the mosaic resulted in an enhancement of the largest linear features.) Several of the east-west lineaments show an excellent correspondence in geo- graphic position with the great east-west fracture zones of the Pacific such as the Murray, Pioneer, and Mendocino suggesting they are in fact single con- tinuous structures of global significance. In detail, most of the major features recognized are composed of a connected series of structural phe— nomena along their length such as fault zones, frac- ture zones, linear folds and fold belts, and edges of regional features such as downwarps and upwarps, all constituting narrow, linear zones of deformation when considered at the viewing scale. A few linea- ments are marked by what appears to be a single well-defined suture along their length. There is a gen- erally systematic distribution for all orders of linear features. As a generalization, one can say they are distributed azimuthally in sets following approximately NE, NW, E—W, and N—S directions over very large regions. At this point one can take the data derived from this first level interpretation of the imagery and use it directly to pinpoint areas which, on the basis of previous experience, promise to be prospective. One may, however, wish to proceed to the second level of interpretation in which the observed data are con— REGIONAL LINEAR ANALYSIS, GUIDE TO MINERAL EXPLORATION sidered, not only in the context of previous experi- ence, but also within the context of one or more hypotheses concerning the origin and significance of linear structural features and the manner in which they may exert control on the location and genesis of ore deposits. Thus, at the first level, the interpreter must decide on the nature and extent of the features which he has detected in the imagery and present the data in some form of direct use such as a map. At the second level, he may apply systematic analysis to the data and interpret this within the context of some hypothesis. As a result, the theoretical views and ex- perience of the interpreter have a strong influence on the nature of the second level interpretation. The observed systematic areal pattern of the major lineaments is reasonably compatible with that pre- dicted by theories which postulate large-scale sys- tematic fracturing in the lithosphere as a result of global deformation caused by systematic and non- systematic changes in the rotation of the Earth or the position of its axis of rotation relative to the litho- sphere. The several theories that postulate crustal fracture patterns similar to those we observe on the Landsat imagery have been reviewed in detail in the literature by Sonder (1947), Vening-Meinesz (1947), and 167 Belousov (1962). As a result of the investigations out- lined here, it can be concluded that the tectonic forces which first produced the great Earth lineaments are global in nature and have to do with tidal forces and changes in the Earth’s shape caused by changes in the rate of rotation of the Earth. These forces can be considered to act today at essentially the same level they did in the beginning, and the major lineaments have existed since then and have maintained a con- tinuous low level of tectonic activity. This hypothesis, if true, is significant in that it states that linear features have always existed in the Earth’s crust as we see them today and, therefore, could have acted as channelways for mineralization at any time in geologic history. If such uniform structural conditions for mineraliza- tion persisted through geologic time, has there been a uniform distribution of mineralization, or has there been any preferential mineralization along lineaments having certain azimuths? Are there ore belts along which we might expect certain types of minerals to be more plentiful than others? These possibilities have been considered, of course, and figure 16 shows a map by Nikolai Thamm in which he has marked the Silver Line of Spurr, noting incidentally that it fol- lows closely a great circle. Thamm has also noted that FIGURE 16.—-Spurr's Silver Line and the platinum—nickel line of Thamm. (After Thamm, 1976.) 168 V FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE ‘l7.—Zonal distribution of platinum—nickel deposits along a Great Circle in eastern Africa. Note correspondence in azimuth with fracture zones and ridges of the Indian Ocean. (After Thamm, 1976.) there appears to be a preference for the occurrence of platinum and nickel along a similar line or belt through Africa and the Middle East, Middle Europe and Siberia. This too follows a great circle. Figure 17 shows the southern portion of this zone in Africa where Thamm has noted a remarkable coincidence in trend and extent between the ore belt and the great fractures traversing the Indian Ocean. Thus, it seems that great ore channels may exist and follow the azimuths of great Iineament belts. Within these belts we may find that actual mineral deposits are prefer— entially associated with a particular class or order of linear feature as suggested diagrammatically in figure 18. They may occur preferentially along a given set of Iineaments within an ore belt, at or near Iineament intersections, et cetera. As a working hypothesis, we might start with Thamm’s statement in his last paper (1976) that “All such zones, lines or belts seem to originate at con— siderable depths and more often than not, bear no relation to any surface structures. Just as with joint- ing, the existence of mineral-bearing zones is a permanent phenomenon, i.e., it re-appears with its identical mineralization after long periods of time and and after structural disturbances in an independent manner.” REGIONAL LINEAR ANALYSIS, GUIDE TO MINERAL EXPLORATION PLANETARY FRACTURE GENETICALLY RELATED SECONDARY FRACTURES 169 REGIONAL FRACTURE FRACTURE ZONES JOINTS FIGURE 18.—Sectional diagram showing various orders of fractures which may be genetically related and which may control the location of mineral deposits. CONCLUSIONS In conclusion one can say that Landsat imagery offers a new and powerful tool for the delineation in detail of regional linear structures. It serves as a guide to pinpoint areas of direct interest to minerals ex— ploration which can then be further investigated with conventional means. The synoptic aspect of Landsat imagery, coupled with the amount of detail shown, largely eliminates the need for the statistical approach in mapping re- gional linear features. Their identification is direct and, in many examples, unequivocal. Systematic analytic methods should be directed primarily toward enhancing or abstracting some particular aspect of interest from the total observed data content. The remarkably straight course of regional linea- ments over tens to hundreds of kilometres shows they are vertical and must extend far into the lithosphere if not entirely through it in some cases. They may serve as channelways for mineral producing solutions from the greatest depths. Persistence of regional linear structures across all other structures suggests some level of continuous structural activity along the lineaments. The possibility of periodic increases in the level of such activity must be considered. It may be, however, that such activity, while pro- moting the upward movement of mineralizing solu- tions from great depths, may inhibit the formation of mineral deposits along the lineament itself near the surface. More favorable environments might be found along associated secondary orders of linear structures as suggested by Wertz (1976) and as diagrammed in figure 18. We must consider that geologic features occurring along the same lineament, although separated by tens or hundreds of kilometres may, in fact, be genetically related and time equivalent. Mineralization could occur in connection with the lineaments at any time in geologic history, at succes- sive intervals along the same lines or belt and, involve the same mineral suites each time. Detailed information on the location and nature of existing mineral deposits over large regions is needed to test the various existing hypotheses of the regional associations of mineral deposits and regional linea- ments. Once such associations are established, it will increase our ability to extrapolate favorable condi- tions for mineralization into new areas and so de- crease the costs of regional exploration by making the exploration programs more efficient. SELECTED REFERENCES Belousov, V. V., 1962, Basic problems in geotectonics (Eng. ed., J. C. Maxwell, ed): New York, McGraw-Hill, 816 p, 170 Blanchet, P. M., 1957, Development of fracture analy- sis as exploration method: Assoc. Am. Petroleum Geologists Bull., v. 41, p. 1748—1759. Gay, 8. P., Jr., 1973, Pervasive orthogonal fracturing in Earth’s continental crust: Am. Stereo Map Co. Tech. Pub. no. 2, 121 p. Haman, P. J., 1961, Lineament analysis on aerial photographs exemplified in the North Sturgeon Lake area, Alberta: West Canadian Research Pub., sec. 2, no. 1, 20 p. 1975, A lineament analysis of the United States: West Canadian Research Pub. Geology and Re— lated Sciences, Series 4, no. 1, 27 p. Hobbs, W. H., 1904, Lineaments of the Atlantic border regions: Geol. Soc. America Bull., v. 15, p. 483— 506. 1905, The correlation of'fracture systems and the evidence of planetary dislocations within the Earth’s crust: Wis. Acad. Sci., Arts and Letters‘ Trans, v. 15, p. 15—29. 1911, Repeating patterns in the relief and in the structure of the land: Geol. Soc. America Bull., v. 22, p. 123—176. Hodgson, R. A., 1961a, Regional study of jointing in Comb Ridge-Navajo Mountain area, Arizona and Utah: Assoc. Am. Petroleum Geologists Bull., v. 45, p. 1—38. 1961b, Reconnaissance of jointing in Bright Angel area, Grand Canyon, Arizona: Assoc. Am. Petroleum Geologists Bull., v. 45, p. 95—97. 1965, Genetic and geometric relations between structures in basement and overlying sedimentary rocks, with examples from Colorado Plateau and Wyoming: Assoc. Am. Petroleum Geologists Bull., v. 49, p. 935—949. Isachsen, Y. W., Fakundiny, R. H., and Forster, S. W., 1973, Evaluation of ERTS~1 imagery for geo- logical sensing over the diverse geological ter- ranes of New York State, in NASA Goddard Space Flight Center Symposium on Significant Results Obtained from Earth Resources Techno- logy Satellite—l, 2d, New Carrollton, Md, Mar. 1973, Tech. Presentations, v. 1, sec. A, p. 223— 230. Kjerulf, Theodor, 1880, Die Geologie des sudlichen and mittleren Norwegen: Bonn, Gurlt, 350 p. Krebs, Wolfgang, 1975, Formation of southwest Pa- cific Island arc-trench and mountain systems: Assoc. Am. Petroleum Geologists Bull., v. 59, p. 1639-1666. Kutina, Jan, 1975, Tectonic development and metal- logeny of Madagascar with reference to the FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM fracture patterns of the Indian Ocean: Geol. Soc. America Bull., v. 86, p. 582—592. Lillestrand, R. L., and Hoyt, R. P., 1974, The design of advanced digital image processing systems: Photogrammetric Eng, v. 40, p. 1201—1218. Mollard, J. D., 1957, Aerial mosaics reveal fracture patterns on surface materials in southern Saskat- chewan and Manitoba: Oil in Canada, v. 9, p. 26—50. Pariiskii, N. N., 1960, Earth tides and the Earth’s inner structure: Akad. Nauk SSSR Vestnik, no. 6, p. 61—69. Pilger, A., 1976, The importance of lineaments in the tectonic evolution of the Earth’s crust and in the occurrence of ore deposits in Middle Europe, in Internat. Conf. on the New Basement Tectonics, lst, Salt Lake City 1975, Proc.: Utah Geol. Assoc. Pub. no. 5. Polyakov, M. M., and Trukhalev, A. I., 1975, The Popigay volcanotectonic ring structure (Eng. ed): Internat. Geology Rev., v. 17, p. 1027—1034. Prucha, J. J., Graham, J. A., and Nickelsen, R. P., 1965, Basement controlled deformation in Wyoming province of Rocky Mountains foreland: Assoc. Am. Petroleum Geologists Bull., v. 49, p. 966—992. Sonder, R. A., 1947, Discussion of shear patterns of the Earth’s crust by F. A. Vening-Meinesz: Am. Geophys. Union Trans, v. 28, p. 939—945. Spencer, E. W., 1959, Geologic evolution of the Bear— tooth Mountains, Montana and Wyoming, pt. 2, Fracture patterns: Geol. Soc. America Bull., v. 70, p. 467—508. Stovas, M. V., 1957, The irregularity of the Earth’s rotation as a planetary geomorphological and geotectonic factor: Geol. Jour. of the Ukrainian S.S.R., v. 3, p. 58-69. Thamm, Nikolai, 1969, Great circles—The leading lines for jointing and mineralization in the upper Earth‘s crust: Geol. Rundschau, v. 58, p. 667—696. 1976, The distribution of oil and natural gas deposits in the Earth’s crust in relation to the mineralization with Ni, Pt, and Cr along Great Circles (abs): Internat. Conf. on the New Base— ment Tectonics, lst, Salt Lake City 1975, Proc.: Utah Geol. Assoc. Pub. no. 5, p. 617. Geological Survey and American Association of Petroleum Geologists, 1961, Tectonic map of the United States, exclusive of Alaska and Hawaii: US. Geol. Survey, 2 sheets, scale 1:2,500,000 (1962). US. REGIONAL LINEAR ANALYSIS, GUIDE TO MINERAL EXPLORATION Vening-Meinesz, F. A., 1947, Shear patterns of the Earth's crust: Am. Geophys. Union Trans, v. 28, p. 1—61. Wertz, J. B., 1976, Detection and significance of linea- ments and lineament intersections in parts of the North Cordillera, in Internat. Conf. on the New Basement Tectonics, lst, Salt Lake City 1975, Proc.: Utah Geol. Assoc. Pub. no. 5, p. 42—53. Wise, D. U., 1958, The relationship of Precambrian and Laramide structures in the southern Bear- 171 tooth Mountains, Wyoming: Billings Geol. Soc, 9th Am. Field Conf., Guidebook, p. 24—30. 1964, Microjointing in basement, middle Rocky Mountains of Montana and Wyoming: Geol. Soc. America Bull, v. 75, p. 287—306. 1976, Sub-continental sized fracture systems etched into the topography of New England, in Internat. Conf. on the New Basement Tectonics, lst, Salt Lake City 1975, Proc.: Utah Geol. Assoc. Pub. no. 5. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA The Application of Remote Sensing Technology to Assess the Effects of and Monitor Changes in Coal Mining in Eastern Tennessee By A. E. Coker, A. L. Higer, and R. L. Rogers, US. Geological Survey, Tampa, Florida; Miami, Florida; Roswell, New Mexico ABSTRACT In response to significant increases in coal mining activity expected in the Appalachians, an area with limited available data and rugged terrain, remote- sensing techniques are used to monitor and assess coal mining changes for investigating the effect of mining on water quality, sedimentation, and stream Flow in this region. Digitally rectified and rescanned Landsat data were integrated with existing maps at a scale of 124,000. The images, acquired between February 1973 and March 1975, were used to analyze mining activities during this period. The processed data are also used to delineate land-water cover type including vegetation species and agriculture and water areas. These data are useful for updating maps of strip mining activity, updating assessments of coal reserves, and surface mine detection and monitoring. The mapping techniques utilized the Bendix MDAS (Multispectral Data Analysis System) system. Thermal imagery data are used to delineate ground—water out- flow, acid mine drainage, ponding on strip—mining benches, storm runoff, and surface-water flow. These remote-sensing data provide information to aid devel- opment of a digital and pictorial data base. 173 PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Exploration by Petroleum Independents Using Imagery and Photos from EROS and Manned Space Surveys By R. W. Worthing, Consulting Geologist, The Reserve Petroleum Company, Oklahoma City, Oklahoma My early evaluation of the Landsat program was that it would be a high»cost throughput for the inde- pendent. I was wrong. The U.S. Geological Survey and Goddard Space Flight Center have made it available for the “poor-boy” oil company. Nevertheless, problems exist. For example, the lack of ability to extract information from this wonderful aid to exploration results in a “hit and miss” approach, due to lack of training of independent geologists and engineers. This will soon disappear as user education increases. I hope soon that small company manage- ment or funders quit saying that “this erudite program is not for us.” This paper is concerned largely with lineaments as they relate to brecciation and improved carbonate petroliferous reservoirs, such as chalk, oolitic lime, dolomite, and clay-silica limes. The existence of these lineaments, containing hairline fractures, joints, or megashears, increases the rate of oil production from reservoirs of these types that account for more oil, over the world, than the sandstones. The Landsat program and related underfiights have been directly responsible for renewed interest in shear patterns on the Earth’s surface because they have revealed anomalies that had been forgotten or per- haps had not been previously revealed. The extent of such shear patterns as seen on Landsat data has been noted frequently. In many areas, matching sets or pairs of shears are present at approximately right angles. These sets of shears at right angles are often recognized although one set may be more obscure than the other. What is less well known in areas, such as Oklahoma and the Illinois and Michigan basins, is that there are sometimes secondary and tertiary rectilinear joint sets, which are only clear locally and which are not widely recognized or accepted, although they are sometimes described as joint patterns. What is far more important to this presentation is the fact that new spectral bands of surveying have picked up secondary and otherwise very obscure joint sets, even sets that have been missed in previous geological data gathering. Let us examine the newly recorded evidence of these phenomena which may be discerned at least faintly across the major portion of Oklahoma’s Anadarko basin (Landsat image 1347—16462).1 To me, these are indications of previously undetected re- gional tectonic directions. The well-known primary fracture sets in Oklahoma are N. 70°W. and N. 14°E. It is also significant that the newly detected linea- ments reveal a graben 5 miles wide and approximately 35 miles long not previously known and possibly of considerable importance in western Oklahoma. Only the previously known Seneca graben in northeastern Oklahoma approaches this in size. Needless to say, I intend to pay close attention to “Landsat” lineaments in the future and I believe their study will become a major chapter in the basic structural geology texts of the future. In a different application, impressive work has been coming recently from West Coast studies of faulting and (or) earthquakes. The Landsat color print of San Francisco Bay (image 1075—18173) shows parts of the Calaveras, Hayward, and San Andreas (the Grand- daddy) faults. On an image produced by combining bands 5 and 7 in black and white and enlarging the 1Copies of referenced images may be obtained from the EROS Data Center, Sioux Falls, South Dakota 57198. 175 176 image to approximately 1:125,000 scale, shear fea- tures are shown. I am told that Landsat has recorded many lineaments along which strain and (or) faulting has occurred in the California mobile belts. These are of considerable importance as a contribution to new earthquake knowledge. Incidentally, the association of this tectonic belt to second order anticlines containing very valuable oilfields has been pointed out in the literature by Moody and Hill (1956) and others. Studies applying this type of tectonic analysis to the Michigan basin suggest that the Scipio-Albion field is a “baby” San Andreas structure. Scipio was dis- covered with the help of a crystal ball gazer. If this embarrasses the geologists, think of the evaluation engineers who said the wells would not pay—only to watch 140,000,000 bbl flow out between 1959 and 1968. The decline curve is still straight and very flat. Some geologists agree that the Scipio—Albion field lies over a basement shear zone and more agree that the Trenton Limestone reservoir is not broken by faulting. A possibility of rock flow associated with strain or a strain lineament has been noted. The possi- bility of post—glacial movement of perhaps 1 mile does coincide with a lineament detected on Landsat imagery. The economic importance attached to linea- ment tectonics in the Michigan basin is impressive. Every oilfield in the Michigan basin with an accumula- tion of 10,000,000 bbl of crude oil (actually $100,000,- 000 of products at present prices) can be closely associated with such tectonic patterns. Landsat images of the Saginaw Bay area (image 1320—15525) show three important geologic controls which mark the Michigan Thumb fault running from the Bay City area and controlling the southeast side of this fresh-water basin. On these images, the Arenac shelf can be seen protruding into Saginaw Bay as a result of movement along the Bay Rifle River linea- ment. The Arenac shelf lineaments come to the sur- face and are known to cause fracturing in the Bayport Limestone of Mississippian age; far more importantly, these lineaments have been associated with produc- tivity in Arenac County amounting to as much as 1,600,000 bbl per 40-acre well and recoveries in the Deep River Pool are at least 40,000 bbl per acre. These are extremely high productivities for the US. oil industry. Cores from reservoirs developed in fractured shear zones are typical of secondary dolomitization. The sharp black areas are high—permeability through-put channels that make these reservoirs excellent, and there is a coincidence of these zones with linear fea- tures expressed on Landsat data. FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM The Nisson anticline in the Williston basin is sim- ilar to the Scipio feature and has a string of pools with an expected production of 500,000,000 bbl. I think it is very important that there are two or three features that appear to be of similar mechanical origin in the Illinois and Michigan basins, and they are un- tested. In an entirely different mineralized situation, let’s look at the southern Illinois and Kentucky fiuorspar districts which produce approximately 150,000 tons of fiuorspar per year. The economic importance of this mineral may be measured by prices of $109 per ton and up for calcium fluoride (wet cake) and indica- tions that prices may rise in a tight market. It now appears that a half dozen mining companies are active in the area and producing fluoride valued at up to $15,000,000 per year from an area with most unusual lineaments. These producers and possibly some of those prospecting, in the west, might take stock of these patterns and look for similar structures else- where. There are at least six different directions of shear disturbances visible on Landsat data of the Illinois basin. One set, at least, appears to be important to the accumulation of highly productive oil pools found in the pre-Pennsylvanian carbonate rocks and seems to be closely allied with a similar pattern of major shear zones throughout the mid-continent area as well as the southern, middle, and northern Rocky Moun- tains. Oil and gas fields, in this area, are spatially coincident with these shear zones, and the same type of shear lineaments appear to be significant in the State of California where second order anticlines as- sociated with these shear patterns are, quite obvious- 1y. controlled by the regional pattern. The major shear belts of the Rockies also seemingly control second order tectonic features which in turn may partly con- trol the orebodies of the Colorado mineral belt. The gross continental orientation of lineations, such as those shown in an analysis of major lineations of the United States, undertaken by W. D. Carter (in Fischer and others, 1976), may have far reaching sci- entific implications. In Carter‘s analysis, the orienta- tion of the primary pair of shears appears to rotate clockwise as they approach the western edge of the continent; is this a reflection of continental mobility? I would like to emphasize that the science of ex- ploration may have experienced a new dawning in July 1972 and that as yet the new sun has not come up over the horizon. This is only a personal opinion, but I would like to restate that early skepticism on the part of individuals in the oil industry as to the value of the new imagery was premature, often resulting EXPLORATION BY INDEPENDENTS USING PHOTOS AND SPACE SURVEYS from (1) poor images or mosaics, (2) lack of under— standing, and (3) incorrect experimentation. Further, the independent exploration geologist lacks experi- ence in putting the computer to work to realize the full value of the data. There is perhaps much disillu- sionment based upon not seeing that “particularly favorite crustal structure or geomorphic anomaly." Take heart; there can be a reason for this! REFERENCES Fischer, W. A., Angsuwathana, Prayong, Carter, W. D., Hoshino, Kazuo, Lathram, E. H., and Rich, E. I., 1976, Surveying the Earth and its environment 177 from space: Am. Assoc. Petroleum Geologists Mem. (in press). Moody, J. D., and Hill, M. J., 1956, Wrench-fault tec— tonics: Geol. Soc. America Bull, v. 67, p. 1227. POSTSCRIPT The ABC’s of Discovery A for Alluvial Deposition, Areal Geology, etc. B for Below Surface Mining & Drilling Data, etc. C for Color Photos—Computer Printouts, etc. We can now say: ABCD’S of Discovery—ADD “D” D for Determinations and Discovery from Space PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Tectonic Deductions from Alaskan Space Imagery By Ernest H. Lathram and Robert G. H. Raynolds, US. Geological Survey, Menlo Park, California ABSTRACT Inductive analyses (reasoning from specific data to general conclusions) based on compilations of de- tailed mapping postulate numerous tectonic and metallogenic lineaments. Landsat, NOAA, and Nimbus images provide successively more regional and gen- eral views, which permit direct observation of significant lineament patterns. Deductions based on these patterns are in accord with many inductively derived patterns and solutions and provide new in- sights into tectonics and metallogenesis. Most major space image lineaments reflect zones of weakness, many of which have persisted since Pre- cambrian time, and most of which reflect major tectonic discontinuities. These lineaments display co- herent patterns throughout the North American Cordillera suggesting a consistent regional stress field. Subparallel sets of lineaments 1,000 km or more in length trend dominantly northwest, northeast, east and north. The pattern of these lineaments is con- sidered to represent the surface trace of the bound— aries of a mosaic of crustal blocks. In Alaska, some lineaments (e.g., northwest-trending Yentna lineament) reflect differential movement of crustal blocks in response to contemporary forces of plate convergence. Others coincide with portions of sutures that reflect earlier plate interactions, and still others traverse successively accreted belts of conti- nental crust, suggesting that continental masses and materials accreted to them react to the stresses of plate convergence by accommodation along lines of weakness preexisting in the continents. Many inductively derived metal “province” outlines reflect lineament trends, and some areas of metal fecundity occur at lineament intersections. The use of space image lineament to define crustal block boundaries and the analysis of the metallogenic char- acter within the area of single blocks may be a fruit- ful exploration tool. INTRODUCTION Lineament studies of continental masses have a common foundation in the work of Sonder (1947) pro- posing a “global regmatic shear" pattern for Earth, and of Katterfel’d and Charushin (1970) for Earth and other planets as well; a pattern resulting from stresses induced by the rotation of a planet on its axis. Moody and Hill (1956) modified this thesis and developed “wrench fault tectonics," later applied by Moody (1973) to petroleum exploration. Alinements of mineral deposits led Heyl (1972) and Snyder (1970) to postulate lineaments along the 36th, 38th, and 4lst parallels in the United States and Kutina (1969) to formulate an empirical “shear stress net” of mineral- izing lineaments in the western United States. Many others, among them Thomas (1974), Roberts (1964), Sikabonyi and Rodgers (1959), Tweto and Sims (1963), Maughan (1966), Haites (1960), and Ozoray (1972) point to more local systems of fractures, believed to be crustal in origin and to have controlled the site or origin of oil or mineral deposits. The most fascinating lineament studies in the past have been those dealing with features of the ocean floor. The combined work of many authors, summar- ized in the maps of Pitman, Larson, and Herron (1974), display a lineament system of oceanic ridges and nearly orthogonal fractures or transform faults. These have become a cornerstone of the new' global tec- tonics. The discovery of concentrations of metal-rich brines and sediments along spreading plate boundaries has led to metallogenic concepts in these areas sim- 179 180 ilar to the concepts basic to lineament studies on the continents. All of the above studies have one element in com— mon—the derivation of general hypotheses through the synthesis of specific details, i.e., inductive reason~ ing. Even the quantum-jump from the hypothesis of sea—floor spreading to the theory of plate tectonics is inductive, resting as it does on the synthesis of scat- tered data developed through oceanographic study. The advent of polar-orbitting satellite imagery, par- ticularly that of Landsat, has for the first time per- mitted application of deductive methods through lineament study. For the first time, lineaments can be observed directly over large areas of the Earth, their pattern defined, and their geologic and metallogenic significance examined and amplified by specific study. Lineament studies from Landsat imagery have been pursued by many researchers and are reported primarily in Proceedings of various Landsat symposia (Freden and others, 1973, 1974; Smistad, 1975), the First International Symposium on the New Basement Tectonics (Hodgson, Gay, and Benjamins, 1976) and the First William T. Pecora Memorial Symposium (this volume). Most of these Landsat studies have concen- trated on areas the size of several states or smaller, and many have been concerned with analyses of the number and concentration of linears, i.e., “linear popu- lations.” To our knowledge, only three studies (Carter, 1974; Haman, 1975; and Hodgson, oral commun., 1975) have treated the distribution of Landsat linea- ments throughout the conterminous United States. Metallogenic modeling has occupied the attention of many earth scientists. Earlier models based on the classical geosynclinal theory generally related exo— genic deposits to crustal downwarping, plutonism, and associated hydrothermal activity, primarily in mobile zones. Most recent models (see Guild, 1972), based on contemporary plate tectonic theory, associate exogenic deposits with subduction or mantle upwell— ing along plate margins. Repeated tectonism or plu- tonism, changes in the rate of subduction, or the development of new plate margins are believed to generate differing epochs of mineralization. As Kutina (oral commun., 1975) and others point out, however, many deposits occur in continental interiors distant from mobile zones or plate boundaries, under condi- tions difficult to reconcile with either tectonic theory, despite application of “intracratonic,” “platform,” ”intraplate,” “small-plate,” and “migrating plume” theories. Lineament studies of continental masses have gen— erally presumed exogenic deposits to be related to crustal fractures, or zones of weakness, which have FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM provided pathways for mineralizing agents to rise from a general source in the lower crust or upper mantle and form deposits wherever host conditions were propitious. Our approach to metallogenic modeling combines the new dimension of space image lineament study with tectonic knowledge and theory to develop new insights into continental tectonics. Through these addi- tional parameters, we attempt to identify viable new rationales for mechanisms of mineral concentration and favorable locales for mineral exploration. We have chosen to concentrate our attention, not on linear populations, but on the location and pattern of continental-sized lineaments—those 1,000 km or more in length—and their geologic, tectonic, and ultimately metallogenic significance. We have also chosen not to limit our study to Landsat imagery. If our conclusions (Lathram and Albert, 1976) are cor- rect, that abundant, short, sharp linear features reflect the youngest deformation of the upper lithosphere and tend to mask less abundant, longer, and more diffuse lineaments marking deeper crustal structures, and the conclusions of Shapiro (1971) that all imagery contains comparable quantities of data and only the level of generalization changes with altitude and scale of viewing, imagery from other satellites will assist in regional lineament study. This is analogous to the practice in airborne magnetometry in which the flight altitude is increased to observe deeper magnetic fields. Accordingly, we have analyzed imagery taken by Nimbus (1,100—km altitude, ~3.7-km resolution) and NCAA (1,500-km altitude, 1-km resolution) as well as Landsat (900-km altitude, 80-m resolution) satellites. All analyses have been by visual examination of photo- graphic prints and mosaics, primarily the Landsat mosaics of the US. Department of Agriculture and of the Canadian Department of Energy, Mines and Resources. We have examined geologic, tectonic, magnetic, and gravity maps and reports, and all the lineaments represented can be shown to coincide with known alinements or discontinuities in the character of lithospheric materials. We have concentrated on space image lineaments in Alaska. However, because data on Precambrian and Paleozoic geologic history is sparse in Alaska, we have made a general examination of imagery of the western parts of Canada and the conterminous United States, where such data exist. We believe that con- clusions as to the antiquity and recurrent nature of structures whose trace is marked by lineaments in these areas can be extrapolated directly to similar lineaments in Alaska along whose trace geologic, tec- TECTONIC DEDUCTIONS FROM ALASKAN SPACE IMAGERY tonic, and geophysical anomalies or discontinuities indicate significant changes at depth. Our studies suggest that a pattern of continental- sized lineaments, recognized in Alaska, extends over the entire North American Cordillera and probably over the entire continent. These lineaments reflect major crustal boundaries, and higher or more synoptic satellite views seem to reveal deeper and more funda- mental structures. We believe that knowledge of the nature and history of these lineaments is not only important in unraveling the tectonic history of a re- gion but is critical to understanding how continental masses react internally to the stresses of plate move— ment. This knowledge can lead to the development of even more fundamental tectonic theories than those of today, theories that reconcile both mobilistic and stabilistic evidence of lithospheric deformation. To us a lineament network as a guide to a “plumbing sys— tem” is too simplistic; we view the lineament pattern as a plan of a mosaic of crustal blocks (similar to that of Garson and Krs, 1976) whose horizontal and verti- cal jostling through geologic time has guided tectonic relief to Earth stresses and, concomitantly, has guided the emplacement of orebodies. LINEAMENT PATTERN Study of a 1971 Nimbus IV image resulted in iden- tification of a number of previously unknown very long lineaments in Alaska (Lathram, 1972). Subsequent study of Landsat images and mosaics (Lathram and Albert, 1976; Lathram and Raynolds, 1975) revealed not only most of these lineaments and others as well, but also a possible geometric pattern. Further exam- ination of the Nimbus and Landsat images and mosaics, and of images from NOAA 3 satellite has disclosed even more lineaments (fig. 1). The lineaments identified occur as alinements of surface geologic structures, linear valleys or ridges, and linear Changes in tonal contrast marking differ- ences in soil type, soil moisture, or vegetation. Most of the major lineaments are composed of combinations of some or all of these. Most are broad and diffuse, some being poorly identifiable for short stretches along their length. Parts of the trace of many of the lineaments coincide with the trace of known major faults in Alaska of Cretaceous or younger age (Grantz, 1966), which are primarily strike-slip in character. The number of large lineaments identifiable on space imagery of Alaska (fig. 1) increases from \Iimbus (fig. 1A) to NOAA (fig. 18) to Landsat (fig. 1C) Iiews, i.e., from smaller scale, low-resolution images 0 larger scale, higher resolution images. Most linea- nents visible on Nimbus and NCAA images can be 181 identified on Landsat images, but on the smaller scale images they tend to be more diffuse and can be less accurately located. The more synoptic view of the smaller scale images, however, permits recognition of a greater length for some of the lineaments than that shown on the Landsat mosaics and also shows that some shorter, individual, alined lineaments are probably segments of a longer lineament that is only intermittently discernible. On the other hand, the sharpness of features in the higher resolution images may lead to selection of unrelated younger tectonic elements as representatives of the lineament, and cause its mislocation. Some lineaments visible on the low-resolution images are less easily recognized on Landsat mosaics. Examples are the curvilinear and circular features shown on the NOAA image (fig. 18). They are recognizable on the Landsat mosaic but were identified only after they were recognized on the NOAA image. It is also apparent that, at the scale and resolution of the Landsat imagery, the number of lineaments discernible, and their varied trends, tend to obscure a regular pattern (fig. 1C). This suggests that for the purpose of broad, regional tectonic study, the op- timum scale and resolution may lie between that of the NCAA and Landsat imagery. This observation is supported by our increased ability to discern broad lineament systems in the conterminous United States by examining a lithograph print of the Department of Agriculture 1:5,000,000-scale Landsat mosaic (US. Geological Survey, 1974) and comparing it to the original 121,000,000-scale mosaic. Carter (1974) also had good success in identifying many of these linea- ments by direct study of the 125,000,000-sca1e mosaic. Identification of a systematic pattern of lineaments in Alaska (fig 1D) required the elimination of a large number that constitute a concealing background. The lineaments shown on the Nimbus and NCAA images display a pattern, and we have found that they are marked by significant geological or geophysical anom— alies or discontinuities along much of their length. These lineaments were used as guides in identifying the lineament pattern. We are aware of the degree of subjectivity involved in the selection of individual lineaments forming the resulting pattern. Many other lineaments are parallel, or nearly so, to those selected, and through future study may be shown to be of greater tectonic signi- ficance. The number selected was determined by the scale of this study, to illustrate the pattern clearly and its existence elsewhere as well. A broadly com- parable spacing is apparent (fig. lD); Albert (oral commun, 1975) has noted in more detailed studies in 120° 130° 150° 160° 170° 182 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 120° 130“ 140° 150‘I 100° 170° \ °% 3 y \\\ n W \l ‘1 0° ‘é’ \\ ‘ 3 w °o 32/ =-‘,E/: . n-Jgfj; ll» ".5: 31:. \\ 28‘! g i . H - g ’ — TA $> z :4): ' 1‘ A c» :5 § / - , ': a g E o u g \I‘n\\ 'f‘! O N k 9 ~- w «@x o s .\ \ i , t ‘N g l o o 3 . \‘ %&o 8 a: K g \ E g 1‘ l9»,— a! A A‘ k 3“» m ' “is, a o \‘ ‘ 9 § < 5 *V‘ a, g < ’ 5 «5 ° % slugs ’ l 9‘ < 954v / ‘5 '- é 533% ”L ' 59 "’ 915$ “L o e " / \ ‘ t 2 on ’9 / 1 °" , M I "a" on o\\ s '2. ' Q j/ \ 8 i n // \ . é L‘ i n % \ o o \ 1" / P \ , ,. 5a \ \\ ° /jo‘g \\ v v?!’ \ " 0° . “ \“VI V I \ \ U, 7U ‘\ ‘ -“ ’ tn \ g g/ l ' [A . 5 3 °8 - E’I‘J .— Mix“ E/ /, “°" STATES a Q“ «’3 Vii-iii}? ' ‘ 3 “L—rr’ U z * ‘ -. g“ v7.4"; a * E g :5 § ‘ 4’ . 3‘"? i- 3 § 2 ~.\ g g 2, § S .Q, ’17“; 0 § § \ ~ g a \WQ 1 ‘~ g go I 0 ‘\fi.;<\“3r§&‘ 3 ,~,, a o m x - o o 3 i E \I l \V” §‘ W o l = rfiiw“ w 2 , ° 2 ‘ \ leA‘VI‘ ‘ " r» ‘5 A‘leug \: 5‘ I .§ . g I ‘1? "A‘ '1' g 914.43 a / 5 a s /\ i. d“. ‘g‘ D§sn\5.u~n Q o [I n a! a 1% " ,' ‘1. ‘ (See text for explana- NOAA satellite. C, Landsat satellite. D, Landsat lineament pattern. tion of symbols on figure 10.) FIGURE 1,—Space image lineaments in Alaska. A, Nimbus satellite. B, TECTONIC DEDUCTIONS FROM ALASKAN SPACE IMAGERY Alaska that major lineaments parallel to those shown occur with a general spacing of multiples of about 30 km. It is tempting to assign firm compass measurements to the trends of these lineaments; however, the trend of most is not sufficiently consistent, and the orienta- tion of the overall pattern seems to change slightly from area to area. This may be partly due to problems of map projection, but, if these lineaments do reflect crust-a1 block boundaries and the blocks have moved differentially, one would not expect sharp, straight boundaries. Further, no persuasive model has yet been proposed to explain the mechanism of propagation of crustal features through overlying younger material. Hence, we are unsure how later tectonic movements would affect the trends of lineaments marking deep crustal features. For example, a number of the linea- ments are transected by faults known to have sig- nificant lateral displacements, i.e., the Denali fault system, but are not themselves offset, suggests that later tectonic movements either had little effect or some mechanism exists whereby the lineament re- adjusts its trace. In order to permit some comparison of these lineaments with those in other distant areas, their trends in central Alaska are approximated to be roughly N. 40°E., N. 60°E., N. 50°W., N. 80°W., and N. Previous study (Lathram and Albert, 1976) showed that lineaments having trends parallel to those of the pattern in Alaska also characterized the western United States. Examination of Nimbus and NCAA imagery, the new Landsat mosaics of western Canada and the Landsat mosaic of the western United States substantiates this conclusion and shows that the pat- tern recognized in Alaska extends throughout the North American Cordillera (fig. 2). GEOLOGICAL AND GEOPHYSICAL EVIDENCE It is impractical to discuss in this paper all of the geological and geophysical features that characterize the lineaments identified in the pattern (figs. 1D, 2). Consequently, we shall discuss briefly some selected lineaments in Alaska, mention some others elsewhere in the North American Cordillera for which evidence of antiquity and persistent movement is known, and compare the space image lineament pattern to linea- ments or linear crustal fractures recognized in some previous inductive studies. It is also impractical to include adequate geo- graphic, geologic, or geophysical detail in the figures. Most of our data appear on the Tectonic Map of North America (King, 1969) and the Bouguer Gravity Map 183 of Alaska (Barnes, 1976), which should be used in conjunction with this paper. Data that do not appear on these maps are referenced to the appropriate paper. Examination of the supporting data will show that, whereas most lineaments are discernible throughout their length, many do not coincide with linear changes or disruptions in surface geology and/or subsurface geophysical data throughout their length. Portions of lineaments marking significant tectonic or geophysical changes are adjacent to portions that show none. We have no explanation for this, other than that the sensi- ble adjustment on the boundary occurred prior to emplacement of the earth materials yielding the geo- logic or geophysical data. Therefore some mechanism must exist for the propagation of these lineaments through overlying strata. Lineament A, the Tanana lineament of this paper (fig. 1D), trends along the Tanana River and northwest to Cape Lisburne in northern Alaska. At the Canadian Border, it separates the dominantly polymetamor— phosed Precambrian(?) and Paleozoic Yukon-Tanana terrane from the Permian and younger arc-trench rep- resentatives to the southwest. In the Fairbanks area, it transects the southern margin of the metamorphic terrane but coincides with a marked change in struc- tural trends within the terrane, a series of saddles in otherwise north-trending magnetic anomalies (Brosgé and others, 1970), and a zone of shallow microseisms (Gedney and others, 1972). Throughout this area the lineament separates two regimes of significantly dis- parate gravity anomalies. Through the Ray Mountains, between Fairbanks and the Koyukuk River area, structural trends are nearly orthogonal to the linea- ment, and no disruption in them or change in Permian and younger tectonic elements is evident, but the lineament follows a line marking a regional change in gravity measurements. In the Koyukuk area, it coincides with, and may have governed the location of, an anomalously northwest-trending Cretaceous basin in the Yukon-Koyukuk Volcanic Province of Patton (1973). Between the Koyukuk area and Cape Lisburne, the lineament marks a structural warp in mid-Paleozoic Brooks Range strata, a line dividing distinctive gravity anomalies under the Brooks Range from disparate anomalies to the south and southwest, and a line of change in structural style in Jurassic and Cretaceous formations northwest of the Delong Mountains (Lathram, 1974). If extended to the north- west under the Chukchi Sea, the lineament would coincide with the fault boundary between the exten- sion of the thrust-faulted Paleozoic strata of the 184 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM \ \ {4/ TECTONIC DEDUCTIONS FROM ALASKAN SPACE IMAGERY Brooks Range orogen and the Mesozoic basin to the northeast (Grantz and others, 1975). Lineament B (fig. 1D) coincides with the Tintina fault in Yukon Territory and extends northwest to Icy Cape. In the southeast, the lineament separates the unmetamorphosed Yukon stable block (Lathram, 1973) and MacKenzie Mountain thrust belt from the Yukon-Tanana metamorphic complex, and is followed by a series of magnetic lows (Brosgé and others, 1970) and an alinement of excursions in gravity con- tours. To the northwest, it marks the eastern termina- tion of the Kobuk trench, the point of inflection of the Paleozoic schist and ophiolite rim of the Yukon- Koyukuk Volcanic Province, and a pronounced offset in the low gravity anomaly underlying the Brooks Range. Between the Range and the coast, the linea- ment parallels Mesozoic facies lines (Detterman, 1973) and bisects a broad gravity low. Lineament C (fig. 1D), described by Lathram (1972) and Lathram and Albert (1976), may represent an early Paleozoic continental margin throughout much of its length, and is characterized by a series of gravity highs and alined gravity gradients. Geologic and tectonic features along lineament D (fig. 1D) were detailed by Lathram and Raynolds (1975). Briefly, in the east, the lineament coincides with a fundamental fault separating Paleozoic and Mesozoic strata. In the Cook Inlet area, it marks significant changes in physiographic and magnetic features, is the northern terminus of the Aleutian Volcanic Arc and the northern boundary of the Jurassic intrusive core of the Alaska Peninsula. In the west, a young pluton is anomalously foliated parallel to the lineament on the Bering Sea coast. A pro- nounced line along which gravity contours change trend and strong linear anomalies terminate or seem to be offset follows the lineament through much of its length. Elsewhere in the North American Cordillera, the lineaments are characterized by features whose na- ture and history are better known. Lineament E (fig. 2) is marked by the MacDonald or East Arm fault system and coextensive Hay River fault and Skeena Arch believed to have exhibited persistent movement since Precambrian time (Haites, 1960; Sikabonyi and Rodgers, 1959; Sutherland-Brown and others, 1971). Lineament F (fig. 2) coincides with the Athabasca- Peace River lineament, active since the Precambrian and particularly in the late Paleozoic (Sikabonyi, 1967). Lineament G (fig. 2) lies along and is a con- tinuation of the Lewis and Clark line which, in the eastern Washington and Idaho area, has been per— sistent since the Precambrian (Sales, 1968). Linea— 185 ments H and I (fig. 2) correspond to postulated basement structures Eaton and others (1975) believe controlled the emplacement of the basalts of the Snake River Group. Lineament J (fig. 2) marks the line described by Yates and others (1966) and Taube- _ neck and Armstrong (1974) as the western boundary of Precambrian basement. A compilation was made of the lineaments and crustal fractures proposed by a number of other authors through inductive analysis (fig. 3). The pattern of space image lineaments correlates well with many of these features, in general orientation, if not in spe- cific compass trend, and many individual space image lineaments correspond to mineralizing crustal frac- tures. Although many of the long lineaments of the compilation (fig. 3) are based on physiographic ex- pression, some of the long and most of the short linear elements are based on geological, geophysical, or tectonic evidence. TECTONIC CONSIDERATIONS The existence of a systematic pattern of space image lineaments 1,000 km or more in length and evidence that many (1) coincide with faults known to have been persistent since Precambrian time, (2) cor— respond to systems of mineralizing crustal fractures recognized locally, (3) mark significant changes in gravity and magnetic fields, and (4) bound funda- mental tectonic elements suggest to us that these lineaments reflect boundaries between crustal blocks. Vertical and horizontal adjustments along these boundaries in response to Earth stresses have re- sulted in each block possessing different properties and a different history. This accords with the principal conclusions reached by Garson and Krs (1976) who recognized crustal block tectonics in the Red Sea area, reasoning primarily from geophysical data but also noting the coincidence of some Landsat linea- ments with geophysical lineaments. Recognition of the possible existence of a mosaic of crustal blocks in Alaska, each with a varied history of vertical or horizontal adjustment, helps to explain the known varied character and disjunctive distribu- tion of basement rocks in the State. A number of the lineaments evident in the North American Cordillera transect boundaries that have been considered to coincide with former ocean/ continent interfaces without notable deviation in trend. Garson and Krs (1976) discussed the rifting of a continent to generate a spreading ridge (e.g., the Red Sea and Atlantic Ocean) and concluded that ancient crustal fractures determine the form and posi- 186 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM ‘\,/ {7,1 (I ' i 5. (5 Q \\ magi) @QQ? I. J / _ FIGURE 3.—Lineaments and crustal fractures suggested in selected inductive analyses. Compiled from: Moody (1966) and Ozoray (1972), (most long Iineaments, fine line); and (a) Grant (1969), (b) Zietz and others (1969, 1971), (c) Thomas (1974), (d) Roberts (1964), (e) Shoemaker and others (1974), (f) Muehlberger (1965), (g) Cook (1957), (h) Heyl (1972), (i) Stewart and others (1975), (j) Yates (1968), (k) Maughan (1966), (I) Landwehr (1967), (m) Eaton and others (1975), (in western Canada) Haites (1960). TECTONIC DEDUCTIONS FROM ALASKAN SPACE IMAGERY tion of the ridge and of the transform faults that transect it. In consequence, continental crustal frac- tures are traceable across passive continent/ocean boundaries into major transform faults. This conclu- sion does not hold for active continent/ocean margins, in which the two are decoupled, and for members of different plates. Indeed, in the Pacific Ocean, ringed by active margins, few transform faults have been demonstrated to be coextensive with either conti- nental fractures or space image lineaments. Our studies suggest that oceanic crust with active margins, whose spreading is not directly associated with continental rifting and whose structures may trend at angles to continental structures, will react with continental crust in a manner such that failure will occur along or parallel to lines of weakness pre- existent in the continental crust. Accordingly, the preexisting lines of weakness in the continents will be reactivated, and extensions of these lines of weakness will develop in the products of the interaction as they are accreted to the continents. Sympathetic parallel lines of weakness will also develop in the accreted materials. This mechanism helps to explain the propa- gation of lineaments near active plate margins but does not explain their vertical propagation along passive margins nor within continental interiors. From the foregoing, it is evident that the study of space image lineaments can provide insight into the nature of ocean/continent interactions and their effects within continental interiors. The Yentna linea- ment is an instructive example. The Yentna (Y, fig. 1D) and parallel Tanana linea- ments (A, fig. 1D) played a significant role in the interaction between the Pacific (oceanic) plate and continental Alaska in late Cenozoic time. The Yentna lineament forms the northeast boundary of a zone of pronounced change in trend of structures in the con- tinental shelf (Von Huene and Shor, 1969), the north- east boundary of the segment of ocean floor warped in the 1964 Alaska earthquake (Plafker, 1969), the northeast boundary of the major interarc basin of Cook Inlet (Grantz and Kirschner, 1975), and the northeastern termination of the late Cenozoic volcanic arc of the Aleutian Islands and Alaska Peninsula. The lineament parallels the direction maintained by the oceanic plate vector in the Gulf of Alaska over the last 40 million years (Grow and Atwater, 1970). In the Alaska Range, the lineament coincides with the Yentna lineament of Van Wormer and others (1974) along which a warp occurs in the northwest—dipping Benioff zone, or zone of inclined earthquake foci. According to these authors, northeast of the Yentna ineament the Benioff zone is more steeply inclined 187 and dips more northerly. The Benioff zone continues northeast beyond Mt. McKinley and terminates near the Tanana lineament. These data suggest that movement of the Aleutian Arc-Trench system in the last 40 million years has been essentially translatory northwestward past the more resistant main continental mass of Alaska. The stress of plate interaction has been successively ac- commodated along crustal block boundaries parallel to and southwest of the Tanana lineament. The change in the Benioff zone at the Yenta lineament is sugges- tive of drag resulting from dextral shear. Southwest- ward of the Yentna, the volcanic chain of the Alaska Peninsula is segmented in plan, segment boundaries coinciding with lineaments at Lake Iliamna and at Controller Bay which trend northwest parallel to the Yentna and Tanana lineaments. To the west, a similar segmentation of the Aleutian Volcanic Arc has been noted (Jordan and others, 1965). The Tanana linea— ment also coincides in part with an earlier suture of Permian-Triassic age (Richter and Jones, 1973). From the above phenomena, we conclude that ad- justments in continental crust as a result of oceanic plate convergence have occurred along zones of weak- ness already existing in the continental crust. Con- tinued adjustments have occurred successively along other zones of weakness parallel to these preexisting zones of weakness. We believe that the tectonized remnants of the Permian to early Cenozoic arcs and basins owe their present shape to accretion against a northern buttress formed of a northeast-trending early and middle Paleozoic continental shelf on the northwest and a remnant northwest-trending Pre- cambrian and Paleozoic remnant volcanic arc com- plex in the northeast (Yukon-Tanana Upland). Northward compressive stresses which successively collapsed these arcs and basins against the continental mass (Richter and Jones, 1973; Jones and others, 1970) were relieved by decoupling along northeast- and northwest-trending fractures already existing in the continental crust, parallel to the trends of the buttress. Two of these fractures are now visible as northeast- and northwest-trending lineaments which mark the trace of parts of the Denali fault system in eastern and western Alaska. Short, convex-east curvilinear features were noted in the area south of the Denali fault system on ex- amination of the NOAA images (fig. 18). They are less apparent, but present, on Landsat images and ap- parently are enhanced by the distortion of the NOAA images. These curvilinears do not accord with any of the lithologic or tectonic trends in the area, except the slightly convex-east arcuate pattern of volcanoes 188 FIRST ANNUAL PECORA from Lake Iliamna to the Yentna lineament. The Denali fault system exhibits a history of dextral move- ment since Cretaceous time, but there is also evidence of vertical movement in the segment west of the Nenana River (Grantz, 1966). The vertical movement in this area would accord with the presumed north- west shortening in response to crowding from the Tertiary arc complex, but the dextral movement is less easily understood. Clague and others (1975), studying the age of volcanic rocks dredged from the Hawaiian Ridge and Emperor Seamounts, recapitu- lated the case for the Hawaiian-Emperor bend form- ing as a result of a change from north to northwest in the movement of the Pacific plate over a melting spot in the mantle and added supporting evidence of pro- gressive changes in age of volcanism that would ensue were this hypothesis valid. They conclude the age of the bend is 41 to 43 my. If such a vector change oc- curred, it would have impressed a counterclockwise rotation on the area south of the Denali fault system, engendering a dextral stress release along it. The curvilinear features may reflect this distortion and, therefore, be young and temporary. METALLOGENIC SIGNIFICANCE In an earlier study, Lathram and Gryc (1973) re- lated some of the space image Iineaments to the distribution of mineralized areas, and compared. this with the conclusions of Sutherland-Brown and others (1971) that areas of mineral concentration in British Columbia were related to the intersection of major through-going orthogonal structures. Preliminary com- parison of the full pattern of Iineaments developed in this paper to the distribution of mineral deposits in Alaska has as yet revealed no significant additional relationships. However, comparison of this pattern of Iineaments to the empirical “shear-stress network” developed by Kutina (1969) in the western United States reveals that a large number of Iineaments cross a number of highly mineralized areas, a large number of lineament intersections coincide with metal-rich areas (as do Kutina’s “shear-stress” lines), and there is a striking correlation between the two patterns in orientation and spacing of individual ele- ments. Further, the trends and spacing of Iineaments of the space image pattern correspond closely to the trends and spacing of “deep—seated tectonic zones" recognized by Carson and Krs (1976) in the Red Sea area. Carson and Krs showed these zones to be re- lated not only to mineral deposits in the continental crust but also to brine pools in the Red Sea. Further MEMORIAL SYMPOSIUM study of the pattern in Alaska, or modifications to it, should be fruitful for mineral exploration. Many metallogenic experts, especially Clark and others (1974) in Alaska and Sutherland—Brown (1974) in British Columbia have also pointed to broad areas or zones of like mineralization which seem uncon- trolled by major structures. Sutherland-Brown (1974, p. 54) in particular studied the regional distribution of metals as shown by geochemical sampling and concluded that the similarity between the pattern of mineral deposits and metal background means that the deposits owe their origin to the level of back— ground in the terrane they inhabit——“the rocks have the ore deposits they deserve.” On the other hand, Sillitoe (1974) has examined the known linear zona- tion of mineral deposits in the Andes and, using geologic discontinuities transverse to the length of the zones, has subdivided the Andes into tectonic seg- ments, each with a metallogenic character different from that of its neighbors. From the foregoing, we suggest that mineral con- centrations may be expected along or at the inter- section of space image Iineaments that mark signifi- cant crustal block boundaries. We also suggest that the identification of major crustal block boundaries through the study of space image Iineaments and the analysis of the tectonic history, geochemistry, and mineralization within each block (as opposed to broader regions or zones) may also be fruitful in mineral exploration. In fact, the differing history ex- perienced by individual blocks as the result of relief of internal as well as external plate stresses may be a more significant determinant of the mineral fecun- dity of an area. A possible example is the lower Kus- kokwim area, rich in mercury deposits as compared to the rest of Alaska. Although individual deposits seem controlled by local faults, the area of minerali- zation is not related to a single lineament but is largely confined between two northeast—trending Iineaments (Lathram and Gryc, 1973) and two north- west-trending ones, the Yentna and one parallel to it south of Lake Iliamna. CONCLUSIONS We conclude that: 1. Most space image Iineaments 1,000 km or more in length are the surface expression of major changes in the character of the lithosphere re- sulting from persistent movement, vertically and/or horizontally along zones of weakness in the crust. TECTONIC DEDUCTIONS FROM ALASKAN SPACE IMAGERY N A systematic pattern exists in these lineaments and represents a systematic pattern of bound- aries of a mosaic of crustal blocks. 3. Differing vertical and/or horizontal adjustments from block to block in response to plate stresses result in differing tectonic histories and, hence, differing geological and geophysical characteristics in the lithosphere of each block. 4. The response of continental crust to forces of plate convergence or divergence and of oceanic crustal materials accreted to the continents dur- ing plate convergence is guided by accommo- dation along these preexisting crustal block boundaries. 5. Metal concentrations may not only be controlled by these crustal block boundaries and their intersections, as felicitous channels for the mi- gration of mineralizing agents but may also be governed by greater or lesser mineral fecundity within individual blocks, as determined by dif— ferences in tectonic history. 6. The study of very long space image lineaments and their geologic significance provides a powerful new tool for tectonic and metallogenic analysis. SELECTED REFERENCES Barnes, D. F., 1976, Bouguer gravity map of Alaska: US. Geol. Survey Open-File Rept. 76—70, 1: 2,500,000. Brosgé, W. P., Brabb, E. E., and King, E. R., 1970, Geologic interpretation of reconnaissance aero- magnetic survey of northeastern Alaska: US. Geol. Survey Bull. 1271—F, p. F1—F12. Carter, W. D., 1974, Tectolinear overlay of the United States: US. Geol. Survey Open-File Rept., 1: 5,000,000. Clague, D. A., Dalrymple, G. B., and Moberly, Ralph, 1975, Petrography and K-Ar ages of dredged volcanic rocks from the western Hawaiian Ridge and the southern Emperor Seamount chain: Geol. Soc. America Bull, v. 86, p. 991-998. Clark, A. L., Berg, H. C., Cobb, E. H., Eberlein, G. D., MacKevett, E. M., Jr., and Miller, T. P., 1974, Metal provinces of Alaska: US. Geol. Survey Misc. Inv. Map I—384, 1:5,000,000. 300k, D. R., ed, 1957, Geology of the East Tintic Mountains and ore deposits of the Tintic mining district: Utah Geol. Soc. Guidebook to the ge— ology of Utah, no. 12, 183 p. 189 Detterman, R. L., 1973, Mesozoic sequence in Arctic Alaska, in Pitcher, M. (3., ed., Arctic geology: Am. Assoc. Petroleum Geologists Mem. 19, p. 376—387. Eaton, G. P., Christiansen, R. L., Iyer, H. M., Pitt, A. M., Mabey, D. R, Blank, H. R., Jr., Zietz, Isidore, and Gettings, M. E., 1975, Magma beneath Yellowstone National Park: Science, v. 188, p. 787—796. Freden, S. C., Mercanti, E. P., and Becker, M. A., eds, 1973, Symposium on significant results ob— tained from the Earth Resources Technology Satellite~1, v. 1: Technical presentations: Natl. Aeronautics and Space Admin. Spec. Pub., SP 327, 2 sections, 1,730 p. —~——1974, Third Earth Resources Technology Satellite—1 symposium, v. 1: Technical presenta- tions: Natl. Aeronautics and Space Admin. Spec. Pub., SP 351, 2 sections, 1,994 p. Garson, M. S., and Krs, Miroslav, 1976, Geophysical and geological evidence of the relationship of Red Sea transverse tectonics to ancient frac- tures: Geol. Soc. America Bull, v. 87, p. 169— 181. Gedney, L., Shapiro, L. H., Van Wormer, J. D., and Weber, F. L., 1972, Correlation of epicenters with mapped faults, east—central Alaska, 1968— 1971: US. Geol. Survey Open-File Rept. 1,768 p. Grant, A. R., 1969, Chemical and physical controls for base metal deposition in the Cascade Range of Washington: Washington Div. Mines and Geol. Bull. 58. Grantz, Arthur, 1966, Strike-slip faults in Alaska: US. Geol. Survey Open—File Rept. 852, 82 p. Grantz, Arthur, Holmes, M. L., and Kososki, B. A., 1975, in Yorath, C. J., Parker, E. R., and Glass, D. J., eds, Canada’s continental margins and offshore petroleum exploration: Canada Soc. Petroleum Geologists Mem. 4, p. 669—700. Grantz, Arthur, and Kirschner, C. E., 1975, Tectonic framework of petroliferous rocks in Alaska: US. Geol. Survey Open-File Rept. 75—149, 29 p. Grow, J. A., and Atwater, Tanya, 1970, Mid-Tertiary tectonic transitions in the Aleutian Arc: Geol. Soc. America Bull, v. 81, p. 3715—3722. Guild, P. W., 1972, Metallogeny and the new global tectonics, in Internat. Geol. Cong, 24th, Mon- treal 1972, Proc., Sec. 4, Mineral deposits, p. 17— 24. Haites, T. B., 1960, Transcurrent faults in western Canada: Alberta Soc. Petroleum Geologists Jour., v. 8, p. 33—79. 190 FIRST ANNUAL PECORA Haman, P. J., 1975, A lineament analysis of the United States: West Can. Res. Pub., ser. 4, no. 1, 27 p. Heyl, A. V., 1972, The 38th parallel lineament and its relationship to ore deposits: Econ. Geol., v. 67, p. 879—894. Hobbs, W. H., 1968, Geologic setting of metallic ore deposits in the northern Rocky Mountains and adjacent areas, in Ridge, J. D., Ore deposits in the United States: Am. Inst. Min. Eng, p. 1352— 1371. Hodgson, R. A., Gay, S. P., and Benjamins, J. Y., eds., 1976, Proceedings of the First International Conference on the New Basement Tectonics: Utah Geol. Assoc. Pub. 5, 636 p. Jones, D. L., MacKevett, E. M., and Plafker, George, 1970, Speculations on late Mesozoic tectonic his- tory of part of southern Alaska (abs): Am. Assoc. Petroleum Geologists Bull, v. 54, p. 2489. Jordan, J. N., Lander, J. F., and Black, R. A., 1965, Aftershocks of the 4 February 1965 Rat Island earthquake: Science, v. 168, p. 1323—1325. Katterfel’d, G. N., and Charushin, G. V., 1970, Global fracturing on the earth and other planets: Geo- tectonics, no. 6, 1970, p. 333. King, P. B., comp, 1969, Tectonic map of North America: US. Geol. Survey, scale 125,000,000. Kutina, Jan, 1969, Hydrothermal ore deposits in the western United States: a new concept of struc- tural control of distribution: Science, v. 165, p. 1113—1119. Landwehr, W. R., 1967, Belts of major mineralization in western United States: Econ. Geol., v. 62, p. 494—501. Lathram, E. H., 1972, Nimbus IV view of the major structural features in Alaska: Science, v. 175, p. 1423—1427. 1973, Tectonic framework of northern and central Alaska, in Pitcher, M. G., ed., Arctic geology: Am. Assoc. Petroleum Geologists Mem. 19, p. 351-360. 1974, Geologic applications of ERTS imagery in Alaska, in Freden, S. C., Mercanti, E. P., and Becker, M. A., eds, Third Earth Resources Tech- nology Satellite-1 Symposium Volume I: Tech- nical Presentations: Natl. Aeronautics and Space Admin. Spec. Pub., SP 351, Addendum-Techni- cal Presentation G3—1, 12 p. Lathram, E. H., and Albert, N. R. D., 1976, Signifi- cance of space image linears in Alaska, in Hodg- son, R. A., Gay, S. P., and Benjamins, J. Y., eds., Proceedings of the First International Confer- MEMORIAL SYMPOSIUM ence on the New Basement Tectonics: Utah Geol. Assoc. Pub. 5, p. 11—26. Lathram, E. H., and Gryc, George, 1973, Metallogenic significance of Alaskan geostructures seen from space: Internat. Symposium on Remote Sensing of Environment: 8th, Ann Arbor, Mich. 1973, Proc., p. 1209—1211. Lathram, E. H., and Raynolds, R. G. H., 1975, Final report on significance of giant linears in Alaska, pt. 11: Geologic differences along selected giant linears: U.S. Geol. Survey Open-File Rept., p. 29—43. Maughan, E. K, 1966, Environment of deposition of Permian salt in the Wulliston and Alliance Basins, in Northern Ohio Geol. Soc. Symposium on Salt, 2d, Cleveland, Ohio 1965, Proc., v. 1, p. 35—47. Moody, J. D., 1966, Crustal shear patterns and orogenesis: Tectonophysics, v. 3, p. 479—522. 1973, Petroleum aspects of wrench-fault tec- tonics: Am. Assoc. Petroleum Geologists Bull, v. 57, p. 449~476. Moody, J. D., and Hill, M. J., 1956, Wrench-fault tec- tonics: Geol. Soc. America Bull, v. 67, p. 1227. Muehlberger, W. R., 1965, Late Paleozoic movement along the Texas lineament: New York Acad. Sci. Trans, v. 27, p. 385—392. Ozoray, G. F., 1972, Tectonic control of morphology on the Canadian interior plains, in Adams, W. P., and Helleiner, F. M., International geography 1972: Res. Council Alberta, Canada Contr. 547, p. 51-54. Patton, W. W., Jr., 1973, Reconnaissance geology of the northern Yukon-Koyukuk province, Alaska: US. Geol. Survey Prof. Paper 774A, 17 p. Patton, W. W., Jr., Lanphere, M. A., Miller, T. P., and Scott, R. A., 1976, Age and tectonic significance of volcanic rocks on St. Matthew Island, Bering Sea, Alaska: US. Geol. Survey, Jour. Research, v. 4, p. 67—73. Pitman, W. C., III, Larson, R. L., and Herron, E. M, 1974, The age of the ocean basins: Geol. Soc. America, 2 charts and summary statement. Plafker, George, 1969, Tectonics of the March 27. 1964, Alaska earthquake: U.S. Geol. Survey Prof Paper 543—1, p. 11—174. Richter, D. H., and Jones, D. L., 1973, Structure anc stratigraphy of eastern Alaska Range, Alaska, i1 Pitcher, M. G., ed., Arctic geology: Am. Assoc Petroleum Geologists Mem. 19, p. 408—420. Roberts, R. J., 1964, Economic geology, in Minera and water resources of Nevada: Nevada Bur Mines Bull. 65, p. 39-45. TECTONIC DEDUCTIONS FROM ALASKAN SPACE IMAGERY Sales, J. K, 1968, Crustal mechanics of cordilleran foreland deformation: a regional and small-scale approach: Am. Assoc. Petroleum Geologists Bull, v. 52, p. 2016—2044. Scholl, D. W., Buffington, E. C., and Marlow, M. S., 1975, Plate tectonics and the structural evolu- tion of the Aleutian-Bering Sea region, in Forbes, R. 8., ed., The geophysics and geology of the Bering Sea region: Geol. Soc. America Special Paper 151, p. 1-32. Shapiro, L. H., 1971, Comparison of information con- tent of space photography and low altitude aerial photography: U.S. Geol. Survey Interagency Rept. USGS—229, 40 p. Shoemaker, E. M., Squires, R. L., and Abrams, M. J., 1974, The Bright Angel and Mesa Butte fault systems of northern Arizona, in Karlstrom, T. N. V., Swann, G. A., and Eastwood, R. L, eds, Geology of northern Arizona with notes on archeology and paleoclimate, pt. I: Geol. Soc. America Rocky Mt. Sect. Mtg, Flagstaff, Ari- zona, p. 355—391. Sikabonyi, L. A., 1967, Major tectonic trends in the Prairie region of Canada: Alberta Soc. Petroleum Geologists Jour., v. 5, p. 23—28. Sikabonyi, L. A., and Rodgers, W. J., 1959, Paleozoic tectonics and sedimentation in the northern half of the West Canadian Basin: Alberta Soc. Petro- leum Geologists Jour., v. 7, p. 193—216. Sillitoe, R. H., 1974, Tectonic segmentation of the Andes: implications for magmatism and metal- logeny: Nature, v. 250, p. 542—545. Smistad, Olav, coord, 1975, Proceedings of the NASA Earth Resources Survey Symposium, June, 1975, v. I: Technical session presentations: Natl. Aeronautics and Space Admin. Tech. Mem., TM X-58168, 2 sections, 1,497 p. Snyder, F. G., 1970, Structural lineaments and mineral deposits (abs): Mining Eng, v. 22, p. 36. Sonder, R. A., 1947, Discussion of “Shear patterns of the Earth’s crust” by F. A. Vening Meinesz: Am. Geophys, Union Trans, v. 28, p. 939—945. Stewart, J. H., Walker, G. W., and Kleinhampl, F. J., 1975, Oregon-Nevada lineament: Geology, v. 3, p. 265—268. Stone, D. B., 1974, Platetectonics and Alaska-crustal motions: Univ. Alaska, Geophys. Inst., Ann. Rept. 1973—74. 191 Sutherland—Brown, A., 1974, Aspects of metal abund- ances and mineral deposits in the Canadian Cor- dillera: Canada Inst. Min. Metal. Trans, v. 77, p. 14—21. Sutherland-Brown, A., Cathro, R. J., Panteleyev, A., and Ney, C. S., 1971, Metallogeny of the Cana- dian Cordillera: Canada Inst. Min. Metal. Trans, v. 74, p. 121—145. Taubeneck, W. H., and Armstrong, R. L., 1974, The Precambrian continental margin in western Idaho (abs): Geol. Soc. America Abs. with Program, v. 6, p. 477—478. Thomas, G. E., 1974, Lineament-block tectonics: Wil— liston-Blood Creek Basin: Am. Assoc. Petroleum Geologists Bull, v. 58, p. 1305—1322. Tweto, O. L., and Sims, P. K, 1963, Precambrian ancestry of the Colorado mineral belt: Geol. Soc. America Bull, v. 74, p. 991—1014. Van Wormer, J. D., Davies, J., and Gedney, L., 1974, Seismicity and plate tectonics in south central Alaska: Seismol. Soc. America Bull, v. 64, p. 1467—1475. Von Huene, Roland, and Shor, G. 6., Jr., 1969, The structure and tectonic history of the eastern Aleutian Trench: Geol. Soc. America Bull, v. 80, p. 18894902. Yates, R. G., 1968, The trans-Idaho discontinuity, in Internat. Geol. Cong, 23rd, Prague, 1968, Proc., sec. 1, Upper mantle (geol. processes): Prague, Academia, p. 117—123. Yates, R. G., Becraft, G. E., Campbell, A. B., and Pearson, R. C., 1966, Tectonic framework of northeastern Washington, northern Idaho, and northwestern Montana: Canada Inst. Min. Metal. Spec., v. 8, p. 47—59. Zietz, Isidore, Bateman, P. 0, Case, J. E., Crittenden, M. D., Jr., Griscom, Andrew, King, E. R., Rob- erts, R. J., and Lorentzen, G. R., 1969, Aero- magnetic investigation of crustal structure for a strip across the western United States: Geol. Soc. America Bull, v. 80, p. 1703—1714. Zietz, Isidore, Hearn, B. C., Jr., Higgins, M. W., Robin- son, G. D., and Swanson, D. A., 1971, Interpre- tation of an aeromagnetic strip across the north- western United States: Geol. SocfAmerica BulI., v. 82, p. 3347—3372. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Computer-Enhanced Landsat Imagery as a Tool for Mineral Exploration in Alaska By Nairn R. D. Albert, US. Geological Survey, Menlo Park, California, and Pat S. Chavez, Jr., US. Geological Survey, Flagstaff, Arizona ABSTRACT Recent work in the Nabesna and McCarthy quad- rangles, Alaska, indicates that computer-enhanced Landsat imagery shows many of the known mineral deposits and can help in the prediction of potential mineral occurrences. False color, “simulated natural color,” and color ratio‘ techniques, were used suc- cessfully in conjunction with a black and white, single band image mosaic of Alaska. Computer techniques involved two stages of digital image processing (1) atmospheric and Sun-elevation corrections, noise re— moval, computer mosaicking, and change of the data format and (2) image enhancement, involving data manipulation for maximum discrimination of surface materials and structure. Application of a new tech- nique called a “sinusoidal” stretch gave information not available in other products having standard con- trast stretches. We identified several orthogonal sets of linears and found parallel linears to be regularly spaced at approximately 30 to 35-km intervals. The locations of known mineral occurrences correlate well with the linears. Extensions of known faults and possible locations of hidden intrusive bodies were identified. Analysis of numerous areas of anomalous light reflectance showed that whereas most are as- sociated with known mineral occurrences, altered zones, or geochemical anomalies, some are not and may represent unexplored altered zones or mineral— ized areas worthy of future exploration. INTRODUCTION Recent work carried out in the Nabesna and Mc- Carthy quadrangles, Alaska, under the Alaskan Min- eral Resource Assessment Program (AMRAP) indi- cates that Landsat data can significantly help detect and predict mineral occurrences, when used in con- junction with geologic mapping, geochemical sampl- ing, and aeromagnetic and gravimetric data. Landsat data can provide additional information on geology and structure relevant to mineral exploration that cannot be or may not have been acquired by other methods. This report presents the results of Landsat data interpretation in the Nabesna quadrangle, Alaska (Albert, 1975), and the preliminary results of similar interpretations in the McCarthy quadrangle, Alaska, and their significance for mineral exploration. LOCATION The Nabesna and McCarthy quadrangles are in south-central Alaska (fig. 1) bounded on the north and south by the 63°N. and 61°N. latitudes and on the west by 144°W. longitude. The eastern boundary is the Alaska-Canada border (141°W. longitude). PHYSIOGRAPHY Topographic extremes in the area studied range from a low of about 305 m (1,000 ft) in the western part of the McCarthy quadrangle to a high of more than 5,000 m (16,400 ft) in the southeastern part. The area can be divided into three types of terrain: (1) Plains and lowlands from about 305 m (1,000 ft) to 1,220 m (4,000 ft) with heavy to moderate vegeta- tion and few outcrops; (2) low mountains from about 915 m (3,000 ft) to 1,830 m (6,000 ft), with heavy to light vegetation and numerous outcrops in the higher 193 194 ALASKA CANADA I | | I l l +NAIESNA +HCCARTHV L, 1000 KM I FIGURE 1,—Map of Alaska showing location of the Nabesna and McCarthy quadrangles. elevations; and (3) high rugged mountains from about 915 m (3,000 ft) to peaks of 5,000 m (16,400 ft), with moderate to no vegetation, numerous glaciers, heavy snow cover and extensive outcrops. Temperatures in these quadrangles range from highs above 38°C (100°F) during the summer to lows below —57°C (—70°F) during the winter. Much of the area is underlain by discontinuous permafrost. BRIEF GEOLOGIC DESCRIPTION The Nabesna and McCarthy quadrangles are under- lain by three general geologic terranes that are sepa- rated by the Denali and Border Ranges faults (fig. 2). North of the Denali fault are mainly regionally meta- morphosed lower Paleozoic sedimentary and subordi- nate igneous rocks. South of the Denali fault and north of the Border Ranges fault, there are three as- semblages of extensive andesitic volcanic and asso- ciated sedimentary and intrusive rocks: Upper Paleo- zoic, Upper Mesozoic, and Upper Cenozoic. South of the Border Ranges fault are extensive Mesozoic fiysch—type sedimentary deposits. Most mineral production in the Nabesna and Mc- Carthy quadrangles, chiefly copper, gold, and silver (Richter and others, 1975; MacKevett and Cobb, FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 62°N .31.?“ .13" EXPLANATION Lower Paleozoic Metasedimentary Rocks 1 nd Subordinate Igneous Rocks ’ Upper Paleozoic,_ Mesozoic, end Cenozoic Andesitic Volcanic and Associated Sedimentary nd Intrusive Roe . Mesozoic FIysch-Type Deposits 0 100 KM L 1 FIGURE 2.—Map of the Nabesna and McCarthy quadrangles showing generalized geologic terranes. 1972), has been in the area between the Denali and Border Ranges faults. INTERPRETATION OF IMAGERY The study of three kinds of computer-enhanced Landsat imagery in conjunction with a single-band, black and white Landsat image mosaic of Alaska, re— sulted in (1) the identification of linear and circular features and their correlation with known geology, mineral occurrences, and geophysical data and (2) the identification of mineral occurrences and numerous potential targets for future exploration. LINEAR FEATURES One of the types of imagery used in this study was the Alaskan Landsat mosaic constructed in 1973 by COMPUTER-ENHANCED TOOL FOR the Soil Conservation Service, US. Department of Agriculture, using band 7 images generated without computer enhancement. Because of the synoptic per- spective and low Sun-angle of the non-summer images used, the mosaic was most useful for identifying lin- ear and circular features. Additional linear and cir- cular features were identified on the computer- filtered imagery discussed below. Three nearly orthogonal sets of linears were iden- tified in the Nabesna and McCarthy quadrangles (fig. 3). The predominant set trends approximately N. 43° W. and N. 48° E., the subordinate sets approximately N. 72° W. and N. 20° E. and N. 87° E. and north. Parallel linears trending N. 43° W., N. 48° E. or N. 87° E. are spaced approximately 30—35 km apart. East-trending linear features seen clearly on Land- sat imagery (figs. 3 and 4) suggest the existence of major east-trending concealed structures in the Nabesna quadrangle south of the Denali fault. Aero- magnetic (Griscom, 1975) and gravimetric (Barnes 63°N NABESNA 62°N 61°N EXPLANATION 141°W . Linear feature (dotted where uncertain) 0 Known mines 0 100 KM l 1 FIGURE 3.—Map of the Nabesna and McCarthy quadrangles showing linear features and mine locations. MINERAL EXPLORATION IN ALASKA 195 and Morin, 1975) data support the Landsat interpre- tations. These east—trending structures can also be seen in the McCarthy quadrangle (figs. 3 and 5) on Landsat imagery and should be substantiated by future aeromagnetic and gravimetric studies. The correspondence of known mineral occurrences to linear features in the Nabesna and McCarthy quad- rangles is very good. Of 257 known mineral occur- rences (Richter and others, 1975; MacKevett and Cobb, 1972; MacKevett, written communication), 141 (55%) occur within 1 km of linear features. One hundred twenty-four occurrences are prospects or mines, 78 (63%) of which occur within 1 km of linear features. Of the 17 mines, 14 (82%) occur within 1 km of linear features (fig. 3). These correlations sug- gest a strong relation between linear features and the more significant mineral occurrences. Statistically, the different mineral commodities in the McCarthy quadrangle appear to be related to the trend of specific orthogonal linear feature directions (table 1). For example, 34 of 85 copper prospect and mine deposits occurring within 1 km of a linear fea- ture are associated with northwest-trending linears. Subordinately, 18 copper deposits are associated with northeast-trending linears. This orthogonal relation appears to apply for silver, molybdenum, and, pos- sibly, antimony. Gold, on the other hand, occurs pri- marily near linears with north-northeast and north- east directions and subordinately near linears with northwest trends. Chromium, iron, zinc, and nickel, not shown, also appear to have preferred orthogonal linear direction associations. - Evaluation of significant mineral occurrences in each linear direction reveals three groupings: (1) northwest and northeast—trending linears to which most mineral occurrences are related, (2) north- northwest, north-northeast, and east—trending linears, and (3) west-northwest, east-northeast, and north- trending linears to which the fewest mineral occur- rences are related. Although the information is pre- liminary and incomplete, these trends are suggestive of the relations resulting from the Moody and Hill (1956) model for primary, secondary, and tertiary fault and fold development in regions of wrench-fault tectonics. CIRCULAR FEATURES Numerous circular features were identified in the Nabesna and McCarthy quadrangles (figs. 4 and 5). North of the Denali fault, these circular features can be correlated with aeromagnetic and geologic data suggesting that they may be related to concealed in- 196 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 144°W EXPLANATION ......... Linear and circular features (dotted where uncertain) 0 Color anomaly 0 50 100 KM L I I FIGURE 4.—Map of the Nabesna quadrangle showing linear and circular features and color anomalies observed on computer- enhanced Landsat imagery. (From Albert, 1975.) TABLE 1.—Number of significant mineral occurrences (mines and prospects) in the various linear directions in the McCarthy quadrangle, Alaska Linear direction E WNW NW NNW N NNE NE ENE Antimony ________ ___ -__ 3 ___ ___ ___ ___ ___ Copper __________ 11 5 34 5 2 6 18 4 Cold ____________ ___ ___ 3 1 ___ 6 5 2 Molybdenum ______ _-_ ___ 2 1 ___ ___ ___ ___ Silver ____________ ___ ___ 9 3 ___ ___ 3 1 Total ______ 11 5 51 10 2 12 26 7 trusive bodies. Those circular features observed be— COMPUTER-ENHANCED IMAGERY tween the Denali and Border Ranges faults, some nearly 200 km in diameter, appear to be related to Three types of computer-enhanced Landsat im- volcanic activity. The relation of circular features to agery were used successfully to locate mineral oc- mineralization is not yet clear. currences in the Nabesna quadrangle: (1) color ratio, COMPUTER—ENHANCED TOOL FOR MINERAL EXPLORATION IN ALASKA 197 62°N EXPLANATION “1"” _______ ....... Linear and circular features (dotted where uncertain) 1?OKM 0 50 l l FIGURE 5.—Map of the McCarthy quadrangle showing linear and circular features. (2) standard false color, and (3) simulated natural color. An additional enhancement technique called a “sinusoidal" stretch is being used on imagery of the McCarthy quadrangle. Analysis of these “sinusoi- dally” stretched images is still in progress, but pre- liminary observations indicate that they can supply information not available on other types of images. Digital image processing can be separated into two stages, the preprocessing or “clean-up” stage and the actual image enhancement stage. The output of the clean-up stage is called a data base and is used as input to the different image—enhancement techniques. The preprocessing stage includes noise (striping) removal, haze and Sun—elevation corrections, and a geometric correction (Chavez, 1975). The final step in the preprocessing stage is to mosaic by computer the different images needed to cover the area of in- terest. For the 3—degree quadrangles in Alaska, it usually means mosaicking parts of three or four images. Four data-base tapes are then generated, each containing one of the four Landsat bands that cover the entire area of interest. The several enhancement techniques used in this study can be separated into the following types: con- trast stretches, color ratios, structural and linear en- hancements, simulated natural color, and a ”sinusoi- dal” stretch that maximizes the color variations within an image. The first type of enhancement, contrast stretch, is a standard false color composite made by using three of the four Landsat bands to which linear stretches have been applied. For this study, band 4 was filtered with blue light, band 5 with green, and band 7 with red. The second type of enhancement, color ratios, involves the division of spectral values of one band by those of another band and the application of linear 198 FIRST ANNUAL PECORA stretches to the subsequent ratios. In the Nabesna quadrangle, the color-ratio image was generated by filtering bands 5/4 with red light, 6/4 with green, and 4/5 with blue. Structural and linear enhancement is generated by the use of a high-frequency filter that removes most of the albedo information in order to bring out struc- tural and linear information. In this study, the struc- tural and linear information obtained by this method was added to that observed on the Alaskan image mosaic. Another type of enhancement is the simulation of natural color (Eliason and others, 1974). In this tech- nique, a pixel (picture element) is classified into one of three general categories, vegetation, rocks and soils, and water, using the ratio of band 5 to band 6 after haze removal. Once the pixel has been classified, a different algorithm is used for each category to extrapolate a theoretical value for the blue region of the spectrum. This new band is then used to generate a color composite with colors approximating those that might be seen without atmospheric effects from satellite altitude (fig. 6, p. XIX). A recently developed enhancement technique, the “sinusoidal” stretch, is applied to any three bands used to generate a color composite having maximum color variation (fig. 7, p. XX). Most dissimilar materials show up as dissimilar colors in the composite except where materials have the same spectral response in all three bands selected. The sinusoidal stretch ex- tends multiple-input spectral reflectivity values over the entire output spectral range such that the color changes are gradual for small differences of gray levels. This new stretch enhances not only large spectral differences within the image but also very subtle differences not usually enhanced by other methods. Color anomalies identified on Landsat imagery were found to correspond well with known mineral deposits and geochemical anomalies. A color anomaly is con- sidered to be a variation in tone or color observed on the images that differs significantly from the local background color, indicating a reflectivity difference on the ground. In the Nabesna quadrangle, 120 color anomalies were identified on the standard false color and simu— lated natural color images. Of these, 72 and 69 were identified respectively on the standard false color and simulated natural color images. Twenty-one were identified on both. No color anomalies were identified on the color—ratio image. Of the 120 color anomalies, 39 correspond to known mineral occurrences. Of the 81 that do not, 17 corre- MEMORIAL SYMPOSIUM spond to geochemical anomalies (fig. 8). Of the 64 color anomalies that correspond to neither known mineral occurrences nor geochemical anomalies, 27 were in areas that were not geochemically sampled. Thus, at least 56 (47%) of the color anomalies ob- served on Landsat imagery of the Nabesna quadrangle appear to be related to mineralization. (Many others are thought by Richter [oral comm, 1975] to corre- spond to unmapped altered zones). In addition, 27 color anomalies may be related to unknown mineral occurrences and should certainly be targets for future exploration. Studies of computer-enhanced Landsat imagery in the McCarthy quadrangle are still in progress. Three types of imagery are being used: (1) simulated natural color, (2) color ratio with bands 5/4 in blue, 6/7 in green, and 7/4 in red, and (3) false color using sinu- soidal stretches, with the various bands expressed in different color combinations. One of the advantages of the sinusodal stretch is that snow, a major ground cover in Alaska, does not have to be displayed in white, as it does in standard |00 0/0 120 Color anomalies 67 °/o 81 No mineralized zones 5 3 °/o 64 No geochemical anomalies Not related to 3!? geochemical 3 anomalies or mineralized zones Geochemical anomalies . . 330/0 '4 /o A Mineralized zones 39 1 7 22% No geochemical 27 samples (TARGETS FOR EXPLORATION) RELATED TO MINERALIZATION OR GEOCHEMICAL ANOMALIES FIGURE 8,—Breakdown of color anomalies seen on computer- enhanced Landsat imagery and their association with mineral- ized areas and geochemical anomalies in the Nabesna quad- rangle. 4 7 0/0 COMPUTER-ENHANCED TOOL FOR false-color and simulated natural-color images (fig. 7, p. XX). Photographically, the white, snow7covered areas tend to “bleed” over into adjacent non-snow areas, reducing reflectivity discrimination of rock, soil, or vegetation differences. By superimposing mapped faults in the McCarthy quadrangle over a sinusoidally stretched image with bands 5 in green, 6 in blue, and 7 in red, it was possi- ble to draw numerous extensions of these faults based on linear and curvilinear features visible in the image (fig. 9). The true nature of these extensions and their relation to mapped faults is uncertain. Their actual surficial expressions have not been observed by field geologists (MacKevett, oral comm, 1975) and may be detectable only by high-altitude imaging systems. Additional data on the existence and nature of these MINERAL EXPLORATION IN ALASKA 199 extensions may soon be available ‘upon completion of an aeromagnetic map of the McCarthy quadrangle. Further studies of the Landsat imagery of the Mc- Carthy and other Alaskan quadrangles now in pro- gress will make possible a more thorough evaluation of sinusoidally stretched imagery and its usefulness in mineral exploration. CONCLUSIONS Landsat data have furnished significant geologic and structural information to mineral resource ap- praisal studies of the Nabesna and McCarthy quad- rangles, Alaska. Information of this type, when used in conjunction with geologic, geophysical, and geo— chemical data, is a modern tool that should be con- 62°N U...'-.,I"-I e ‘ '\ \._, l . \ -\./... .. . .}...\ 0.. .0... e. .0. .. . .\ \' . . Q .I‘\ ... . \ "fl. .0 \. .e \ \ Q. \ 0 e. .0 . \‘\ “K ‘ e e '0. '. § \ \ O O .. 0 Q‘; \ Vy) \ I .. '0’. \ \\\__.l / \‘l.°.£ eeee I " \\.e O k \' '°' ,‘%00 O . .. O Y‘V'T'TV‘V) . ‘00 / no 4 '0 . . o. .e _. .e 3. e e O \\ Fr .°e.:° ........ / 4 A) .0. 0 e e \ ‘ r o, ¥ >2 0. . '. : '0. \vN T‘x \ ‘f. e. '0..'. E ‘x’... \ l O. .53.... ‘ .‘ o O. '. '~ 2‘ 90.. I 0.; e e o. 4‘ “at 0 0" ° '0 . . ..3 O 9. 0.... .... .0 \- O.. .. '\\ '3 '51 0.} \_ 'e. ‘ f. 0. \—>‘\v.. W35. 5‘ °. 0. ' K A;\ ‘e \ 0:0 °. \ e ’e. e. \ O \\ “>1 0. \\ EXPLANATION “IV. . 61°N 144°W 141°W _ __ _ . . .. Mapped faults (dotted where uncertain) T T _'_ ...... Mapped low-angle faults (dotted where uncertain) ....,.. . . . Possible extensions of faqu 2 530 1:30 KM FIGURE 9.—-Map of the Mc&rthy quadrangle showing known faults and their possible extensions as determined by linear and curvilinear features visible in the sinusoidally stretched fake-color Landsat image of the McCarthy quadrangle. 200 sidered essential to effective regional mineral explora- tion. REFERENCES CITED Albert, N. R. D., 1975, Interpretation of Earth Re- sources Technology Satellite imagery of the Nabesna quadrangle, Alaska: US. Geological Survey Misc. Field Studies Map MF—655J, 2 sheets, scale 1:250,000. Barnes, D. F., and Morin, R. L., 1975, Gravity map of the Nabesna quadrangle, Alaska: US. Geological Survey Misc. Field Studies Map MF—655I, 1 sheet, scale 1:250,000. Chavez, P. 8., Jr., 1975, Atmosphere, solar, and MTF corrections for ERTS digital image (abs): American Society of Photogrammetry, Fall Con- vention, Phoenix, Ariz. Oct. 1975, Proc., pp. 69- 69a. FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM Eliason, E. M., Chavez, P. 8., Jr., and Soderblom, L. A., 1974, Simulated true color from ERTS: Geology, v. 2, no. 5, pp. 231—234. Griscom, Andrew, 1975, Aeromagnetic map and inter- pretation of the Nabesna quadrangle, Alaska: US. Geological Survey Misc. Field Studies Map MF—655H, 2 sheets, scale 1:250,000. MacKevett, E. M., Jr., and Cobb, E. H., 1972, Metallic mineral resources map of the McCarthy quad- rangle, Alaska: US. Geological Survey Misc. Field Studies Map MF—395, scale 1:250,000. Moody, J. D., and Hill, M. J., 1956, Wrench—fault tec- tonics: Geol. Soc. America Bull, v. 67, pp. 1207— 46. Richter, D. H., Singer, D. A., and Cox, D. P., 1975, Mineral resources map of the Nabesna quad- rangle, Alaska: U. S. Geological Survey Misc. Field Studies Map MF—655K, 1 sheet, scale 1:250,000. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Evaluation of Improved Digital-Processing Techniques of Landsat Data for Sulfide Mineral Prospecting By R. G. Schmidt, US. Geological Survey, Reston, Virginia and Ralph Bernstein, International Business Machines Corporation, Gaithersburg, Maryland ABSTRACT A relatively simple method of digital computer classification of multispectral scanner data was tested at an ideal porphyry copper deposit in a very arid part of Pakistan and was then successfully applied to mineral exploration in an adjacent region. The surface expressions of the already known porphyry copper deposit and the five new prospects discovered in this experiment are all characterized by abundant light- toned sulfate minerals and do not seem to contain much pigmentation by iron oxides. Digital multispectral classification was performed at the IBM Digital Image Processing Facility by using reformatted computer-compatible tapes of one scene. A test area of 55 km 2 was selected in the Saindak, Pakistan, area, which included a well—mapped por- phyry copper deposit. A “supervised” classification table was prepared of high and low limits of accept- able reflectance in each of the four spectral bands for each surface type by using control areas selected from a geologic map. The first classification table was revised several times until it was deemed capable of giving information useful in mineral exploration, even though we expected that many of the rock identifica- tions would be incorrect. This revised table was ap- plied to evaluation of 2,100 km ’-’ in the Chagai District, Pakistan. Of 50 sites classified as “mineralized” by the digital-processing program, 23 were selected as possibly similar to the control area and therefore deserving of inspection in the field. Nineteen of these were visited in October 1974; five of the sites, consti- tuting a total area of 4.7 km 2, contain sulfide—bearing hydrothermally altered rock that is mostly quartz- feldspar porphyry. Parts of all five sites are believed to represent parts of porphyry copper systems. The presence of copper beyond trace amounts has been established at two points, and the prospects are under investigation by the Government of Pakistan. It may be important when using remote-sensing methods in some geologic environments to seek altered parts of porphyry copper deposits that are not specifically enriched or pigmented by iron oxides. Our experiment indicates that simple methods of digital classification of Landsat data can aid in the location of mineral deposits in desert terrain. These methods can complement conventional methods of mineral exploration. A major problem in identifying the sulfide-bearing areas is the overlap of reflectance values of mineral- ized areas with the reflectance values from other high-albedo areas, such as parts of drywashes and, most commonly, areas of windblown sand. In an effort to solve this problem, new classification tables using several classes for each rock type have been pre- pared. Each class has relatively close high/ low limits; several classes are therefore needed to cover the full range of albedo that can be expected from each major surface type, as for example, from similar rock occurring on slopes of different orientation to the Sun. The narrowed reflectance limits increase the possi- bility of discrimination between materials of generally similar reflectance values. Recent tests of these more sophisticated classification tables suggest that we can now achieve better discrimination of mineralized areas than we did in the evaluation study in 1974. Future satellite-borne sensors having higher sensitivity, wider 201 202 spectral range, narrower spectral bands, and greater resolution should help to alleviate the reflectance- overlap problem. Our experiment used high—albedo areas associated with intense quartz—sericite alteration as guides to sulfide mineralization although some of the control areas do contain some hematitic and some jarositic pigmentation. Attempts to use strongly red stained areas in the outer pyrite zone as control areas resulted in no satisfactory delineation of the area of mineraliza- tion. All prophyry copper deposits may not include areas of light—toned sulfate and/or clay alteration; also, not all known deposits have “red-thumbs” or areas of strong red iron oxide pigmentation derived from decomposition of pyrite. INTRODUCTION The concept of using remote—sensing methods at the Saindak porphyry copper deposit predated Land— A FGHANISTAN 66" FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM sat—1 by almost 10 years, for it was discussed by the geologists during field mapping from 1961 to 1966, and some simple spectral tests were made at that time using ordinary color photography and colored filters (Schmidt, 1968, p. 59). Two investigations that have sought a simple method of identifying sulfide deposits using Landsat multispectral scanner data have used the Saindak deposit as a test site because the deposit is large, has well-developed alteration zones, and scant vegetation and is well exposed and undisturbed (fig. 1). The deposit has been mapped in detail at a scale of 116,000 (Khan, 1972), and the senior author was familiar with the deposit and re- gion. Both investigations used image 1125—05545 (Nov. 25, 1972) and the first experiment also used image 1090—05595 (Oct. 21, 1972) and 1124—05491 (Nov. 24, 1972). In the first experiment, false-color composites were visually examined to locate possible favorable areas; in the second, favorable areas were v I [/W Lahore © ) I l /62. . 30°-—*” I ) @Quetm . Saindak I Zahldan© t~ , 7-/“1€19 Miriowu © Mashki Chuh (S'VY‘/ \‘ N°k© bGPAo/v ,1 Km; c“ . x o “ PAKISTAN / ' S IRAN \ “ ch‘ ,/\/ :7 (I ' ‘~ r 7 \/C K-- W Kamcm \\ o 100 200 MILES J Q—A o 200 400 KILOMETRES “/I" L_______—__.J FIGURE ‘l.—Inde>< map showing the Chagai District and Saindak, Pakistan. IMPROVED DIGITAL-PROCESSING DATA FOR SULFIDE PROSPECTING selected by digital computer classification. The near- winter solstice date of these images and the resultant low Sun illumination angle was probably an advantage in the first experiment and a disadvantage in the sec- ond. Both experiments were designed to test relatively simple methods of utilizing Landsat multispectral scanner (MSS) data in mineral exploration. Simplicity is considered important because we are looking for a method that may find wide usage in both govern- ment and private sector mineral exploration applica- tions. For the first experiment, false-color composites were made using NASA system corrected MSS images of three scenes over a very arid and vegetation-free part of the western Chagai District, Pakistan. Later, a color composite was also prepared from MSS bands 4, 5, and 7 by computer processing of the taped data to produce an image that was geometrically corrected for systematic errors and radiometrically adjusted (fig. 2, p. XXI). These composites were visually ex- amined for tonal features resembling the known porphyry copper deposit at Saindak (Schmidt, 1974). Seven areas were selected for prospecting, and three of these were field checked. None of the three con- tained significant mineralization. Later analysis of the results indicated that too few areas were selected by visual examination and that this method must be con- siderably modified for greatest effectiveness in min- eral exploration. Some further testing in which more light-toned areas are selected should be made before the results of this experiment can be considered final. In the second experiment, machine processing was used. Digital multispectral classification was experi- mentally performed on the reformatted com‘puter- compatible tapes of one scene. The IBM Digital Image Processing Facility was used for image correction, image enhancement, and multispectral classification (fig. 3). A 55-km2 area in the Saindak vicinity was extracted and displayed in photographic form using the imagery printer, and a test area was selected which included known porphyry deposits. Areas rep- resenting specific rock types were identified, and their reflectance values were extracted for MSS bands 4, 5, 6, and 7. Provisional classification tables were pre- pared for given geological units; each table was used to classify the test area, and the tables were modified upon comparison of these results with the known geo- logy. The cycle was repeated five times until a classi- fication table was developed that could, it was hoped, provide information useful in mineral evaluation. High accuracy of identification was not considered to be 203 necessary for success of the method, and was not expected. Then the multispectral classification program, using the classification table already prepared, was used to evaluate 2,100 km 2 east of the test area, and a partial field check of the results was conducted. By evaluating the resultant digital classification map, 23 prospecting targets were selected as being similar to the Saindak altered rock area, and 19 of these were visited in the field. Of these 19, 5 localities constituting a total area of 4.7 km 2 contain hydrothermally altered rock, mostly quartz feldspar porphyry. A program for evalu- ation of these new prospects has been prepared by the Government of Pakistan. Despite the apparent success of the 1974 applica- tion experiment, it used classification tables that were relatively simple. While the number of sites falsely classified as mineralized was not unreasonable when compared to other standard field methods of mineral prospecting, we had reason to believe that our method could be refined. This would result in more distinct delineation of mineralized areas, fewer areas falsely classified as mineralized, and perhaps improvement of quality of mapping of other surficial materials. This refinement has been the goal of our continuing experimentation. Acknowledgments—The original color composite experiment conducted by the US. Geological Survey was partly supported by the National Aeronautics and Space Administration. Financial support for part of the digital processing experiment and for the travel expenses of the field checking was provided by the EROS Program, and logistical support in the field was furnished by the Resource Development Corporation, a Pakistan Government corporation. The extraction of reflectance data for several sites in the experiment area by Jon D. Dykstra of Dartmouth College, using computer facilities and algorithm of the NASA God- dard Institute of Space Studies, New York, is grate- fully acknowledged. GEOLOGY OF THE EXPERIMENT AREA The general geology of the western Chagai District is portrayed on 1:253,440 photogeologic maps (Hunt- ing Survey Corporation, Ltd., 1960). The rocks of the region consist of marine and terrestrial sedimentary and volcanic rocks of Cretaceous to Holocene age, and a few shallow intrusive bodies. The influence of widespread volcanism is continuous in the geologic record from Cretaceous time to the present. The high- est mountains in the area are the dormant volcanoes Koh-i-Sultan in Pakistan and Kuh-e-Taftan in nearby Iran. FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 204 .EoEtwqxw 9: E «65.5th mirmcm 22v vcm wcfimwuoa ———4———— 3&5 B 52:53 gisola $50: _ _ _ _ _ _ # Sac mw_>_ «0 5050:;me .mbomamEDE o mmHDmEOU meEm 3am mmzmmczowamoi 0. mmzm 882mm I 22”. Sflwm I x850 22“. a 3.3mm c058: _ 4320 $3.3m o mmtm _ Pm>4n__mc< Ucm cosmocamfio 3:032:35. _ 33mg. c028: _ $35 2mech a 33.4 m: Err _ 3:53 o _ (5.5 momDOm ._.w>._._8m=u 33m: 98 25:80. umprsc of. .ufliam mm? “.052: cozmuEmmflu 45R? 9: 2ch we; 9: “En xmvEmm .3: 35 3.5.8 «5 9:305 63min; 5:35 Emmcu 5983 m5 *0 ~98 :20 UEmmEIxmuEmm 9:.I.v 550: 9.8a meu:w._ X momm=_> © ZOFc wetsofi cu .wcv :_mD-_-50v_ .0 «8.5 new ca .0 92: cozmuEWmEUILm mEDUE II In... III-III: .I. .232:- z. 1—. ._x;:- I: -x IIII:I ._xxx .1: .4. ..x I I“: II: 3. ._.n.__:.:__..I.._...._._....._.._..._.._.,._..III-1.... o.I-x: .IIIx III: .I .I... : o : xx: . xx xx x 1...: xxx. . xx I _II .1. .z. z _. __. . _ :_._. __.; 1.2.... . _ .22... .I 3.... I 32.7.: Ix I.._I . 7" 3. I. x :- .III_ .1 II. _ I _ I: _ : .; 5:... . x I 1.7.1.4... .2: 1... x I. z :22: I I. I. I fix: 1 .III: _-,.I 1:; __. ._ 2:... . : T: I ._ .- I... 17 __x_ _xnx 1:1 3.: u z. I —: o o: _z .. I... ._ 2. _ _ _ _ v... . _ _. I 1... I II II. 5._x.:.:x:.. urioIII _ III: I .:..:_IIqu it I a. .3. 1:7. :_u_ ~ 2...... I. r __ _ ..__::....zx....._: . 1.15.1: 5...: 1... . f I... _I x. :21... .1... .5. :31: 1.... __.. .. Z! — o _. L11... 2. . I x 2:: __..i::, .o z :5??? 4 I 1- . III... . .... ......x. ._ h. ... .. I. .2. :::.:: _ II III 2 I I .. In .. l. . I: _ .: z 2; . ._ . .. .. : 1:... 52...... .5 I : 3.x... .7 :7 _ __ . 1. II III I. _. o.._:.:.._. . I... _. z. 1: ~:......:::.. I .I. _ _ _I......:. . x... . I. n: - ° . 2.... 3:33:21 I..IIIIII.::. raf- I I. .2 2.... $2.1: . .. III-II. II o o... 5...; .I.................... ....._ 3. . IIIIIIIII o. x_::.x.:... ._ ... ... IIIIIIII .5 .._._...x. .. __ ._ .2 .22.: L: . ”.Z_.__: w IIIIIIII L I .. T I .IIIII .. .I .2. Z. I.III o. .IoI . — I III-II I ~ I _ __.._I .III III-I Io oooI . H I . _. ti... . I.IIIII.I _ _ Z I . . _ I :_. I III I I IIII I .. . . . — 7o... IoIoZIIII .oo .......u_ .a. ,0 00.0 .9 0.0600000000 coo. o no . II IIIII. I I; . .. . I u IIIoovo I a II. . .2 u I n. _ .I :— o . __. . . I: I I I I. .. I...I.¢¢I¢I.:o .IIIIIxIIL : _ _. . Io..:::I.: :. Iaxxon. . xv. I :1. . I. IIIu. .. __. IxxI o II ....._...t.. O. _ —>.__Z_.- :_;III II. IIuII III. I. .2 . . . 11:13... 1:... .II: .I II... In... IIIIIII I III ..:. I . . I‘voo..: x xx I . III... .I I I... . I IIII~.III nun I _ I . 5. u . I I III o o _ . . I III I Z. . _ oo .o . II... "I, n o I. on. I _ . . I. II. 7—1... :III: 1 7:31... .— I .I 5.2.:- I... v.21: I _ it. I __.: _I..._I..::IIIII _o .1: . I . .21.: I... . . :1::.... :II .o. L I .. . .v.k1 I . h ‘o.XtY :1 u .:.va?__2 I .33.... 1.51.. I run...» I ._ IVE: XI... I 2x... __. I .XIIIII :1. 1.. 7 I . .I... .2: 2...: . x: . _ .I _ IxI II v..._ 2 .3. Ix: 2: ::~.:.. III-:33: . _. II II .I . I z I .x: .u... xxx; x . . I. T.- x.x _ __.... I I I .- ... 3.1L. .— x. I .x 33- I. 7...... I. z . __.— _¢I I 22-: _ _ I. _ _ LIX—xxx; I _ IIIZxIXI .. 1:5}: 2 :III 0. :u.z..xx _ . :x I... x x x _o .I- I . I _ L: ..I I. u _ II II: _ I _ I. o: I .5.- xx.. Xxx: . . . _ . x... :. Y... I:— xx... I .2. _ .. :1..." . I . I .13... x x:- o I X: .I... . 2.x..leII x...-x XoIII. III. III-III: .2 :21. v... 3. . I... o... .. ¢.IIIu.x:. :- 22... xx 2. I I x. 1...... II o .. I III..II...I.: 1.x... 5;: :1: I: I: o _ III coo. 4 II. o. :20. :oII.m . I .x x. .8:- x. x x 5...: a. . III. I Inn-Io o. o 210 FIRST ANNUAL PECORA graphs and in the field. As part of the field check all anomalous areas were first examined on stereoscopic pairs of 1:40,000-scale aerial photographs. At this point, it was possible to reject seven more areas as probably related to windblown sand. Nineteen sites were examined in the field, and four desirable sites were not reached in the field checks. Five sites were found to be extensive outcrops of hydrothermally altered sulfide—rich rock (figs. 4 and 5). Two additional sites contain altered rock with some sulfide but seem less attractive for prospecting at this time. EVALUATION OF RESULTS Classification using table “B”.—In the evaluation using table “B," mineralized rock is believed to have been generally identified correctly, but much other light-toned surface was also classified as mineralized. Alluvium was mostly classified accurately, but some gravel fans derived from light-toned rocks were not. Perhaps only 25 percent of dune sand was correctly identified by means of table “B," most sand having been classified as mineralized rock. Classification of areas of dark rock outcrops and several types of lag gravels yielded different combinations of classes 10, 11, and 12 and made it possible to differentiate among some of these materials, and some of these differentia- tions were made among surfaces having surprisingly subtle differences. In classification table “B,” four of the main surface types are each represented by a high and low relia- bility class. The high reliability or more restrictive class has reflectance limits in at least one band that made it exclusive from all other high reliability classes; the secondary class has reflectance limits that overlap the limits of at least one other class. The classes of two reliability levels have not been used in later tables. The classification map made by using table “B” is compared with a geologic map (fig. 6). In building table “B,” as the acceptable reflectance limits of a class were narrowed, fewer matches were found. In classes 1 and 2 (mineralized quartz diorite, table 1), for example, if we chose limits that included most pixels in the training areas and also many pixels in all the areas of known mineralization, we also got false classifications in dry-wash material. The limits were subjectively adjusted wider or narrower so that enough points were classified correctly in the areas of known mineralization to call attention to those areas and the false classifications were reduced to an ac- ceptable level. Because revisions to the tables are time-consuming and successive revisions generally MEMORIAL SYMPOSIUM achieve diminishing improvement, table “B” as used in this experiment was revised only a few times. The three dark-surface classes were established on the basis of small samples of particular rock forma- tions in training areas, but the classes were very un- satisfactory in discriminating among the three formations. The classes would have been abandoned except that various combinations of the three classes delineate areas of dark rock outcrops and lag gravels. The printout of class 11 (symbol #) in particular re- sulted in abundant pixel groups just inside the peri- meter of the propylitic zone at Saindak (no similar propylitic zone in volcanic sedimentary rocks, as at Saindak, was found near the new prospects). Classification using table “S”.—The data collected in the field check made it possible to evaluate the accuracy of identification of all of the surface types, not the mineralized rock alone, and to revise the tables to improve accuracy. A new style of table evolved, having several relatively narrow classes spanning the expected albedo range for each rock type but no high and low reliability classes, resulting in classification table “SSD” (table 2) and finally classification table “8” (table 3). The limiting values for each class span relatively narrower ranges than those in table “B” and more classes are required. The use of several classes for a range in albedo from similar surface types is designed as a simple way to compensate for variations in reflectance angles, espe— cially those caused by southeast- and northwest-facing slopes, and also albedo-related systematic changes in color balance of similar surfaces, regardless of their cause. Classification using table “SSD” was performed to reevaluate part of the application area. An area of 170 km2 was tested that includes four of the new prospects. The results are very encouraging. The area tested using table “SSD,” included 13 of the groups of “mineralized”-class pixels originally evaluated using table “B.” Of the 13 pixel groups, 7 had been selected as possibly mineralized rock and worth field checking, 5 were subsequently visited during the field check, and 4 were found to contain some mineralization. Evaluation of the same area using table “SSD” resulted in the delineation of only 5 areas as “mineralized,” 4 of these were known from earlier field checking of table ”B” targets to have mineralization, 1 was identified in the earlier evaluation but not chosen as deserving a field check, and one was not delineated by the table “B” evalua- tion. (The latter two sites, not chosen for checking in the earlier study, have not had field examinations). One mineralized site identified using table “B” was omitted by the table “SSD” evaluation. Table “SSD” IMPROVED DIGITAL-PROCESSING DIGITAL CLASSIFICATION MAP DATA FOR SULFIDE PROSPECTINC 211 GEOLOGY MAPPED IN FIELD =xxxx=xn ll“? >e= =:—==Ké.x_*j~-==-=— — -_‘_-== = x: = _ _________ xxxxxxx == === ._‘__‘_‘_‘_ ::__ x _____ — _"— _: —— = —= — = - 7 “m p.- _— f _ — _= 9 === == — \(XX‘WW XXXX _ \xxxxxxx‘ . __f-__ XXXXXX \m --—_-: = _ XXXX £55: *- \xxxx ==x== xxxx :== \XXX \XI := i‘ = I, = i: ‘32:: N XX = a: = X X= o X XX: g: ==X-= =XXX=X= = ==-XX =X== = == =X X = X = == = :Xllflfl = x ==: === =XX X X 'x = X X Ix = X |\x = — -== X = i: “\X = ;= Limit of pyrite zone = Geology adapted from S. N. Kuhn, I972 EXPLANATION DIGITAL CLASSIFICATION GEOLOGIC 0 5000 FEET MAP MAP I I ff: AIIuvium 0 2 KM .. _ I J Ell! fl Minerallzed I! I” quartz diorite Unpatterned areas are #X *ng x x x x Pyrite-rich rock other rock types ##XX it X X X X FIGURE 6.—Comparison of digital-classification map made by u sing table ”B” and geology mapped in field, Saindak porphyry copper deposit, Pakistan. does a much more effective job of separating mineral- ized areas from other light-toned materials, largely eliminating the need for human evaluation of a large number of “false” tonal anomalies. Omission of one of the best areas of mineralization points to the need for slight revision of the mineralized rock classes. Revisions to table “SSD” have yielded table “S" which includes classes to cover sands of higher albedo and dry-wash material of lower albedo than were in- cluded in table “SSD,” classes for dark pebble desert and unmineralized felsic rock (rock types not included in earlier tables), and modified reflectance limits for part of the mineralized rock classes. Only a few pre- liminary tests have been made of this new table. SUMMARY Visual evaluation of false color composites and a relatively simple method of digital classification were tested as aids to mineral exploration by using areas of known hydrothermal alteration and mineralized and altered intrusive stock as control areas. Both methods seem to be useful exploration tools, and greatest advantage is probably gained by using them together. The simple classification method was applied to evaluate 2,100 km 2 of area regarded as having a high potential for deposits of the porphyry copper type, leading to the discovery of five sizable areas of hydrothermally altered rock containing abundant sulfide, disseminated or in stockwork veins. Although 212 the classification method used was relatively simple and unrefined, and 14 prospecting targets proved to be false leads, the number of sulfide-bearing areas identified was outstanding, and the falsely classified areas were not so many as to require an unreasonable amount of field checking. Further experimentation in- dicates that better discrimination of mineralized areas than we used in the field test can be achieved. Our study indicates that simple methods of digital classi— fication of Landsat data can aid in exploration for mineral deposits in desert regions, and these methods can complement conventional exploration techniques. REFERENCES CITED Ahmed, Waheeduddin, Khan, S. N., and Schmidt, R. G., 1972, Geology and copper mineralization of the Saindak quadrangle, Chagai District, West Paki-' stan: US. Geol. Survey Prof. Paper 716—A, 21 p. Bernstein, Ralph, 1974, Scene correction (precision processing) of ERTS sensor data using digital image processing techniques: Third Earth Re- sources Technology Satelliteel Symposium, v. 1: FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM Technical Presentations Section B, NASA SP— 351, Dec. 10—14, 1973, p. 1909—1928. Hunting Survey Corporation, Ltd., 1960, Reconnais- sance geology of part of West Pakistan; a C0- lombo Plan Cooperative Project: Toronto, 550 p. (Report published by Government of Canada for the Government of Pakistan). Khan, S. N., 1972, Interim report on copper deposit of Saindak, Chagai District (Baluchistan), Paki- stan: Pakistan Geol. Fdrvey, Saindak copper report no. 1. Lowell, J. D., and Guilbert, J. M, 1970, Lateral and vertical aIteration-mineralization zoning in por- phyry ore deposits: Econ. Geology, v. 65, no. 4, p. 373—408. Schmidt, R. G., 1968, Exploration possibilities in the western Chagai District, West Pakistan: Econ. Geology, v. 63, p. 51—60. 1974, The use of ERTS—1 images in the search for large sulfide deposits in the Chagai District, Pakistan: US. Dept. of Commerce, Natl. Tech. Inf. Service, E74—10726, 38 p. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA A Deeper Look at Landsat—1 Images of Umiat, Alaska By Andre F. Maurin Compaignie Francaise des Pétroles, Paris, France, and Ernest H. Lathram, US. Geological Survey, Menlo Park, California A BSTRACT Discussion of subjective evidence for visual identi- fication of previously unrecognized lake lineations shown on Landsat—1 (formerly ERTS—1) image 1004— 21395 (Fischer and Lathram, Oil and Gas Journal, May 28, 1973) led the senior author to undertake more precise geomorphic mapping. The use of a new statistical tool, the Texture Analyser, has provided a mathematical control for lake lineations as they were first sketched by Fischer and Lathram. The structural pattern has been developed into a full map of the area covered by thaw lakes. Enlarged pictures of an almost similar image taken by Landsat—1 earlier in the spring confirm the meteorological origin, but with an underlying tectonic influence for the most con— spicuous (north) trend in lake shores. A second trend, east-northeast, is demonstrated to result from a com- plex elliptical pattern caused by numerous closed structures. Further details are given on the potential of the Texture Analyser for cuesta mapping and fault enhancement through discrimination of certain values of grays, although this last method is not as accurate as other computerized programs such as gray stretch- ing. Its instant reading makes it very valuable for geomorphic applications in numerous domains. INTRODUCTION The Arctic Coastal Plain of Alaska is characterized by thousands of thaw lakes. Many are oriented; elongated rectilinear shorelines trend approximately N. 9° W., nearly perpendicular (~850) to the prevail- ing wind direction. Most authors ascribe the orienta- tion of these lakes to elongation of shorelines through wind—induced erosion (Black and Barksdale, 1949: Carson and Hussey, 1962; Price, 1968; Black, 1969). In many of the elongated lakes, short shorelines are straight and orthogonal to long ones, suggesting to Rosenfeld and Hussey (1958) a possible structural control. In addition, in the Umiat area, individual lakes are aligned in a direction perpendicular to the elonga— tion, as are linear areas betweeen lakes, and some groups of lakes, lines of lakes and lines of interlake areas form oval or circular cluster-shapes. These ob- servations led Fischer and Lathram (1973) to suspect that the lake orientation results from control by con- cealed structural patterns. The features noted above are consistent with pat- terns noted in other areas and ascribed to block- faulting (Bostock, 1948; Price, 1968). In particular, the alignment of individual lakes and the linear pattern of interlake areas seem more logically to be the result of structural control, as wind influence generally ini- tiates a random distribution pattern, acting only on individual shapes. With respect to oval or circular cluster shapes, it is interesting to note Bostock’s illus- tration of an “oval basin shape developing” within an area of oriented lakes strongly affected by block faulting (Price, 1968, p. 794). The purpose of this paper is to present the method and preliminary results of a mathematical analysis of lake distribution and orientation in northern Alaska (fig. 1) designed to provide an objective and rigorous basis for geomorphic interpretations and for infer- ences of structural control. The method combines the regional (34,000 km 2) view and orthoplan representa- tion of Landsat images with the statistical analysis capability of the Texture Analyser (Leitz-Ecole des Mines de Fontainebleau, France), originally developed 213 214 ARCTIC OCEAN I Coastal Brooks Range FIGURE 1.—lndcx map of northern Alaska showing physio— graphic provinces, the area covered by Landsat image 1004— 21393, and the area of oriented lakes previously studied by Carson and Hussey (1962). for analysis of microscopic textural fabrics. Landsat—1 image 1004—21395 (fig. 2), the basis for Fischer and Lathram’s interpretations, is admirably suited for the purpose. All of the geomorphic features described above are included on a single scene. Further, the band 7 image (0.8—1.1 ,lm) presents a high contrast between land and water and a sharp definition of the form and distribution of lakes, because this part of the light spectrum is absorbed by water. The method assists in geomorphic analysis by as- sessing the statistical value of each character and provides a more firm basis for naturalistic interpreta- tions but, like other such computer—like systems, should not be substituted for human interpretations. Computer enhancement of the quantitative values produced by Landsat multispectral sensors has pre- viously been employed for several purposes. Values of light reflectance, grouped by computer as to type of natural or manmade feature having a character- istic, reflectance or group of reflectances, have been used to determine land cover by class, and type and species of vegetation (Ellefsen and others, in press; Hall and others, 1974). Ratios, or augmentations, of reflectances in different wavelength bands have been utilized to identify rock types, alteration products, and soil types (Rowan and others, 1974; Albert, 1975). In addition, gratings, with parallel or concentric ruling, FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM have been used to augment the appearance of linear or curvilinear features and to determine qualitatively the trends and patterns of Earth features. The method here employed permits the objective measurement of the trend of linear or curvilinear features and of the form and shape of Earth features and of their quanti- fication by ADP methods. In short, it provides a sta- tistical basis for geometrical description of Earth features, a fundamental step in the study of structures and their significance in the exploration for mineral and mineral fuel resources. MATHEMATICAL APPROACH The mathematical basis for and geometric rationale of the use of structural elements in texture analysis is given in Serra (1972). The description of a prototype Texture Analyser and its use is given by Klein and Serra (1972). Since their paper, an operational model has been built from the prototype. Most applications for the Texture Analyser and previous quantifying instruments have been directed toward study at micro— scopic scales (grain distribution in cement and in ground ore, ore mineralogy and metallurgic study in polished section, porosity distribution, cytology, etc). The application reported here represents a quantum jump in the scale of study. For clarity some mathematical terms employed herein require a simplified definition: 1. Erosion—an elementary geometrical operation for filtering, resulting in the “smoothing” of bound- aries (grain contours. patches, etc.) such as sup- pression of peaks. (Presumably high band-pass spatial filtering, ed.—) 2. Dilatation.—complementary process to erosion, addressing intergrain geometry (e.g., porosity). (Presumably low band-pass spatial filtering, ed.—) 3. Opening—the result of both erosion and dilatation processes designed to improve, through simpli- fication, the structural meaning of a geometrical pattern. Because of the possibility of varying the size of the hexagonal element (the geometric figure employed to analyse patterns), such a pattern provides great flexibility in observation. 4. Closing—the reverse function of opening; e.g., a closing to simplify grain pattern is an opening with respect to the porosity pattern. GRAY DISCRIMINATION This is the basic step in use of the Texture Analy- ser. As the analyser makes statistical counts of white A DEEPER LOOK AT LANDSAT—1 IMAGES OF UMIAT, ALASKA 215 FIGURE 2.—Landsat image 1004—21395, band 7, of the Umiat area. Lakes are in black. or black dots, contrast in the image must be high, ever, increased contrast was obtained by making a and intermediate gray levels suppressed. The un- simple ozalid print of the positive transparency (fig. 3) enhanced positive transparency from Landsat band 7 and by making a high—speed Ektachrome slide for provides sufficient contrast for a direct read-out. How- microscopic use. 216 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 3.—Ozalid print of positive transparency of Landsat image 1004—21395. band 7, showing contrast enhancement and location of detailed figures. Two dashed circles represent the areas in which average lake lengths were measured automati- cally. Three small heavy circles are the three zones studied to produce the direction diagram in figure 4. AVERAGE LENGTH OF LAKES puting the average length of linear groups of uninter- rupted black dots recognized by the analyser along An automatic survey of the length of lakes was selected diameters of circular fields of observation. made on the analyser. This was accomplished by com- Two large circular areas were surveyed providing a A DEEPER LOOK AT LANDSAT—1 IMAGES OF UMIAT, ALASKA count of more than 100 lakes. Trend A (fig. 3) was chosen arbitrarily parallel to the major trend in lake shores and dry paths (linear interlake areas) with a similar direction. The average length of lakes along Trend A is 1.426 km. A measurement perpendicular to Trend A yields 1.406 km, while counts at 20°, 40°, and 600 from Trend A give 1.102, 1.147, and 1.246 km, respectively. (Because this computation involves a complicated statistical function, requiring nearly 2 hours for hand calculation of these five measurements, a hardware connection is under development to permit instant computation.) TRENDS OF PREFERENTIAL ALIGNMENTS A reduced photographic slide of the lake area was observed through the microscopic apparatus using programmed rotations of the “electronic rotating stage.” Six circles were analysed at high speed. Each circle was studied by “Hit or Miss Transformation" (Serra, 1972) of black versus white lengths, every 2°. The computation for one circle required about 60 seconds. The direction diagram (fig. 4) shows these measurements integrated in 4° segments. Out of the six circles, three give apparently random diagrams (not represented in fig. 4), but three seem to give rather significant results. Figure 4 is a diagram composed from mixing the three last results. This 217 diagram is qualitative and provides two families of trends. The first family is composed of two clear trends which form a narrow angle bisected by Trend A (figs. 3 and 4). The second family is composed of a number of lesser trends, subtending a greater angle (40°), bracketed by two major trends; the bisectrix of the second family we designate as Trend B. It should be stressed that in no way is this rose diagram influenced by the length of lakes. Following Serra’s (1972, p. 95—96) consideration, these trends represent the pro- portion of black (lake) versus white (land) segments in a given direction. By the same token, this measure- ment provides an indication of preferential rectilinear alignments either of lakes or of paths. At this point in the investigation, it is clear that it is not possible to apply to the Umiat area the conclu- sions of Carson and Hussey (1962) resulting from their field study near Point Barrow, north of the Umiat area. Normal winds could initiate a narrow forked pattern such as that of Trend A by interference in wind directions (Carson and Hussey, 1962, figs. 3, 4, and 5), but Trend B and the anomalous length com- puted in average length counting must originate through another process. It is logical to expect a greater influence of tectonics in the Umiat area as compared to the Point Barrow vicinity, as it is com- mon for structures to decrease in intensity from the FIGURE 4.—Direction diagram computed from the three heavy small circles of figure 3. A and B are significant trends dis- cussed in the text. FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 218 @352: m U55 26:3 3 CcoEamccz .3250; Coucmzcocs V0 m_m3mc< .EEmQQm m_ Homto cm can sic: copra: >2 :32? 9:. 4:3: :5:___>:C :33: or: £32; >2 CUEiCE i 53?; LP: r: < £53355 .0 A DEEPER LOOK AT LANDSAT—1 IMAGES OF UMlAT, ALASKA inner to outer zones of foothills and toward the plains, as in northeast British Columbia. Observing trends along the Ikpikpuk River extend- ing some 15 miles southward from the Point Barrow area, Rosenfeld and Hussey (1958) inferred a struc— tural influence, and suggested an additional east- northeast lineation (i.e., Trend B). It is obvious that the smaller scale and wider observation offered by Landsat images allow instant perception of a greater span of anomalous trends. HEXAGONAL TRANSFORMATIONS Klein and Serra (1972) describe the Texture Analy- ser as an apparatus able to “measure forms, sizes, and patterns.” Sizes and forms have been partly investi- gated in the two preceding sections. Hexagonal logic allows the analysis of the structural pattern (this logic may also be introduced into the first two steps). The circular areas shown on figure 3 were analysed through hexagonal transformation (figs. 5 and 6). Each figure exhibits the control image, after gray discrimina- tion (electronic suppression of intermediate levels of gray) of the exact area of survey. Then, this image is transformed into a set of patches made of individual hexagons (fig. SB). These patches are the result of a mathematical erosion (Serra, 1972) determined by the size of individual hexagons (notice that the hexagon is the most complex polygon which allows a complete coverage of a bidimensional surface without overlap; such an overlap would be embarrassing for statistical purposes). The next processing is opening of black patterns. Klein and Serra (1972), and Serra (1972) demonstrate the structural value of such a mathemati- cal process, and complete references to earlier works are contained in their papers. Figures 5C and D exhibit an ENE—WSW trending linear set and a near-circular feature made of two crescentic anomalies of possible morphological value. Such circular or elliptical lineations of lakes deserve a lot of attention. Coupled with crescentic lineations, they are believed to be the subtle expression of a mini-cuesta system, e.g., erosional breaks formed on very gently sloping beds within the Pleistocene Gubik Formation. Figure 6, converted into the hexagonal pattern and opened (comparable to 5D), shows a crescentic imbri- cation of prime significance in geomorphic mapping. It is located close to the most northerly Foothills anti- cline axis (Lathram, 1965; Fischer and Lathram, 1973, fig. 3) and provides a basis for geomorphic mapping. 219 FIGURE 6.—-Another area studied using the erosion and opening process. A, control screen without gray discrimination. B, eroded and opened cluster zone. Analysis of unenhanced positive transparency of Landsat band 7 image. CEOMORPHIC MAPPING We believe the foregoing mathematical data indi- cate a structural origin for both Trends A and B (figs. 3 and 4). With respect to B, the angular array of high values, broader than the array for Trend A and with many intermediate trends, may originate from a differ- ent tectonic cause than that of A. The Trend A array may be caused by intercepts with narrowly diverging rectilinear paths or lines of lakes, whereas the B array may reflect numerous broadly divergent intercepts controlled by the curved outline of the ellipses. Logi- cally, intercepts with a rectilinear zone would be nar- row, but intercepts tangent to the outline of an ellipse would be broadly divergent. By the same reasoning, 220 the more narrow angular display of Trend A may represent both a wind—induced preferential shore direction and a structural element. This element may have lesser impact on morphology than the ENE— WSW trending ellipses. Unfortunately, the Texture Analyser is not designed at present to analyze and compare curved elements, but only to compare and measure straight ones. A pro- gram to analyze and compare the radii of curved image features is being developed and will be opera- tional in late 1975. Accordingly. conventional methods of geomorphic interpretation, guided by the foregoing data and rea— soning, were used to develop a structural map of the area (fig. 7). Using the surface anticline (Lathram, 1965) closest to the Coastal Plain as a standard, de- flections and bends of the north—flowing Ikpikpuk River are considered to be produced by anticlinal plunges, with apexes located east of the river. West of the most southerly bend in the river on the Coastal Plain, three sets of lakes arranged in three crescentic patterns appear to be the remnants of the bend as the river was displaced erosionally to the east. ENE—WSW lineations between lakes (paths) are clearly bent and follow lines of elliptical anomalies. These lines are considered to be anticlinal axes; the chaplets of apexes create elliptical patterns. The FIGURE 7.—Structura| interpretation by conventional geomor— phologic study using outcropping anticline as a standard and the elliptical "path” concept. FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM three more northerly axes are drawn on less sure grounds but are suggested by the lines of ellipses. The long clear lineament along the east bank of the river (fig. 7) is one of several linear anomalies lying parallel to Trend A. However, its special bearing on tectonic interpretation deserves some discussion. Axes on the west, aligned along the centers of cres- cents, are displaced from those on the other side of this lineament. The lesser number of lakes and ellipti- cal anomalies on the west side may indicate that this is a down—thrown block. However, this feature is in— terpreted herein as a transverse fault. Other linea- ments parallel to this fault or Trend A could also represent minor fracturing. We note that the linea- ment shown on figure 7 is an extension of the trans— verse trend of the Maybe Creek “low" recognized by Brosgé and Whittington (1966, p. 583). ADDITIONAL OBSERVATIONS FROM THE LANDSAT IMAGERY Additional observations have been made which substantiate our interpretations and provide addi- tional support for the mathematical approach. At the same time, the incredible amount of information dis- played on Landsat imagery is demonstrated. OBSERVATIONS ON METEOROLOGIC EFFECTS ON THAW-LAKES About 20 percent of the lakes investigated in this study are oriented according to the definition of Car- son and Hussey (1962) and Black (1969). This permits a thorough test of their concepts. On Landsat—1 image 1004—21395, the east-west banding of clouds provides an instant meteorological indicator of the wind direc- tion they derived from many historical records. Fur- ther, an almost similarly located image (1345—21344, band 7) taken by Landsat—1 on July 3, 1973, (almost 1 month earlier than the previous one) exhibits par- tially frozen lakes (fig. 8). Some of the elongated lakes contain mushroom-shaped ice cakes which substan- tiate the observations of Carson and Hussey (1962) as shown on their figures 10a, b, and c (Ikroavik Lake) and support their meteorological interpretation. The symmetrical pattern of the ice cakes and the bulbous shape of the distal wings indicate a dominant wind blowing westwardly and countervrotational lake cur- rents as they suggest. The repetitive views of Landsat provided a time dimension they were not afforded in the Field, i.e., Landsat observations every 18 days and their field work once per summer. A DEEPER LOOK AT LANDSAT—l IMAGES OF UMIAT, ALASKA FIGURE 8,—Two selected areas of Landsat image 1345—21344, band 7, with common orientation showing mushroom ice cakes (arrows) in black lakes. APPLICATION OF THE TEXTURE ANALYSER TO STRUCTURE IN THE FOOTHILLS Although the topic of this paper is centered around the thaw-lakes area, some observations in the Foot- hills south of the Coastal Plain will’help understand the anticline-transverse faulting pattern in the Coastal Plain. Once again, information is derived from Land— sat—1 image 1004—21395, proving the wealth of in- formation provided by such small-scale imagery. Near the big bend of the Ikpikpuk River, the map of Lath- ram (1965) indicates a number of anticlines within the area of figure 9 (location on fig. 3). After gray discrimination, a further process elim- inates, through thresholding, some specific grays from the control scale (right of fig. 9B). This helps in de— lineating dip slopes and displacements by enhancing shadows and triangular facets. Hexagonal transfor- mation with minimal erosion strongly enhances such displacements. Cross faults appear very sharply (fig. 9C). It is interesting to note that most of these faults are parallel to Trend A. Analysis of the Foothills zone on image 1004— 21395 exhibits two groups of faults (with dextral and sinistral offsets) belonging certainly to a transverse fault system. These observations support the struc- tural map (fig. 7) of the thaw—lake area. 221 FIGURE 9.—Experimental processing of part of the Northern Foothills shown on Landsat image 1004—21395. A, Control image See figure 3 for location of surface anticlinal axes. 8, Same image after thresholding of penul- timate gray level of the analyser, scale shown on right margin without level 1 (white). C, Hexagonal transforma- tion of the black and white information of figure B, with minimal erosion. Note that in C shadows which may be cuestas are more strongly enhanced. Faults are enhanced and clearly shown by linear disruptions (see arrows); in some cases offsets along the faults can be quantified by the analyseri 222 FIRST ANNUAL PECORA CONCLUSION The morphologic value of both Landsat imagery and the use of the Texture Analyser have been dem- onstrated. In such an area as the Arctic Coastal Plain, where technical problems of resource exploration (ecology, costs, etc.) are very stringent, the method can pro- vide a maximum of information without environmen- tal stress, can be evaluated with a minimal amount of conventional checks, and provide guides to the opti— mum location of such checks. The structural map may be greatly improved and become an actual opera- tional tool. These results are sufficiently encouraging to pursue the same approach west and eastward. To the east the Prudhoe Bay area will provide the type of check- ing and control required. Precise angular measure- ments are needed in order to separate tectonic ef— fects from those of dominant winds in those areas where both directions are not parallel or perpendicu- lar, and work on these is proceeding. The use of only two sequential Landsat images of the same scene has permitted observations to be made in three domains, the conciliation of the two main genetic theories for the origin of thaw-lakes, and a suggestion of even more fundamental ideas: 1. The Foothills domain in which faulting is normal to anticlinal axes. 2. The thaw-lake area in which nearly perpendicular linear trends point toward a structural pattern, and ice cakes in many of the lakes conform to meteorological theories of the formation of oriented lakes. 3. The cloudy zone in which banding is parallel to the general direction of the mountains. From these observations we believe that meteoro- logical conditions, mainly dominant winds over the Coastal Plain, are a topographic consequence of the Range and Foothills system. Further, the structural pattern within the Plain is a weak but subsequent image of the tectonic style in the Foothills. If these conclusions are valid, then a perfect similarity should exist between wind-induced shore patterns and cross faulting through an anticlinal area of great petroleum potential. ACKNOWLEDGMENTS Publication of this paper has been authorized by Compagnie Francaise des Pétroles (Paris and Cal- gary). We thank Messrs. Klein and Serra for kindly providing mathematical precision and Centre de Mor- MEMORIAL SYMPOSIUM phologie Mathématique of Ecole des Mines, in Fon- tainebleau, France (Co-owner with IRSID, of the patent on the Texture Analyser, manufactured by Leitz, Germany), for experimental access to the Tex- ture Analyser. J. M. Fonck and M. Riguidel, CFP, Paris, provided much information on previous and future use of the machine. REFERENCES Albert, N. R. D., 1975, Interpretation of Earth Re- sources Technology Satellite imagery of the Nabesna quadrangle, Alaska: US. Geol. Survey Misc. Field Studies Map MF—655J, 2 sheets, scale 1:250,000. Black, R. F., 1969, Thaw depressions and thaw lakes, a review: Biul. Peryglacjalny, no. 19, pp. 131— 150. Black, R. F.,‘ and Barksdale, W. L., 1949, Oriented lakes of Northern Alaska; J. Geol, v. 57, pp. 105—118. Bostock, H. S., 1948, Physiography of the Canadian Cordillera, with special reference to the area north of the fifty—fifth parallel: Geol. Survey Can. Memoir 247, 106 p. Brosgé, W. P., and Whittington, C. L., 1966, Geology of the Umiat—Maybe Creek region, Alaska: US. Geol. Survey Prof. Paper 303—H, 638 p. Carson, C. E., and Hussey, K. M., 1962, The oriented lakes of Arctic Alaska: J. Geol, v. 70, pp. 417— 439. Ellefsen, R., Gaydos, L., Swain, P., and Wray, J., New techniques in mapping urban land use and moni- toring change for selected US Metropolitan areas: an experiment employing computer—as- sisted analysis of ERTS—l MSS data, in Internat. Soc. of Photogrammetry, Commission VII Sym- posium, Banff, Alberta, Oct. 1974, Proc. (in press). Fischer, W. -A., and Lathram, E. H., 1973, Concealed structures in Arctic Alaska identified on ERTS—l imagery: Oil and Gas J., v. 71, no. 22, pp. 97— 102. Hall, F. (3., Bauer, M. E., and Malila, W. A., 1974, First results from the crop identification tech- nology assessment for remote sensing: West Lafayette, Indiana, Purdue University, Laboratory for Application of Remote Sensing, LARS Infor- mation Note 041874. Klein, J. D., and Serra, J., 1972, The Texture Analyser: J. Microsc., v. 95, pp. 349—356. A DEEPER LOOK AT LANDSAT—‘l IMAGES OF UMIAT, ALASKA 223 Lathram, E. H., 1965, Preliminary geologic map of Rowan, L. C., Wetlaufer, P. H, Goetz, A. F. H., Bil- Northern Alaska: US. Geol. Survey open-file lingsley, F. C., and Stewart, J. H., 1974, Discrimi- map, scale 1: 1,000,000. nation of rock types and detection of hydrother— Price, W. A, 1968, Oriented lakes, in Fairbridge, R. mally altered areas in South-central Nevada by W., ed, The encyclopedia of geomorphology: the use of computer~enhanced ERTS images: Reinhold Book Corp, pp. 784—796. U.S. Geol. Survey Prof. Paper 883, 35 p. Rosenfeld, G. A, and Hussey, K. M, 1958, A con- sideration of the problem of oriented lakes: Iowa Serra,‘J., 1972, Stereology and structuring elements: Acad. Sci. Proc, v. 65, pp. 279—287. J. Microsc, v. 95, p. 93—103. «mwflhayfi . PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Why Landsat? A Management View By J. Robert Porter, Jr., President, Earth Satellite Corporation, Washington, DC. Earth Satellite Corporation has provided substan- tive geological consulting services to over 100 proj- ects in which Landsat imagery has been utilized. To the best of my knowledge, this is considerably more exploration experience with Landsat than any other commercial or governmental group has had to date. Today, I thought you would be interested in some of the conclusions we have made, considering these projects as “case studies” from which one can gen- eralize about the learning curve corporations and in— dividuals experience in working with Landsat data. For the purpose of this presentation, I am only going to address the oil and gas applications of EarthSat and exclude our hydrological and geothermal activ- ities. Two years ago, the initial decision to utilize Land- sat (then ERTS) imagery was generally made either at the top or the bottom of the corporation. The rea- son for this was quite simple, and can be related to the time frame in which a person's job is evaluated. A corporate president or vice president of explora- tion in a medium to large company is primarily focus— ing his attention on attaining his corporate goals over a 2- to 10-year period. He was interested in under- standing, first and foremost, how Landsat and/or other remote-sensing techniques may affect his cor- porate strategy. Further, he was interested in intro- ducing new technology to his exploration groups from the standpoint of developing a “training” familiarity which may pay off in a few years. In brief, he ex- pected very little in the short term but felt a “little. money ought to be put into this to see what’s there.” At the bottom of the organization are the geol- ogists and engineers who are basically interested in preparing themselves and their company for the fu- ture, even if the new technology is not of proven worth initially. In the middle sits the exploration manager for a district or a region, who has a very tight, well-defined exploration budget and who must obtain results within a short period of time or be considered a failure. Understandably, he was initially not interested in Landsat because there was no one in his organization, or elsewhere, who could convince him 2 years ago that this would save him time or money and help him meet his corporate objectives with more certainty. During the last 2 years a significant change has taken place and Landsat is becoming an exploration tool which will soon be routinely a part of the budget con- trolled by this exploration manager. The use of Land— sat imagery has moved from research to operations. The reason for this is simple. The technology works! There is no question but that a new “quality” of in- formation is available. I say “quality” because its contribution to the investment decision is quite differ- ent, for example, than aerial photography. The way a corporation utilizes Landsat imagery varies considerably with its familiarity and experience. Almost every group I know went through an initial “gee whiz” phase in which they bought or were shown a few photographs from Sioux Falls and noted with interest that several features they were familiar with could be seen in the imagery. Alternatively, they may have been impressed by a false-color rendition of some salient or subtle feature exhibited by a com- mercial vendor. The second phase is characteristically much more professional but not yet commercial. At this point, geologists and engineers are looking at Landsat imagery as though it were a low resolution aerial 225 226 FIRST ANNUAL PECORA photo. Naturally they became more interested in technique development which, in a sense, from their view at this time, is a means of compensating for the “limited resolution” inherent in Landsat. During this phase, several important mental limitations still exist. Primarily, they do not really understand what is differ- ent in the scene they view which wouldn’t be better perceived by an aerial photograph if they had it. The principal attraction is low cost and newness. None- theless, at this point, Landsat imagery starts to be a tool in which features are identified and related to one another so that it becomes incorporated, somewhat traditionally, into ongoing operational programs. In the third phase. by combining a model of re- source occurrence (based on theory and previous work) with the large area relationships visible in orbital imagery, data from satellite platforms become a powerful exploration tool. The unique ability of satellite systems to acquire imagery that portray large portions of the Earth‘s surface enables geologists to characterize large, wide- ly separated regions (such as shield areas or extensive sedimentary sequences) and compare known mineral- ized areas with as yet unexplored areas as an impor- tant first step in the search for mineral deposits. Moreover, once having located an area of general interest (a proper geologic environment), investigators have mapped the tectonic style of the area and de— lineated those subareas which generally favor eco- nomic accumulations of minerals—for example, sedimentary basins, greenstone belts, and fold moun- tain belts. In several instances, the Landsat imagery enabled the geologists to proceed a step further and identify and map structural features as lithologic associations within the subareas which might serve as keys to economic accumulations of minerals, petroleum, ground water, or geothermal resources—— for example, salt domes, anticlines, intrusions, ultra- mafic rocks, and fracture intersections. At this level of detail, the geologist is able to correlate known economic deposits with lithologic and structural fea- tures perceived in the imagery and draw additional inferences about factors controlling and localizing economic deposits, thus refining his initial exploration model. There is some evidence that space imagery can lead to the direct detection of certain geologic re- sources by virtue of recognizing such features as gossans, peculiar tonal-textural features that are strongly correlated with oil and gas production, de- pressions under the icecaps that mark possible geo— thermal heat sources, or terraces along streams known to carry gold. However, most often the contribution MEMORIAL SYMPOSIUM of space-acquired data is based on inferences drawn from experiences or from a particular exploration model. In these circumstances, the strength of space data as an exploration tool is greatly increased by the inclusion of ground data in combination with data from other exploration devices such as airborne mag- netometry. Several exploration groups are now doing just this. The space imagery is used to locate the favorable structural setting, and then airborne electro- magnetic and gravity data are used to refine the de- tails of the model. Targets thus identified are followed up on the ground using geochemistry, resistivity, con- ventional mapping techniques, and, finally, drilling. Several geologists have used Landsat and Skylab imagery to tentatively locate structural features of interest to petroleum exploration. In Alaska, near the Umiat oilfield, Fischer and Lathram (1973) spotted a peculiar arrangement of lakes that geophysical data corroborated as an attractive structural feature for petroleum exploration. Further, Saunders and others (1973), related lineaments seen in Landsat imagery to oil and gas occurrences in west Texas. Based on the relationships they saw in the imagery, they have iden- tified several prospective areas, In the Anadarko basin, we found that many of the known oil and gas producing structural features could be identified on Landsat and Skylab imagery. In addi- tion, we reported that, as yet unexplained, tonal- textural features, called “hazy anomalies,” are strongly correlated with oil and gas production. Upon examining the pervasive pattern of lineaments seen in the space imagery and correlating these with geo- physical and well data, we concluded that faulting has played a much larger role in the origin and devel- opment of the basin than was previously suspected. Based on the study of Landsat imagery in the Basin and Range province of the western United States, we noted a complex interrelationship between faulting, intrusion, and mineralization. Interrelationships noted can be used to guide further exploration even in this relatively well explored area. We also noted that there seems to be a regular spacing of lineaments (fractures) that control mineralization and that the highest po- tential for mineralization is at the intersection of two such regularly spaced sets. The savings in time and money realized from these applications is already substantial and will increase as they are used more widely and as these tools are more completely integrated into exploration programs. Phase three could be called the “commercial phase” in which the technology is better understood and the product is both incorporated into ongoing com- mercial strategy and becomes a stimulus to the change WHY LANDSAT? A MANAGEMENT VIEW in corporate strategy. In my opinion, to equate Landsat imagery with aerial photography is to miss the point almost completely and to lose much of the enormous commercial edge Landsat can provide. Landsat sees geological features in a way aerial photography doesn’t. First, its perspective is different. It is not a substi- tute. It is different. Not always better. But always different. It avoids the distortion of wide-angle view- ing and permits the application of computerized tech- niques to areas of meaningful geological size where the illumination levels and Sun angle are uniform. The net effect of this is that in Landsat imagery features can be detected and often identified which are totally unrecognizable in conventional aerial photography. Further, it has been our experience, and the experi- ence of our clients, that in virtually every case, even those in which extensive ground work and aerial photography had been previously done, significant new geological features were revealed through the utilization of Landsat imagery. Second, its spectral coverage is different from any existing film system and its output format on digital tape provides analysts with many more levels of spec- tral intensity variation to evaluate and manipulate. Third, traditionally companies have been restricted as a practical matter strategically to areas in which they already had a substantial base of geological infor- mation. The use of Landsat as a means to evaluate rapidly new areas changes the cost equation dra— matically and makes it possible for a company to consider an entirely new exploration strategy. For example, in a number of projects in which we have been associated, prospective concession areas were rapidly and effectively evaluated so that the explora- tion costs of alternative prospects could be evaluated for less money and with greater certainty than was ever possible previously. Consequently, the companies were not dependent on limited photography or sec— ondhand geological interpretations based on limited information. Further, a consistency was added to the evaluation not previously possible, and new options have been made available at very modest cost in dollars and in time. So what is next? There is no question in my mind that within 5 years Landsat will be a standard medium used by the oil and mining industries. Unquestionably coincident with this would be greatly improved en- hancement techniques taking advantage of digital tapes and repetitive looks during different seasons to identify different features. (One of the great mis— conceptions during the second phase of development mentioned above is the approach that one picture does 227 it. This is just not so and repetitive coverage under different seasonal and weather conditions makes a significant difference.) Further, one must consider the context of the times. The advent of Landsat technology is certainly timely: First, because the demand factor for new oil, gas, and mineral reserves is inordinately high and great pres- sure is being put both from the economic and a nationalistic standpoint on finding new reserves. Sec- ond, associated with this, companies are very much of a mind to spread their bets and diminish their risk, in part because they recognize that the political climate of the world is such that it is not possible to predict which situations will be stable from an eco- nomic standpoint over the period required to attain a reasonable return on investment. Further, this eco— nomic unpredictability does not correlate simply with traditional indicators of political stability as is shown by the fact that even well—established and “stable” governments have become increasingly aggressive in regulation and taxation under the pressures of the energy crisis. Finally, of great consequence, is the fact that new exploration models must be developed. The ability of satellite systems to provide an uninter— rupted view of the regional geology of large areas, coupled with the new concepts of plate tectonics, provide many opportunities for perceiving interrela— tionships that previously had gone unrecognized. These types of perceptions have far—reaching conse— quences, both for geologic exploration and the development of geologic hypotheses on the origin and evolution of major portions of the Earth’s crust. For example, in some of our work using Landsat imagery, we have observed that a constant spacing in major fractures in the Earth’s crust appears to be character- istic of specific crust blocks. Geologic theory has long held that the emplacement of intrusive bodies of igneous rocks, some of which may carry economically valuable minerals, is controlled largely by pre-existing fractures. Thus once the fracture spacing pattern is determined and a few economic deposits are found in a particular area, one has a powerful tool for searching for other deposits in the same region. Another promising area of research and one which will undoubtedly receive increasing attention as min- eral exploration increases over the next few years, is the field of geobotany. While this work certainly pushes the state-of-the-art of Landsat resolution, nevertheless Landsat affords a needed opportunity to obtain repetitive coverage at low cost so that correla- tion can be established and new geobotanical explora- tion models developed. 228 Finally, a significant edge will be gained by those organizations which are able to restructure their ex- ploration strategy to take advantage of this technology over the next 3 years. All pictures and all people are not equal, and the merging of high-quality enhance- ment techniques coupled with high-quality geological minds will be a telling combination. On the other hand, I would like to close with a note of caution. We are just now at the beginning of a commercial phase where we are starting to understand critical subtleties of the technology with which we work. Oil and min- eral exploration is a damn tough business. Answers are not spewed out of black boxes run by aerospace “whiz kids.” Rather, valuable information is provided which must be evaluated within the context of‘a sound exploration model. As was the case with all other new exploration techniques, the state-of-the-art will always be stretched and challenged beyond its reasonable limits, and the ability of black boxes to provide an- swers, as always, will be oversold. Nonetheless, this FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM is an exciting time because what we are seeing is a revolution in access and perception which will tax us all over the next 5 years in just applying the current state-of—the-art. On the other hand, there is a lot of profit to be made, and a lot of losses which could be avoided, if we get on with it aggressively. REFERENCES CITED Fischer, W. A., and Lathram, E. H., 1973, Concealed structures in Arctic Alaska identified on ERTS—l imagery: Oil and Gas Jour., May 28, 1973. Saunders, D. F., Thomas, G. E., Kinsman, F. E., and Beatty, D. F., 1973, ERTS—l imagery use in reconnaissance prospecting—Evaluation of the commercial utility of ERTS-l imagery in struc- tural reconnaissance for minerals and petroleum: Type III Final Report to Natl. Aeronautics and Space Admin. Available from US. Dept. of Com- merce Natl. Tech. Inf. Service as E74—10345. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Regional and Global Geological Studies Using Satellite Magnetometer Data By Robert D. Regan, US. Geological Survey, Reston, Va. ABSTRACT A global magnetic anomaly map derived from satel- lite measurements offers a new perspective in regional and global geologic studies. Preliminary correlation with tectonic and geologic maps indicates that in addi- tion to reflecting obvious large—scale features such as continental shields, the map also reveals many unex- plained structures. Particularly striking are the lack of anomalies in the Pacific Ocean, with the exception of a pair (+6 gammas and —6 gammas) near the 180° meridian in the area where the Emperor Seamount and the Hawaiian Island chain intersect, a broad mag- netic low over the Gulf of Mexico, and several highs in the central United States. Also evident is a major magnetic anomaly in central Africa that does not directly correlate with any known tectonic feature. This anomaly, termed the Bangui anomaly, was discoveredvin the satellite data and confirmed by aeromagnetic measurements. The anom— aly is over an area of thick continental crust and has been interpreted as a massive crustal intrusion that may be related to crustal thickening. Many major mineral deposits are associated with this body. Although much more work is needed to completely interpret the map and to determine the causes of all the anomalies, the initial results show the utility of a satellite magnetometer as a geological/geophysical tool. The global magnetic anomaly map and a complete discussion of its derivation are published in the Journal of Geophysical Research, v. 80, no. 5, p. 794. 229 PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA AlRTRACETM—An Airborne Geochemical Exploration Technique By A. R. Barringer, Barringer Research Limited, Rexdale, Ontario, Canada, M9W 5(32 ABSTRACT The AIRTRACET-‘I system is a new airborne geo- chemical technique for collecting and analysing atmos- pheric particulate material and relating this to the underlying geology and geochemistry. The technique depends upon the existence in the atmosphere of a dispersion of particulate material which covers a broad range of particle sizes. The coarser material above 10 ,t in diameter tends to be localized to such an extent that if this size fraction is used the effects of manmade pollution are not a problem in most exploration areas. In temperate and tropical regions vegetation is a prime source of the particulates utilized in the system and laboratory research with radioactive traces as well as field surveys has demonstrated the important role played by vegetation in releasing particulates, carrying heavy metals and other trace elements. Special techniques have been developed for sepa- rating particulates of local derivation from the gen— eral ambient background by specifically isolating the material in rising thermal plumes. Good results have been obtained in test surveys over orebodies covered by glacial overburden and heavy vegetation. Elements detected include Cu, Zn, Ni, Mn, Fe, Cr, Cd, Al, Mg, Ca, Ti, Si, and C. Techniques for measuring uranium are under development. Tests have also been carried out over an oil and gas field, and it has been demon- strated that airborne anomalies correspond with the location of the field as well as surface rock and soil anomalies in carbon isotopes and manganese. The equipment is being flown in helicopters and fixed winged aircraft. lNTRODUCTION In the conventional approach to geochemical ex- ploration samples of soil, stream sediment, rocks, or plant material are gathered on a reconnaissance basis or on a close grid in order to determine the regional or detailed distribution patterns of mineralization. The methods of sampling and analysis have been studied over a period of more than 25 years and have led to the establishment of geochemical exploration as a prime mineral exploration method (Hawkes and Webb, 1962; Levenson, 1974). Considerable research has also been carried out in hydrocarbon geochemistry; however, the techniques of geochemical exploration for oil and gas appear to have found widespread application only in Russia (Kartsev and others, 1954). In the work reported in this paper investigations have been made of the use of atmospheric particulate geochemistry as an exploration tool for both minerals and hydrocarbons, with the aim of supplementing ground geochemical methods and providing techniques for low-cost, large—scale airborne reconnaissance. An early proponent of the use of atmospheric par- ticulates for airborne geochemical applications was Weiss (1967). Weiss initially employed a filter method for sucking in air and an X-ray analytical technique for analysing particles on the filters. Subsequently, he has used a system of towed filaments that are dragged through the air in order to collect particles on their surfaces. These filaments are periodically wiped, and the material subsequently analysed. Weiss has referred to the atmospheric aerosols as containing mineral particles and has stated that the method is not applic- able where the terrain is covered by heavy vegetation (Weiss, 1971). In the present investigations, attempts have been made to develop new particle collection techniques that are capable of separating and collecting particles that have risen from underlying terrain and' rejecting particles that have travelled substantial distances 231 232 FIRST ANNUAL PECORA laterally. Investigations have also been carried out on the contribution of vegetation to the generation of meaningful geochemical responses in the atmosphere, and, in addition, there has been substantial analytical chemical research to develop appropriate methods for use in atmospheric geochemistry. SOURCES OF ATMOSPHERIC PARTICULATES AND AEROSOLS The sources of atmospheric particulates and aero— sols are both manmade (anthropogenic) and natural, the former being a source of interference in airborne geochemical studies, Anthropogenic pollutant sources associated with urban and industrial areas generate very high loadings of atmospheric particulates, the effects of which tend to be localized in the giant particle sizes (plus 10 p) but carry great distances under some meteorological conditions in the small particle size range. In urban environments the mean residence time in the atmosphere of submicron parti- cles in the absence of precipitation is in the order of 100 to 1,000 h (Esmen and Corn, 1971). Pollutant particles of greater than 10 ,t in diameter will, there- fore, have mean residence times of less than 10 h and will be generally localized to the vicinity of urban and industrial areas. Studies of easily recognized car- bon and flyash spherules over the North Atlantic confirm that these pollutant particles do not exceed 6 [L in diameter over open ocean and reach larger sizes in only a very small percentage of the total mass of particulates in ocean areas close to pollutant sources such as the Bay of Maine (Parkin and others, 1970). It has been estimated that the mean residence time of particles in the lower troposphere is about 4 days (Poet and others, 1972), and it is, therefore, clear that in theiabsence of precipitation there can be wide- spread migration of the mean particulate burden over distances that can on occasion amount to hundreds of miles or more. The fine particle component is the fraction that migrates, and it is this fraction which contains the highest concentration of trace metals (Lee and others, 1967). Approximately 45 percent of the mass of particles in nonurban air close to the ground lies in the size range above 10 ,u. in diameter (Noll and Pilat, 1971), and it is essential to consider only this size range if the effects of urban and indus- trial pollution are to be minimized when studying the atmospheric geochemistry of nonurban areas. The existence of high concentrations of heavy metals in atmospheric particulates is observed even in regions remote from civilization (Rahn and Win- chester, 1971); this is frequently cited as evidence of MEMORIAL SYMPOSIUM an all-pervading influence of anthropogenic pollution. However, the presence of high concentrations of heavy metals in atmospheric particulates is by no means a proven indicator of the influence of pollu- tion, since it has been found that particulates rich in heavy metals can be released from vegetation by a phenomenon that parallels the release of water by transpiration (Beauford and others, 1975). Once again the effects of large-scale migration of particulates from vegetation sources can be minimized by con— :idering on‘y particles of greater than 10 ,1 in diam- eter. The release of heavy metals from vegetation and particle localization will be dealt with in greater detail later. Additional natural terrestrial sources of particulates and aerosols include the weathering of geological materials to produce soil particulates at the surface, the effects of wind erosion (Hilst and Nickola, 1959), the formation of aerosol particles derived from naturally occurring hydrocarbons produced by plants (Went, 1964, 1967), the generation of condensation aerosols by homogeneous gas reactions (Walter, 1973), the formation of condensation nuclei by evaporation processes in semiarid terrains (Twomey, 1960); and the production of organic particulate ma- terial by the decay, biodegradation, and weathering of plant material. The oceans also provide a significant source of par— ticulates in the atmosphere. Ocean derived particu- lates are generated by dehydration of windblown spray and the bursting of air bubbles entrained by wave action or rain (Woodcock and Gifford, 1949; Blanchard and Woodcock, 1957; Junge, 1972). Partic- ulates of marine origin can contain up to 20 percent of organic carbon (Hoffman and Duce, 1974) and in some areas can contain even higher percentages of organic material. This anomalously high concentration of organic substances in marine particles is derived from the organic layer on the ocean surface (Garrett, 1967; Barger and Garrett, 1970). Aerosols generated by bursting bubbles are coated with the organic material (Blanchard, 1963, 1964, 1968; Garrett, 1967; Williams, 1967). Microorganisms can also be injected into the atmosphere from the ocean surface by the bubble—bursting mechanism (Carlucci and Williams, 1965; Blanchard and Syzdek, 1972), and airborne marine microorganisms have been observed by the author and his colleagues as well as other researchers (ZoBell and Mathews, 1939; Stevenson and Collier, 1962). AIRTRACE'I‘31—AN AIRBORNE GEOCHEMICAL EXPLORATION TECHNIQUE RELEASE OF HEAVY METALS IN PARTICULATE FORM FROM NATURAL SURFACES The transfer of elements into the atmosphere from vegetation is of special interest in relation to the application of atmosphere geochemical methods in vegetated terrain. Movement of elements into the atmosphere from coniferous trees has been studied (Curtin and others, 1974), and it has been suggested that volatile exudates are the carrier. It has also been suggested that aerosols may be formed subsequently by agglomeration processes from the volatiles and that they could be used for geochemical sampling purposes. Recent work, however, with laboratory plants using radioactive zinc 65 and lead 210 as tracers in the plant nutrients has indicated that the initial release takes place in particulate form (Beau- ford and others, 1975). Some of this work, as yet un— published, has shown that particulates in small laboratory plants are released dominantly in the sub- micron range. However, in the natural environment particles labelled with radioactive tracers can be released through a size range that is collected through all stages of a multistage particles impactor. It is of particular significance to the present study to note that radioactively labelled particles in size ranges up to at least as large as 10 ,i in diameter can be released under free convection even in the absence of any wind. The release of radioactive materials from vegeta- tion that has been sprayed with a solution containing strontium 89 has been reported; it has been shown that there is a steady loss of radioactivity to the atmosphere (Moorby and Squire, 1963). It was sug— gested that the release might be taking place during the shedding of cuticular waxes, but the work was not continued by Moorby and Squire to verify this concept. Cuticular waxes abound on the leaves of most plants in the form of small wax platelets which provide the leaves with a water shedding surface. These platelets are continuously released by abrasion and the rubbing of leaves against each other and, in many instances, are able to renew themselves (Eglin- ton and Hamilton, 1967; Martin and Juniper, 1970). The use of various solvents on particulates that have been released from vegetation, clearly indicates that they carry only a small wax content and an analysis of the solvent extracts also shows that the waxes do not appear to be the major carrier of metals. In considering the mechanisms of elemental re- lease from plant surfaces it is important to note that such surfaces contain salts that can be readily leached out by rain or dew (LeClerc and Breazeale, 1908; 233 Long and others, 1956; Clement and others, 1971). Whereas most studies are concerned with the leach- ing of major elements such as K, Na, and Cl, the writer and his colleagues have confirmed (to be pub- lished) that the heavy metal trace elements such as copper, lead, and zinc may also be leached from the surface of leaves with simulated dew or rain. It seems highly probable that the deposition of elements at the surface of leaves is related to the transpiration pro- cess in which soil solutions carrying dissolved salts are taken up by the plant roots, but the evaporation from the leaf surfaces is confined to the release of essentially pure water. Some of the salts in upward migration through the plant are utilized in the normal metabolic processes of the plant; however, certain elements may be enriched in the vicinity of the leaf surfaces. A preliminary study is being carried out using newly available equipment on the distributions of 19 ele- ments in soils, plants, and vegetative particulates on two sites in the Bancroft area of Ontario, Canada. The region is one of Pre-Cambrian geology, and one of the sites carries pegmatitic uranium mineralization while the other has only background values in uranium. Analyses of the A and B soils, bark, needles, and branches of conifers and deciduous trees were studied as well as the particulates generated by this vegetation, Particulates were sampled by filling large plastic bags with vegetation material and collecting the particulate material released by agitating the bags. High-volume suction sampling methods were used employing a small cyclone thereby providing sufficient material for precision multielement analysis. Results are shown in table 1 and indicate certain interesting features. For example, the concentration of many elements in the particulates released from branches and needles is comparable with the concen- tration of these elements in soil. However, in the case of Fe, Al, Ti, and U, these elements are substantially depleted in the plants with respect to the soil concen— trations, but the particulates released by the plants in general are enriched in comparison with the plant concentrations. It seems as if the vegetation minimizes the uptake of these elements and at the same time tends to further eliminate them by shedding enriched particulates. These elements belong to the nonbio- genic class that occupies a certain restricted area in an ionic radius/ionic charge diagram (Hutchinson, 1943; Brook, 1972). It will also be observed that uranium mineralization is more readily detected in the particulate material than in the vegetation from which it is derived. FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 234 .meE 9 boomanU 8:232th E Dozutco ucm m__om 9 82?: 5:3 Sofia E Ugo—awn 3:25.? to 23m NSN 33 mm? 88 SE was 8.2 «m; QNN 3 MN oi 3 2.; WE 8.3 .3 ...... 83853 5:: meow. wV 8.wa 5 R 3.3 88 $8 98 38 N8 .88 .3 f 3 «8 oz 5 0.8 3 ................. 5:: :28 8 $3 NAN E 28 8% NE 82 SA Rm 8 3 2 w? 8.8 mam mm 53 8.2 - SEBEE 68$ maoancmo NV Ea 3 oz 8.2: 82 8 08 3 oz 5 3 3 3 M: oz 2: :3 3 ......... 865cm msoancmo to E: QM a: 2 88 58 R2: «.3 RN 8 E E 3 E Na oz 2; 8.0 l- 8832:: gm Sozguwo No 0.8% 3 8m 8.: 88 im BE 2“ 5+ 5 3 A: 99 E mg 3: o: 3 ............. {mm £6358 no ~on «.8 «8 33 oz 9: NS 8.0 88 mm 2 MN m3 md 99 NS #8 “.2 --- $.28Eé m.632 @Eou NV 93m oz m2 8.8 8% 8m 83 2: em :8 2: f 2 3 oz oz 8.2 m8 ............. 3.802 588 FV :8 I: 2: 2K 88 v: 88 QR 5: mm 2 to 9m 3 3 E, 0.8 3 ...... 353:5 v23 5:80 _.V $8 3 mt oz 88 N3 88 EN oi o? 3 oz 3 oz oz oz 8.8 3 ................ {mm 5::8 NV 38. $8 8% SE 88 82 8: o.“ 352 2: new we 3 H: 3m 1: N8 3 ................. BEN m :8 I. .28 $8 82 0.8m 88 88 8: ma 3ri 3: E o; E 3 8.8 mm N8 Wm ................. 25M < :8 5 a E *_< z v. w: 3 .m *3 £2 92 w< 5 8 5 _z :N 3 $5 ozoomuxuélm ohm o32 I I | | l l I l l J l l I I j I I T l I fl 0 IKM 2.0 I KM go lI KM ; 30 SEC GOSEC soSEc GOSEC BOSEC GOSEC ——-> -———_> —_> AIRCRAFT TRAVERSE AT 15 METRES TERRAIN CLEARANCE FIGURE 5.—Effect of mixing activity on particulate loading. 242 FIRST ANNUAL PECORA use of fast response thermal detectors at the air scoop inlets in conjunction with a suitable electrical filter, a digital delay line system and solenoid valve switching. Delays are adjusted appropriately so that switching of the particle stream can be accurately timed in relation to the transit delay for the particles passing through the piping of the system. This technique en- ables very small air parcels to be sampled with a resolution in the order of 5 ft. Samples are integrated into hot and cold spots on the tape using collection intervals that can be selected at either 5 or 10 s. The effectiveness and precise resolution of this plume sampling technique is a function of the choice of the lower level of particle size collected, the selec- tion of the wave filter functions applied to the temperature sensor, and the sensitivity settings of the thermal switching thresholds. The optimization of FIGURE 6.—AIRTRACE“‘ equipment mounted on a 2063 let Ranger. MEMORIAL SYMPOSIUM these parameters has to be based upon extensive experimentation. The equipment is usually mounted on a helicopter (fig. 6), and surveys are carried out at altitudes of 20 or 30 ft above the treetops. This altitude turns out to be quite practical and is in fact used very commonly in crop spraying operations. Navigation of standard grid surveys at a quarter-mile spacing, for example, can be achieved in most terrains using conventional navigation from a photomosaic, providing that the navigator is skilled. Alternatively electronic navigation can be employed. The equipment is also being operated in fixed-wing installations where surveys are carried out at 200 ft altitude. At this altitude the equipment is well suited for climates where there is very strong mixing such as in South Africa or Australia. AIRTRACETM—AN AIRBORNE GEOCHEMICAL EXPLORATION TECHNIQUE The basic AIRTRACE system is employed in con- junction with other geophysical equipment such as magnetometers in helicopters and an EM system in the case of a fixediwing installation. Inflight monitoring is provided by a multichannel analog recorder for optical measurements of the particle stream, thermal monitoring, etc. ANALYTICAL SYSTEMS One of the problems of airborne geochemistry is the very small amount of material that is collected. This makes it necessary to use extremely sensitive analytical techniques. Relatively simple equipment has been developed that provides infiight analysis of ab- sorbed hydrocarbon and mercury, and there are some valuable applications for this approach. However, as research on airborne geochemistry has proceeded, it has become obvious that a system capable of analys- ing for a large number of elements would provide the optimum. Specifications for such a system include multielement capability, very high sensitivity, wide dynamic range, and high speed of operation. These requirements are extremely difficult to meet in real- time analytical equipment due to the problems of weight, power consumption, complexity, and the diffi- culties of operation and maintenance in an aircraft. It was, therefore, decided to develop a post-flight analytical system. In considering possible systems, one approach that at first seems obvious would be the use of X-ray fluorescence analysis. Unfortunately, however, even the most advanced of these systems have inadequate sensitivity for the particulate levels that are en- countered under some climatic and terrain conditions, and, furthermore, the time taken for sample analysis tends to be long in comparison with the time taken to collect samples. The solution that has been employed is the use of laser vaporization in conjunction with inductively coupled plasma spectroscopic analysis. These two techniques used together provide a very wide dynamic range, high sensitivity, simultaneous multielement capability for more than a dozen elements, good linearity, and an analysis time of seconds per sample. The operational equipment is capable of functioning in a fully automatic mode, and the output is on mag- netic tape where it is directly available for use in the preparation of maps by computerized methods. All readings on the equipment bear a fixed relation- ship to the mass of each element present on the analytical tape so that any pair of elements can be ratioed. The linearity of the system is demonstrated by figures 7 and 8 which show the plots of silicon 243 versus titanium and zinc versus titanium in a minus 200 mesh soil sample through a 40:1 range of loadings. This degree of precision is duplicated for most of the elements that are recorded and applies to both organic and inorganic matrices. Elements currently being analysed include Cu, Zn, Ni, Cr, Mn, Fe, Ti, Mg, Ca, Na, K, Cd, Si, and C. It is hoped to expand this list shortly and to include uranium as well as certain volatile elements. With regard to data handling, in a simple approach it is possible to plot the ratios of a given element pair such as copper and titanium to look for copper anom- alies. However, in practice a more sophisticated method is used in which linear regressions are em- ployed to plot the variations of elements such as copper against six other preselected elements. This provides a very satisfactory approach that is giving good results, but a number of other possibilities re- main to be tested. These include statistical methods such as factor and cluster analysis. FIELD RESULTS An interesting example of the early research work carried out with the system was in a truck-mounted version of the equipment which was traversed across a small orebody at Limerick, Ontario. This mineral— ization was substantially covered by trees with very little exposed soil present. An important feature of the area was the presence of a seepage zone located to one side of the mineralization and which provided a strong cold extractable anomaly in the soils. The truck traverse results, which were obtained with an atomic absorption analytical system, are presented in figure 9. It will be noted that the main AIRTRACE anomaly occurs over the seepage zone and that the anomalies were present both in the absolute concen- trations of metals in the atmosphere and also in the copper/zinc metal ratio. Since the area is heavily vegetated, it seems almost certain that the anomaly must be mainly biogeochemical in origin. Bulk col— lections of aerosol that were subject to sink and float separation in carbon tetrachloride (density 1.60) indi- cated a very high proportion of organic material being present. This substantiated the biogeochemical nature of the particulate anomaly. An example obtained with the current equipment operating under very difficult geochemical conditions is shown in figures 10 and 11. This survey was flown over a deposit recently discovered with the INPUT airborne EM system by Selco Mining Corporation and Pickens Mather Incorporated in Brouillan township in Northern Quebec. The deposit is reported to comprise three separate zones with an unofficial combined 244 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 1.20 x104 4 l ' ' l 1.08— 0.96 0.84 0.72 5 0.60 0.48 0.36 0.24 — 0.12 0.00 I I I I I ++ I I I I 0.00 0.23 0.46 0.69 0.92 1.16 TI X104 4 1.39 1.62 1.85 2.08 2.31 FIGURE 7,—A test of linearity of laser analytical system for variations in loadings of U568 soil (XRC—-6), silicon versus titanium. Date: 7/10/75. tonnage in excess of 50 million tons. The first zone discovered contains an estimated 35,400,000 tons of ore averaging 0.39 percent copper, 2.30 percent zinc, 1.04 02 silver, and 0.009 02 gold and lies beneath a minimum of 30 ft of glacial overburden. Ore grades in the other two zones are reported to be much higher in copper than the first zone, and they occur at greater depths. The anomaly map was derived by a multiple regres- sion technique carried out on a line by line basis for copper and zinc, and, in addition, spatial filtering was employed to smooth out noise and generate a con- toured map. The area was flown twice, both times under far from ideal weather conditions, since in this region weather conditions are frequently rather poor. Never- theless, clear—cut delineation of an anomalous zone was obtained on both flights, in each case somewhat displaced but clearly relating to the location of the airborne EM conductor. These results taken in con- junction with airborne geophysical data provide a definite indication of a drilling target in a region which contains large numbers of barren EM conductors that normally have to be disproven by expensive drilling. An interesting example of the AIRTRACE system applied to a hydrocarbon target is shown in figure 12. This traverse was chosen to cross 'a geochemical anomaly in carbon isotope ratios identified by T. Donovon of the US. Geological Survey (Donovon, 1974). The survey covered the Cement field in Okla- homa which is named for the cementation by carbon- ate of the surface rocks over part of the oilfield. These surface rocks not only show anomalous C“"/C1‘-’ values but also a manganese anomaly. A ground truth survey carried out by collecting fine particulate material from the surface on a traverse line indicated a definite anomaly in the manganese/iron ratio across the oil- field, and this was subsequently duplicated in airborne AIRTRACETM—AN AIRBORNE GEOCHEMICAL EXPLORATION TECHNIQUE 245 400.00 | l l 1 360.00 320.00 — 280.00 240.00 — 5, 200.00 — 160.00 I- 120.00 I + + 80.00 I- + 40.00 — + 0.00 I l 1 ' X10¢4 I J l | l 0.00 0.23 0.46 0.69 0.92 1.16 1.39 1.62 1.85 , 2.09 2.31 Tl FIGURE 8.—A test of linearity of laser analytical system for variations in loadings of U565 soil (XRG—6), zinc versus titanium. Date: 7/10/75. traverses with the AIRTRACE system. Additional tests have been carried out over other oilfields using the AIRTRACE technique for hydrocarbon volatiles Bar- ringer, 1973a), and strong, highly localized anomalies have been observed—some apparently relating to faults and other structures over oilfields. CONTINUING RESEARCH A programme of research on botanical aspects of airborne geochemistry is continuing in the Botany Department of Imperial College, London, under a post-doctorate fellowship agreement with W. Beau- ford, a post-graduate scholarship with M. Luton, and with consulting assistance from J. Barber, Reader in the department. Current work is on the factors affect- ing the release of particulates from vegetation in situ as related to seasonal and diurnal factors, meteoro- logical conditions, plant growth conditions, etc. Research is also continuing on the atmospheric geo- chemistry over oilfields and the application of the AIRTRACE techniques to hydrocarbon exploration. On the analytical side, new methods of fingerprinting using spectro-fiuorometric and mass spectrometric approaches are being studied in the laboratories of Barringer Research. OPERATIONAL PLANS The AIRTRACE equipment is now being used operationally in parallel with continuing testing over an increasing range of environments. Certain promis- ing applications remain to be tested, including the use of the system for uranium, coal, and kimberlite exploration. The system is being operated by Minsearch Surveys Limited for minerals and Barringer Hydrocarbons Limited for hydrocarbon exploration. Both companies 246 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM so - 66 I 50 - 4o - o g 30_ MORNING . g AFTERNOON ‘ A E A EVENING I < l (I I: 20 - / m g A A A IO - / \ A\-\\ 4:/\\ / ‘/‘ l‘ l \ ” \\ / ll/.: \I/KL— —I—\Olo/ "”<: L\ 8W 7w 6W /5+50w/ 5w E //3' IW IE 25 35 4 /// see page 20 neO ROAD TRAVERSE ////// mineralization FIGURE 9.—Limerick prospect, near Bancroft, Ontario, AIRTRACETM Mk Ill—copper. are joint ventures between Anglo American Corpora- tion and Barringer Research Limited. CONCLUSIONS The AIRTRACE system for airborne geochemical prospecting has been evolved over a period of years in a series of parallel research programmes on par- ticulates collection and sampling techniques, meteoro- logical controls near the ground, analytical methods, biogeochemical studies, and ground truth geochemical programmes. Some of the earlier field studies and case histories programmes have been made obsoles- cent by the later development of greatly improved collection methods, thermal switching techniques, multielement analyses, etc. There is an immediate necessity therefore to gather a great deal more data with all of the new technology. Nevertheless a sci- entific basis has been established for the development of multielement particulate anomalies over areas that are partly or completely covered by vegetation as well as in the more obvious regions of soil exposure. Sur- veys carried out over known mineral deposits in areas covered by glacial overburden and vegetation have yielded positive results. Furthermore the possibilities for atmospheric geochemistry being of value not only for mineral exploration but also for hydrocarbons has been demonstrated and anomalous effects have been seen in both organic and inorganic parameters over known oil and gas fields. One of the more obvious potential applications of atmospheric geochemical exploration could be in the areas having difficult [access such as tropical rain forests. It will be a great interest for example to see the effects of deep taproots and high transpiration rates on the overlying atmospheric geochemistry. Some recent preliminary field data indicate that the principles of biogenic enrichments and depletions AlRTRACETM—AN AIRBORNE CEOCHEMICAL EXPLORATION TECHNIQUE 247 SCALE-MILES 0 1 2 , dd' r \ wmd direction wm "'90 IOI‘I O _ 3600/15—20mph N320 ”0 25mph N l \ _ I 4 _ ‘ a of . ‘ torted///////// @230, ‘W/ Area of ng ,‘I . h Re rted .z. Dri ling Jt " i Input maly Anomaly Scale 1 2 I_i__l_.i_l “ 5 MILES 5 FIGURE 11.——F|ight B AIRTRACETM activity contours Cu+Zn Bl regression. FIGURE 10.—Flight A AIRTRACETM activity contours Cu+Zn regression. release. A full understanding of this phenomenon should in the future allow a more fundamental ap- between plants and soil have a parallel in the enrich- proach to the interpretation of data gathered over ment factors between plants and the particulates they vegetated terrain. 75 4 III .. ... .55... . . . . . . . I I I I II IIIIIIIIIHIIIIIIIIIIHIII . . 5......u.u .. . . ... ... . . . .. .. . .... ... .. .. . . . ....u...... . . ”...”...2. .... ......... 5...... .....5.... .5 ...........5... .55....5. ....... ... ...........5.......... ............. ......5. I IIIIIIIIIII IIIIIIIII IIIIIIIII III ...........5........ . . . ... . .... .......... 5.....5 ... ............... ... . . ... .55 . ... .55.... . ...... . . . ......5. ...... . .... ...... . .....5 .. ... ...u....... .. .. 5... 5555”.» . . ...... .. . . . ... .... ... ...... ... . .. ............ ... ... .50}... . . .......... 55.5.5555 ... ... . ..................5I..... ..... .55.. . ....5 . ... .... . . ...... .. I I I IIIIIIII I IIIIIIII I IIIIII I I l ,I III I III. .......I.n......5.. ,. . ... .. ... . I I IIIII 5 .... ...... . I ... . 555...... 555...... ................. .... ............ EEIIIIIIIIIIIIIIIIIII I I I I IIIIII II I II III. I ...................5 . ......5 . .I... .......u..u. 5&. .. . . . . . . . . . II I II II .55.... .. ......4..... . I III I 15... ...... III IIIII IIIIII .... ... .. . . .. ..... .. ...... . ...... .... . .... . ....... . . ... ...... . ....... .. ... . 5......5..... . .... . III II IIII IIIII . .... ... . .55... . .. . ......v 5 . ...I.. .55..» I III I IIIIIII 5.5. III . .. .555... . II III I III-II I I I II I I I I I I I II IIIIIJ. IIIII?II II . ......u.. ... ... .... . .... ............ II I III . . . . ... . . . . 555... . .. . .. . . .. ....5. . ... .... . . 5.... ............ . . .....5. . . . . . . . ... . ......n... .... .55.“... . I KI I n......n..u.. .u...... . . I'I I". . . . ... II I IIIIII .5..n...... .....5 . I. .....5. ".... . .. ......I.......... . .......55.... . .. ... . ... I I I I III I II II II IIIIII III I II I I I III I I II II III I III I I I I IIII I If I I IIIII .. . .. . . . ... 5 5.55....v. ............. . .. 1...... ......n... . . I... . ... . . or. II I I .... . ...... a . .... . ... 5.... .... . .. .... . . .....5. .. {II IIIIIIII IIIIII 5...... ....I..... . .. . . . ......5... ....... ... . . . . .. . 555... 3.5“” 5 III I IIIIIII . ... . . . ....5 .. .. . . .r. ...... :...... .. . ... . ..........5.nuu. .....I+.. . .. ...... ...”...5... ..n............. 2...... .55... ..................... .... ....n.u.... ..uu.......55.»...9...3......~5...x....5.... 5.3" .I IIIIIIIII I I III. II . ..... .5...) ... . . . .. ...... . . ...? 5 .5: . . . $55.... 55'- . . . 5... ... . .5 . ...... III IIIJIIHIIIIIII I II I I I I I ... . ............. II II III IIIIIIII ...... ... ...... . .x. .... ............... ...... .. . 5.55.... . ........... . ... . . IIIIIIIIII III} . .55.... .5555 ... ... +1.. . . . . ... ... ... . . .... .. .... .5 .. .. ... ... scALE IN MILES T 0 tr I I I I I I I III-IISKI I I I I III II III-II . 55 .....5.... 55.”... .. ... IIIII II IIII IIIIIIIII ... .51... ....I . ...... . 5...... . ...... ...... .... 5...... . .. .........I.. I......I........... ....... .. :...... ......r. ....5...... I .... ..:... . ... ...... . .... . . I IIII ....5 I III. .. «New... 55...... ...u...... ... . . ... .. IIIIIfIIII I IIIIIIII I I I III Ix. III II II ...... . .... . . ....5 II. ....x..........u..u .... . ......fi.....¢. . ..........5....u..... . . I I III. WIND DIRECTION ROM 3‘5 Q40 KNOTs ‘IA. .5 . . I I I I I I I II I I I I II II... I III I I I .I. II ..:.......... 5................."..... ..............~ . ...5 ...n... . 5...... .........n............. . . . . ...u5 .5 I. I.II.Iu..m....u.I.nnN..I 5va III...” . ...... I . : II III.I III.III IIII IIII I I I I I I I I I I I I IIIIIIII IIIISII IIIII .fi........5.... .. . . .. 555...... .v... .....5........... 5+5 ... ..... .. . .......... ...5. II... .5 . ...... ...... 5 .u. ...... an. .. .. .. . .. ...NI. .4- 55' II I “if-:- . "I “I :95: ..... ...... ....... a .. ...... . 5%.... I. ......u........... I... IIII II I I I II .a. . .. . . 5 . . _ .....Iwn.u.....n...n....5 .........u.. I I I I I I III I . . ......5. .I . . . . .. ..I...I..In..5. . ... 5.x 5:........ ... . . .. ..:. . 5. . . 5.5.1:... .... ... ... . ....n. . ... . I I I I I I I I k. I I I I I)? I I I I I I ... . . . . . . . 5. . . .....I. .5555...“ IVs? .9¢r .55”... .. ........... I . .5”. .a I I I 1'5 I” *I I ’I" % IIIIII IIIIIII . . . . . . . . I . . . . . . I . . . ......5 ...... . . . . . . . . I .............n I. ...I.............. .. .... I ... . 5.5.5....555 . . . . .5. . . . 5... . . I x:- *5". ..II I\VI Mn/ Fe RatiC QROUNDTRUTH Cyril UM SYMPOSI UAL PECORA MEMORIAL N FIRST AN I.O . . ..II.1 . ... .... . ,. ...wx. : Mfinwflmnmw 5%.... ..5 I... . 55 . ...... 5.. .. a. .I....... my :.....fl... .... .........w.... . ....flx. ....H..... 5... . “W A ...ow ... . . ... 5... $5.5... ...... . ...wmwfififixfiuw 5...»... . 55...”... III .- I‘ c'.’ I .5‘ II I 'I I I I BORNE. I‘AAlnR/ Fe RO‘I’IO ...5 . . . I... I “65...... . . .. ,‘. . . . .~. 5%.: _._.- ._-.5- :5: ,._.5 I II I} ”#5. I III 248 ._-.- ._. ,. ._-. . «:3: I, . .. 555?... .. . .......... .55... . . ....r “5.. ... ... .55 .. ... ..5. . ...... «......I. . I.I ... IIIIIIIII .IIIII III I ......I II und truth. TM data and gro parison of AIRTRACE t com 2.—Cemen FIGURE 1 AlRTRACETM—AN AIRBORNE GEOCHEMICAL EXPLORATION TECHNIQUE ACKNOWLEDGMENTS The writer has been associated with a number of colleagues within Barringer Research Ltd. who have contributed substantially to the research reported in this paper and to the development of the operational AIRTRACE system. These persons include the follow- ing. Geochemical investigations: P. M. Bradshaw, B. Smee, 1. Thompson. Development of operation hard- ware and electronics: M. Failes,* M. Paul, F. Lanza. Analytical developments: M. Silvester, F. Aber- crombie, H. Zwick. Microbiological research: A. Murray. All of the above are members of Barringer Research except as indicated. The writer is especially indebted to Mr. D. Macourt** who has played a very important role in many phases of the AIRTRACE development pro- gramme from its inception. Research on botanical and biological aspects has been carried out in the Botany Department, Imperial College, by J. Barber and W. Beauford. The AIRTRACE development has only been possi- ble through the joint efforts of all of the above together with the combined funding of Barringer Re- search Ltd., the Anglo American Corporation of Canada, and Hudson Bay Mining and Smelting Co. Ltd. SELECTED REFERENCES Barker, D. R., and Zeitlin, H., 1972, Metal-ion concen- trations in sea-surface microlayer and size- separated atmospheric aerosol samples in Hawaii: Jour. Geophys. Research, v. 27, p. 5076— 5086. Barger, W. R., and Garrett, W. D., 1970, Surface ac- tive organic material in the marine atmosphere: Jour. Geophys. Research, v. 75, p. 4561—4566. Barringer, A. R., 1973a, US. Patent No. 3759617. Method of apparatus for geochemical surveying. 1973b, US. Patent No. 3768302. Method of apparatus for sensing substances by analysis of adsorbed matter associated with atmospheric particulates. 1974a, US. Patent Pending. Method and ap- paratus for collecting atmospheric samples. 1974b, Canadian Patent No. 954564. Method and apparatus for transferring particles from one fluid stream to another. 1975, US. Patent No. 386822. High resolution geochemical prospecting method and apparatus. *M. Failes, formerly Barringer Research Limited, now Inde- pendent Consultant, Toronto. **D. Macourt, independent Consultant, Sydney. 249 Beauford, W., Barber, J., and Barringer, A. R., 1975, Heavy metal release from plants into the atmos— phere: Nature, no. 256, p. 35—37. Blanco, A. J., and McIntyre, R. G., 1972, An infra-red spectroscopic view of atmospheric particulates over El Paso, Texas: Atmospheric Environment, v. 6, p. 557—562. Blanchard, D., 1963, The electrification of the atmos- phere by particles from bubbles in the sea: Prog. in Oceanography, v. 1, p. 71—202. Blanchard, D., and Woodcock“ A. H., 1957, Bubble formation and modification in the sea and its meteorological significance: Tellus, no. 2, p. 145— 158. Blanchard, D. C., 1964, Sea-to-air transport of surface active material: Science, v. 146, p. 396—397. 1968, Surface active organic material on air- borne salt particles, in Internat. Conf. on Cloud Physics: Toronto, Proc. Blanchard, D. C., and Syzdek, L. D., 1970, Mechanism for water to air transfer and concentration of bacteria: Science, v. 170, p. 626—628. 1972, Concentration of bacteria in jet drops from bursting bubbles: Jour. Geophys. Research, v. 77, p. 5087—5099. Brooks, R. P., 1972, Biobotany and biogeochemistry in mineral exploration: New York, Harper and Row. Bussinger, J. A., 1972, Remote sensing of the tropo- sphere: U.S. Dept. of Commerce Natl. Oceanic and Atmospheric Admin. (Univ. Colo). Carlucci, A. F., and Williams, P. M., 1965, Concentra- tion of bacteria from sea water by bubble scavenging: J. Cons. Perm. Int. Explor. Mer., v. 30, p. 28—33. Clement, C. R., Jones, L. H. P., and Hopper, M. J., 1971, The leaching of some elements from herb- age plants by simulated rain: Grassland Research Inst, Hurley, Berkshire. Curtin, G. C., King, H. D., and Mosier, E. L., 1974, Movement of elements into the atmosphere from coniferous trees in subalpine forests of Colorado and Idaho: Jour. Geochem. Exploration, v. 3, p. 245—263. Donovon, T., 1974, Petroleum microseepage at Ce- ment, Oklahoma: Evidence and mechanism: Am. Assoc. Petroleum Geologists Bull., v. 58, no. 3, p. 429—446. Eglinton, G., and Hamilton, R. J., 1967, Leaf epicuticu- lar waxes: Science, v. 156, p. 1322—1335. Esmen, N. A., and Corn, M., 1971, Residence time of particles in urban air: Atmospheric Envir., v. 5, p. 571—578. 250 FIRST ANNUAL PECORA Garrett, W. D., 1967, Stabilization of air bubbles at the air-sea interface by surface-active material, I, Sea surface: Deep Sea Research, v. 14, p. 661— 672. Graedel, T. E., and Franey, J. P., 1974, Atmospheric aerosol size spectra: Rapid concentration fluctua- tions and bimodality: Jour. Geophys. Research, v. 79, p. 5643-5650. Hawkes, H. E., and Webb, J. S., 1962, Geochemistry in mineral exploration: Harpers Geoscience Series. Hilst, G. R., and Nickola, P. W., 1959, On the wind erosion of small particles: Am. Meteorological Soc. Bull, v. 40, p. 73—77. Hoffman, E. J., and Duce, R. A., 1974, The organic carbon content of marine aerosols collected on Bermuda: Jour. Geophys. Research, v. 79, p. 4474. Hoffman, G. L., Duce, R. A., and Hoffman, E. J.,1972, Trace metals in the Hawaiian marine atmos— phere: Jour. Geophys. Research, v. 77, p. 5322— 5329. Hutchinson, C. E., 1943, The biogeochemistry of alum- inum and of certain related elements: Quart. Rev. Biology, v. 18, p. 1—29, 129—153, 242—262, and 331—363. Junge, C. E., 1972, Our knowledge of the physico- chemistry of aerosols in the undistributed marine environment: Jour. Geophys. Research, v. 77, p. 5183—5200. Kaimal, J. C., and Bussinger, J. A., 1970, Case studies of a convective plume and a dust devil: Jour. Applied Meteorology, v. 9, p. 612—620. Kartsev, A. A., Tabasarankii, Z. A., Subbota, M. 1., and Mogilevski, G. A., 1954, Geochemical meth- ods of prospecting and exploration for petroleum and natural gas: Univ. California Press. LeClerc, J. A., and Breazeale, J. F., 1908, Plant food removed from growing plants by rain or dew: US. Dept. of Agriculture Yearbook, p. 389—402. Lee, R. E., Jr., Patterson, R. K., and Wagman, J., 1967, Particle size distribution of metal compon- ents in urban air: Am. Chem. Soc. Mtg. Levenson, A. A., 1974, Introduction to exploration geo- chemistry: Calgary, Applied Publishing. Long, W. G., Swell, D. V., and Tukey, H. B., 1956, Loss of nutrients from foliage by leaching as in- dicated by radio—isotopes: Science, v. 27, p. 1039— 1040. Martin, J. T., and Juniper, B. E., 1970, The cuticles of plants: New York, St. Martin’s Press. MEMORIAL SYMPOSIUM ~Moorby, J., and Squire, H. M., 1963, The loss of radio— active isotopes from the leaves of plants in dry conditions: Rad. Botany, v. 3, p. 163—167. Noll, K. E., and Pilat, M. J., 1971, Size distribution of atmospheric particles: Atmospheric Envir., v. 5, p. 527—540. Neumann, G. H., Fonseluis, S., and Wahlman, L., 1959, Measurements on the content of nonvolatile or- ganic material in atmospheric precipitation: Internat. Jour. Air Pollution, v. 2, p. 132—141. Parkin, D. W., Phillips, D. R., and Sullivan, R. A. L., 1970, Airborne dust collections over the North Atlantic: Jour. Geophys. Research, v. 75, p. 1782— 1793. Piotrowicz, S. R., Ray, B. J., Hoffman, G. L., and Duce, R. A., 1972, Trace metal enrichment in the sea-surface microlayer: Jour. Geophys. Research, v. 77, p. 5243—5254. Poet, S. E., Moore, H. E., and Martel], E. A., 1972, Lead 210, Bismuth 210, and Polonium 210 in the atmosphere: Accurate ratio measurement and application to aerosol residence time determina- tion: Jour. Geophys. Research, v. 77, p. 6515— 6527. Rahn, K. A., and Winchester, J. W., 1971, Sources of trace elements in aerosols: An approach to clean air: US. Atomic Energy Commission. Available from US. Dept. of Commerce, Natl. Tech. Inf. Service as COO—1705—9. Stern, A. C., 1968, Source control by centrifugal force and gravity, in Caplan, K. J., Air pollution: New York, Academic Press, v. 3, p. 359—395. Stevenson, R. E., and Collier, A., 1962, Preliminary observations on the occurrence of airborne ma- rine phytoplankton: Lloydia, v. 25, p. 89—93. Szekielda, K. M., Kupferman, S. L., Klemas, V., and Polis, D. F., 1972, Element enrichment in organic films and foam associated with aquatic frontal systems: Jour. Geophys. Research, v. 77, p. 5278-— 5282. Taylor, R. J., 1958, Thermal structures in the lowest layers of the atmosphere: Aust. Jour. Physics, v. 11, p. 168-176. Twomey, S., 1970, On the nature and origin of natural cloud nuclei: L’Observatoire du Puy de Dome Bull, January-March, no. 1, p. 1—19. Walter, H., 1973, Coagulation and size distribution of condensation aerosols: Aerosol Science, v. 4, p. 1—15. ‘ Weiss, 0., 1967, US. Patent No. 3309518. Method of aerial prospecting which includes a step of analys- ing each sample for element content, number and size of particles. AIRTRACETM—AN AIRBORNE GEOCHEMICAL EXPLORATION TECHNIQUE 251 1971, Airborne geochemical prospecting. CIM Woodcock, A. H., and Gifford, M. M., 1949, Sampling Special Vol. 11: Geochem. Explor., p. 502—514. atmospheric sea-salt nuclei over the ocean: Jour. Went, F. W., 1964, The nature of Aitken condensation . .01: Marine Research, V. 8’ no. 1_3’ p 177—191 . . . \ Wilhams, P. M., 1967, Sea surface chemistry. Organic nuclei in the atmosphere: Botany, v. 51, Natl. . . . . Acad Sci Proc p 1259_12 67 carbon and inorganlc nitrogen and phosphates in ' ' " ' ' surface films and sub-surface waters: Deep Sea 1967, Formation of aerosol particulates derived Research, v. 14, p. 791—800. from naturally occurring hydrocarbons produced ZoBell, C. E., and Mathews, H. M., 1939, A qualitative by plants: Air Pollution Control Assoc. Jour., v. study of the bacterial flora of sea and land 17, p. 579—580. breezes: Natl. Acad. Sci. Proc., v. 22, p. 567-572. ‘ PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA An Application of Satellite Imagery to Mineral Exploration By Mark A. Liggett and John F. Childs, Cyprus Georesearch Company, Los Angeles, California 90071 ABSTRACT The application of satellite remote-sensing tech- niques to mineral exploration is based on the ability to recognize a variety of geologic features character— istically associated with hydrothermal alteration and related mineralization. Such features can include favorable structural settings, lithologic associations, and alteration color or topographic anomalies. Suc— cessful application of satellite imagery, however, is dependent upon the size and expression of these characteristic features and their predictive value for narrowing the area of exploration to a practical size for economical evaluation using other exploration techniques. Landsat—1 MSS imagery has been effec- tively used in studying the regional tectonic controls of Cenozoic volcanism, plutonism, and related gold, silver, and base metal mineralization in part of the Basin and Range province of southern Nevada, east- ern California, and northwestern Arizona. Within a volcanogenic province aligned along the Colorado River south of Lake Mead, Nevada, the locations of known mineral deposits appear in satellite imagery to be spatially associated with generally east-west struc- tural trends, transverse to the north—south structural grain typical of the province. Field reconnaissance in this area has confirmed a temporal as well as spatial relationship between mineralization and the anom— alous east-west structural trends. Assuming a genetic relationship between structure and mineralization, systematic analysis of Landsat imagery and subsidiary data has provided a basis for selecting new targets for ground-based exploration reconnaissance. INTRODUCTION Applications of satellite remote-sensing techniques to mineral exploration have generally been based on the ability to observe characteristics of mineralization, associated alteration or other related features in synoptic imagery of relatively low resolution. Such characteristic features can include alteration color and topographic anomalies, distinctive lithologic associa- tions, favorable structural settings, and vegetation changes caused by soil geochemical anomalies. The relative value of these features as exploration guides often varies with the genetic type of mineralization, geologic terrane, and climatic setting. The economic advantages of using satellite imagery depend both on the success of detecting these features and the costs of obtaining comparable information using alternate methods. In the western United States, direct surface expres- sions of alteration and mineralization are likely to have been recognized long ago by conventional means. In such well-explored areas, it is necessary that the exploration methods used lead to the detection of blind or poorly exposed mineralization. One approach to reconnaissance exploration for such ore deposits is the selection of preliminary targets on the basis of favorable structural settings or other regional geologic associations. This paper summarizes a study in which Landsat—1 imagery was used in studying the regional structural settings of known gold, silver, and base metal deposits in part of the Basin and Range province of southern Nevada, eastern California, northwestern Arizona and southwestern Utah as shown in figure 1. The synoptic overview of regional geology and the identification of specific structural features expressed in the Landsat—1 imagery has led to the development of a regional model for the tectonic control of Cenozoic volcanism, plutonism, and genetically related epithermal mineral- ization within a portion of the study area. 253 254 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM SAN DIEGO l I MILES \ , ’ ’ ___________ \\ o 25 so 75 .— ______ , F——F— “- f ,/ r; \ I \ h I'I N.\ E V I o 25 50 75 100 m \ ‘/ \' \ I KILOMETERS 1* I WALKER LAA’E\ I \I. \ I M N E R A L \ I I \_ J 'l-lmrflmnn \\ I /\ v _ ,1 V \g \ T A H / L \ / _ ____ O/TUOLUMNE\\ ”fflg\ / _I l O / 38°N J ’K x x \ I N Jr E \ ,, \‘fl,’ \‘\) MONO / ESMERAL l- G A R Fl \ 2’ q //‘('\\"‘) \\ N C O / N CIdarC/Iy- I ~ I L ____—1 _I __________ IMARIPOS A 1/ /»\ a,“ \ comm- '— ‘— — l" K ,»—’< ,’ \I'llf““—\ II ( K A N \ ,z I, k ”vi/Map \ \, \I IWASHINGTO I / 1—), NI 1 K I ‘ j | SI 610:9: "’ ' T“ ,— on . ,- __ ,< MADERA/r / ‘\ I, _I___I__———- r / \ / t\ I II “I l H T U L A it E St I\ .Pflru’" Les cha: r 1/ \ ' \-. x / I I I ' If \’ \..f \\ // I / /I -\ I \I\ I. In,‘ '. g I I \ I \ o" 9' c o c o N I was}; 7.1;? ———— \ . \ I ~’ ~ \ I | Z O N A / I.'\\ 6 I. . ‘~\ BLAC - OLCANIC .8110, I L1 I‘I I l ‘ 9‘; PROV I L___1 /' I 4.54ng I P \‘-A Ir- S/A N a E R N A R D I I I A v A ,\ I 'i I I x X \ \ {Xx I f ”'_ ‘-\Vr L \“I ,_J’ K v \\ l ‘9‘ 0’1 ‘l I r— 5011 Elma/dim: \- /’Parhr II o — ‘- '_ I ________ I . ____————-—--—————' 34N\ I , w , \‘ l 3 _.’ I I '2 l I I .- . R I V E R S I D E I I M A R I C O P A J I «I I Phoenix. _I'-\~ _______ d/ i _____ ‘ I l‘~‘______ ALro/vsa 5 Y U M A I I l l ) I PERIAL \_ I FIGURE 1.—-Location map for the area of study. The generalized positions of the Black Mountains and Nye County volcanic provinces are shown with the stippled pattern. The area of the Landsat mosaic of figure 2 (p. XXII) is outlined with a heavy line. LANDSAT_1 IMAGERY from black and white MSS transparencies using a A false-color mosaic of Landsat—1 multispectral 32;;ng developed by MacGalliard and LIggett scanner (MSS) composites which covers the area of ' study is shown in figure 2 (p. XXII). The individual Landsat—1 frames used in this mosaic are identified in figure 3. The false-color composites were prepared The false-color composites used in figure 2 were produced with the following combination of M88 bands and printing filters: AN APPLICATION OF SATELLITE IMAGERY TO MINERAL EXPLORATION 255 177°30’W O 2550 FIGURE 3.—|ndex map showing the locations and identification numbers of the Landsat—1 MSS frames used in the false—color mosaic of figure 2 (p. XXII). Area of study is outlined. MSS band Spectral range Printing filter 4 ______ Green and yellow (0.5—0.6 um) ____ Blue 5 ______ Red (0.6—0.7 um) ________________ Green 7 ______ Near-infrared (0.8—1.1 pm) ________ Red As a result of this MSS band-filter combination, green vegetation appears as intense red clue to the high reflectance of vegetation in the near-infrared portion of the spectrum. Most buff-to-brown rock or soil units in the composites reproduce in approximately their natural colors. Neutral colored igneous and sedi- mentary rocks appear with gray to slightly blue color- ing; dark igneous and metamorphic rocks reproduce with dark-brown to steel-gray coloring. Red, iron-rich sedimentary rocks and iron oxide staining associated with hydrothermal alteration appear as yellow color- ing, sometimes varying between greenish yellow and orange, depending on the mineralogy and moisture content of the surface material. Water bodies appear dark blue or black due to water’s high absorption in the spectral range recorded by the three MSS bands. Individual Landsat~l images in the form 1:500,000- scale false-color composites and spectral ratio images (Liggett and research staff, 1974) were studied in comparison with the geologic and structural maps referenced in the bibliography. Analysis of the satellite imagery focused primarily on structural lineaments considered to be possible faults of Cenozoic age. These structures were systematically compared with available geologic maps or checked in the field. The tectonic map of figure 4 is a compilation of those lineaments identified as faults or fault zones which have undergone Cenozoic movement. For the purpose of clarity, faults having traces less than 5 km in length have been eliminated from the compilation. A map of Cenozoic volcanic and plutonic rocks (fig. 5) was compiled from the referenced published IIEIS°W l — l . I I I I w comma /. a ll \"_3"' I“- I '! v-4. .u ~_- . night 34°N_ : n7°w FIRST ANNUAL PECORA MEMORIAL "\ loam; cu, SYMPOSIUM r : ‘ .3333”... i- \j‘. ~ ~35°N 'l \ \'. [—34° N ”gow u3°w FIGURE 4.—Generalized map of the major Cenozoic fault systems visible in the Landsat—1 imagery of figure 2 (p. XXII). Sense of movement on major strike-slip faults is indicated by arrows; bars and balls are on the down-thrown sides of major normal faults. sources and from reconnaissance mapping guided by analysis of Landsat—1 imagery. The mapped distribu- tion of volcanic rock types has been generalized to distinguish between those of predominantly silicic to intermediate composition (rhyolite and trachyte through andesite) and those of predominantly basaltic composition (olivine andesite and basalt). Plutonic rocks shown in figure 5 range in composition from granite to gabbro although most are of intermediate composition. REGIONAL GEOLOGIC SETTING The area shown in the Landsat—1 mosaic of figure 2 (p. XXII) is located along the border between the AN APPLICATION ll7°W 0 34°N— rometeu II7°W uéow Basin and Range province to the west and the C010- rado Plateau to the east. The physiography of the Basin and Range province, as expressed in the Land- sat—1 imagery, is characterized by systems of north- erly trending mountain ranges separated by deep alluvial valleys. The province forms a distinct physio— OF SATELLITE IMAGERY TO MINERAL EXPLORATION \ Lo: Vow: . 257 l|5°W II4°W |l3°W I 5, W I v I—38°N . o a an ,9 III % 5 <3 0 V D i a) ‘l \ t "J ‘I. ‘ " —37°N —36°N l—34°N WW I II4°W FIGURE 5.—Generalized map of the Cenozoic volcanic and plutonic rocks in the study area. Solid black areas are plutonic bodies of predominantly intermediate composition; the outlined areas are volcanic rocks of generally siIicic to intermediate com- position; and the stippled areas are volcanic rocks of predominantly basaltic composition. I ||5°W graphic and structural terrane that can be traced from southern Oregon into northern Mexico. Much of the Basin and Range province is underlain by Precambrian, Paleozoic, and Mesozoic rocks which were deformed during several orogenies of late Paleozoic and Mesozoic age (Armstrong, 1968). The 258 formation of the characteristic Basin and Range phy- siography began in mid-Tertiary time with the onset of regional normal faulting and widespread volcanic and plutonic activity which followed a 60-million-year period of relative crustal stability. The Cenozoic structure of the province is charac- terized by northerly trending systems of complex grabens, horsts, and tilted blocks bounded by normal faults (Stewart, 1971). Major range-front faults typi- cally dip at approximately 60° toward the basins with the angle of dip decreasing at depth (Hamilton and Myers, 1966). However, range-front faults having large displacements and dips of less than 25° are not uncommon in the Basin and Range province (Long— well, 1945; Anderson, 1971; Liggett and Ehrenspeck, 1974). Within the area of study, much of the late Cenozoic volcanic activity is concentrated within two general- ized areas referred to here as the Black Mountains and Nye County volcanic provinces (figs. 1 and 5). The volcanic rocks of these provinces are generally silicic to intermediate in composition and were erupted from fissure systems and caldera structures which are closely associated with the normal faults that characterize the Basin and Range province. Within the Black Mountains volcanic province, ero— sion has exposed several large, late Cenozoic intru- sive bodies which are genetically related to the wide- spread silicic and intermediate volcanic rocks (fig. 5). In several areas, massive dike swarms are exposed which cut both the crystalline basement and overlying volcanic rocks. These swarms are believed to have been feeders for at least some of the volcanic rocks, and their emplacement can be shown to be both temporally and spatially related to Basin and Range normal faulting (Liggett and Childs, 1974). Systems of right- and left-lateral strike-slip faults are mapped within the Basin and Range province, generally striking at high angles to the northerly trends typical of the province. The estimates of lateral displacements on these fault systems range up to tens of kilometres. An example is the Las Vegas Shear Zone which can be traced from the southern margin of the Nye County volcanic province to the northern margin of the Black Mountains volcanic province, a distance of nearly 150 km (fig. 4). Various theories which have been proposed to ex- plain the origin of Basin and Range structure are discussed in excellent summaries by Nolan (1943), Gilluly (1963), Roberts (1968), and Stewart (1971). Most concepts can be separated into the following three categories: FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 1. Basin and Range structure has resulted from the collapse of the upper crust caused by such mechanisms as lateral transfer of lower crustal material (Gilluly, 1963) or eruption of huge volumes of volcanic rocks (Le Conte, 1889; Mackin, 1960). 2. Basin and Range structure has formed en echelon to movement on deep-seated, conjugate sets of right- and left-lateral strike-slip faults (Shawe, 1965; Sales, 1966). 3. Basin and Range structure is the result of regional crustal extension in an east-west direction (Cook, 1966; Hamilton and Myers, 1966; Rob- erts, 1968; Stewart, 1971). This process is be- lieved to have occurred by plastic extension of the lower crust, perhaps accompanied by in- trusion of plutons beneath Basin and Range grabens (Thompson, 1966). The net amount of crustal extension has been estimated to be as great as 300 km, or 100 percent of the former width of the province (Hamilton and Myers, 1966). Most current theories of Basin and Range structure presume net crustal extension within the province during late Cenozoic time. Although the causes, mech- anisms, and amounts of extension in the province re- main controversial, evidence of extension has been documented by geologic mapping and geophysical studies. TECTONIC MODEL Based on analysis of Landsat—1 imagery and data published by Fleck (1970a), a tectonic model was pro- posed by Liggett and Childs (1974) in an attempt to explain the temporal and spatial relationships ob- served for strike-slip deformation on the Las Vegas Shear Zone and normal faulting, plutonism, volcan- ism, and related mineralization in the Black Moun- tains and Nye County volcanic provinces. Figure 6 shows by means of a simplified model how the Las Vegas Shear Zone is believed to have functioned as an intracontinental “ridge—ridge trans- form fault” (as defined by Wilson, 1965) which formed in response to east-west crustal extension in the Black Mountains and Nye County volcanic provinces. The geology and chronological development of the Las Vegas Shear Zone and the two areas of inferred crustal extension are summarized below. LAS VEGAS SHEAR ZONE A major zone of right-lateral strike-slip deforma- tion passing through Las Vegas Valley was first pos- AN APPLICATION OF SATELLITE IMAGERY TO MINERAL EXPLORATION 259 x 100 MILES 100 KILOMETERS Y FIGURE 6.——Diagrammatic model relating right—lateral strike—slip movement on the Las Vegas Shear Zone to crustal extension in the two volcanic provinces (stippled patterns). The crustal extension is expressed in the model by major low—angle normal faulting which has caused anomalous thinning of the crust beneath the volcanic provinces. The amount of extension is repre- sented by the increase in length of X—Y to X’—Y’. tulated by Gianella and Callaghan (1934) in their dis- cussion of the regional implications of the Cedar Mountain earthquake of 1932. The existence of this fault zone was supported by detailed mapping and named the Las Vegas Valley shear zone by Longwell (1960) (see fig. 4). Indirect evidence suggesting more than 40 km of right-lateral strike-slip on this structure has been cited by several workers. These estimates have been based on displacements of stratigraphic isopachs and sedi- mentary facies (Longwell and others, 1965; Fleck, 1967; Stewart and others, 1968) and offset of dis- tinctive thrust faults of the Sevier orogenic belt (Long- well and others, 1965; Fleck, 1967). At the scale of the Landsat—1 imagery, evidence for the shear zone is expressed in the fiexure of the range trends im- mediately north and south of the zone. Compensat- ing for the effect of flexure along the shear zone, Fleck (1967) estimated approximately 70 km of right- lateral displacement of features across the deformed belt bordering the shear zone. Of this total, a net slip of 30 km has been estimated for features along the trace of the structure. The eastern termination of the Las Vegas Shear Zone is thought to lie in the vicinity of Lake Mead where major right-lateral slip on the shear zone gives way to a system of left-lateral oblique-slip faults (fig. 4). Immediately north of Lake Mead, Anderson (1973) has mapped two halves of a Miocene stratovolcano which are displaced left-laterally a distance of ap- proximately 19 km along a northeast-striking fault zone. This structure, called the Hamblin Bay Fault, strikes at a low angle to the easternmost mapped branch of the Las Vegas Shear Zone. East of the Hamblin Bay Fault, the Gold Butte, and Lime Ridge Faults of Longwell and others (1965) are considered by Anderson (1973) to be left-lateral strike-slip faults. This hypothesis is supported by the fiexure of hogback ridges observed adjacent to these faults in the Landsat—1 imagery. The amount of left- lateral strike-slip deformation in this region is poorly known, although it is believed to exceed 19 km (Anderson, 1973). The spatial relationship of these structures to late Cenozoic normal faulting, volcan- ism, and plutonism in the Saint George—Cedar City area of Utah, suggests that the strike-slip deforma- 260 tion may be mechanically related to localized crustal extension in that area (figs. 4 and 5). An alternate model which involves differential subcrustal flow of the mantle has been proposed by Anderson (1973). The duration of movement on the Las Vegas Shear Zone has been estimated (Fleck, 1967) from radio- metric age determinations of rock units exposed along its trace. A radiometric age date of 15 million years from deformed beds of the Gale Hills Formation sug- gests that significant movement has occurred on the shear zone since that time. Undeformed basalts of the Muddy Creek Formation along the shear zone have been dated at 10.7 million years. From these dates and field evidence, Fleck (1967) concluded that most strike-slip movement on the Las Vegas Shear Zone probably occurred during the period from 17 to 10 million years ago. BLACK MOUNTAINS VOLCANIC PROVINCE The Black Mountains volcanic province is a dis- tinctive igneous and structural terrane which extends southward along the Colorado River from Lake Mead, Nevada, to near Parker, Arizona (figs. 1 and 5). Re- connaissance maps of portions of this region have been published by Longwell (1963), Longwell and others (1965), Wilson and others (1969), and Volborth (1973). Detailed studies of mining districts within the province have been published by Schrader (1917), Ransome (1923), Callaghan (1939), Hansen (1962), Anderson (1971), and Thorson (1971). The Black Mountains volcanic province is charac- terized by thick deposits of ignimbrites, flows, and volcaniclastic sediments of generally silicic to inter— mediate composition, although thin basalt fiows are locally widespread. Near Nelson, Nevada, the com— posite thickness of the late Tertiary volcanic sequence is estimated to be over 5 km (Anderson and others, 1972). The volcanics were deposited on an erosional sur- face developed on a crystalline basement of Precam- brian gneiss and granite. In parts of the province, the Precambrian basement may have been subjected to metamorphism during a Jurassic orogeny (Vol- borth, 1973). Both the pre-Tertiary crystalline base- ment and the overlying volcanic rocks have been in- truded by plutons of Miocene age (Anderson and others, 1972; Volborth, 1973). The plutons are gen- erally elongate north-south, and range in composition from leucocratic granite to gabbro, although granite, quartz monzonite, and quartz diorite predominate (Anderson and others, 1972). Structurally controlled, northerly striking dikes of rhyolite, andesite, and diabase cut both the crystalline FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM basement and the volcanic cover throughout much of the province. In the Newberry Mountains, 25 km southeast of Searchlight, Nevada, a massive swarm of dikes is especially well—exposed forming a belt over 10 km wide. These dikes form a bimodal suite consisting of porphyritic rhyolite and hornblende diabase. Near Nelson, Nevada, dikes of similar com- positions are exposed in the lower portions of the volcanic cover, generally decreasing in number up- ward in the stratigraphic section. It is probable that these dike swarms fed much of the volcanic cover and were in part synchronous with plutonism (Lausen, 1931; Volborth, 1973; Liggett and Childs, 1974). A close genetic relationship between Tertiary intru- sive rocks and chemically equivalent volcanic facies was suggested as early as 1923 by Ransome in a re- connaissance study of the Oatman mining district, Arizona. This conclusion has been supported by more recent mapping and geochemical studies in that dis- trict by Thorson (1971). Callaghan (1939) suggested a genetic relationship for an intrusive body and adja- cent volcanic rocks in the Searchlight district, Ne- vada. Based on detailed geochemistry and radio- metric age date analysis, Volborth (1973) has pro- posed a genetic interrelationship for plutonism, hy- pabyssal dike emplacement and volcanism in nearly the entire western half of the Black Mountains vol- canic province. Most of this igneous activity occurred during the period from about 18 to 10 million years before present (Thorson, 1971; Anderson and others, 1972; Volborth, 1973). The structure of the Black Mountains volcanic province is dominated by northerly striking normal faults, many of which dip at angles of 10° to 20°. This style of deformation has been mapped in detail near Nelson, Nevada, by Anderson (1971) who at- tributed the low-angle faulting to extreme east-west distension of the upper crust. Individual normal faults within the Black Moun- tains volcanic province are estimated to have dis- placements of 2 to 5 km (Anderson, 1971; Anderson and others, 1972). Within many of the uplifted blocks, erosion has removed the volcanic cover, and expo- sures of crystalline basement are commonly jux- taposed against thick sequences of late Tertiary vol- canic rocks. Although there is insufficient stratigraphic evidence upon which to base accurate estimates of dip-slip on most of the faults in the Black Mountains volcanic province, the frequency of major low-angle normal faults suggests that crustal extension may exceed 70 km. Estimates of similar magnitude have been made for other portions of the Basin and Range province AN APPLICATION OF SATELLITE IMAGERY TO MINERAL EXPLORATION by Hamilton and Myers (1966); Proffett (1971), and Davis and Burchfiel (1973). Based on a seismic-refraction profile recorded across the Las Vegas Shear Zone from near King- man, Arizona, toward the Nye County volcanic prov- ince, Roller (1964) has suggested that an anomalously thin crust, 27 km thick, underlies the Black Mountains volcanic province. North of the Las Vegas Shear Zone, the thickness of the crust increases rapidly to 32 km. The seismic-refraction pattern is supported by the existence of a northerly trending Bouguer gravity high (US. Air Force, 1968) which is aligned with the trend of the Black Mountains volcanic province. The gravity anomaly suggests an upward bulge of the mantle beneath the volcanic province, possibly the result of isostatic compensation for a thin, distended crust. The high gravity anomaly, like the volcanic province, terminates north of Lake Mead along the Las Vegas Shear Zone and Hamblin Bay Fault sys- tems. The southern end of the Black Mountains volcanic province is complex and indefinite. The pattern of volcanism, plutonism, and normal faulting within the volcanic province appears to terminate in the vicinity of Parker, Arizona, against a broad zone of north- westerly striking faults. Although this fault system is poorly mapped, field reconnaissance by the authors near Vicksburg, Arizona, and geologic mapping of the Quartzite quadrangle, Arizona, by Miller (1970) indi- cate right-lateral strike-slip on many of the northwest— erly striking faults. The total amount of displacement on this fault system is unknown. NYE COUNTY VOLCANIC PROVINCE The Nye County volcanic province lies northwest of the Las Vegas Shear Zone in southern Nye County, Nevada (figs. 1 and 5). This province contains 10 known volcanic centers, at least 5 of which are be- lieved to have formed caldera structures (Ekren, 1968). Detailed geologic studies have been conducted in several mining areas including Rhyolite and Beatty (Cornwall and Kleinhampl, 1964) and Goldfield (Ran- some, 1909; Cornwall, 1972; Ashley, 1974). Detailed mapping, stratigraphic, and geophysical studies have been conducted by the US. Geological Survey in the US. Atomic Energy Commission’s Southern Nevada Test Site (Eckel, 1968). The Tertiary ignimbrites and flows recognized within the Nye County volcanic province are esti— mated to have a composite thickness of approxi- mately 9 km and a volume of over 11,000 km3 (Ekren, 1968). Most of these rocks range in composition from dacite to rhyolite, although andesite and basalt flows 261 are locally abundant (Anderson and Ekren, 1968; Ekren, 1968). The volcanic units unconformably over- lie a basement of Paleozoic carbonate rocks, which were folded, thrust faulted and metamorphosed dur- ing pre-Tertiary time. Several Mesozoic plutons have intruded this Paleozoic basement. Numerous small plugs, domes, and dikes which range in composition from rhyolite to andesite with minor diabase have intruded the Tertiary volcanic cover. These intrusives are similar in composition to the volcanic rocks and many are believed to have been feeders (Ekren and others, 1971). Most of the volcanism and plutonism within the Nye County vol- canic province was synchronous with major normal faulting in a time span from approximately 26.5 to 11 million years ago (Ekren and others, 1968). The Tertiary deformation within the Nye County volcanic province is dominated by Basin and Range normal faults which have produced a complex horst and graben structure similar to that of the Black Mountains volcanic province. This structural pattern is locally interrupted by caldera subsidence structures, domes, and radial faults related to separate volcanic centers in the province. Arcuate faults which rim cal— dera structures have displacements estimated on the basis of gravity anomalies and drill hole data to be as great as 2 km (Orkild and others, 1968). Much of the Basin and Range structure is masked by the youngest volcanic rocks, and the full extent of nor— mal faulting is unknown. On the basis of gravity data, several basins are estimated to be filled with as much as 4.8 km of volcanic rock (Healey, 1968); structural control of these basins is suggested by the pro- nounced northerly trends of the gravity anomalies. The southern margin of the Nye County volcanic province is formed by the western end of the Las Vegas Shear Zone. Within this area of complex struc- tural intersection, several short northeast-striking faults have been mapped which are believed to have undergone left-lateral strike-slip movement of from 3 to 5 km (Ekren, 1968). No continuation of the Las Vegas Shear Zone has been recognized west of the Nye County volcanic province. CHRONOLOGY The structural model proposed here for late- Tertiary deformation in the southern Basin and Range province is supported by the synchronism of strike- slip movement on the Las Vegas Shear Zone and volcanism, plutonism, and normal faulting in the two areas of inferred crustal extension. The chronology of these events is summarized in figure 7. 262 STRIKE-SLIP DISPLACEMENT ON THE LAS VEGAS SHEAR ZONE BASIN & RANGE NORMAL FAULTING: Black Mtns. volcanic province Nye Co. volcanic province SILICIC TO INTERMEDIATE VOLCANISM & PLUTONISM: Black Mtns. volcanic province Nye Co. volcanic province PRECIOUS & BASE METAL MINERALIZATION: Black Mtns. volcanic province Nye Co. volcanic province BASALTIC VOLCANISM: Black Mtns. volcanic province Nye Co. volcanic province FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 54:05 l_lllll l—l I’ll, lllllll 30 25 l fi 20 15 10 5 0 x 106 million years before present FIGURE 7.—Generalized chronology of strike-slip movement on the Las Vegas Shear Zone and normal faulting, igneous activity, and related mineralization in the Black Mountains and Nye County volcanic provinces. Within the area of study, the folding and thrust faulting of the Sevier orogeny is believed to have been confined to a relatively brief time span between 90 and 75 million years ago (Fleck, 1970b). The Sevier orogeny appears to have been followed by a long period of relative stability and moderate erosion which by mid-Tertiary time had resulted in a broad terrane of subdued topography. During mid-Tertiary time the area south of Lake Mead appears to have been the site of a broad arch which shed arkosic conglomerates and fanglomerates containing fragments of Precambrian rock toward the northeast. These sediments overlie an erosional sur- face cut in the Paleozoic rock of the western Colorado Plateau northeast of Kingman, Arizona. The sedi- ments are conformably overlain by the Peach Springs Tuff of Young (1966) which has been variously dated at 18.3i0.6 and 16.9:04 million years (Young and Brennan, 1974). These relationships indicate that by Miocene time erosion had unroofed Precambrian basement in the arch south of Lake Mead and that major normal faulting had not yet separated deposi- tional areas on the Colorado Plateau from source areas in the ancestral Basin and Range province. Following deposition of the Peach Springs Tuff, Basin and Range faulting disrupted the northeast-flowing drainage, leading to formation of a new pattern of isolated structural depressions filled by basin deposits (Lucchitta, 1972). According to Anderson and others (1972), in the core of the Black Mountains volcanic province the oldest volcanic rocks overlying the Precambrian AN APPLICATION OF SATELLITE IMAGERY TO MINERAL EXPLORATION crystalline basement are tuff units believed to be 18.6 million years old; the youngest volumetrically significant volcanic units in the area consist of tuffs and flows dated at 12.7 million years; and most of the epizonal plutonic rocks in the Black Mountains vol- canic province range in age from 17 to 12 million years. In the Nye County volcanic province, the chronol- ogy of volcanism and structural deformation is similar to that in the Black Mountains volcanic province. An erosional surface of low relief developed on the folded and thrust faulted Paleozoic basement in early Tertiary time. This surface was covered by a widespread welded tuff unit dated at 26.5 million years (Ekren and others, 1968). Shortly after the eruption of this tuff, normal faults with northeast and northwest strikes began to develop. The typical north-trending Basin and Range normal faults first began to form in the Nye County volcanic province after deposition of a tuff breccia dated at 17.8 million years. The present mountain ranges are believed to have been well de— fined prior to eruption of the Thirsty Canyon Tuff dated at 7 million years (Ekren and others, 1968), although minor normal faulting has continued to the present. Albers and Kleinhampl (1970) have studied the genetic relationships between precious metal mineral- ization and spatially associated volcanic centers of Cenozoic age throughout Nevada. On the basis of radiometric age determinations these workers suggest that mineralization was at least temporally related to the late stages of igneous activity in the associated volcanic centers. The available evidence suggests similar relationships in the Black Mountains volcanic province. In summary, the first major normal faulting, vol— canism, and plutonism in the Black Mountains and Nye County volcanic provinces began in the period from approximately 26 to 18 million years ago. Right- lateral strike-slip movement on the Las Vegas Shear Zone appears to have begun about 17 million years ago and to have continued with synchronous igneous activity and extensional normal faulting in both vol- canic provinces. Major strike-slip movement, igneous activity, and related mineralization ended by approxi- mately 10 million years ago, although Basin and Range normal faulting and intermittent basaltic vol- canism have continued to the present time. STRUCTURAL CONTROL OF MINERALIZATION IN THE BLACK MOUNTAINS VOLCANIC PROVINCE The gold, silver, and base metal mineralization of Cenozoic age in the Black Mountains and Nye County 263 volcanic provinces is both spatially and temporally related to the igneous activity and structural deforma- tion of these two areas. The structural control of mineralization in the Black Mountains volcanic prov- ince is of particular interest since both the regional structural setting of the province and the key struc— tural trends which have localized mineralization are visible in the Landsat—1 imagery. The major ore production in the Black Mountains volcanic province has come from the Eldorado and Searchlight districts in southern Clark County, Nevada, and the Oatman and Katherine districts east of the Colorado River in Mohave County, Arizona (fig. 8). Small mine workings and prospects are found through- out the province. The local geology and structural controls of mineralization in these districts are sum- marized below. ELDORADO MINING DISTRICT The Eldorado mining district is located in southern Clark County, Nevada, approximately 35 km south of Lake Mead. Most of the mines in the district are located within a 10-km radius of the town of Nelson (fig. 8). The host rocks for mineralization include Pre- cambrian gneiss and Miocene plutonic and volcanic rocks of silicic to intermediate composition. The structure of the Eldorado mining district is dominated by closely spaced, northerly striking normal faults, which commonly have dips of less than 30°. Anderson (1971) has attributed this system of low— angle normal faults to tensional rifting and distension of the upper crust, possibly related to the intrusion of a granitic pluton at shallow depth. In spite of the strong north-south structural grain, mineralization in the district is in almost all cases controlled by a sys- tem of steeply dipping, east-west striking faults. This transverse structural trend is well expressed in the Landsat—1 imagery. The mineralization occurs in veins of fault breccia cemented with quartz and calcite gangue. The gold and silver are believed to have been contained in dis- seminated pyrite with minor galena, sphalerite, and chalcopyrite; however, the original sulfide minerals are now largely oxidized (Hansen, 1962). The veins typically show evidence of multiple episodes of fault movement followed by cementation and mineraliza- tion; many of the transverse faults along which the veins were formed are known to have undergone strike-slip movement (Anderson, 1971). Total production from the Eldorado mining district during the period 1907—1945 is cited by Hansen (1962) as 101,522 02 gold and 2,339,353 02 silver, with minor production of copper and lead. 264 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM miles 25 5 O 25 50 kilometers O O “5 W “4 W FIGURE 8.——Areas of late Cenozoic precious and base metal mineralization (stippled) in the Black Mountains volcanic province in relation to the Cenzoic fault systems visible in the Landsat imagery. AN APPLICATION OF SATELLITE IMAGERY TO MINERAL EXPLORATION A relatively small mining area known as the Weaver district is located in the northern Black Mountains of Arizona due east of the Eldorado mining district. Like the Eldorado district, precious metals and minor copper occur in quartz and calcite vein systems which generally strike from N. 45° to 100° E., transverse to the structural grain of the province. Host rocks include Precambrian gneiss, Miocene volcanic rocks, and granites of probable Tertiary age. Although exact production figures are unavailable, total production from the Weaver district is believed to be small. SEARCHLIGHT MINING DISTRICT The Searchlight mining district is located at the southern end of the Eldorado Mountains, approximate- ly 30 km south of the Eldorado mining district (fig. 8). The most prominent geologic feature of the district is a large quartz monzonite pluton of Miocene age‘ which intruded a crystalline basement of Precambrian gneiss and granite and Tertiary volcanic rocks of inter- mediate composition (Callaghan, 1939). Although the shape of the pluton is generally controlled by the northerly structural grain of the Black Mountains vol- canic province, the pluton is terminated on the south by a major east-west striking fault (Volborth, 1973). Gold, silver, copper, and lead mineralization in the Searchlight district occurs in structurally controlled vein systems in and adjacent to the quartz monzonite pluton. Approximately 25 of the 34 productive veins in the district strike between N. 80° and 123° E.; most of these clip steeply to the south (Callaghan, 1939). The veins contain clasts of brecciated country rock usually cemented with quartz gangue although calcite and adularia occur locally. As in the Eldorado district, the veins show evidence of multiple episodes of fault movement followed by cementation and mineraliza- tion. Although the sense and amount of movement on the east-west striking faults is not known, these trans- verse structures contain the principal mineralization of the district. Production from the Searchlight district for the period 1902—1934 is reported by Callaghan (1939) to be 207,570 oz gold, 219,596 oz silver, 650,550 lb cop- per, and 1,675,560 lb lead. An area of numerous small gold and silver mines is located in Tertiary volcanic rocks east of the Colo- rado River, approximately 30 km east-northeast of the Searchlight district. The largest producing mine in this area is known as the Golden Door mine. The major mineralized quartz-calcite veins in the area strike generally eastward with generally steep dips; sub- ordinate veins are irregular in strike and have shallow 265 dips. Recent development and exploration work in the Golden Door mine area has concentrated on flat- lying volcanic units which have been preferentially silicified and mineralized. OATMAN MINING DISTRICT The Oatman mining district is located east of the Colorado River on the western slope of the Black Mountains, approximately 115 km south of Lake Mead (fig. 8). Most of the production from the Oatman dis- trict has come from vein systems within a thick se- quence of late Tertiary volcanic rocks of silicic to intermediate composition. These volcanic rocks un- conformably overlie a crystalline basement of Pre- cambrian granite and gneiss. Both the crystalline basement and overlying volcanic rocks were locally introduced by dikes and small granitic plutons which are believed to be genetically related to the volcanic cover (Thorson, 1971). The mineralized veins were formed within trans- verse fault zones, most of which range in strike from west to northwest and dip steeply toward the north or northeast. Several of the northwest—striking faults have undergone right—lateral strike-slip movement. The veins range in width from a few centimetres to several metres, and typically contain clasts of country rock cemented with a gangue of coarsely crystalline quartz and calcite with smaller amounts of micro‘ crystalline adularia and fluorite. The veins commonly show a pronounced banding in cross section, produced by successive episodes of fault movement, cementa- tion, and mineralization. Lausen (1931) reported five distinct stages of quartz deposition in the mineralized veins of which the later stages generally contain the highest ore values. Host rocks adjacent to the veins commonly show evidence of propylitic alteration. A small amount of gold and silver has been pro- duced from several mines in an area known as the Katherine mining district, located approximately 25 km north-northwest of Oatman. The veins in the Kath- erine district are similar in mineralogy to those at Oatman, although unlike the Oatman veins, they occur primarily within the Precambrian granitic basement which has been locally intruded by dikes and plugs of rhyolite (Lausen, 1931). The mineralized veins in the Katherine mining district generally strike at high angles to the northerly structural grain of the Black Mountains volcanic province. The combined value of gold and silver production from the Oatman and Katherine mining districts is reported by Lausen (1931) as $35,417,926 for the period from 1897 through 1928. This is estimated to represent a metal production of approximately 266 1,714,000 02 t gold and 870,000 oz t silver from ore averaging approximately 0.6 oz of gold and 0.3 oz of silver per ton. DISCUSSION Field reconnaissance and study of geologic litera— ture guided by analysis of Landsat—1 imagery have led to a model which relates strike-slip deformation on the Las Vegas Shear Zone to normal faulting, vol- canism, and plutonism in the Black Mountains and Nye County volcanic provinces. The Las Vegas Shear Zone is believed to have functioned as a ridge-ridge transform fault, which separated these two areas of east-west crustal extension. Geochronology and field evidence indicate that the right-lateral strike—slip move- ment on the Las Vegas ’Shear Zone was synchronous with normal faulting, igneous activity, and related mineralization in both volcanic provinces. The east—west crustal extension in the Black Moun- tains volcanic province is represented by northerly trending systems of normal faults, dike swarms, and shallow plutons. The pronounced north-trending struc- tural grain of the province is locally crossed by transverse fault systems, several of which are known to have undergone strike-slip movement. These trans- verse structures are well expressed in the Landsat—1 imagery and have played an important role in local— izing late Cenozoic precious and base metal mineral- ization within the volcanic province. In several areas within the Black Mountains volcanic province a complex interrelationship has been ob- served between the normal and strike-slip fault systems. Many of the northerly striking normal faults terminate against transverse strike—slip faults without apparent offset. In addition, transverse strike—slip faults are frequently observed to end by turning abruptly in strike to merge with north-striking normal faults. Examples have been mapped by Anderson (1971) and Volborth (1973) in parts of the Boulder City and Nelson 15’ quadrangles. This interrelation- ship of strike-slip and normal faults suggests that these structures may be mechanically related as illustrated in the diagrammatic model of figure 9. This model suggests that the transverse strike-slip faults in the Black Mountains volcanic province may have formed as minor transform faults which sepa- rated areas of differential crustal extension. The mechanisms by which the transverse faults have controlled the location of mineralization are not well understood. It is possible that the steep dips of these structures provided effective conduits for ascending hydrothermal solutions. Because each transverse fault is mechanically linked to a system of normal faults, it FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 9.—Diagrammatic model illustrating the mechanical in- terrelationship of strike-slip and normal faulting believed to exist in the Black Mountains volcanic province. Note that movement on the strike-slip fault zone is mechanically linked to slip on the system of low-angle normal faults. would remain active, and thus open to ascending solu- tions, over prolonged periods of extensional deforma- tion. In contrast, individual gently dipping normal faults could be easily sealed by a variety of processes and are not likely to have afforded direct channelways for mineralizing solutions. The model proposed here for the regional structural setting of the Black Mountains volcanic province uni- fies many of the temporal and spatial relationships recognized for the Cenozoic tectonics, igneous activity, and mineralization in this part of the Basin and Range province. Although many details of the structural de— formation and the physicochemical controls of magma genesis and mineralization are not understood, the structural model and specific structural features ex- pressed in the Landsat—1 imagery provide an effective basis for extrapolating mineralized trends and for selecting blind or poorly exposed targets for further investigation. The application of satellite remote-sensing tech— niques to mineral exploration is not likely to replace the use of conventional methods. However, the synoptic perspective of satellite imagery can provide a valuable framework for synthesizing diverse geo- logic data and for guiding field research to test work- ing hypotheses. Integrated with other exploration AN APPLICATION OF SATELLITE IMAGERY TO MINERAL EXPLORATION techniques as part of a systematic program, satellite imagery can be a valuable tool in reconnaissance exploration. ACKNOWLEDGMENTS This investigation benefited greatly from the work of R. E. Anderson, R. J. Fleck, C. R. Longwell, A. Volborth, and the other individuals cited in the refer- ences. Wally MacGalliard prepared the false-color composites of Landsat—1 MSS imagery used in this investigation, and H. E. Ehrenspeck compiled much of the data presented in figures 4 and 5. The authors, however, are solely responsible for errors in fact or interpretation. This research was supported jointly by Cyprus Mines Corporation and the National Aeronautics and Space Administration under Contract NAS5-21809. REFERENCES CITED Albers, J. P., and Kleinhampl, F. J., 1970, Spatial rela- tion of mineral deposits to Tertiary volcanic centers in Nevada: US. Geol. Survey Prof. Paper 700—C, p. C1—C10. Anderson, R. E., 1971, Thin skin distension in Tertiary rocks of southeastern Nevada: Geol. Soc. Amer- ica Bull, v. 82, p. 43—58. 1973, Large-magnitude late Tertiary strike-slip faulting north of Lake Mead, Nevada: US. Geol. Survey Prof. Paper 794, 18 p. Anderson, R. E., and Ekren, E. B., 1968, Widespread Miocene igneous rocks of intermediate composi— tion, southern Nye County, Nevada, in Eckel, E. 3, ed, Nevada Test Site: Geol. Soc. America Mem. 110, p. 57—63. Anderson, R. E., Longwell, C. R., Armstrong, R. L., and Marvin, R. F., 1972, Significance of K—Ar ages of Tertiary rocks from the Lake Mead re- gion, Nevada-Arizona: Geol. Soc. America Bull, v. 83, p. 273-288. Armstrong, R. L., 1968, Sevier orogenic belt in Nevada and Utah: Geol. Soc. America Bull., v. 79, p. 429—458. Ashley, R. P., 1974, Goldfield mining district, in Guide- book to the geology of four Tertiary volcanic centers in central Nevada: Nevada Bur. of Mines and Geology Rept. 19, p. 49—66. Callaghan, Eugene, 1939, Geology of the Searchlight district, Clark County, Nevada: US. Geol. Survey Bull. 906 D, p. 135—185. Cook, K. L., 1966, Rift system in the Basin and Range province, in The world rift system: Canada Geol. Survey Paper 66-14, p. 246—279. 267 Cornwall, H. R., 1972, Geology and mineral deposits of southern Nye County, Nevada: Nevada Bur. Mines and Geology Bull. 77, 49 p. Cornwall, H. R., and Kleinhampl, F. J., 1964, Geology of Bullfrog quadrangle and ore deposits related to Bullfrog Hills caldera, Nye County, Nevada, and Inyo County, California: US. Geol. Survey Prof. Paper 454—J, p. J1—J25. Davis, G. A., and Burchfiel, B. C., 1973, Garlock fault: An intracontinental transform structure, southern California: Geol. Soc. America Bull., v. 84, p. 1407—1422. Eckel, E. 8., ed., 1968, Nevada Test Site: Geol. Soc. America Mem. 110, 290 p. . Ekren, E. B., 1968, Geologic setting of Nevada Test Site and Nellis Air Force Range, in Eckel, E. B., ed., Nevada Test Site: Geol. Soc. America Mem. 110, p. 11—19. Ekren, E. B., Anderson, R. E., Rogers, C. L., and Noble, D. C., 1971, Geology of northern Nellis Air Force Base Bombing and Gunnery Range, Nye County, Nevada: US. Geol. Survey Prof. Paper 651, 91 p. Ekren, E. 8., Rogers, C. L., Anderson, R. E., and Orkild, P. P., 1968, Age of Basin and Range normal faults in Nevada Test Site and Nellis Air Force Range, Nevada, in Eckel, E. B., ed., Nevada Test Site: Geol. Soc. America Mem. 110, p. 247— 250. Fleck, R. J., 1967, The magnitude, sequence, and style of deformation in southern Nevada and eastern California (Ph.D. thesis): Berkeley, Univ. Cali- fornia, 92 p. 1970a, Age and possible origin of the Las Vegas shear zone, Clark and Nye Counties, Nevada: Geol. Soc. America, Abs. with Programs (Rocky Mountain sect), v. 2, no. 5, p. 333. 1970b, Tectonic style, magnitude, and age of deformation in the Sevier orogenic belt in south- ern Nevada and eastern California: Geol. Soc. America Bull, v. 81, p. 1705—1720. Gianella, V. P., and Callaghan, E., 1934, The earth- quake of December 20, 1932, at Cedar Mountain, Nevada, and its bearing on the genesis of Basin Range structure: Jour. Geology, v. 42, no. 1, p. 1—22. Gilluly, J., 1963, The tectonic evolution of the western United States: Geol. Soc. London Quart. Jour., v. 119, p. 133-174. Hamilton, W., and Myers, W. B., 1966, Cenozoic tec- tonics of the western United States: Rev. Geo- physics, v. 4, no. 4, p. 509—549. 268 Hansen, S. M., 1962, The geology of the Eldorado mining district, Clark County, Nevada (PhD. thesis): Rolla, Univ. Missouri, 262 p. Healey, D. L., 1968, Application of gravity data to geologic problems at Nevada Test Site, in Eckel, E. 8., ed., Nevada Test Site: Geol. Soc. America Mem. 110, p. 147—156. Lausen, Carl, 1931, Geology and ore deposits of the Oatman and Katherine districts, Arizona: Arizona Bur. Mines and Geol. Series no. 6, Bull. 131, 126 p. Le Conte, J., 1889, On the origin of normal faults and of the structure of the Basin region: Am. Jour. Sci., third series, v. 38, no. 226, p. 257—263. Liggett, M. A., and Childs, J. F., 1974, Crustal exten- sion and transform faulting in the southern Basin Range Province: Argus Exploration Company, NASA Rept. Inv., NASA—CR—137256, E74—10411, 54 p. Liggett, M. A., and Ehrenspeck, H. E., 1974, Pahran- agat shear system, Lincoln County, Nevada: Argus Exploration Company, NASA Rept. Inv., NASA—CR—136388, E74—10206, 12 p. Liggett, M. A., and research staff, 1974, A reconnais- sance space sensing investigation of crustal structure for a strip from the eastern Sierra Nevada to the Colorado Plateau: Argus Explora- tion Company, NASA Final Rept. Inv., NASA— CR—139434, E74—10705, 156 p., plus 16 appen- dices, 17 figures, and 8 plates. Longwell, C. R., 1945, Low-angle normal faults in the Basin Range Province: Am. Geophys. Union Trans, v. 26, p. 107—118. 1960, Possible explanation of diverse structural patterns in southern Nevada: Am. Jour. Sci. (Bradley Volume), v. 258—A, p. A192—A203. —1963, Reconnaissance geology between Lake Mead and Davis Dam, Arizona-Nevada: US. Geol. Survey Prof. Paper 374—E, p. E1-E51. Longwell, C. R., Pampeyan, E. H., Bowyer, B., and Roberts, R. J., 1965, Geology and mineral de- posits of Clark County, Nevada: Nevada Bur. Mines and Geology Bull. 62, 218 p. Lucchitta, I., 1972, Early history of the Colorado River in the Basin and Range province: Geol. Soc. America Bull., v. 83, p. 1933—1948. MacGalliard, Wally, and Liggett, M. A., 1973, False- color compositing of ERTS—l MSS imagery: Argus Exploration Company, NASA Rept. Inv., NASA—CR—135859, E74-10018, 5 p. Mackin, J. H., 1960, Structural significance of Tertiary volcanic rocks in southwestern Utah: Am. Jour. Sci, v. 258, p. 81—131. FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM Miller, F. K, 1970, Geologic map of the Quartzite quadrangle, Yuma County, Arizona: US. Geol. Survey Map GQ—841, scale 1:62,500. Nolan, T. B., 1943, The Basin and Range province in Utah, Nevada and California: US. Geol. Survey Prof. Paper 197-D, p. D141—D196. Orkild, P. P., Byers, F. M., Jr., Hoover, D. L., and Sargent, K. A., 1968, Subsurface geology of Silent Canyon caldera, Nevada Test Site, Nevada, in Eckel, E. 8., ed., Nevada Test Site: Geol. Soc. America Mem. 110, p. 77—86. Proffett, J. M., Jr., 1971, Late Cenozoic structure in the Yerington district, Nevada, and the origin of the Great Basin: Geol. Soc. America, Abs. with Programs (Cordilleran sect), v. 3, no. 2, p. 181. Ransome, F. L., 1909, The geology and ore deposits of Goldfield, Nevada: US. Geol. Survey Prof. Paper 66, 258 p. 1923, Geology of the Oatman gold district, Arizona: A preliminary report: US. Geol. Survey Bull. 743, 58 p. Roberts, R. J., 1968, Tectonic framework of the Great Basin: Rolla, Univ. Missouri Res. Jour., no. 1, p. 101—119. Roller, J. C., 1964, Crustal structure in the vicinity of Las Vegas, Nevada, from seismic and gravity observations: US. Geol. Survey Prof. Paper 475—D, p. D108—D111. Sales, J. K., 1966, Structural analysis of the Basin Range province in terms of wrench faulting (PhD. dissert.): Reno, Univ. Nevada, 289 p. Schrader, F. C., 1917, Geology and ore deposits of Mohave County, Arizona: Am. Inst. Mining Engi- neers Trans, v. 56, p. 195—236. Shawe, D. R., 1965, Strike-slip control of Basin-Range structure indicated by historical faults in western Nevada: Geol. Soc. America Bull, v. 76, p. 1361— 1378. Stewart, J. H., 1971, Basin and Range structure: A system of horsts and grabens produced by deep- seated extension: Geol. Soc. America Bull, v. 82, p. 1019—1044. Stewart, J. H., Albers, J. P., and Poole, F. G., 1968, Summary of regional evidence for right-lateral displacement in the western Great Basin: Geol. Soc. America Bull, v. 79, p. 1407-1414. Thompson, G. A., 1966, The rift system of the western United States, in The world rift system: Canada Geol. Survey Paper 66—14, p. 280—290. Thorson, J. P., 1971, Igneous petrology of the Oatman district, Mohave County, Arizona (PhD. dissert.): Santa Barbara, Univ. California, 173 p. AN APPLICATION OF SATELLITE IMAGERY TO MINERAL EXPLORATION US. Air Force Aeronautical Chart and Information Center, 1968, Transcontinental Geophysical Sur- vey (35°—39° N) Bouguer gravity map from 112° W. longitude to the coast of California: US. Geol. Survey Misc. Geol. Inv. Map I—532—B. Volbroth, Alexis, 1973, Geology of the granite com- plex of the Eldorado, Newberry and northern Dead Mountains, Clark County, Nevada: Nevada Bur. Mines and Geology Bull. 80, 40 p. Wilson, E. D., Moore, R. T., and Cooper, J. R., 1969, Geologic map of Arizona: US. Geol. Survey, scale 12500000. Wilson, J. T., 1965, A new class of faults and their bearing on continental drift: Nature, v. 207, p. 343-347. Young, R. A., 1966, Cenozoic geology along the edge of the Colorado Plateau in northwestern Arizona: Dissert. Abs., sec. B, v. 27, no. 6, p. 1994. Young, R. A., and Brennan, W. J., 1974, Peach Springs Tuf‘f: Its bearing on structural evolution of the Colorado Plateau and development of Cenozoic drainage in Mohave County, Arizona: Geol. Soc. America Bull., v. 85, p. 83—90. ADDITIONAL REFERENCES USED IN COMPILING FIGURES 4 AND 5 Albers, J. P., and Stewart, J. H., 1972, Geology and mineral deposits of Esmeralda County, Nevada: Nevada Bur. Mines and Geology Bull. 78, 80 p. Anderson, R. E., 1969, Notes on the geology and paleohydrology of the Boulder City pluton, south- ern Nevada: US. Geol. Survey Prof. Paper 650—B, p. B35—B40. Bassett, A. M., and Kupfer, D. H., 1964, A geologic reconnaissance in the southeastern Mojave Desert, California: California Div. Mines and Geol. Spec. Rept. 83, 43 p. ‘ Bechtold, I. C., Liggett, M. A., and Childs, J. F., March 1973, Regional tectonic control of Tertiary min— eralization and Recent faulting in the southern Basin-Range Province: An application of ERTS—l data, in Freden, S. C., Mercanti, E. P., and Becker, M. A., eds., Symposium on significant results ob- tained from ERTS—l, New Carrollton, Maryland, v. 1, sect. A, paper G—21, NASA—SP—327, p. 425-432. Bishop, C. C., 1963, Geologic map of California, Needles sheet, Olaf P. Jenkins edition: California Div. Mines and Geology, scale 1:250,000. Bowen, 0. E., Jr., 1954, Geology and mineral deposits of Barstow quadrangle, San Bernardino County, California: California Div. Mines and Geology Bull. 165, 208 p. 269 Carlson, J. E., and Willden, Ronald, 1968, Trans- continental Geophysical Survey (35°—39° N) geo- logic map from 1120 W longitude to the coast of California: US. Geol. Survey Misc. Geol. Inv. Map I—532—C, scale 1:1,000,000. Childs, J. F., 1973a, The Salt Creek Fault, Death Valley, California (abs): Argus Exploration Com- pany, NASA Rept. Inv., NASA—CR—133141, E73—10774, 6 p. 1973b, A major normal fault in Esmeralda County, Nevada (abs): Argus Exploration Com- pany, NASA Rept. Inv., NASA—CR—135859, E74—10018, 6 p. Clary, M. R., 1967, Geology of the eastern part of the Clark Mountains Range, San Bernardino County, California: California Div. Mines and Geology Map Sheet 6. Cook, E. F., 1957, Geology of the Pine Valley Moun- tains, Utah: Utah,Geol. and Mineralog. Survey Bull. 58, 111 p. 1960, Geologic atlas of Utah, Washington County: Utah Geol. and Mineralog. Survey Bull. 70, 124 p. Dibblee, T. W., Jr., 1967, Areal geology of the western Mojave Desert, California: US. Geol. Survey Prof. Paper 522, 153 p. Gillespie, J. B., and Bentley, C. B., 1971, Geo- hydrology of Hualapai and Sacramento Valleys, Mohave County, Arizona: US. Geol. Survey Water-Supply Paper 1899—H, p. H1—H37. Gregory, H. E., 1950, Geology of eastern Iron County, Utah: Utah Geol. and Mineralog. Survey Bull. 37, 153 p. Hall, W. E, and Stephens, H. G., 1963, Economic geo- logy of the Panamint Butte quadrangle and Modoc district, Inyo County, California: Cali- fornia Div. Mines and Geology, Spec. Rept. 73, 39 p, Hamblin, W. K, 1970, Structure of the western Grand Canyon region, in Hamblin, W. K, and Best, M. G., eds. Guidebook to the geology of Utah: Utah Geol. Soc. no. 23, p. 3—19. Hewett, D. F., 1931, Geology and ore deposits of the Goodsprings quadrangle, Nevada: US. Geol. Survey Prof. Paper 162, 172 p. Heylmun, E. B., ed., 1963, Guidebook to the geology of southwestern Utah: Intermountain Assoc. Petroleum Geologists, 12th Annual Field Confer- ence, Salt Lake City, Utah, 232 p. Hintze, L. F., 1963, Geologic map of southwestern Utah: Utah Geol. and Mineralog. Survey, scale 1:250,000. 270 FIRST ANNUAL PECORA Jahns, R. H., ed., 1954, Geology of southern Cali- fornia: California Div. Mines and Geology Bull. 170. Jennings, C. W., 1958, Geologic map of California, Death Valley sheet, Olaf P. Jenkins edition: California Div. Mines and Geology, scale 1:250,000. 1961, Geologic map of California, Kingman sheet, Olaf P. Jenkins edition: California Div. Mines and GeolOgy, scale 1:250,000. 1972, Geologic map of California, south half (preliminary): California Div. Mines and Geology, scale 1:750,000. Jennings, C. W., Burnett, J. L., and Troxel, B. W., 1962, Geologic map of California, Trona sheet, Olaf P. Jenkins edition: California Div. Mines and Geology, scale 1:250,000. Kupfer, D. H., 1960, Thrust faulting and chaos struc- ture, Silurian Hills, San Bernardino County, Cali- fornia: Geol. Soc. America Bull, v. 71, p. 181— 214. Liggett, M. A., and Childs, J. F., 1973, Evidence of a major fault zone along the California—Nevada state line 35°30’—36°30’ north latitude: Argus Exploration Company, NASA Rept. Inv., NASA— CR—133140, E73—10773, 13 p. Malmberg, G. T., 1967, Hydrology of the valley-fill and carbonate-rock reservoirs, Pahrump Valley, Ne— vada-California: U.S. Geol. Survey Water—Supply Paper 1832, 47 p. Maxey, G. B., and Jameson, C. H., 1948, Geology and water resources of Las Vegas, Pahrump, and Indian Spring Valleys, Clark and Nye Counties, Nevada: Nevada Water Resources Bull. 5, 121 p. McAllister, J. F., 1952, Rocks and structure of the Quartz Spring area, northern Panamint Range, California: California Div. Mines and Geology Spec. Rept. 25, 38 p. 1955, Geology of mineral deposits in the Ube- hebe Peak quadrangle, Inyo County, California: California Div. Mines and Geology Spec. Rept. 42, 63 p. 1956, Geology of the Ubehebe Peak quad- rangle, California: US. Geol. Survey Map GQ—95, scale 1:62,500. MEMORIAL SYMPOSIUM 1970, Geology of the Furnace Creek borate area, Death Valley, Inyo County, California: Cali- fornia Div. Mines and Geology Map Sheet 14, scale 1124,000. McKee, E. H., 1968, Geology of the Magruder Moun- tain area, Nevada-California: U.S. Geol. Survey Bull. 1251—H, 40 p. Noble, D. C., 1968, Kane Springs Wash volcanic cen- ter, Lincoln County, Nevada, in Eckel, E. 8., ed., Nevada Test Site: Geol. Soc. America Mem. 110, p. 109—116. Norman, L. A., Jr., and Stewart, R. M., 1951, Mines and mineral resources of Inyo County: California Jour. Mines and Geology, v. 47, no. 1, p. 17—223. Rogers, T. H., 1967, Geologic map of California, San Bernardino sheet, Olaf P. Jenkins edition: Cali- fornia Div. Mines and Geology, scale 1:250,000. Ross, D. C., 1965, Geology of the Independence quad- rangle, Inyo County, California: US. Geol. Survey Bull. 1181—0, 64 p. Schrader, F. C., 1909, Mineral deposits of the Cerbat Range, Black Mountains, and Grand Wash Cliffs, Mohave Co., Arizona: US. Geol. Survey Bull. 397, 226 p. Smith, A. R., 1964, Geologic map of California, Bakersfield sheet, Olaf P. Jenkins edition: Cali- fornia. Div. Mines and Geology, scale 1:250,000. Smith, G. 1., Troxel, B. W., Gray, C. H., and von Huene, Roland, 1968, Geologic reconnaissance of the Slate Range, San Bernardino and Inyo Coun- ties, California: California Div. Mines and Geology Spec. Rept. 96, 33 p. Stokes, W. L., and Heylmun, E. B., 1963, Tectonic history of southwestern Utah, in Heylmun, E. 3, ed, Guidebook to the geology of southwestern Utah: Intermountain Assoc. Petroleum Geologists Guidebook 12, p. 19—25. Strand, R. G., 1967, Geologic map of California, Mariposa sheet, Olaf P. Jenkins edition: Cali- fornia Div. Mines and Geology, scale 1:250,000. Threet, R. L., 1963, Structure of the Colorado Plateau margin near Cedar City, Utah, in Heylmun, E. B., ed., Guidebook to the geology of southwestern Utah: Intermountain Assoc. Petroleum Geologists Guidebook 12, p. 104—117. Tschanz, C. M., and Pampeyan, E. H., 1970, Geology and mineral deposits of Lincoln County, Nevada: Nevada Bur. Mines and Geology Bull. 73, 187 p. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Mineral Exploration Applications of Digitally Processed Landsat Imagery By R. J. P. Lyon, Stanford Remote Sensing Laboratory, Stanford University, Stanford, California 94305 ABSTRACT Enhancement processing of Landsat CCT digital data can markedly increase its application in mineral exploration in two principal ways: (1) by producing enhanced images with higher contrast and resolution, and hence higher information content for the photo- geologist and (2) by pattern-recognition techniques applied by computer directly to the four-band digital spectra. Of these, by far the greatest use will be the photo- geological usage of enhanced images because of the simplicity of analysis in requiring only the mind of a creative photogeologist. The use of computers in geo— logical analysis of these images is complicated be- cause of the inability of the creative photogeologist to specify mathematically the decisionmaking process by which he arrives at an analysis. Algorithms can be created for some of the steps only, but by using an interactive computer display system, under the control of the geologist personally familiar with the problem and area being studied, the man-machine interaction can extract the benefits of both. This paper describes the use of such a system (STANSORT) to mineral exploration problems, in zero vegetation cover (Yerington; Goldfield, Nev.), mixed cover of pifion pine and juniper (Pine Nut Mtns., Nev), heavy birch forest (Karasjok, Norway), and full tropical cover (Tifalmin, New Guinea). Clearly the in- creasing vegetation cover makes the recognition pro- cess more complex. The inclusion of geobotanical (and biogeochemical) information in exploration is essential. Landsat data are an excellent base for this purpose, but ground information must be secured for their full application. INTRODUCTION Over two-thirds of the land surface of the world is covered with vegetation, of which 42 percent is for- est, 24 percent is grasslands, and 21 percent comprises a cover of desertic shrubs and grasses in semiarid terrains (Draeger and Lauer, 1967). In any mineral exploration program, therefore, to neglect this ob- scuring cover is naive. Many of the applications of digitally enhanced Landsat data unfortunately have been made in areas devoid of vegetation cover (Paki- stan, Peru, etc.) and consequently the effect of such cover has not been considered. This paper speaks to this point and shows the application of this emerging technology to areas of sparse desertic cover (Nevada), moderate cover of pines and juniper (Nevada), and full cover of birch forest (northern Norway), and rain forest-jungle (central New Guinea Highlands). We must consider vegetation in our Landsat analy- sis or we will be dealing essentially with only one part of the total exploration problem. The principal meas- urement problem is a familiar one—sampling—and revolves around making sure that the coverage of the field equipment adequately expresses what is seen by the satellite in one pixel (0.4 ha or 1 acre) resolution. Clearly one must evolve techniques which include soils and vegetation in the right proportions, and this is difficult at ground level. A secondary, but very real, problem is that no one technique of data enhancement (stretching, ratioing, etc.) works all the time in every area, mainly because of the variable effect of differing vegetation cover. One must use a group of techniques to obtain the maximum benefits from the computer compatible tapes (CCT’s) from Landsat. 271 272 FIRST ANNUAL PECORA BACKGROUND The principal attraction of the Landsat satellite sys- tem to a modern mineral exploration program is, that for the first time, a calibrated and quantitative spec- trally filtered photographic system is available with coverage of almost any part of the land surface of the Earth. Unlike traditional aerial photography, one does not have to question the film type, filters, the develop- ment practices which were used, or by what means were numerical data extracted from the film—base materials. We can directly compare spectra. The spec- tral response is fixed and constant, the gray-scale characteristics identifiable, no matter whether we are looking at Norway or New Guinea, Australia, or Argentina. In addition, as an important corollary, any software algorithms developed to process the digital data from the satellite apply equally to any other geo- graphic locality. Spatially with Landsat we have to modify our think- ing somewhat, as each resolution cell (the highest resolution unit possible) covers 0.4 ha (1 acre) and is seen at scales smaller than those with which we have had most of our prior experience. This can be looked at in at least two ways. Normally we use a higher resolution element-say 10 cm‘~’ and a scale of 1240,000—because our decisionmaking process in photointerpretation requires this degree of spatial resolution for shape identification. Even when using Trees .l _] i Radiance M O 4 -III WM ITIII IIIIIIIII MEMORIAL SYMPOSIUM color film we base most of our photogeological analy- sis on shape and neglect the discriminant power of the spectral color. This conditioned reflex has come about because our color reproduction in printing leaves much to be desired, and we have not been able to state (or remember) precisely what spectral colors we were viewing. Landsat with its quantitative re- producibility has changed much of that. Furthermore, although the pixel is 0.4 ha in area, there are still 250/km'-‘ (640/mi‘-’). If decisions can be made on the spectral information content of each pixel, then we can consider this resolution as giving 250 decisions/ km2 (640 decisions/mi”)—a very different way of looking at data. Each Landsat scene contains about 7 million pixels covering 34,000 kmg, and any set of pixels in that or adjoining tapes can be rapidly eval- uated by the same algorithm, thus markedly broaden- ing the usefulness of this search approach in explora— tion. In our work at Stanford, we have continually worked at high resolution with the Landsat data, often using only 100 pixels or so to perform analyses re- lating ground measurements to those of the satellite. Our basic study areas are 20 km‘-’ (7.6 mi”) presented on a video-display and on paper printout at about 120,000 scale. For larger coverages, at the district level we prepare film images of 300—360 km2 (125 miz) at a scale of about 1220000, to which we have Freq Water I Ill) I) I will) lWater K. \\ a V v v 510% 620 can 550 sec Column TRAVERSE A- A' (Row I234) Water L Serpentine _l I 280J Serpentine I Tract l ’I T T | I s. WM) ”m 82 Mill) D v v v v 670 530 390 700 Colu mn voir (water); TRAVERSE B—B' FIGURE 1.—ChanneI-by-channe| spectral plots for picture elements (pixels) along a raster line. T:trees; sterpentine; F:freeway (l—280); (Row I248) K:Crystal Springs reser- thousing tracts. From Levine (1976). MINERAL EXPLORATION APPLICATIONS OF LANDSAT IMAGERY made the required geometric correction. We thus continually remind ourselves that the field problems to which we are applying these techniques require detailed answers for 20—300 km? areas at scales be- tween 1:24,000 and 12100000 to be of use in the typical exploration program. RATIOING CONCEPTS AND ADVANTAGES Of all the concepts presently used in enhancing Landsat digital data that of ratioing any pair of the four channels has proved the most significant. This has several immediate advantages—it tends to normalize the spectral data, removing “brightness” contrasts be tween any two pixels and enabling a better compari- son of their hue or color. Such ratioing has been 273 channel and a negative of the other, but many as- sumptions as to linearity of response of the two film records have to be made. By far the more positive and simpler task is to use the digital data and perform the single ratio step in the computer. At the same time one often stretches the data in each channel to make maximum use of the full dynamic range (0—255) before ratioing. A small exercise is useful at this point to elucidate further the physical meaning of Landsat spectra and the ratioing process and, at the same time, to show the high resolution capabilities of spectral analysis of this system. To illustrate these points figures 1 and 2 have been taken from a more detailed analysis (Levine, 1976) of the concept initially developed by Honey and others (1974). performed optically before (Whitaker, 1965) by The upper traverse (A) in figure 1 represents a plot simultaneously printing a positive transparency of one of 50 pixels along a single raster line (pixels 610—660; 7'} 5H 0 .. 2 5' a m 40- .- ’,_‘ u —-\ -‘\ m 30‘ ~\ \ \ ‘ r: w \ \ \‘ 2»- €/\ ax \ ‘ ~ \ ‘ I ‘ III‘ ‘ S u m m e r O U C (V a O 0 7. O I v —--1 2 ~~ ‘ 1‘ ~ ________ = I, / / m J / / ,l E z’ I l’ u / I I E II iOct 72 4Jan ”May 28 May ZlMay 3Jul 3 Jul 21 Jul 21 Jul 26 Aug 26 Aug 73 FIGURE 2.——Time series of averaged spectra and radiance (upper) and normalized reflectance (lower). Spectra for averages of 10 pixels. Dashed spectra are for burned—over areas (soil areas). From Levine (1976). 274 FIRST ANNUAL PECORA row 1234) for a San Francisco Bay scene (1075— 18173). The lower traverse (B) represents a similar traverse of 40 pixels (pixels 660—700; row 1248) slightly farther to the southeast. Each small zigzag pattern is a radiance plot of all four Landsat channel values for a single pixel, with the digital numbers joined to give a low resolution spectrum. The spectra can be easily classed into sim- ilar shapes, by passing the eye rapidly from left to right along the raster line, using the well-developed shape-recognition features of the human eye-brain system (Step 1). As Step 2, we can correlate these pattern classes precisely with mapped land-water in- terfaces by locating them on large-scale topographic maps (124,000 scale) or, preferably, if available, on orthophoto maps at the same scale. The pattern (T) could be identified with tree-covered slopes, (K) with the western arm, and (K2) with the eastern arm of Crystal Springs reservoir. Similarly (SI) and (82) were identifiable with patches of serpentine outcrops cut by Interstate 280 (F), and (H) as a housing tract which lay to the east. To adapt this eye-recognition method to machine operation an automatic, unsupervised clustering algorithm was written by Honey which simulates Step 1 in the computer and forms the basis of our STANSORT program (Honey and others, 1974). Steps 1 and 2 together when performed manually help the worker gain confidence in the credibility of Landsat spectra and in possible pattern recognition processes. This part of the exercise has important teaching appli- cations. Atmospheric corrections, which remove the effects of back scattering of sunlight by the atmospheric column, were performed (Step 3) by subtracting the observed apparent brightness in each channel from over a large area (45 pixels) of zero-reflectivity car- bon-black refinery waste lying on the south shore of Carquinez Bay (northeast of San Francisco but still on the same Landsat scene and tape; see (Lyon and others, 1975). Reduction of the atmospherically corrected values of reflectance is performed in Step 4 by observing the apparent brightness of a known reflectance target—the concrete parking aprons of Moffett Field, on the southwest edge of San Francisco Bay, in the same Landsat tape. As the bidirectional reflectance of the concrete has been measured by us (28 percent, 31 percent, 30 percent, 32 percent in each channel, 4—7) a linear interpolation between the zero— reflectance target (< 0.5 percent in all channels) and the concrete “bright” target enables us to convert any other pixel spectrum to reflectance. MEMORIAL SYMPOSIUM Figure 2 (upper) shows radiance spectra (averaged over 10 pixels) from another nearby grassy/ soil area, in a time-series (plotted as the abscissa) from fall (October 6, 1972) through summer (August 26, 1973). Only the solid curves, like (1) and (2), need concern us; the dashed curves (3) are for burned-over areas of significance later in the year. By performing the ra- tioing (Step 5) much more familiar spectral shapes appear. Here in figure 2 (lower), Channels 5, 6, and 7 have been divided by Channel 4, although equally well any other channel, the total of all channels, etc., could be used, and particularly so if Channel 4 is noisy. Ratioing removes the varying absolute levels of brightness (see within pattern T in fig. 1) which are from the sunlit and shaded sides of hills covered with a constant terrain material—in this case, trees. Steps 3 and 4 may be omitted if one does not require at- mospheric corrections and reflectance data and com- parable shadow—free “flat earth” spectral images may be made. Ratioing thus enhances the spectral-color information (Whitaker, 1966) in Landsat data by sup- pressing the relative brightness due solely to topog- raphy. Obviously in structural analysis this would not be an asset where shadows materially aid in the analysis, but the later discussion of applications will show the merit of ratioing for lithological differentia— tion. We retain both structural and lithological infor- mation by making three-color imagery, with two of the three color-guns assigned to ratio images (say Ch 7/4*=green and Ch 5/4=blue) and with the red gun assigned to our “edge-enhancement” procedure, which produces nondirectional symbols at spectral- contrast edges in the four-channel data matrix (Honey, Prelat and Lyon, 1974). The human eye again is used to connect the edge dots into meaningful lineaments. These may or may not parallel lithologic contacts as seen in the other two ratio images in the three-color print. APPLICATION CASE HISTORIES Where the surface materials over a mineralized area are residual and contain alteration minerals— clays, iron oxides, etc—typical of that style of min- eralization, the techniques for use of Landsat digital data in a search mode are relatively straightforward. These have been well documented by Goetz and others (1975), Vincent (1973), and by ourselves (Lizaur, 1975; Ballew, 1975a, 1975b; Lyon and others, 1975; Lyon, 1976). However where the mineralized targets of search are covered—either by glacial debris, wind- *Ratios may be shown as Ch 7/4 or R74, etc. MINERAL EXPLORATION APPLICATIONS OF LANDSAT IMAGERY blown sand, alluvial outwash, etc., or more often, by even minimal amounts of vegetation—their discovery is much more complicated. This selection of our case histories is oriented to these points and is arranged in order from essentially zero vegetative and surficial cover (Yerington, Ne- vada, pit and clumps), through light desertic cover 275 (Singatse Range, and Goldfield, Nevada), to moder- ate pifion pine and juniper forest (Pine Nut Moun- tains, Nevada), and, finally, heavy forest of birch in an Arctic environment with a full cover of glacial till (Karasjok, Norway), and jungle at Tifalmin, New Guinea. The localities are tabulated in table 1. TABLE_1.—Localities and results of case histories Very low to zero cover: Yerington copper pit, dumps, and surrounding Singatse Range, Nevada. Oxidized and nonoxidized ore (sulfide) may be differentiated in the pit and the sulfide rock ”dis- covered” in the tailings pond using Landsat data in a search mode (STANSORT/ SEARCH). Coldfield, Nevada. Alteration zones and gossans surrounding gold-alunite mineraliza- tion. Outline of alteration zone can be mapped by Landsat image enhancements. Moderate cover: Pine Nut Mountains, Nevada. A molybdenum-bearing skarn with a biogeochemical anomaly in the pifion pine and juniper, independently located by color-ratio images prepared from Landsat data. Heavy cover: Karasiok, Norway. A known copper-bearing biogeochemical anomaly was relocated by the Landsat data and now has been extensively studied in the field. Tifalmin, New Guinea. A known set of copper—bearing intrusives under full jungle cover. So far very little success in determining any vegetative anomaly. VERY LOW TO ZERO COVER YERINGTON COPPER PIT AND DUMPS, MASON VALLEY, NEVADA General description—The simplest example which can be chosen for analysis, and which is still of sig- nificance to mineral exploration, is that of an open-pit mine and its surrounding clumps. Here, single rock type materials cover enough area as to be statistically significant (>8—10 pixels) in the pit itself (generally a single type of intrusive rock with one or two super- posed alteration or oxidation mineralogies). The clumps have been carefully segregated during mining and represent waste rock, low-grade ore, and tailings from the metal extraction. Yerington pit in Lyon County, Nevada, about 80 km south-southeast of Reno, satisfies all of these criteria. This pit, while small (2X 1 km), may be considered to have an oxidized upper portion and a nonoxidized sul- fide-bearing lower portion (fig. 3). The southern dumps are made up of alluvial overburden and waste rock (largely granodiorite). The northern clumps (from northwest to northeast) are respectively oxide tailings from leach ore, sulfide tailings pond (T) and low-grade mixed oxide-sulfide leachable ore. All the northern clumps represent varying grades of oxidation of the quartz monzonite and granodiorite. From the pit, to the south, the dumps are alluvial outwash and cover. The housing settlement of Weed Heights (W) and the irrigated river-bottom lands of the Mason Valley (M) and Walker River show patterns typical of vegetation (gardens, lawns, fields, etc.) The Singatse Range (8—8) has a typical Nevadan Basin and Range topography bounded on the east by several range-front faults. The geology of- the range is complex and is composed of Jurassic granodiorites and sediments to the southwest of the pit, with pre- dominantly Tertiary volcanics (ignimbrites, etc.) to the west and northwest of Weed Heights. The vegetated cover of the pits and dumps is zero. The cover on the Singatse Range is sparse (mainly sagebrush and other desertic shrubs) typically 3 to 5 bushes per 10 m2. From above, less than 10 percent of the ground would be covered. Analysis—Single band images prepared on a line- printer, with each pixel value being encoded as a sym- bol (lighter toned for higher value=brighter areas, darker for lower value=darker areas) with 14 steps or gray tones, are shown in figure 4. These resemble over-enlarged photographs but are very useful as “road maps” for locating specific pixels or areas for detailed analysis. Figure 4 shows a typical output stage of data each for 20 km2 of Channel 5 (left) and Channel 7 (right). 276 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 2 KILOMETERS FIGURE 3.—Yerington pit area, Nevada, aerial photograph from 20 km. T:tailings pond; Fezacidified ferric sulfate leach liquor; D:dumps, D1:oxide leach waste, D2:leach dumps, low-grade oxide and sul- fide; D3 & D4:mixed alluvial and waste rock, mainly granodiorite; M:Mason Valley; W:Weed Heights housing area for the mine; f—fzrange-front fault; S—S:Singatse Range. In pit: ‘l:sulfide, 2: oxide, 3:mixed, 4:!eached ore. Original copied from a color photo (RBS7—MX248) taken August 11, 1973. MINERAL EXPLORATION APPLICATIONS OF LANDSAT IMAGERY 277 m.mmusssux:.musmsssunuuuuum ....... ms" nanusun-nussuumsssmsssm .umsmu mum FIGURE 4.—Raw lineprinter output for channels 5 (left) and 7 (right), Yerington pit area, Nevada. t:taiIings pond; dzdumps, oxide waste; szerington pit; w:Weed Heights housing area; m:Mason Valley; a=agricu|ture along the Walker River; S-S:Singatse Range. Reduced about 5 times; approximate scale, 1124,000. Each figure represents a (5: 1) photoreduction of raw lineprinter output. Vegetation in the Mason Valley (m) is dark in Chan- nel 5 (left) and lighter in Channel 7 (right), if growing Well (as at a) where some field shapes can be seen. The eastern flanks of the Singatse Range occupy the left edge of each image. Weed Heights housing area (w) for the mine is also dark in Channel 5 and light in Channel 7. The pit area (p), clumps (d) and tailings pond (t) ap- FIGURE 5.——Effect of varying the tolerance level in STANSORT/ CLUSTER. Data for the immediate Yerington pit area, showing the decreasing number of classes (letters) of materials as the tolerance is increased, from 6 percent to 9 percent, thereby simplifying the pit classifications. Landsat CCT 1397- 18051, smoothed. pear in varying shades of gray in each channel. Some water (black in Channel 5 and Channel 7) lies at the north (uppermost) end of the tailings pond inside the lighter toned dikes. This represents the simplest form of Landsat en- hancement—stretching and density encoding—and is presented at an original scale of about 1 :24,000. In figure 5, the next step of classification is shown. The clustering algorithm used in STANSORT/CLUS— TER (Honey, Prelat, and Lyon, 1974) operates with a gating technique to make a “like/unlike" decision on the four—band spectra of each pixel, commencing at the upper left corner of the matrix and proceeding to the lower right. The gating value is an input variable and may be modified at will by the user, in an inter- active mode. Wider tolerances (say 15 percent) allow most spectra to pass into a few classes—tighter toler- ance produces many classes. The system uses the 26 alphabetical symbols and then blanks for all above 26. The clustering step is very fast and takes 10—30 5. If the pattern produced on the video screen has too few symbols one closes up the tolerance—if there are too many blanks one opens the tolerance. Up to 20 km2 may be classified at once. FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM The key to the method is that the operator decides which pattern best fits his knowledge of the training area he has selected. In the case shown in figure 5, of ‘the area immediately surrounding the Yerington pit, we felt that T (tolerance) = 8 or 9 percent gave the best fit, permitting segregation of class ‘B’ for oxidized quartz monzonite and ‘P’ for unoxidized (“sulfide- bearing”) quartz monzonite. Class ‘A’ correlates with the surface dumps, while class ‘C’ represented the alluvial outwash from the Singatse Range to the west. With this best—fit decision made, the spectra are “frozen," and can be used to search the rest of the 34,000 km‘-‘ on the tape. Figure 6 shows the subsequent nununuk ununntgt m “(Kiwi-MU“ ' inn :étw .( w ‘5 \KI‘JJQL 5.1 J‘JQL ung gag N“RQC A.“ Out-u “U“ H M m .4” N “,1 M. “t ,t my my Aummccfl U CCCC I. tLfiZ‘ {KnCCC C _ ‘,X vggyrcns ant FIGURE 6.——Clustering analysis using STANSORT/SEARCH, toler- ance:8 percent after training on the immediate pit area. This map starts at row 1140/col. 707, Landsat CCT 1397—18051, training area was as shown in fig. 5. T=tailings pond; P: pit; W:Weed Heights housing area; MzMason Valley irri- gated land; Azagriculture in the valley along the Walker River. Pit classification symbol “P” appears only in the tailings pond area, with symbol "B” only in the pond dikes and on some of the dumps. Most of the southern dumps have a symbol “A" which matches some of the leach residues. MINERAL EXPLORATION APPLICATIONS OF LANDSAT IMAGERY patterns developed in STANSORT/SEARCH using the frozen spectra in a supervised mode of pattern recognition but retaining the same tolerance (T=8). The following points are significant: 1. The vegetated areas of Weed Heights housing area (W) and the Mason Valley (M) and (A) are simi- larly identified. 2. The nonoxidized, sulfide-bearing monzonite (P, also labelled with a #5 arrow only occurs else— where in the tailings pond (T)—exactly where one would expect to find it after the copper- fiotation step. Only a few other isolated pixels in the entire Singatse Range are so classified and are considered as noise. 3. The oxidized monzonite (B symbol), on the other hand, appears mainly on the dikes around the tailings and in patches between the pond and the pit. Elsewhere in the Singatse Range this symbol occurs principally at the clumps of the Ludwig mine off to the southwest. 4. The “low-oxide copper” and other dump materials (symbol A) occur immediately to the south of the pit, to the north near the leach liquor pond (blank), and on the east flank of the tailings pond. Thus patterns selected in the pit (training area) can be used to classify and segregate similar materials nearby on the mine property or in the rest of the Singatse Range itself. Potentially the system can search the whole Landsat scene, covering 3.3 million ha (8.2 million acres). . A slightly different technique is shown in figure 7 (p. XXIII). Black and white enhanced images of Chan- nels 5, 6, and 7 of Landsat data over the Yerington- Singatse Range area were combined into a color print using the Dicomed image-forming process. Digital data preprocessed for enhancement were fed sequentially into the Dicomed unit using a blue filter for Channel 5, a green filter for Channel 6, and a red filter for Chan— nel 7. The resulting false color infrared image is shown as figure 78 (p. XXIII), which can be directly compared with an aerial color photograph (fig. 7A, p. XXIII) taken from 20 km (65,000 ft) 13 days earlier. A close match geometrically is evident, and the increased con- trast more clearly reveals features like roads and desert washes. To make figure 7D (p. XXIII), which is a color ratio print, the ratio channel data were prepared for the Dicomed unit. This time ratio R54 was coded with a blue filter, R64 with a red filter, and R74 with a green filter. This color ratio print (fig. 7D, p. XXIII) differs mark- edly in information content from the color-infrared 279 print (fig. 78, p. XXIII), although pixel-by-pixel the data locations are the same. The ratioing “smooths out" the topography, remov- ing the shadowed-side effects, and emphasizing the spectral content of the data. Now irregular patches of color replace the topographic-dominance in figure 78 (p. XXIII) and in figure 7A (p. XXIII), too. Careful com- parison with the geological map (fig. 7C, p. XXIII) (Moore, 1969) reveals that the ratioing has enabled a close correlation of similarly colored areas with the regional geology to be prepared using Landsat digital data alone. We are planning a much more detailed ground measurement program in this area to relate better the geology with the ratio images, but even the initial re- sults shown here are most significant of the role of Landsat digital processing in regional geology. COLDFIELD, SOUTHWEST-CENTRAL NEVADA The old mining camp of Goldfield had its heyday around 1906 but is now becoming something of a celebrity again as a test site for Landsat analysis tech- niques (Rowan and others, 1974; Goetz and others, 1975; Lizaur, 1975; Ballew, 1975b). These several ef- forts are based upon the painstaking studies of Roger Ashley who mapped the 45 km’-‘ (17 mi!) area in the 1968—1971 period under the US. Geological Survey’s Heavy Metal Program. Ashley’s geological maps (Ash- ley, 1971, 1974) and especially the map of the argillic alteration, which is so pervasive at Goldfield, form the ground calibration data for most of the studies. Figure 8A shows the typical low vegetative cover of the desert area. Figure 9 compares three map products to emphasize their similarities. Figure 9A is taken from Ashley’s cal- dera study (Ashley, 1974) and shows the preminerali— zation structural pattern with the dominant ring frac- ture systems, central fractures (northeast in the north- ern half and swinging to southeast in the southern half of the caldera), and the tangential “taillike” fracturing extending out to the east-southeast. The premineraliza- tion intrusive centers (hatchured) are strongly clus— tered on the inferred ring—fault system. Figure 9B is an outline map (from Ashley, 1971) of the same area at about the same scale but, with the outcrop zone of the argillic alteration in the volcanics, shaded a darker tone. This bull’s-eye and tail pattern is the key to the search for ore mineralization at Gold- field, showing a strong structural control for both the alteration and mineralization stages. Thefsame spatial pattern (of the argillic alteration zones) can be defined directly from the Landsat digital 280 FIGURE 8.—Comparison of vegetation covers. 8A FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM fl Oldfield, Nevada, with sage- '4 ' .~., ; C brush. BB; Pine Nut Mountains, Nevada, with pinon pine and juniper, and some sagebrush in the foreground. This area is closer to the Sierras and has a higher rainfall, although still a semidesert. data, by ratioing Channel 5/Channel 4, as prepared from CCT 1342—18003 (June 30, 1973, fig. 9C). This ratio (R54) may be used to emphasize iron—oxide—rich areas which show high ratios with these bands. White areas in the caldera ring and east—southeast tail have ratios above 1.20 whereas darker areas, and the Mal— pais basalt to the southwest, have values around 0.90— O 95 (tables 2, 3, 4, and 5). Similar comparisons for other ratios (although not so striking) can be found in the ground-measured Gold- field data, shown in table 4A—Altered samples, and 4B—Unaltered samples. The ratio image (R54, fig. 9C) used to emphasize iron-rich areas, shows the ring fracture (and east- southeast tail) features clearly (inside the solid black arrows), although on this image the central bull’s-eye pattern is very subtle. The white patch to the north- east shows the argillic alteration and gossans of the North Diamondfield district, outside the main struc- ture, again indicating how a pattern developed for one 281 IMAGERY MINERAL EXPLORATION APPLICATIONS OF LANDSAT ._0u\©$ 39. am :2 Eng: E chEwa _mme_ v5. or: ome 9 33.: m9: «Km; .55: 9t 0. «ohm cocflwzm cm 3323:. u E Beta :mQO .925 833m: .2033 ~ cm 9:: .mEN \mocwwlmvg ._.UU $22.3 .EEBE ‘0 E 52:3 .EwcvcoEflo *0 $5 or: U555 cozflmzm *0 mc: 05 EmEtE u :0 mict< .Cnow SwEm< Eta n 3R3 ioEm< Eta m 93>: mbm>wZ fiECECO L3 Q mwmct 9:8 32 Emtcfi .2: CE :8 mQCON .5588? as 833:? ECOEE of :3333 coszzI. _ 3,39 2.6.5 9.8» 859.5 233.: .93 S 1823 3 a 69:25 221,» .355“ 3 V0» £5.25 9.... vat-E 3 o 623.... 22: ‘28.. _. 2:33: v.3 12:3 m $.30: 282 TABLE 2.—Typical ratio R54 values for Coldfield (after Lizaur, 1975) LANDSAT—derived data (1342-18003, June 30,1973) Ratio Iron-oxide-rich areas ______________________ Above 1.20 Malpais basalt ____________________________ 0.90—0.95 FIELD-measurement data Iron oxide gossan soils _____________________ 1.60:0.20 Montmorillonite soils (average) ______________ 1.22:0.13 Montmorillonite soils (selected) _____________ 1.30:0.04 Kaolinite soils (average) ____________________ 1.22:0.16 Kaolinite soils (selected) ___________________ 1.17:0.08 Most altered rocks (average) ________________ 1.30:0.17 Unaltered rocks (average) __________________ 1.10:0.11 TABLE 3.——Ratio comparisons for Coldfield, Nevada (after Lizaur, 1975) R54 Landsat R54 Ground Una/tered rock Malpais basalt ___________ Below 0.92 __ Not measured Milltown andesite ________ ____do ______ 1.00-1.20 Background Gray volcanics ___________ 1.00—1.11 ____ 1.10—1.19 Altered (targets) Traverse areas ___________ Above 1.19 __ 1.18—1.42 Gossan soils __________________________ 1.48—1.74 locality can be used to locate other similar areas elsewhere. A large map of R54 values was made by Lizaur (1975), and several background areas were selected along access roads to establish the character of these areas as seen from Landsat. Tables 2, 3, and 4 have been calculated from his field measurements over areas shown to be anomalous by R54 ratio maps pre- pared from Landsat data (CCT 1342—18003, June 30, 1973) and have been segregated by him to be field checked. In three traverses in sec. 4 he crossed a typical quartz-alunite (silicified) ridge, locally called a “ledge.” These are usually gold ore bearing and sur- rounded by a kaolinite-rich selvage (15—20 m wide), grading outwards into a montmorillonite zone more than 100 m wide. Table 5 lists published data for sev- eral alteration minerals. Of the altered rocks seen in these traverses and at localities in secs. 21, 22, the gossan soils clearly were anomalously high in R74, R65, and R64 as well as R54, although these rocks differed from the unaltered or background samples in almost every ratio. MODERATE COVER Geobotanical anomalies which have been carefully studied and described in the literature may be used as FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM test targets to establish the credibility of Landsat de- tectability (for example, Billings, 1950). We have worked with two such areas (“copper-barrens”) in Norway as well as with a mapped bio-geochemical anomaly (high molybdenum values in the leaves of vegetation) in western Nevada. Our aim was to define the correlation between (1) the anomalously high metal content in the soils, which affects the types and species-proportions of the vegetation populations near such metal deposits, (2) their reflectance charac- teristics as measured with Landsat bandpass equip- ment on the ground, and (3) the Landsat satellite data taken over the same areas. The two areas differ, as the Pine Nut Mountains area in western Nevada has a moderate cover (about 16 trees/ ha) whereas the Karasjok locality in northern Norway has a total cover of birch forest or grass. A third type of area, at Tifalmin, central western New Guinea, also with a total cover (jungle trees and grasses) did not show any evidence in the vegetation, as seen by Landsat, which would enable the detection of the copper mineralization known in the rocks and soils beneath the cover (Press and Norman, 1972). PINE NUT MOUNTAINS, CENTRAL-WEST NEVADA Being closer than Goldfield to the Sierra Nevada mountains of California which lie 30 km to the west, the Pine Nut Mountains enjoy a higher moisture con- tent (mainly snowfall, about 20—25 cm/ yr). The hills lie between 2,100—2,300 m in elevation and are covered with evenly mixed pifion pine (Pinus monophylla) and juniper (Juniperus utahensis), at about 16/ ha (40/ acre). The area between these trees is predominantly sage- brush, 1—1.5 m in height, with some rabbitbrush (Chrysothamnus sp.) (fig. SB). The bio-geochemical anomaly which patchily covers about 1—2 km2 on the eastern flank of a low ridge of intermediate volcanics and limestones has been de- scribed in detail (Lyon, 1976). The anomaly takes the form of anomalously high (from 75—500 ppm) molyb- denum levels in the ash of juniper and pine needles extending down slope from a molybdenum-bearing skarn mineralized zone in the rocks surrounding a small quartz monzonite intrusive (see stippled area, fig. 10) (De Long, 1971; Noble, 1962). A 3-5 times larger anomalous area, covering about 4.5 km2 and located centrally over the bio-geochemical anomaly, was independently found directly from the Landsat digital data, without prior knowledge of the specific location of the molybdenum-bearing vegeta- tion pattern (Lyon and Honey, 1974). The STANSORT interactive computer program was used to identify the 283 MINERAL EXPLORATION APPLICATIONS OF LANDSAT IMAGERY .oo.:HNz+>z+xz Nfi+m+2u5$£un>z 5+m+28$£unxz NXEHSOB 5:53.? we E20580" .83 8E: .32 a: 88: K5120: a5 :5 :5 a5 33 33 :3 s5 :5 3.5 :5 35 mm ........ 339 8.82 mi 3: m2 5: m: a we» :2 ,5, a: Em SN £85 986 Am: 8: So; 3: KO; Re; AND; AN: a: :Nq SN; 3: 8: SN 3: EN 3: Q Eu :2 ha em.“ 2: 2: w -- 8:57: E :8 E38 .5“ $5 $5 :5 :5 35 :3 $3 85 :5 85 $5 85 m2 :1 m: o: 3: mm 3». 3% NS 3m, E Em o .......... $8 852% .m :3 $3 :3 E: E: :3 :3 a5 :5 :5 :5 :5 t: m: 3: mi E 3 NS .E E: me» Q», Em m .......... 8; 852% :5 :5 :5 :5 :5 :5 :3 $3 :5 :5 65 $5 NNF 5; gr m: P: 5 «cm mvm @mm wmm Axum mvw ow lllllllll wcoN 9.2505. .N $3 8: $3 63 53 £3 E: 55 :5 63 83 30.5 om: 9: E: S; we: E a; Em 5 N3 2% mom m .......... 8; 852% :5 :5 :5 :5 :5 s5 :3 E: :5 :5 :5 :5” NNP Nmr @NF 5“: F: ww cow mmm «mm owm .VNN am: No I: QCON 0::2 :oEEOE .— Amxun: USSR _ a. __ ~_ www.mZm; :meZV 31E? v.33? E E S E K OE >22 xzz n e m _ v 5 z 956 EESMEQ SROQSE MFG: 6an SEN: EC: US$328? Sam» mbm>m2 \EchcD \mv‘uog DEER 3:3 ‘55 3:5 bc~€2£ bum mcmmEI. >80 .v a: 8: AN: 63 63 $08 Ame; 2.0.8 Am: £3 E: K: m: w»: c: omw MOP ow vow Nmm @ON NmN mar Kr N xxxxxxxxx mu_cmu_o> >90 .m 8: 83 3.: 8: Q: as Go; Ame; $8 $3 :5 Am: 0: v: NNr NNF r: 08 gm mmm cg wmm mm: NE‘ m 111111 3:93. 5.85:5 .N 8: $0; 888 8“: K: 80.8 K08 an: 8m; 8m; cum; KN; o: mmF mmr cm? c: an Em 8mm Om: VE‘ 3; NmP m IIIIII wzmmtcm :30: 2 .r 950waqu men 3 mozfi «mm :9: >3 «6qu :m \wtmwncm E5852 wEUEuxw meHZV moEEmm UszmcD vm vo mo vm mm mum >2”. XZM n w m w LU Z @390 ~6ng fiEchU 539 ©9832: 3:3 5* we?! uthLEE bzm mcmmEIdv Ema; MINERAL EXPLORATION APPLICATIONS OF LANDSAT IMAGERY TABLE 5.—Published ratio values for some alteration minerals Reflectance Alteration mineral % x100 Ch4 5 6 7 R76 75 74 65 64 54 Montmorillonite ______________ 33 50 60 56 93 112 170 120 182 154 Kaolinite _____________________ 67 72 73 72 99 100 107 101 109 107 Alunite ______________________ 43 61 7 71 100 116 165 116 165 142 Hematitic limonite" ___________ 18 38 44 46 103 121 256 116 244 214 Coethitic limonite ____________ 5 9 12 12 100 133 240 133 240 180 larosite ______________________ 7 10 10 7 74 74 108 100 146 146 Digitized from Rowan and othersl1974, fig. 20) from data originally in several papers. *:samples in place. 23 .MAX X755C770/Y ms...“ . ' 7‘5 zit/mime LY _ FIGURE 10.—Locality map of Landsat-detected anomaly and the Mo-anomaly in the vegetation. Sections are in T. 12 N., R. 21 E., Mt. Siegel quad, Nevada. NW-trending linears are limestone outcrops; NE- trending linears are strongly developed joints (?faults). Quartz monzonite outcrop is in solid black near the Alpine mine mill. Stippled anomaly area is 75 ppm M0 in the juniper needle ash. Landsat anomaly is outlined in heavy dashes. Divide and Cherokee mines are also shown. 285 286 features and ratio images (especially R74) mode which showed the shape and location of the anomaly. We have performed a considerable amount of ground measurements in the field in the area, both with cut branches and with trees in place, to attempt a correlation with tree reflectance and Landsat- derived ratios. The ground data show higher ratios for vegetation inside the anomaly relative to similar trees outside the area (0.5—1 km away). These higher values are offset, however, by a lowered tree spacing per hectare inside the anomalous area (12—14/ha inside versus 16—20/ ha outside). The specific percentages of trees versus sagebrush and soils has not been deter- mined, but clearly increased sagebrush or soil cover- ages, with R74 ratios of 150—180 (instead of 400—500 as with the trees), would lead to a darker area on a ratio image, comparable to that which we found in the Landsat data (Lyon, 1976). FULL COVER KARASJOK, NORTHERN NORWAY A locality in northern Norway in heavily birch cov- ered glacial till areas, contains several copper sulfide bearing strata in the Karasjok group of hornblende schists (fig. 118). These “copper layers” in turn pro- duce copper-rich, ground-water springs which poison the birch trees where the springs issue beneath the till cover (Lag and Bolviken, 1974). This well-docu- mented, poisoned area covering approximately 100 m in width by 1 km in length was successfullyllocated on Landsat image 1365-09430, shown in figures 11A (Channel 5, red) and 11C (Channel 6, infrared). The locality is circled on the 1:250,000-sca1e geological map (fig. 118) cut from the larger Karasjok sheet pre- pared by the Norwegian Geological Survey. Channel 5 data (fig. 11A) show a pronounced light- gray banding to the north and east of the letter “A,” which closely approximates the S-curved stratigraphic strike of the granodiorite/metasediment boundary (gd/m). Bright white patches on the inner side of the river curves are low—lying river gravels, with a sparse tree cover, but having a heavy coverage of reindeer moss, a white lichen form. Channel 6 data (fig. 11C), however, show principally the patterns of vegetation and much less information on the stratigraphy. Strong ENE and SSW fracturing is brought out by thin bright lines along the drainage. Several portions of the river towards the top of the image also show the east-northeast trend. The anomaly of Lag and Bolviken lies inside the open arrows, in figures 11D and E, and inside the open circle in figure 118. Insets 11D and E are enlarge- FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM ments. The ground location for this was defined for us on aerial photographs at about 1:16,000 scale and referenced to topographic maps at 1:50,000 scale. Our role was to see if the patterns observed on the ground, and in large-scale airphotos, could be seen in the Landsat—imaged data. In this we were very suc- cessful. The anomaly is more clear on figure 11E. Ground measurements taken in the summer of 1975 are being interpreted and will be presented in a paper submitted to the International Geological Congress, Sydney, Australia (Bolviken and others, 1976). TlFALMlN, CENTRAL WEST NEW GUINEA Landsat—1 multispectral scanner (MSS) data in CCT format were acquired for scene 1028—00134 which covers an area in Papua, New Guinea, containing a number of porphyry prospects. Only two of these, however, Kennecott’s Tifalmin prospect and Carpen- teria Exploration’s Freida prospect, were sufficiently free of cloud cover during data acquisition to provide adequate data for remote-sensing investigations. Work to date has been confined to the Tifalmin prospect, since basic ground—acquired geological and geochemi- cal data, though limited, are available for that site only (Bamford, 1974). Data manipulations were carried out using the STANSORT computer program. Specific output gen- erated in the current study was Channels 4, 5, 6, and 7 shadeprints; R47, R45, and R75 reflectance ratio prints (with minimum values of ratios subtracted and step intervals selected to maximize print contrast in areas of interest); and cluster analysis prints (Channels 4, 5, 6, and 7 data clustered using a i 14 percent tolerance). A complete set of these data was derived for each of two centers of copper mineralization, the Olgal-Futik and Unfin areas of the Tifalmin prospect. A single line- print page (approximately 1 : 20,000 scale and covering an area about 3.8 by 5.4 km) provided adequate cover- age of each mineralized area and adjacent surround- ings to facilitate definition of reflectance anomalies (if any) within that area (fig. 12). All shadeprints and ratio prints in this study were color coded to aid interpretation using brown to yellow colors of gradu— ated density. Dark colors were used for delineation of higher values. Cluster analyses were coded using pro- gressively darker colors to define progressively small- er (more anomalous) areas. Figure 12 shows part of data generated for the Olgal—Futik and Unfin mineralized areas. Figures 12A and 128 are uncolored M88 7 shadeprints for these areas showing the locations of the mineralized intru- sions and principal drainages. Copper geochemistry 287 MINERAL EXPLORATION APPLICATIONS OF LANDSAT IMAGERY .ooodmmur m_ 96; 5505 mm 28m Jaw—t x3225. de .mu_wo_0w0 >w>5m .mu_mo_ow0 cflmmihoz of E0: UwEvoE m_ 3:: u_wo_omo Swan 30ng Emummm 9: we «Son 9: Eu: 2min *0 ac: :_mu_tw>: 295 m m_ 59an m :0 mp: .Qsota came 95 9: comics in floxi Slop _m>c >Ew m *0 «S: m. 35:95 9: we 53qu m :U 9: .va o :0 5 E59? «39 w_ :5th rmEOCm 9: «m5 :5 daE .mu_wo_oww 2: we 2:953 2232:: mcfiwuim 2: 5:3 ccszEOU :30? m5 @505 m :0 :2: 3:02 .Amr: mmE _mu_mo_0ww or: 5:3 anQEOU 52:8qu m: Ucm Q: 2 355938 é ES, OE c 65:26 E; 3:: m .2ch he 5<§20: 9: $335 5253 288 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM E—MSS 7/5 FIGURE 12.—Examples of Landsat CCT-generated Iineprinter images, compared with the only geological map available (12A), for the Unfin area, Tifalmin prospect, New Guinea. The map shows the approximate location of mineralized intrusives (cross hatched) and the principal drainages. Figure 123 is an uncolored Channel 7 shadeprint, 12C 3 R47 ratio print, 12D a R45 ratio print, 12E a R75 ratio print, and 12F a normalized cluster analysis. All prints cover the same geographic area. over the mineralized intrusions commonly averages Systematic geochemical grid sampling, however, has 1,000 to 2,000 ppm in creek outcrop, rock chip sam- not been completed for these areas. ples, and about 500 ppm in ridgetop soil and rock sam- The MSS raw data shadeprints, band ratio prints, pies which are frequently oxidized and leached. and cluster analyses for the OlgaI-Futik and Unfin MINERAL EXPLORATION APPLICATIONS OF LANDSAT IMAGERY areas do not appear to provide evidence of correla- tion between reflectance and known mineralized areas. Explanations for the lack of detection of reflectance “anomalies" include the possibilities that: (a) reflect- ance modifications may have taken place as suggested by previous work (Press and Norman, 1972) but are of insufficient magnitude to be resolved in Landsat—1 MSS data and (b) vegetation assemblage over the mineralized area may have adapted to the base metal- anomalous soils by survival of tolerant species only and thus may not be significantly more stressed than surrounding vegetation. Work to test these tentative explanations is desirable but not practical at present due to the remoteness of the test locality. It should also be noted that atmospheric corrections have not yet been made, hence reflectance calculations have not been attempted. Such refined data could possibly reveal reflectance anomalies not resolved by the raw data. Lack of information on areas of zero reflectance or other calibrated reflectances precluded determina- tion of an atmospheric correction factor. 289 EFFECT OF VEGETATION ON INTERBAND CORRELATIONS AND RATIOS In tables 6 and 7 various refiectances and interband ratios for 575 ground-measured spectra of four main data sets (A=Goldfield, B=3 Spanish sites, C=Pine Nut (horizontal), and D=Norway) are compared with their cross correlations (table 6) and their individual variabilities, expressed as coeFficient of variability (COV) =o/x. The progression A—D represents increas- ing numbers of vegetation spectra in each total set, either as pure vegetation-only spectra (Pine Nut) or increasing leaf-cover in the field of view (Norway). Each locality is mineralized, and they form typical ex- ploration targets. All sets contain some soils and rock spectra. Table 6 shows two partial sets of the 144 cross—correlations calculated—those of band correla— tions (Channels 4, 5, 6, and 7) are those of the inter- band ratios (R76, 75, 74, 65, 64, and 54). Solid con- tours serve to emphasize the higher positive value, dashed lines enclose the higher negative sets. TABLE 6.—Correlation between variables in data sets with increasing vegetation content [COV values derived using BMDOZD program. Solid lines enclose higher positive values; dashed lines enclose higher negative values.] A. GOLDFIELD Ch 5 6 7 83 80 77 4 77 77 S 87 6. B. SPANISH TEST SITES (3) Ch 5 6 7 96 83 77 4 86 78 5 98 6 (29% vegetation in set; N:229) (0% vegetation in set; N:113) R75 74 65 64 54 C. PINE NUTS (83% vegetation in set; N:151) Ch U1 0‘ \l 9:“ mm 88 D. NORWAY (85—90% vegetation in set; N282) Ch 5 6 7 87 85 4 83 g: 5 6 l 34 30 (:4: 1371\\06 R76 58 70" 363C163 75 34 81 62 74 60 —28 65 59 64 R75 74 65 64 54 62 5 .. 36 ——48 R76 95 98 9o —37 75 94 97 ——11 74 93 —33 65 —01 64 R75 74 65 64 54 86 88 82 84 {—53\ R76 98 99 98 —66\ 75 97 99 —54 I 74 98 I—69 i 65 \55/ 64 R75 74 65 64 54 09 37 — 03 26 R76 70 ' —38 75 59 28 74 7o —44 65 21 64 290 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM TABLE 7.—Means and variability in data sets with increasing vegetation content Set A Set B Set C Set D Spanish sites Pine Nuts Goldfield (3) (horiz) Norway Percentage vegetation in set ___________________ 0% 29% 82% 85—90% N ______________________ 113 229 151 81 0'14 ____________________ *212(.33) *120(.70) 96(.88) 54(.31) Ch5 ____________________ 260(.36) 143(.78) 115(.96) 62(.65) Ch6 ____________________ 317(.32) 208(.60) 214(.44) 159(.76) Ch7 ____________________ 285(.34) 240(.59) 283(.36) 212(.76) NX ____________________ 335(.08) 307(.17) 237(.23) 267(.17) NY ______________________ 403(.05) 344(.20) 253(.30) 264(.15) R76 ____________________ 900(.10) 115(.13) 137(.12) 134(.12) R75 ____________________ 111(.13) 204(.65) 368(.45) 350(.35) R74 ____________________ 136(.16) 226(.57) 408(.41) 413(.30) R65 ____________________ 124(.13) 171(.55) 258(.38) 261(.36) R64 ____________________ 152(.17) 191(.47) 287(.34) 306(.28) R54 ____________________ 123(.13) 115(.19) 113(.10) 122(.30) *Croup mean values; COV in parentheses. (Ch 4—Ch 7, x10; rest, x100); coefficient of variability (COV):0/x; if >O.40, data are widely spread. Calculated using BMDOZD out- put, ungrouped within sets. NX=Ch 4/(Ch 4+R5+R7); NY:Ch5/(Ch4+R5+R7); NX+NY +NZ:1.00. Data sets presumed to be Gaussian populations. Table 7 lists the group mean values for each of the 12 variables used and shows the variability (COV=o/x)* in parentheses (where COV is >040 the data are widely spread). Analysis—Table 6: Cross-Correlations 1. Channel correlations are high (>O.70) in all four data sets, but increase with vegetation content. 2. Correlations between Channels 6 and 7 are higher with increasing vegetation content. 3. Correlations between Channels 5 and 4 are high even where vegetation is dominant. 4. Interband correlations markedly change in value with even a slight increase (29 percent) in vege- tation content (see absolute values in table 7), in some cases doubling in value. 5. High (>80 percent) vegetation sets (C and D) show varying correlation patterns, which may mean each relationship is specific for one data set only. This is being investigated. Analysis—Table 7: Mean (Absolute) Levels of Variables 1. Variability (COV) of the values must be considered for data dispersions, when comparing means. *The COV values were derived using the BMDOZD (UCLA Biomedical Division) statistical package, which examines all the spectra as a set in an ungrouped mode and differs slightly from the BMDO7M discriminant program, which also develops a cor— relation matrix, but uses grouped data leading to consistently higher values for U and for the linear correlation coefficient, r. Data sets were presumed to have normal distributions. For example, even though mean values for each channel vary sympathetically with increasing vegetation, their COV‘s are very high. 2. The normalized color index (NX, NY, NZ, Where NX+NY+NZ=1, NX=Ch 4/(Ch 4+R5+R7), etc., is sensitive to vegetation, being high (330 NX: 400 NY) in altered rock (Set A) and declin- ing to lower values (260 NX: 260 NY) with in- creasing vegetation. (Vigorous, green plant ma- terials would have values as low as 150: 170 re- spectively, for “pure” fields of view). This index is a version of the International CIE color index used for color standardization in the visible region, adapted by us for use in the near-infra— red, e.g., Ch 4+Ch 5+Ch 7 are used (Lyon and others, 1975). 3. Ratios R75, 74, 65 and 64 are vegetation-sensitive (Lyon, 1976) and rapidly change with vegeta- tion content, especially R75 and R74. 4. Ratio R54 usually drops with increasing vegetation, but also rises rapidly with bare exposures of iron-rich soils. The fact that all four sets contain these soils, as well as vegetation, somewhat ob- scures this effect with these data sets. 5. Criteria of “usefulness” (e.g., which variable is the most useful, as in table 7) such as that used suc- cessfully by Goetz and others (1975) (a high COV of a set of group means for a variable) must also take into account the dispersion of each population from which the means were de- MINERAL EXPLORATION APPLICATIONS OF LANDSAT IMAGERY rived. Following their practice, however, one would decide that R74 with the highest COV of 0.46 was the best ratio, although the COV for the R74 mean in Set B-Spanish was 0.57 indicat- ing widely varying data in that total set. CONCLUSIONS Digital processing of Landsat data is of great signifi- cance to mineral exploration. It may be used to pro- duce enhanced images of mosaics of large areas which by computer manipulation now have better definition or increased contrast which will aid the interpreta- tional analyses of photogeologists. Of much greater import is the fact that Landsat is a four-band set of radiometers which in addition to pro- ducing imagery also produce quantified and calibrated measurements of any terrain in the world. As the data are readily available on digital tape, any number of algorithms may be used to remove atmospheric scat- tering effects or stretch densities (both to increase contrast and, hence, interpretability) or to prepare computational products like ratios. Statistics may be calculated and search programs (like STANSORT) written to automatically locate comparable reflect- ances elsewhere. At Stanford our approach to the mineral exploration applications has been concentrated in high resolution studies of small 1—100 km2 areas testing the ultimate limits of the Landsat system because this scale of rep- resentation (124,000) and areal size is that most often of use in exploration. Vegetation must become a part of mineral explora- tion/Landsat studies. Very few areas of interest are devoid of vegetated cover and that cover must be con- sidered in defining the “average Landsat spectrum” of any 1 pixel area (0.4 ha). The principal problem facing us at this stage of the studies is sampling, or how we can integrate enough terrain in our “ground” measure- ments to be equivalent to the Landsat pixel size. Clear- ly the next step involves helicopter-sampling. A second real problem is that no one technique works all the time and a group of approaches must be used. Landsat with its four-quantified channels ma- terially broadens the usual data base available to the explorationist used to working with black and white film. As a corollary to this, an interactive, video-display system, “driven” by the geologist who is personally familiar with the area and problem is essential to ex- tracting the most from the Landsat system. For a while we called our program H—ERTS, because it put the geologist in the “driver’s” seat. Man-machine interac- 291 tions are essential in Landsat analysis, preferably the man should be the field geologist himself. SELECTED REFERENCES Ashley, R. P., 1971, Preliminary geological map of the Goldfield mining district, Nevada: US. Geol. Sur- vey Open-File Map. 1974, Goldfield mining districts: Nevada Bur. Mines and Geology Rept. 19, p. 49—66. Ballew, G. I., 1975, A method for converting LAND— SAT—1 MSS data to reflectance by means of ground calibration sites: Stanford Remote Sens- ing Lab. Tech. Rept. 75—5, p. 90. 1976, Correlation of LANDSAT—1 multispec- tral data with surface geochemistry, in Internat. Symposium on Remote Sensing of Environment, 10th, Ann Arbor, Mich. 1975, Proc. (in press). Bamford, R. W., 1974, Band reflectance ratio maps from ERTS—l data over two copper prospects in New Guinea: Stanford Remote Sensing Lab. Tech. Rept. 74—13, 4 p. Billings, W. D., 1950, Vegetation and plant growth as affected by chemically altered rocks in the west- ern Great Basin: Ecology, v. 31, no. 1, p. 62—74. Bolviken, 8., Honey, F. R. Levine, S. L., Lyon, R. J. P., and Prelat, A., 1976, Detection of naturally heavy- metal poisoned areas by LANDSAT-1 digital data (abs): Internat. Geol. Congress, 6th, Sydney, Aus- tralia 1976, Proc. sec. 10 B (geochemistry). Draeger, W. C., and Lauer, D. T., 1967, Present and future forestry applications of remote sensing from space: Am. Inst. Aeronautics Astronautics, 4th Mtg, Anaheim 1967, Paper 67—765. Goetz, A. F. H., Billingsley, F. C., Gillespie, A. R., Squires, R. L., Shoemaker, E. M., Lucchitta, I., and Elston, D. P., 1975, Application of ERTS images and image processing to regional geologic prob- lems and geologic mapping in northern Arizona: Pasadena, Calif. Inst. Tech. Jet Propulsion Lab., Tech. Rept. 32—1597, p. 1—188. Honey, F. R., Prelat, A., and Lyon, R. J. P., 1974, STANSORT: Stanford Remote Sensing Labora- tory Pattern Recognition and Classification Sys— tem, in Internat. Symposium on Remote Sensing of Environment, 9th, Ann Arbor, Mich, 1974, Proc., p. 857—905. Howard, J. A., Watson, R. D., and Hessin, T. D., 1971, Spectral reflectance properties of Pinus Pon- derosa in relation to copper content of the soil— Malachite mine, Jefferson County, Colorado, in Symposium on Remote Sensing of Environment, 7th, Ann Arbor, Mich. 1971, Proc., p. 285—296. 292 FIRST ANNUAL PECORA Lag, J., and Bolviken, B., 1974, Some naturally heavy- metal poisoned areas of interest in prospecting, soil geochemistry, and geomedicine: Norges Geol. Unders, v. 304, p. 73—96. Levine, S., 1976, Correlation of ERTS spectra with rock/soil types in California grassland areas, in Internat. Symposium on Remote Sensing of En— vironment, 10th, Ann Arbor, Mich. 1975, Proc. (in press). Lizaur, P., 1975, Study of the alteration zone around Goldfield mining district by means of ERTS satel- lite multispectral scanner data: Stanford Remote Sensing Lab. Tech. Rept. 75—3, p. 284. Lyon, R. J. P., 1975, A comparison of observed and model-predicted atmosphere perturbations on tar- get radiances measures by ERTS, in Inst. Elec- trical and Electronics Eng. Symposium on Appli- cation Remote Sensing Digital Imagery to Min- eral and Petroleum Exploration, Houston, Tex. 1975, Proc., 6 p. 1976, Correlation between ground metal anal- ysis, vegetation reflectance, and ERTS brightness over a molybdenum skarn deposit Pine Nut Moun- tains, western Nevada, in Internat. Symposium on Remote Sensing of Environment, 10th, Ann Arbor, Mich. 1975, Proc. (in press). Lyon, .R. J. P., and Honey, F. R., 1974, Multispectral signatures in relation to ground control signature using a nested sampling approach (abs): Natl. Tech. Info. Service (NTIS), NASA Earth Re- sources Survey Prog., Weekly Govt. Abs, E74— 10690, p. 132. MEMORIAL SYMPOSIUM Lyon, R. J. P., Honey, F. R., and Ballew, G. I., 1975, Evaluation of ERTS multispectral signatures in relation to ground control signatures using a nested sampling approach: Final Report, NAS 5—21884, June 1975, 600 p. (contains 16 separate reports). Moore, J. G., 1969, Geological map of Lyon, Douglas and Ormsby Counties, Nevada: Nevada Bur. Mines Bull. 75, pl. 1. Noble, D. C., 1962, Mesozoic geology of the southern Pine Nut Range, Douglas County, Nevada: Stan— ford Univ., Ph.D. thesis, 200 p. Press, N. P., and Norman, J. W., 1972, Detection of ore bodies by remote—sensing the effects of metals on vegetation: Inst. Mining and Metall. Trans, Sec. B, v. 81, p. 166—168. Rowan, L. C., Wetlaufer, P. H., Goetz, A. F. H, Bill- ingsley, F. C., and Stewart, J. H., 1974, Discrimi- nation of rock types and altered areas in Nevada by the use of ERTS images: U.S. Geol. Survey Prof. Paper 883, 35 p. Vincent, R. K, 1973, Spectral ratio imaging methods for geological remote sensing from aircraft and satellites, in Symposium on Management and Utilization of Remote Sensing Data, Sioux Falls 1973, Proc., p. 377—397. Whitaker, E. A., 1965, Colors and the meso-structure of the Maria, in Ranger VII, pt. II, Experimenter‘s analyses and interpretations: Pasadena, Jet Pro- pulsion Lab. Tech. Rept. 32—700, p. 29—39. 1966, The surface on the moon, chap. 3, in Hess, W. N., Mensel, D. H., and O’Keefe, J. A., eds, The nature of the lunar surface: Johns Hop- kins Press, p. 79—98. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Tradeoff Considerations in Utilization of SLAR for Terrain Analysis By Louis F. Dellwig and Richard K. Moore, University of Kansas Center for Research, Inc., Remote Sensing Laboratory, Lawrence, Kansas 66045 ABSTRACT Optimum SLAR system parameters for terrain data extraction have been suggested by previous investi- gators on the basis of comparison of imagery from several different systems with a wide diversity of pa— rameters. For some determinations this proves satis- factory. However, the determination of any optimum system parameter can best be made only on the basis of utilization of a single system, imagery of which could be generated with multiple polarizations, de- pression angles, look directions, or frequencies, or which could be manipulated to alter such parameters. Extraction of data from imagery showing a range of variations in such parameters by discipline-oriented interpreters has enabled us to shed some light on the effect of variations in radar system parameters on ex- traction of geologic, hydrologic, geographic, and bo— tanical data. INTRODUCTION Following the imaging of Panama in 1967 by the US. Army Corps of Engineers, serious evaluation of side-looking airborne radar (SLAR) as a geologic tool was initiated. Earlier numerous small test sites throughout the United States for the NASA Earth Re- sources Survey Program resulted in the generation of imagery over a variety of terrains, and, with the gener- al availability of this imagery to the geoscience com— munity, interest in its evaluation and utilization was accelerated. Following the entrance of Aero Service, Grumman Aircraft Corp, and later Motorola Aerial Remote Sensing, Inc, into the commercial field which had been established initially by Westinghouse Aero— space Corp., imagery of several systems became avail- able to the geoscience community and a not unex— pected evaluation and comparison of these systems began. Now with the reduction of the lower limit of resolution from 35 to 10 ft (11.6 to 3 m), imagery of additional systems is becoming available to the re- searcher and will inevitably be competing with better known systems for the “best in show” award. Recognizing a great diversity of capabilities and limitations, we embarked upon a study under the spon- sorship of the Jet Propulsion Laboratory to determine optimum system parameters for a spacecraft radar. At about the same time we initiated A Demonstration and Evaluation of the Utilization of Side Looking Airborne Radar for Military Terrain Analysis for the Engineer Topographic Laboratories. Through the conduct of these two studies and with the availability of imagery not previously available, it had been our hope that we would be able to determine the optimum SLAR system parameters for terrain analysis. But as the projects progressed we became faced with the problem which had faced us on many occasions, optimum SLAR sys- tem parameters for what? To a large degree the requirements for tactical mili- tary terrain analysis are those of the civilian segment of the population, only two basic requirements being critical in military terrain analysis: (1) Near or real time data acquisition and (2) target detection and identifi- cation. Many civilian uses of airborne monitoring, re- connaissance, or mapping systems also require near real time data and identification of specific or unique features in both urban and rural environments. The optimum parameters for military or civilian op- eration is only one of the choices which must be made. Optimum for the geologist who prefers a wide swath width is not optimum for the city planner who wishes 293 294 to acquire. as 'much detail as possible in a relatively small urban area. Thus recognizing that optimum for the geologist is not optimum for other geoscientists, rather than define a series of hard and fast parameters such as was done by the National Aeronautics and Space Administration (NASA) in the selection of the Landsat Multispectral Scanner (MSS) bands (and satis- fying no one), it seems most reasonable to evaluate variations in parameters and not attempt to define optimum parameters for a single system. With such an approach the reader will essentially have a do-it-Iyour- self kit and can define the parameters (including cost) which best enables him to realize maximum data con- tent for any specified mission. RESOLUTION Few parameters have been devoted the attention which has been received by resolution. In an early geo- logic study (Dellwig and others, 1966), it was stated that there appeared to be some value in the degrada- tion of resolution, but, in a more recent paper by Ryd- strom (1970), it was pointed out that high resolution radar imagery contained not only the same but more data than did imagery of lower resolution systems. It is not the authors’ intention to resurrect this problem, for sufficient data are not available for either author to pursue the point further. Rather, we would prefer to look at the needs of terrain analysts and evaluate systems on their own merit. We can begin by looking at the Phoenix, Arizona, area as imaged by the AN/APQ-152 (fig. 1A) and the AN/APS—94D (fig. 18). There is an obvious strong contrast in resolution, but there are also other perhaps not so obvious contrasts. Both images show full swath width. For the urban planner there is an obvious advantage in the utilization of the AN/APQ—152, but for the geologist who is con- cerned with tracing major linear patterns or in con- ducting reconnaissance studies in heretofore uncharted areas, the wide swath width of the AN/APS—94D (which cannot be obtained with the AN/APQ—152) provides a more desirable format and presents him with adequate data for mapping purposes. On the coarser resolution image no difficulties are encoun- tered in defining the urban areas and adjacent field patterns nor in the separation of industrial from resi- dential areas. In general, swath width cannot be in- creased without suffering some degradation of reso- lution, and thus the user must consider the relative values of fine resolution and wide swath width. In considering resolution, one must look at both spa- tial resolution, which we normally consider, and gray- scale resolution, which is seldom considered in pho- FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM tography because it is usually good, but is always important for radar. The problem is that a radar image produced with the best possible synthetic-aperture resolution has an extremely speckled appearance. This comes about because of the same phase-interference phenomenon that causes radio signals to fade, and that is particularly evident in nighttime standard- broadcast signals and in FM signals received in a car. Shine a laser on a moderately rough surface like a blackboard and you will see this speckle phenomenon dramatically displayed. To reduce speckle and thus improve the ability to determine the average gray scale in a region, one must look at the target element many times and average the results together to make the image. This can be done with a radar by the panchromatic technique (Moore and others, 1969) as in optics, but most radars do not have this capability. A tradeoff must be made between reduced speckle (improved gray-scale resolu- tion) and finer spatial resolution. With properly con- figured systems this is a direct tradeoff: doubling the two linear dimensions of the spatial resolution cell im- proves the standard deviation of the speckle by a factor of two. To determine gray scale within 10 per— cent of the mean at any point requires 100 looks, resulting in 10 times poorer spatial resolution. With 20—percent uncertainty in the gray level of a single cell, the penalty in spatial resolution is only a factor of 5. Even :50 percent uncertainty is often good enough, and in this case the spatial resolution is degraded only by a factor of 2. Probably one would never be satis- fied with this for a single cell, but the eye does the averaging over homogeneous areas to improve the net effect to much better than :50 percent, and having the spatial resolution allows one to determine which areas the eye should average in a way that would not be possible if the averaging were done automatically. The standard deviation of the gray scale turns out to be a relatively poor indicator of the interpretability of an image that has a lot of speckle; a better indi- cator is the ratio of the brightness exceeded 90 per- cent of the time to that exceeded 10 percent of the time (Moore, 1975). With most systems in use today these values can be obtained from a chi-square dis- tribution of 2N degrees of freedom, where N is the number of samples averaged. When this definition is applied, the ratio between the resolutions required for equal interpretability with a fully coherent single-look synthetic-aperture radar and a radar with photo— graphic quality (infinite number of looks) is found to be 4.7. That is, if 30.4 ft (10 m) resolution is required TRADEOFF CONSIDERATIONS IN UTILIZATION OF SLAR FOR ANALYSIS 295 —'—l I KILOMETERS o—r—O , 0 5 10 15 MILES Ifi—LI—‘I—L-‘—I o 5 10 15 KILOMETERS FIGURE 1.——Two radar images of Phoenix, Arizona. A, AN/APQ—152 (X—band) SAR (Synthetic aperature radar) imagery. B, AN/APS—94D (X-band) SLAR imagery. Area outlined indicates aproximate coverage of A. Apparent distortion of outlined area results from the transfer of one image in slant range format to another, also in slant range format. 296 for the single-look system, one can get by with 142.8 ft (47 m) in the photographic-quality system. A recent study at the University of Kansas (Moore, 1975) disclosed that the numerical interpretability of an image can be related to the size of a square pixel by the relation I=I0 exp(—r2/r:) where r is the pixel dimension and re is a critical pixel dimension for the type of application. Figure 2 illus- trates how well this relation (indicated by the solid regression line) fits the experimental data for four different areas and two classes of target. Note that the critical value of resolution for the single-look picture (point of 37—percent interpretability) is at around 30 ft (10 m) for both kinds of application, and that the cor— responding “Photographic” resolution is nearly 150 ft (50 m). Another result of the study was the discovery that the area of the pixel is the key factor in its interpreta- bility, that is, a long, narrow pixel and a square one having the same area turned out to have the same interpretability, This is illustrated, using only the fully coherent image data (fig. 3). The largest length/width ratio used was 10 in this study. From this study, and others like it that remain to be clone, the equivalent resolution required for a given mission can be calculated for any radar whose averag- ing characteristics are known once the single-look or photographic-quality critical resolution has been deter- mined. One interesting result of the study is that the azimuth resolution for a radar that averages 10 looks can be 10 times as great as for a single-look system and achieve the same interpretability, provided the range resolution is the same as for the single-look system. This means that for most aircraft radar appli- cations to earth science a real-aperture radar that is properly designed can be just as effective as a single- look synthetic-aperture system. A 3-look synthetic sys- tem, however, is significantly better than the lO-look system (synthetic or real aperture). Hence, if one real- ly wants to determine whether a particular system can do the job, one must determine the required equivalent photographic resolution and from this calculate the capability of the system under consideration. FREQUENCY Currently plans are being formulated to fly a two- frequency radar aboard a shuttle mission. Defense is being mustered for an X(~3 cm)~ and L(~25 cm)—band system on the basis of its capability for mapping geology. Partly in defense of this has been presented FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM the results of an excellent study by Schaber, Berlin and Brown (1975), who have pointed out that L-band radars rather dramatically effected the separation of coarse (>2.0—3.5 cm) from fine (<2.0—3.5 cm) gravel. In a similar manner, it was pointed out in earlier studies at Pisgah Crater (Dellwig, 1969; MacDonald and Waite, 1972) that the smoothness of playa lake surface and adjacent lavas and alluvial fans in the basin and range country of the Western United States could be defined on the basis of the return from such surfaces on K(1.18—1.67 cm)- or X-band radars. From these studies it becomes quite obvious that the opti- mum frequency for an imaging radar is that frequency which enables the investigator to derive the necessary terrain data for his particular discipline. From the point of view of general geologic mapping, one finds diffi- culty in defending the utilization of an L-band system as opposed to a C(~5 cm)- or even P(77-136 cm)- band system, for geologic studies do not necessarily require the separation of material of a size range smaller than 2.0—3.5 cm from material of a size range larger than that figure. A problem of infinitely more general importance than the separation of materials of varied size ranges is the determination of soil moisture content. Studies concurrently being conducted at the University of Kansas using the Microwave Active Spec— trometer (MAS) system rather dramatically demon— strate that the indication of soil moisture content which one might expect to be reflected in radar back- scatter is strongly frequency dependent (Batlivala and Ulaby, 1975). Whereas the data required by geologists for the most part are not particularly frequency de- pendent, a wide range of earth scientists interested in soil moisture content will undoubtedly show a strong preference for radar imaging in the C-band range. POLARIZATION A parameter never fully investigated, and unfortu— nately not subject to investigation by most earth scien- tists due to a lack of data, is that of polarization. The evaluation of numerous AN/APQ—97 images gen- erated during the NASA Earth Resources Survey Pro- gram fiights failed to reveal any significant difference from the point of view of geologic investigation be- tween parallel-vertically and parallel-horizontally po- larized signal return in Ka(~.86 cm)-band. Compari- sons between like and cross polarized images, how- ever, showed considerable potential for the revelation of data not available in a single image. In studying a variety of rock outcrops in the Western United States, McCauley (1972) identified surface roughness as the cause of polarization reversals in a variety of rock TRADEOFF CONSIDERATIONS IN UTILIZATION OF SLAR FOR ANALYSIS 297 | I I I T T I Combined hard target O Coherent V Noncoherent RELATIVE INTERPRETABILITY N I O O O 1 _ . —4 V0. 0 I 1 . . t 0 1o 20 3o 40 50 60 70 so 90 100 RESOLUTION OF EQUIVALENT COHERENT RADAR row (FEET) o 50 100 150 200 250 300 350 400 450 500 I I I I I I L I I I J RESOLUTION OF EQUIVALENT ”PHOTOGRAPHIC" RADAR reN (FEET) I I I I I I I Combined roads and transportation RELATIVE INTERPRETABILITY N I I I I y 00 10 20 30 40 80 4&6 1 RESOLUTION OF EQUIVALENT COHERENT RADAR reN (FEET) 0 50 100 150 200 250 300 350 400 450 500 I I I I 1 I L 1 I l I RESOLUTION OF EQUIVALENT ”PHOTOGRAPHIC" RADAR reN (FEET) FIGURE 2,—Interpretability vs. equivalent resolution for hard targets, roads, and other transportation—«all data sets combined. types (fig. 4), all of which showed similarity in surface and cross polarized return signal of manmade linear configuration. The studies of cultural features by Lewis features (fig. 5). In the Gulf Coast (fig. 6), MacDonald (1968) also revealed a significant contrast in the like and Waite (1971b) noted strong contrasts in the return 298 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 4 Hl‘llll Ell E 33*- I 3 ._ E] |.I.I I O. 35 3 BIG g E D D .“1— CID. “5 I:I CID Roads and transportation I Square cells C] Rectangular cells 4—“ I I l 7 I I I I I I I I Hardtargets I E 3— u — c7: :5 w - t: n. E 2— _ I'- E E] El E1 Lu I Z I I p— ‘E‘ DI: U El El 0 I I I I I if b I A I I I b 0 50 100 150 LENGTH OF SIDE OF EQUIVALENT SQUARE CELL (FEET) FIGURE 3.-——C0mparison of square and rectangular cells. in the near range between the like and cross polarized signal, a contrast which they attributed to a difference between moist and dry terrain surface. Utilizing the MAS system, Ulaby and others (1974) and Batlivala and Ulaby (1975) have recorded such contrasts in return signal. In order to place a value on dual polarization imag- ing one must first define the cause for tonal reversals, and secondly must be able to utilize this contrast re- gardless of the subtlety of expression. For example, on casual examination of the like and cross polarized imagery of Freeport, Texas (fig. 7), some tonal con— TRADEOFF CONSIDERATIONS IN UTILIZATION OF SLAR FOR ANALYSIS N EAR-v HANG E 4 KILOMETERS FIGURE 4.—Mono Craters, California, AN/APQ—97 (ka-band) SLAR imagery. Flows a and b are blocky lavas and have lower returns on cross polarized (HV) imagery. Flows c and d are covered by ash and appear the same on both poIarizations. 300 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 1 2 3 4 MILES I l I I I 2 3 4 KILOMETERS 0 I I I I I I 0 FIGURE 5.—Bountiful, Utah, AN/APQ—97 (Ka-band) SLAR imagery. Urban and industrial areas are more apparent on HH (like) polarization. Communication networks are more accurately mapped on HV (cross) polarization. TRADEOFF CONSIDERATIONS IN UTILIZATION OF SLAR FOR ANALYSIS 10 MILES ‘IO KILOMETERS FIGURE 6.—Atchafalaya River Basin, Louisiana, AN/APQ—97 (Ka—band) SLAR imagery. Areas of high soil moisture appear dark in near range HH (like) polarization imagery. Dry areas appear light in tone. Soil moisture differences are not detectable on HV (cross) polarized imagery. 302 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM NEAR BANGE‘ Wn‘kak * 5 MILES J l 5 KlLOMETERS o—p—o . FIGURE 7.——Freeport, Texas, AN/APQ—97 (Ka-band) SLAR imagery. Area of figure 8 outlined in HV image. trasts may be obvious but others will escape the inter- LOOK D|RECT|ON AND DEPRESSION preter’s naked eye. However, by adding these two ANGLE images in contrasting colors, the interpreter can easily define the extent of any unique signature and rapidly The need for control of look direction and depres- effect interpretation (fig. 8, p. XXI). sion angle has been well established by numerous TRADEOFF CONSIDERATIONS IN UTILIZATION OF SLAR FOR ANALYSIS 303 0 L I 0 5 10 10 15 MILES I J I 15 KlLOMETERS FIGURE 9.—Ph0enix, Arizona, AN/APS—94D (X-band) SLAR imagery. A, North look of Phoenix and mountains and alluvial fans (middle to far range) to the northeast. B, Northeast look of Phoenix. C, Northeast look of alluvial fans (near to middle range). investigators, among them, MacDonald and Waite (1971a) who pointed to the desirability of selecting specific ranges of depression angles for regions of varying relief. However, relief is not always uniform throughout an area and in areas that in general may be of relatively low relief, occasionally local high relief will result in excessive shadowing if low depres- sion angles (deemed best for relatively flat terrain) are 304 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM zip-May 1972‘ ‘0 20 MILES J 0‘0 7% 2lo KILOMETERS FIGURE 10.——Denver- Colorado Slporings corridor, Colorado SC— 01 (X-band) SAR imagery. used. The significance of control of look direction and the other to the northeast and oblique to the overall depression angle is indicated once again in the Phoenix north—south, east-west street pattern give strongly con- area over which two looks at the city (fig. 9A,B), one trasting views of the urban area. Similarly, two differ- to the north and normal to the overall street plan and ent looks at natural terrain features as is indicated in TRADEOFF CONSIDERATIONS IN UTILIZATION OF SLAR FOR ANALYSIS 305 0 5 10 16 IIIIIIIIIIIIIIIIlllllllllll 1:12,000 _ u 12 /// d v // E ______—.-” m V - a: E‘; V /1:2o,ooo 5’: 8 / _ \ w E // _ II / Notes: Radar 0 1. Photo prices based on single-engine D 4 aircraft. 2. Photo contractor allowed 2 or more seasons for jobs over 9,000 sq. mi. 3. Photo prices do not include mosaic work, _ " radar does 4. Figures apply to the Western United States ollllIIIlJIJLlIIIIIlIIIIIIlIIJ 15 20 SQUARE MILEs(ooo’S) FIGURE 11.—-Price comparisons—photos vs. radar, spring 1975. another pair of Phoenix images (fig. 9A,C) give con- trasting indications of the degree of relief that should be expected in the area covered by alluvial fans. In the one image, because of look direction and depres- sion angle, little contrast is seen between the arroyos and the interfiuve areas, whereas in the second image, due to a change in look direction and depression angle, the contrast between the arroyos and the interfiuve areas is sharp. The extensive shadowing behind iso- lated mountain peaks and the revelation of terrain data in the shadow area through a look from the opposite direction, further points to the desirability of obtain- ing two looks of an area in order to obtain the maxi— mum amount of terrain data. Such was first demon- strated by MacDonald (1969) in the Darien Province, Panama, where it was determined that, with each additional look, additional data were revealed. TIME Numerous investigators utilizing ERTS imagery have been able to document data increase through second and even third or fourth season imagery evaluation. Such should also be true for radar imagery especially in vegetated terrain due to seasonal variations in di- electric properties and surface configuration of the vegetal cover (fig. 10). Not only is the change in data content of winter and spring imagery documented, but also it emphasized the need for selection of imaging season depending on the nature of the desired data. COST Perhaps not a system parameter, but an important factor to be considered in the determination of opti— mum system parameters for specific types of earth science investigations is cost. Mobilization charges and Flight line density control to a large degree the cost of imaging—a cost which is essentially prohibitive for the obtaining of detailed data in relatively small areas. Even assuming that imaging systems of resolutions not presently available would become available in the future, it appears that cost would be prohibitive when compared with aerial photography. Only when real time or near real time (13 seconds between imaging and viewing on a currently operating experimental Dutch imaging radar and instant viewing at the end of a frame on the Kansas University radar [Eichel and others, 1975]) data are absolutely necessary (although certainly to be desired under most conditions) could the utilization of radar for small-scale mapping be justified (fig. 11). Conversely, for mapping large areas, 306 FIRST ANNUAL PECORA unless the resolution now only achieved by aerial pho- tography is absolutely necessary, one finds difficulty in justifying the utilization of any system other than radar. OPTIMUM RADAR-PARAMETERS You would like 1—ft resolution, wide swath width, multiple polarization, controllable depression angles, opposite looks and a range of frequencies and a real time display? Dream away! Even Uncle will have trouble affording that system. REFERENCES Batlivala, P. P., and Ulaby, F. T., 1975, Effects of roughness on the radar response to soil moisture of bare ground: Univ. Kansas Center for Re- search, Inc., RSL Tech. Rept. 264—5, 50 p. Dellwig, L. F., 1969, A geoscience evaluation of multi— frequency radar imagery of the Pisgah Crater area, California: Modern Geology, v. 1, p. 65—73. Dellwig, L. F., Kirk, J. N., and Walters, R. L., 1966, The potential of low resolution radar imagery in regional geologic studies: Jour. Geophysical Re- search, v. 71, p. 4995—4998. Eichel, L. A, Moore, R. K., Weilert, M., and Schlude, F., 1975, An inexpensive side-looking radar with a novel display, in Institute of Electrical and Elec- tronic Engineers Internat. Radar Conf, Arling— ton, Va., April 21—23, 1975, Proc. Lewis, A. J., 1968, Evaluation of multiple polarized radar imagery for the detection of selected cul- tural features: US. Geol. Survey Tech. Letter NASA—130, 52 p. MacDonald, H. C., 1969, Geological evaluation of‘ ra- dar imagery for Darien Province: Modern Geolo- gy, v. 1, p. 1—62. MEMORIAL SYMPOSIUM MacDonald, H. C., and Waite, W. P., 1971a, Optimum radar depression angles for geological analysis: Modern Geology, v. 2, p. 179—193. 1971b, Soil moisture detection with imaging radars: Water Resources Research, v. 7, p. 100— 110. 1972, Terrain roughness and surface materials discrimination with SLAR in arid environments: Univ. Kansas Center for Research, Inc., RSL Tech. Rept. 177—25, 37 p. McCauley, J. R., 1972, Surface configuration as an explanation for lithology—related cross—polarized radar image anomalies, in NASA Manned Space- craft Center, Annual Earth Resources Program Review, 4th, Houston 1972, Proc., v. 2, p. (36—1)— (36—9). Moore, R. K., 1975, SLAR image interpretability— Tradeoffs between picture element dimensions and non-coherent averaging: Univ. Kansas Center for Research, Inc., RSL Tech. Rept. 287—2, p. Bl— 91; p. 11—11. Moore, R. K., Waite, W. P., and Rouse, J. W., Jr., 1969, Panchromatic and polypanchromatic radar: Inst. Electrical Electronic Eng. Proc., v. 57, no. 4, p. 590—593. Rydstrom, H. O., 1970, Capabilities of advanced radar systems for monitoring land resources: Goodyear Aerospace Rept. GERA—1604, Arizona Div., 10 p. Schaber, G. G., Berlin, G. L., and Brown, W. E., Jr., 1975, Variations in surface roughness within Death Valley, California: Geologic evaluation of 25 cm wavelength radar images: US. Geol. Sur- vey Interagency Rept., Astrogeology 65, 63 p. Ulaby, F. T., Cihlar, J., and Moore, R. K., 1974, Active microwave measurement of soil water content: Remote Sensing of Environment, v. 3, p. 185—203. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Worldwide Indexing and Retrieval of Landsat Images By Floyd F. Sabins, Jr., Chevron Oil Field Research Company, La Habra, California 90631 ABSTRACT More than 100,000 Landsat images have been pro- duced since the first satellite was launched in July 1972. In order to select images on a worldwide basis, index maps are necessary. Manual sorting and plotting lists of images onto base maps are tedious, time-con- suming and error-prone tasks. Computers, however, are ideally suited for this work. The Landsat Index Map System described here employs a computer map- ping program and the EROS Landsat digital index tape to produce a series of 22 index maps covering the world at a scale of 1 :5,000,000. Only the highest quality images are plotted. One set of maps is plotted for images with zero cloud cover. Other sets are plotted for 10—, 20-, and 30-percent cloud cover, if necessary, to provide complete cover- age. The index maps are updated periodically from new index tapes. In addition to locating images, the maps indicate areas that lack coverage and image cen- ters located away from the nominal image center points. INTRODUCTION BACKGROUND Since the launch on July 23, 1972, Landsat—1 has transmitted approximately 100,000 images covering much of the land area of the world. Landsat-2, launched on January 22, 1975, is supplying additional images. The 18-day repetition cycle of each satellite results in many areas being imaged numerous times. For users requiring many images of large regions, es- pecially in foreign areas, the plotting of images onto index maps is a time-consuming, tedious, and error- prone operation. This paper describes an index map system developed by Chevron Oil Field Research Company that greatly simplifies and expedites the location and selection of Landsat images throughout the world. Numerous other papers have described the Landsat vehicles and imagery, which are not discussed here. CONVENTIONAL INDEXING METHODS The National Aeronautics and Space Administration provides monthly index listings for domestic and foreign Landsat images. In addition, the EROS Data Center at Sioux Falls can provide, upon request, com- puter printouts of image coverage for specific geo— graphic areas. For the conterminous United States, EROS also provides an index map based on “nominal image” center points. This map does not show specific images but rather the center points at which the repetitive images are positioned. This system origi- nated at the Canada Centre for Remote Sensing. Microfilm browse files of Landsat images are main- tained for public use at various US. Geological Survey offices. Useful Landsat indexes are prepared by the US. Agricultural Stabilization and Conservation Serv- ice aerial photographic laboratory in Salt Lake City. Images for each 18-day cycle are positioned on a US. base map to form a photo index mosaic. In addition to ground coverage, image quality and cloud cover can be determined directly. I understand that the labora- tory is no longer updating these indexes. For foreign areas, the index data consist of lists of Landsat identification numbers which must be sorted and plotted on a base map. Sorting is done on the basis of image quality and cloud cover in order that the best image can be selected for each center point. For mosaic preparation, it is desirable to use images 307 308 FIRST ANNUAL PECORA acquired at the same time of year to minimize tonal changes between adjacent images. This involves a separate sorting step. All of these sorting and plotting operations are tedious and error prone when done manually but are ideally suited for computer process- ing. The balance of this paper describes the system developed at Chevron Oil Field Research Company (COFRC) for producing Landsat index maps on a worldwide basis. ACKNOWLEDGMENTS My role in this work was to recognize the need for a worldwide Landsat index map system and suggest the approach described here. The actual programming and plotting was accomplished by the following COFRC colleagues to whom I am deeply indebted: H. L. Haines, E. 8. Hill, L. C. Bonham, C. D. Boyd, C. C. Davis, W. T. Miller, and G. S. MacKenzie. Per- sonnel at the EROS Data Center were very helpful in explaining the format of their index tape. INDEX MAP SYSTEM One of the truly unique features about the COFRC system is that we have not seen fit to title it with an acronym. The system comprises two essentially sepa- ma MEMORIAL SYMPOSIUM rate operations: (1) preparation of a base map show- ing coastlines, drainages, and political boundaries to- gether with a latitude and longitude grid and (2) plot- ting of Landsat center points and identification nam- bers on the base map. BASE MAP PLOTTING Figure 1 shows the location of the 23 individual maps which we use to index the world. Most of these maps are plotted to the same scale (125,000,000) and projection (Lambert conic conformal) as the Global Navigation Charts (GNC) published by the US. Air Force Chart and Information Center and distributed by the Distribution Division, National Ocean Survey, Washington, DC. 20235. By plotting the index maps on translucent paper, they may be overlain on the GNC to determine coverage of particular mountain ranges, sedimentary basins and other features not noted on the index maps. As shown on the flow chart, figure 2, two digital files are used in base map plotting: (1) world map file containing land outlines and (2) political boundaries and drainage file. Latitude and longitude coordinates are designated for the specific index map, which on figure 2 happens to be Index Map 11 covering southern Wfiflm k '19 FIGURE 1.—Location of COFRC Landsat index maps. Note portion of Index Map 11 that is shOWn on figures 4—9. WORLDWIDE INDEXINC AND RETRIEVAL OF LANDSAT IMAGES 309 LANDSAT PLOT LANDSAT INDEX TAPE, EDITED I RETRIEVE DATA FOR INDEX MAP AREA LISTING OF RETRIEVED DATA _/ CONVERTED LANDSAT SUBFILE SPECIFY % CLOUDS 81 TIME INTERVAL PLOT LANDSAT IDENTIFICATIONS & CENTER POINTS 1408.5871 1407.58701406.5862 + + + 1408.5875 1407.5874 1406.5865 + + + BASE MAP PLOT DRAINAGE& BOUNDARIES FILE WORLD LAND OUTLINE FILE SELECT DATA FOR INDEX MAP AREA I DESIGNATE MAP SCALE PROJECTION 8| ANNOTATION BASE MAP SUBFILE PLOT LAND OUTLINES, BOUNDARIES, DRAINAGES, LAT & LONG COMPLETED LAN DSAT INDEX MAP + FIGURE 2.—COFRC Landsat index map plotting routine. 310 FIRST ANNUAL PECORA Africa. The scale and map projection are designated and the base map is plotted preparatory to adding the Landsat information. LANDSAT IMAGE PLOTTING The first step is to acquire from the EROS Data Center a copy of their Landsat index tape which lists all the images on file at the Data Center. The latest Landsat tape we acquired lists 126,005 images cover- ing the period from launch through April 1975 and cost $135. EROS also provides a description of the tape format. The first step is to edit the tape and delete nonstandard and duplicate entries. Using the edited tape as a source, a subfile tape is created list- ing only the images located within the boundaries of the particular index map (fig. 2). At this stage we elimi- nate images with poor quality or significant cloud coven A listing of all these images in numerical (time) se- quence is also produced. This printout eliminates much of the information on the original index tape (such as image corners, scale and altitude) which is not re— quired for retrieval. On the sample printout of figure 3, an explanation is provided for the various entries. The Landsat identification number (ID. No.) which is unique for each image, is employed on all of our index maps and merits a description. The nine-digit number (ex— ample, 1055.08041) originates as follows: 1—Landsat—1 (This digit is a 5 for 1000 days and more after launch) OBS—Days after launch (July 23, 1972). This day is September 16, 1972. 08—Hour of observation (Greenwich time) 04—Minute of observation 1—tens of seconds On our index maps, the leading zeros are deleted on the time code. This image is plotted as 10558041. The digital tape subfile for the index map area is then converted to a format compatible with our plot— ter. Before plotting the Landsat center points on the base maps, we determine from the listing the approxi— mate number of images in the subfile. Owing to the 18—day repetition cycle there would be much overprint- ing of identification numbers if an entire subfile were printed on one map. By referring to the listing, we designate time intervals, each of which is plotted on a separate map with a minimum of overprinting. For the southern Africa map, the following intervals were plot- ted and are illustrated: Figure Days after launch Dates 4 ____________ 0—73 ____________ 72/07/23—72/10/10 5 ____________ 74—144 ____________ 72/10/11—72/12/14 6 ____________ 145—248 ____________ 72/12/15—73/03/28 MEMORIAL SYMPOSIUM March 28, 1973, was the latest entry on the initial index tape. Updating the maps is described later. For ease of illustration, these maps are the portion of the entire southern Africa index map indicated on figure 1. This series of maps is restricted to images with O-percent cloud cover. Successive map series can be plotted for 10, 20, and higher percent cloud cover if needed to provide complete coverage. On figure 4, note that the center points for paths 1050 and 1068 (18 days apart) nearly coincide. Any ad- ditional images plotted along this path would be over- printed and produce an illegible map. Where minor overprinting occurs (fig. 5), the index numbers are readily resolved by referring to the printout list. On figure 6 note that the center points for paths 1177, 1178, and 1179 are located between the normal posi- tions for these paths. This would present a problem on an index map based on nominal image centers and is discussed later. The 115- by 115-statute-mile outline on figure 4 shows coverage of a Landsat frame. By centering this template over an image point with east and west bor- ders parallel with the orbit path, the ground coverage of each image can be determined. UPDATING THE INDEX MAPS To maintain a current set of index maps, we order new Landsat index tapes at intervals. Each tape is cumulative from the day of launch. We observed that later index tapes commonly list images taken within the time span of an earlier tape, but which are not listed on the earlier tapes. Using the southern African maps (figs. 4, 5, and 6) as an example, our initial tape extended through 248 days after launch (March 28, 1973). A later tape extending through 478 days (No- vember 13, 1973) included 33 images acquired during the first 248 days but not listed on the early tape cov- ering that period. For this reason we always search each new index tape for earlier entries which are plotted. Figure 7 is a retroactive plot of the 13 Cloud- free images that were missing from the initial index tape, but present on the later tape. The other 21 earlier images have higher cloud cover percentages. One explanation is that the EROS Data Center enters images on the tape as they are received from the processing facility at Goddard Space Flight Center, and the images are not necessarily received and en- tered into the data base in a strict time sequence. They are listed sequentially by Landsat index number, how- ever. Another explanation is that the retroactive points may have contained nonstandard characters on the initial tape which would have eliminated them during 311 WORLDWIDE INDEXING AND RETRIEVAL OF LANDSAT IMAGES .F 23m: de xva :0 :39? NA: 922 xobc: 8:2 59:33 2 «NE 2: .meN 5?: 758:3 No SONEE UEOU No co_ton_ll.m 550E u. mm um H R EWE mm M M n.m m ”J .. n cc. .|. mpmmw m w m v mu am so mm o N w ..ms a u n mm m; rll rIL u a m «sum 0 N a m a Mm mm em m m . w M wmmw w m mm mm a mo Nm mu m qu mud Nu unmn m a an mm m vm um mN u m“. “N. am Apnm m m up _ _ . ‘ NoNN mm: choONNN o w NNm Nmm.oNa. Nam.¢Nu m oCquN¢omommo m .Ig-NNNN-mus;‘oN\oo\wN:-::o-,-Jlsmaa!ltiwwm:lawmm.o~n:1.~mm.¢Nn ;.1 n ....... oooNN u:.-Nwflw.mM: LNNboONszI1xb m NNm NcN .NN- wahMan;ia:mnlrbOonx zNNNN- Nw>ou N<2c waxN :NNN are ‘ oz, oz Ncm¢ mm: NN= H83 9: E -mww: SE .533 9: .mw__mEo:m .538 new $=mE0cw .395 mcammm E gnaw? uofimwcofi .mm:mEocm fine of 8 22:33 me: E g: :3 E: new 55: 230% .mcoEIoE :55an -239; : -fioa Ewi .55 of E 2: E .thna .9; 83m 20:: .30qu {av 89 .:E L .LmE €230; 95 @25wa 30. 63:8; .sz 62:8; 2?; £2.50; «Zion LE... :53 @250; \£13 Ew:m -EU 5qu ITS wwSm>< -538 .: w .38 05 E W 73225:. Em: o .33 can L330 >> $955.5: Ucm of E 72.6 uEmwomZ 5:333 >3 .538 m5 mm:mEocm $272: .33 £3.53 .wZEwo: E 35me quE02 -0cm 950E 73:8 of 268220 bEmE :3; fire: 3:3: .9: -02 w>z E 36:8 ucm 2m :5 EMS 9: E :wi -Gom £35 -an £55 .5 3:2qu .35- £39. mcszBE {II 3:50; ............ wcszBZ I--- 3233 -Ewdwflm -Ewiwf -8 ii: .otconEU ........... Eoztoz ._ wEmc uzmcmmE $cmEo>oE mEu: Ucm >=>Ew Em: Em: quUBomc 0:0:me 8 .w.: . m5 5* 28:me 532w Uzm ncw 53me Em: 2w: 6 .9: Emmomxm an E: mEmUEmou 22:3 9:55) Enemoae 3:353 25:me 5320 meEwyoE $.09 25:82:20 v.83 mEON cozflotoU 53> cozflmtou 25688: 9: we ow< uEmEouoi 2: £5) E; 3:3. 56ch .muzao 9: 5:3 EOE—€53 9:30:33. mu:m_._2u§m;u E29380 8303 thEmE 9: v5 muzmtmbflmcw mEEtEmzdlé Ema; 345 COMBINED PROCESSING OF SPACE IMAGE WITH STRUCTURE STUDY .w:_ .352: -v_:_m 53m :5 0.33 4.83 JESS -o:m :zv .o>on~ -:8:_ E3 $558 -53 .w>_:mon_ II 02:5; 523wa 63:..me 52:5; :9: 03 Ron. mm 9:? 9: £85 30-. .30 :93: 2:5 >5> -550m .5 .528 F“ w5 :_ m>5 .mo:wEo:m o .69; 55 53me .w:._v_:_m u .:w_: 55: o5 ARE :3: .305“ -55: 9? 26:2 .555“ :5 m 63:..me :_ o>5~moZ I- 533me $23wa 635me 3 9:3 3: :85 3:9: .5 .:E5 25-. .50 5mg: .29.. -53 .> m .52 55 m :o w:_v_:._m :33: E; .mw__mEo:m v.55 Em: Scum :5 :9: .583 652m 289: o>z w5 :o -023 .Ew: $me 638me II m>=mmoz 52:»va I: 3::wa imE 62:5; mm 583 9: 45a v.83 mEm: >_Ew__m $308390 ZOE «LP -552 .2 60:35:30: 2: 9 _ .855 522.28 m .538 05 m5 9 .23 Ucm : E 35.50; Ezfimansm 3:235": a .538 583 ucm 82.55:: .538 $5 .23 8:5: : m5 9 .533 55: w5 0! 30¢ .EE: 3? :_ 2582 -o:m S538 w .52 £33 533 53“ -55: :_ 8:5: .505th -:B:_ 5 -90 U:m w5 :_ U -55: 95 283 -mww: mwm m5 :_ -o:m o>= .5; >232 5:2 3:05 55: m5 5% .55: :_ 03:..me 63:5: -53.. wagon. :wE 63:5; I- @35me -mwm: 553 .E www5>< 45:53 :_ 239:: 05 :_ Em: ----- ._: 346 FIRST ANNUAL PECORA which corresponds tectonically to the zones of the Umerovsky graben and the Kamyshin projection. It is here where the maximum erosion of ancient deposits has taken place during Quaternary time, reaching 1,000 m in the north, in the area of the Linevsk Eleva— vation. The area lying on the left side of the Volga has, on the contrary, been a zone of subsidence and sedi- ment accumulation during a long time. GRAVITY FIELD The character of Bouguer anomalies is determined chiefly by the internal structure and chemical composi- tion of basement and by the basement surface to- pography, i.e., by factors operating at depth, as well as by structural and lithologic inhomogeneities inside the sedimentary cover, which give rise to density dif- ferentiation of rocks. So far as the gravity field is con- cerned, the territory is divided into two parts: north- western area characterized by a weakly differentiated gravity field and southeastern one—an area of very low, exceedingly separated gravity anomalies. The boundary between these two areas is very distinct and is commonly drawn along a gravitational bench coin- ciding roughly with the right bank of the Volga, where in a band 10 km wide the gravity intensity decreases by a factor of 3 or 4 in the southeastern direction. This boundary corresponds to the flange of the Cis-Caspian depression. According to some concepts, the bound- ary corresponds to the zone of a major fracture and abrupt subsidence of the basement, whereas by other concepts it reflects a relatively gentle sinking of base- ment and the appearance of thick low-density salt- bearing strata in the platform mantle section, which were the cause of the gravity bench formation. In the right—bank territory there are three areas. The northern area bounded by the line running through the Melovatsk-Ilovlensk oil fields and by the Volga bank is characterized by variously trending anomalies of low amplitudes. Here the character of gravity anomalies is the most mosaic. The central area, which is bounded by the lines Melovatsk—Ilovlensk, Severo—Dorozhensk- Yuzhno-Umetovsk and by the Volga bank, displays predominantly northwestern-trending elevated anoma- lies. The most intensive are the isometric anomalies observed south of the Ilovlensk oil—and-gas field and southwest of the Melovatsk gas field. Whereas in the north of the central area northwestern-trending anomalies prevail, in the south submeridional anoma- lies are encountered. According to the results of gravity field recalculation into the upper semi—infinite space (Gilod et al., 1970), a single elevated—density massive was isolated here, in the zone of the south- MEMORIAL SYMPOSIUM eastern slope of the Voronezh anteclise, which was given by these authors the name of the Lower Volga core. The southern area has high-amplitude west-north- west-trending elliptical anomalies extended in a chain. Within the Cis-Caspian depression, east of the gravi- ty bench, there are small-in—area negative anomalies of high amplitude, which are elongated subparallel to the bench trend. MAGNETIC FIELD The character of magnetic anomalies is mainly due to the chemical composition and structural features of the folded basement, as well as to the behaviour of its surface. The differences in shape and amplitude be- tween magnetic anomalies found within the neighbour— ing blocks may be attributed to the different types of rocks, of which these blocks are composed, and to their differing metamorphism. The latter factor may lead, for example, to the secondary enrichment of basement rocks in magnetic minerals, which, in turn, is one of the causes of elevated magnetic anomalies. CORRELATION BETWEEN THE GRAVITY AND MAGNETIC FIELDS A simple visual comparison between the gravity field and the magnetic field does not allow one to es— tablish any relationship between the parameters under consideration for each locality. Accordingly, a scheme showing the correlation between the gravity field and the magnetic field has been constructed (fig. 7) using the computer program “Ankor” and employing the procedure suggested by M. S. Zhdanov and V. I. Shraibman (1973) for the Turan platform. A calculation of correlation between these two geo- physical fields reflecting the integrated characteristic of the density and magnetic properties of the upper layers of the Earth’s crust, which was carried out by the procedure suggested, has made it possible to es- tablish the direction of change in these fields in vari- ous localities. On the scheme, positive and negative areas have a largely mosaic character. But whereas the areas of positive correlation are mainly isometric, those of negative correlation are, as a rule, elongated. Since the character of both the gravity field and the magnetic field in the area under consideration is chief- ly determined by the composition and topography of the basement, which, in turn, was formed under the influence of tectonic processes, one may suppose that the same factors are reflected on the presented scheme, too. COMBINED PROCESSING OF SPACE IMAGE WITH STRUCTURE STUDY 51°11’18"N FIGURE 7.—Scherne of isolines showing the cor- 44°06’08”E 347 46°37’25”E 50°48’07”N I .I'IIII" lIII' lIII IIIIIIIIIIII _ if? 49°17’47"N relation coefficients for the gravity and mag- 49°40'05"” netic fields. A simultaneous analysis of the correlation scheme presented in fig. 7 and of the structural map showing the basement surface has revealed that areas of both positive and negative correlation correspond to posi— tive and negative features of the basement structure. A weak resemblance can be observed between .the patterns of depth isolines and isolines of correlation coefficients. This suggests that the constructed scheme reflects but weakly the basement surface structure. At the same time, when matching the scheme of cor- relation coefficients with the petrographical scheme of Pre- -Riphean basement due to Bogdanova, Lapin- skaya, and Podoba (1971) one can note that the locali- ties of both individual intrusive bodies and fields of occurrence of definite rock complexes coincide. Thus, in particular, over intrusions of gabbonorites in the northwestern part of the photograph under examina- tion one can observe an increase in positive correla— tion up to 0.79 which suggests rather a close relation- ship between gravity and magnetic anomalies in the localities indicated Another extensive area of positive correlation found in the southwestern part of the pho- tograph corresponds to a granulitic block of the base- ment recognized from drilling data (Bogdanova et al, 1973). In the northeastern part of the photographic sheet isolines of positive correlation (correlation co- efficient here is as high as 0.89 or even 0.97) outline a body of granitoids of the granulite facies of meta- norphism (plagiogranites containing garnet and cordi- arite). In the central part, a submeridional zone of 1egative correlation can be observed. 43°25’.02”E 45°51 '53"E SCALE 1:3,369,000 BASEMENT STRUCTURE In regard to structure, the area under study is made up of major elements differing one from another (fig. 8). In the west of the area, there is a periclinal termina- tion of the Voronezh anteclise, where the basement occurs at a depth of 2.5 km; on the southeast, the Cis- Caspian depression slopes down, where the basement has sunk down to 10 or even 11 km. In the boundary zone between these major structural elements, there are several projections and troughs in the basement surface, which are complicated by local structures. Generally, these structures fall within the Saratov off- set of the Pachelmsk aulacogen stretching, according to Nevolin (1971), towards Volgograd. On the north- west, a ramification of the other elements borders on the Cis-Caspian depression. The western and north- western boundaries of the depression are drawn in conformity with a drastic decrease in the amplitudes of gravity anomalies. The basement has been stripped by deep boreholes only in the relatively elevated western part of the terri- tory (fig. 8). Most of core samples have been studied in detail at the department of petrography of the Moscow Institute of Petrochemical and Gas Industry, where a group headed by V P. Florensky and T A. Lapinskaya has systematically processed during many years practically all the core samples taken from the basement of the Volga- Urals region (Florensky, Lapin- skaya, Knyazev, and others, 1956) The above-men- tioned boreholes have stripped a variety of deeply metamorphosed rocks (Lapinskaya, Bogdanova, and 348 44°06’08"E 51°11’18”N 49°4Q‘06”N 43°25'02”E SCALE 1 :3,369,000 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 46°37’25"E 50°48’07”N 49°17’47"N 45°51 ’53”E FIGURE 8.—Basement surface topography (after N. N. Nevolin, 1974). Podoba, 1971). In the southwest of the area, there have been stripped Archean high—alumina gneisses of the granulite facies pierced by granitoids. In the (ex— treme southwest, these rocks are overlain by Protero- zoic phyllites, shales, and metasandstones (Bogda- nova, Podoba, and Serova, 1973). To the north of this area, there has been stripped a field of rocks of the amphibolite facies (biotite-amphibolic gneisses, micro- clinic granites, etc.) containing relics of rocks of the granulite facies. This testifies to a deep diaphthoritic reworking of a part of the granulitic block in the pres- ence of amphibolitic facies. A large massive (massif, ed.) of granitoids of Archean age has been stripped in the north of the area. In the central and eastern parts, where the basement occurs at a great depth and has not been stripped by boreholes, the composition and structure of the basement can be judged only indi- rectly from the geophysical materials available. It has been assumed that in the central part of the area there occur gneisses of decreased density which were re- worked by the Svecofennian-Karellian folding (Gafa- rov, 1973; Nevolin, 1971). The structure of the Pre—Cambrian basement of southeastern area has only been hypothesized tenta- tively. Specifically, east of the zone of decreased- density gneisses R. A. Gafarov (1973) recognizes a stable block made up of rocks that have not experi- enced the secondary reworking. COMPARISON BETWEEN THE SCHEME OF PHOTOANOMALIES AND A COMPLEX OF GEOLOGIC AND GEOPHYSICAL DATA A relief of the studied area is very gentle; being shaded by contrasting spots of vegetation, it is re- flected very weakly in the distribution of optical densities. The surface exposures of sedimentary rocks shown in the geologic map are also weakly reflected in the optical density distribution, which may be associated with the fact that the bedrock composition is monoto- nous, while the bedrocks themselves are covered by soil (mainly plowed). Neotectonic movements shown in isolines of the erosion section (fig. 6) perhaps are best reflected in the optical density distribution (fig. 4). There emerges a single submeridional zone corresponding to a band of the greatest neotectonic uplifts and erosion. The areas of accumulation situated on the left bank of the Volga are lighter in tone; it would appear that the overlying sediments should have concealed the dis- tinctions in this territory; nevertheless, it is differenti- ated photometrically, thus allowing its separation. So far as the pattern of anomalies is concerned, the gravity field has in general a number of features simi- lar to those of the optical density distribution. This similarity, however, manifests itself differently in dif- ferent parts of the area. In order to compare them more objectively, a scheme of two-dimensional corre- lation of the gravity field with the optical density has COMBINED PROCESSING OF SPACE IMAGE WITH STRUCTURE STUDY been constructed (fig. 9) along the direction of the photometric profiles. The values of the gravity field and optical density along the profiles were taken from a sliding interval 45 km long using a step of 30 km. The results of the calculations along the profiles have been extrapolated. The curve showing the correlation of gravity with optical density is not high, but it is not smooth and usually oscillates by :07 or even :09 in some parts. From the sign of correlation, however, it turns out possible to single out only several individual zones. Across the central part of the photograph, there ex- tends a wide submeridional band of positive correla- tion (its width reaches 70 km), which ramifies in the south of the photographed area into two branches, one of them being sublatitudinal and the other submeridi- onal. In some parts of the above band, within the Kamyshin protrusion and northeast of the Kudinovsk elevation, the correlation increases up to 0.98. A sec- ond area of positive correlation occurs in the north- west and roughly corresponds to the eastern slope of the Voronezh anteclise. Zones of negative correlation found in the east are typical of the left-bank area, where in some parts the correlation coefficient reaches —0.95. The aforesaid allows a distinct division of the region discussed into three submeridional zones and, further, into a number of subzones, which is based on the correlation between gravity and optical density. The magnetic field also exhibits a varying relation- ship with the optical density, which enables one to 44°06’08"E 51°11’18"N SCALE 1:3,369,000 49°40’06”N 43°25’02”E 349 isolate in the territory under consideration several zones, which partly coincide with those isolated previously. When comparing the map of magnetic anomalies with that of photographic anomalies, one cannot help seeing that the two are considerably related. In this case, as in comparing the scheme of optical density with the map of gravity anomalies, one may isolate various zones. The central part of the photograph is taken up by a submeridional zone of negative correla— tion coinciding with the zone of positive correlation between the optical density and the gravity field. The other parts of the area display, however, a more com- plicated and differentiated pattern. In particular, the area of positive correlation coincides with the slopes of the Voronezh anticlise, the left-bank region of the Volga is also characterized by the positive correlation between the optical density and the magnetic field. The scheme showing the correlation between the magnetic and gravity fields (fig. 7) makes it possible to single out a submeridional zone bounding the darkest part in the southwest of the area, as does the scheme for optical densities (fig. 4). The principal trends of correlation anomalies coincide with those of optical density anomalies. The area of positive correlation be- tween geophysical fields is relatively dark in the west (D =08 to 1.0), while in the east, the areas of negative correlation between these fields are light (D=0.4). Thus, areas of different correlation of geophysical fields (fig. 7), as well as those between geophysical fields and optical density (fig. 9), yield additional geo- 46°37’25”E 50°48'07”N 49°17’47”N 45°51 ’53"E FIGURE 9.—Scheme of isolines showing the coefficient of correlation of the gravity field with the optical density. 350 logical information, which may be used in territory regioning. Finally, comparison of the scheme of photographic anomalies with a complex of original geologic and geophysical materials enables one to consider the ter- ritory from the standpoint of the character of the relations established. The secondary schemes of cor- relations we have constructed are significantly differ- ent from the original ones. The reality of the new recognized fields is evidenced by the similarity in their patterns as observed in various, almost inde- pendent schemes. When analyzing the results obtained, it should be borne in mind that the photometric char- acteristics of space images of the platform territory reflect not so much surface geological objects, as regional tectonic elements of structure of both the sedimentary strata and the basement. LINEAMENT IDENTIFICATION When separating various fields, we have not dis- cussed the character of their boundaries. However, it is just the geological boundaries between areas of con- trast in some or other properties that are deciphered with the most certainty on space images. In small-scale observations of the Earth’s surface, the existence of the so—called lineaments has been established, i.e., lines of structural significance recog- nized from the gradients of photographic tone. These lines are only partly associated with the geologic and geomorphological structures visible on the surface in field observations. This applies primarily to fractures, whose deciphering on space images has been devel- oped especially well for different tectonic regions and images of the different degrees of generalization (Skaryatin, 1973; Trifonov et al., 1973; Makarov et al., 1974). In mountainous and folded areas with abruptly changing topography or with distinct boundaries be- tween the occurrences of different rocks, the recogni- tion of fractures by conventional geological techniques and of lineaments by deciphering space images re- duces, in effect, to the identification of one and the same structures and structural trends. In the platform regions, however, the methods of lineament recogni- tion are quite different, as are the recognized struc- tures themselves. Indeed, on the platforms, lineaments are expressed as slight (of several metres) changes in topography elevations, as bends in the river and ravine network, or as merely insignificant changes in the composition of outcropping rocks. Such geologic boundaries, even when confined to fractures, have very often stratigraphic rather than tectonical signifi- cance. Thus, the search for the geologic and geo— FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM morphological criteria for the identification of linea- ments in the platform regions is a highly contradictory task. At the scale used in direct geological observa- tions in field work, most of lineaments are unrecog- nizable at all. Thanks to space images of the platform regions, it has turned out possible to recognize es- pecially many lineaments heretofore unknown. It appears that a considerable number of lineaments are not accompanied by anywhere near essential one- directional displacements of blocks, which would im- mediately be reflected on structural maps based on drilling data. One gets the impression that zones of extremely low-amplitude, but, probably, systematically different in direction, displacements, reducing merely to an extended band of increased jointing, are con— fined just to such lineaments. The latter probably inter- sect the Usturt Plateau leading to formation of straight parts of its chinks (border scarps) (Florensky, 1973; Bogorodsky et al., 1974). On small-scale television images transmitted from satellites such as “Meteor” and “Meteor—Priroda” (fig. 1) a submeridional lineament can be seen, which strikes along the general direction of the Volga bed and corresponds geologically to the northwestern flange of the Cis-Caspian depression. There appears also a secondary (with respect to the meridional one) northeastern trend of lineaments. On ERTS—l images (fig. 2) lineaments appear quite differently. The meridional lineament is scarcely trace- able; only straight parts of it on the Volga’s right-bank are present on the images. One succeeds, however, in recognizing a number of other, less significant, linea- ment systems (fig. 10). The best pronounced is the northeast trending lineament traversing the image. In the western and central parts of the image, a bank of clouds is confined to this lineament; the boundary which is known to mark a change in weather in this district coinciding with northeast trending structures also corresponds to this trend. The remaining linea- ments are expressed on the space image as very thin lines of change in photographic tone; only some of them can be identified with straight—line valleys of rivers and streams. Not always can such lineaments coincide with geological boundaries of a locality. After independent identification of the lineaments on both Space images followed by independent identi- fication of them by different geologists (V. P. Buk- hartsev, A. S. Petrenko, P. V. Florensky, and others), an attempt has been made to correlate their position with the gradients of geophysical fields, as well as with the basement surface structure and composition. COMBINED PROCESSING OF SPACE IMAGE WITH STRUCTURE STUDY 44°06’08"E 51°11’18"N 49°40’06”N 351 46°37 ’25”E 50°48 ’ 07" N 43°25’02”E One can observe a definite relationship between the northwest—trending fractures and the gradients of the gravity and magnetic fields. Two lineaments coinciding in trend with the fractures form the southwestern slope of the Saratov ramification of the Pachelmsk trough. Other lineaments striking parallel to them sep- arate blocks of the basement composed of rocks of different compositions, which have already been de- scribed in the paper. Northwest-trending fractures are mainly characteristic of the right-bank region, whereas in the left-bank region northeast-trending fractures are especially prominent, only one of which can be traced from the right-bank area. The zone separating the areas of manifestation of lineaments trending pre- dominantly northwest or northeast corresponds geo- graphically to the Volga bed. Here fractures form something like branches of a fir tree whose trunk is the valley of the Volga, the acute angle of their inter- section opening toward the south. Thus, the lineaments that have been identified on the territory under consideration very weakly control its geomorphological structure. Not all of them affect the structure of the sedimentary cover. Almost all of the most prominent lineaments, however, appear to control the degree of metamorphic reworking of the basement, the structure of its surface and, in places, even the boundaries of variously aged pre-Riphean folding. One gets the impression that the inhomogenei- ties in the basement composition turn out to be just those weak spots along which there have developed the inherited zones of slight displacements taking SCALE 113,369,000 4 FIGURE 10.—Scheme of lineaments (of various degrees of expression on images). 49°17’47”N 5°51 '53"E place in the course of the entire platform stage, up to the recent time. When these displacements are in— herited, singly directed and significant, only then are local structures confined to these lineaments. Where, however, the displacements along fractures have dif- ferent directions, a zone of higher jointing is formed, which controls the character of reservoir. From this standpoint, the identification of such “fractures with- out displacements” is advantageous, as it has allowed a more exact determination of the position of jointy zones possessing improved reservoir properties. Wells that have been drilled in such zones are characterized by high yields. IDENTIFICATION OF BLOCKS AND THEIR CEOLOCIC INTERPRETATION As a result of the integrated analysis of diverse data, it has turned out possible to recognize on the territory under investigation a number of blocks char- acterized by different values of optical density, differ- ent surfaces, different geologic and geophysical pa- rameters, and their different relationships (fig. 11, table 1). The structural elements identified in this way have different amplitudes of neotectonic movements, different characters of gravity and magnetic anoma- lies and varying optical densities of images. But they are recognized especially distinctly when the tech- nique of pairwise correlation of the indicated parame- ters is applied. The method of local correlation of geologic and geophysical, as well as photometric, pa- 352 M°W'WE 51°11’18'N I/fl .11., Ill ."lil W ii FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 46°37’25"E 50°48'07”N lI/h‘ III. I/// I W IIIIIIIII ”II” IIIIII / III III 97 II I” ‘I II} ”"4 I/ I” ll, In. I if“ . .- . I'll! - _- . ~ I. .III 'I II I ’I I), "I I . II V 'u I //// I” I II I//' II I/// // IIIIl/ll Il//////// , , , ym/IIIII/I” ll ', III III 9°17'h7"N 45°51 '53'E I334 (Ex—:15 49°40’06'N ' ’ ’ 43°25'02"; SCALE 123,369,000 I: 1 m 2 m 5 CC (2 5m 7 .5 FIGURE 11.—Zoning scheme for the crystalline basement of the Lower Volga region compiled from a complex of geologic and geophysical data and taking into account the results of deciphering and photometry of TV-space images. 1—Zone of Svecofennian-Karellian reworking of the basement; 2—areas of metamorphic amphibolite facies; 3—areas of metamorphic granulite facies; 4—granitoids; S—basic rocks; 6—boreholes that stripped the basement; 7—structure contours of the basement as shown in fig. 8; and S—Iineaments as shown in fig. 4. rameters has proved to be the most effective in terri- tory regioning. Generally speaking, the whole area under study is divided into three submeridional bands: Western, corresponding to the periclinal closure of the Voronezh anteclise, central, and eastern, corres- ponding to the western flange of the Cis-Caspian syne- clise. These zones, in turn, are divided, according to the internal structure of the basement, into blocks, which, in places, are outlined by fractures. The combined geologic—geophysical and photo- metric characteristics of these blocks are given in the table and, therefore, are not discussed in the text. In what follows, some considerations concerning the internal, block-type structure of the basement are outlined, which are based on a generalization of the whole material investigated. The Western zone, which is the periclinal closure of the Voronezh anteclise, is distinguished by the homo- geneous character of the gravity and magnetic field. Its eastern boundary coincides with a system of meridi- onal and submeridional fractures, identified as linea- ments on space images and coinciding with gradients and local anomalies of the geophysical fields. With re- spect to the orientation of gravity and magnetic anom- alies, the zone is divided into two blocks by a sub- Iatitudinal fracture, to which steep gradients in the gravity and magnetic fields are confined. The Southern block is characterized by sublati- tudinal geophysical anomalies and by an increased optical density. Within the territory of the block, with- in the Korobkovskoye elevation, wells have stripped deeply metamorphozed rocks of the granulite facies (crystalline shales with granite) weakly changed by diaphthoresis. Here, local anomalies correspond to granitoid bodies. In the southwest of the area (within the Kudinovsk structure and in the Panin wells 74, 152, 154, etc.), younger (Proterozoic) phyllite shales spreading farther to the west have been stripped (Bog- danova et al., 1974). The Northernxblock is remarkable for submeridion— ally trending geophysical anomalies and for its lighter colored surface. Within the territory of the block (within the Zhirnovskaya, Linevskaya, Severo-Doroz- hanskaya, and Mishinskaya structures), diaphthorites of the granulite facies have been stripped; for ex— ample, bi-amphibolitic plagiogneisses, granodiorites, and amphibolites. COMBINED PROCESSING OF SPACE IMAGE WITH STRUCTURE STUDY The Central zone is characterized by rather low gravity and magnetic anomalies, as well as by their negative correlation. For this zone, the tendency to uplift and the positive correlation between gravity anomalies and optical density are characteristic. Along its strike, the zone is especially dissected by fractures and is noted for the lowered values of gravity anoma- lies and for the presence of an extensive negative mag- netic anomaly. These features are not at variance with the suppositions that here there occur rocks reworked last by the Svecofennian-Karellian folding (Gafarov, 1973), which resulted in lowered densities of these rocks and in their intrusive injections. If the supposi- tion is correct, the identified zone is a submeridional branch of the Pachelmsk trough, bordering on the southeastern slope of the Voronezh anteclise. As a result of a comprehensive comparison of the geologic and geophysical data available with those of visual deciphering, and as a result of a formalized processing of space images of the area under con- sideration, it has turned out possible to establish that the correlative analysis enables one to make the most complete use of the information one possesses to study the deep-seated structure of parts of the ancient plat- form. Further, the conventional techniques for proc- essing and generalizing the geologic and geophysical information have made it possible to outline, by using the results of direct study of core samples of the basement rocks, some features of the tectonic struc- ture of the basement of the Lower Volga region and to add new details to the existing schemes. Thus, it has been possible to outline one of the sides of the rela- tionship between the character of space images and the features of the internal structure of the region under study realized through neotectonic movements, which have evidently been caused by a compensated uplift of the decreased-density zones in the basement. The presented complex of data on the deep-seated structure of a part of the Lower Volga region permits one to compare the information capacity of a space television image with that of other sources in regard to understanding of the deep-seated structure of the region. First, in recognizing lineaments, fractures, and dis- locations with a break in continuity and in specifying their positions space images have no equal sources of information. It is likely that the introduction of space images into scientific use signifies an essentially new stage in fracture study. This is especially true of the platform regions, in which displacements along the fractures are insignificant and their traces are con- cealed by the sedimentary cover. 353 Second, when taken by itself, the analysis of varia- tions in optical density has shed but little light on the regioning of the studied territory, but when integrated with the analysis of geophysical data, its informative capacity has radically increased. It is likely that, at least for the platform regions, the informativity of space images for. our understanding of the deep-seated structure of the area under consideration and its re- gioning approaches the informativity of regional geo- physical investigations. And only in a combination with the results of regional geophysical studies can the space image information be interpreted. This work has been done at the laboratory for De- ciphering Aerospace Images (Institute of Geology, the USSR Academy of Sciences) under the general guid- ance of V. G. Trifonov, to whom the authors express their deep gratitude. REFERENCES Bakirov, A. A., Present-day concepts of the geologic structure of the crystalline basement of the Rus- sian platform. Trudy Akad. neftyanoy promy- shlennosti, issue 1, 1957. Bogdanova, S. V., Lapinskaya, T. A., and Podoba, N. V., Petrographical characteristics of basement. In: “The study of the geologic structure of the East European platform by geophysical methods,” Moscow, Nedra Publishers, 1971, pp. 69-84, illus. Bogdanova, S. V., Poboda, N. V., and Serova, A. D., On the deep-seated structure and composition of the basement of eastern Russian platform. Izvesti- ya Akad. Nauk SSSR, ser. geol, 1975, no. 12, pp. 19—31, illus. Belyaevsky, N. A., The earth‘s crust within the USSR territory. Moscow, Nedra Publishers, 1974, p. 283, illus. Borisov, A. A., The deep—seated structure of the USSR territory according to geophysical data. Moscow, Nedra Publishers, 1967. Bukhartsev, V. P., Geological prerequisites for the probabilistic forecast of oil- and gas-bearing zones from local ruggedness of relief of platform territories. In: “Mathematics, Computers, and Automatic Control Systems in petroleum geolo- gy." Moscow, Institute of Fuel Mineral Geology! and Prospecting, 1973, p. 130, illus. Bush, E. A., Kuznetsov, V. G., Ryabchenko, F. M., Sokolov, V. I_.., and Shornikov, B. Ya, Geologic structure and objects of search for gas in the flange zone of the cis-Caspian depression in 354 FIRST ANNUAL PECORA southern Volga region. “Geol. nefti i gaza,” no. 5, 1968. Gafarov, R. A, The basement structure of the East European platform and some problems of com- parative tectonics of ancient platforms. In: “Tec- tonics of basement of ancient platforms.” Mos- cow, Nauka Publishers, 1973, pp. 82-94, illus. Gilod, D. A., Vostokov, E. N., and Dabizha, A. 1., On some features of the basement structure of the south-eastem slope of the Voronezh anteclise (ac- cording to geophysical data). Vestnik MSU, ser. geol. no. 5, 1970, p. 115—120, illus. Zhdanov, M. 8., and Shraibman, V. I., A correlation method for separating geophysical anomalies. Moscow, Nedra Publishers, 1973, p, 128, illus. Kuchko, A. 8., Aerophotography. Moscow, Nedra Pub— lishers, 1974. Makarov, V. I., Skobelev, S. F., Trifonov, V. G., Flor- ensky, P. V., and Shchukin, Yu. K, Deep crustal structure as seen on space images. In: ”The study of natural environment from space,” issue 2. Geology and geomorphology. Moscow All-Union Institute of Scientific and Technical Information, 1974, pp. 9—42, illus. Mikhailov, V. Ya, Photography and aerophotography. Moscow, Geodezizdat, 1972. Nevolin, N. V., The main features of the geologic struc- ture of the basement of the East European plat- form. In: “The study of the geologic structure of MEMORIAL SYMPOSIUM the East European platform by geophysical meth- ods.” Moscow, Nedra Publishers, 1971, pp. 87— 91, illus. Skaryatin, N. D., On the study of rupture tectonics from a set of variously-scaled space images of the Earth’s surface (the multistep generalization technique) Geologiya i razvedka, 1973, no. 7, p. 34—60, illus. Trifonov, V. G., Byzova, S. D., Vedeshin, L. A, De- revyanko, O. 8., Ivanova, T. P., Kopp, M. L., Kur- din, N. N., Makarov, V. 1., Skobelev, S. F., and Florensky, P. V., Concerning the procedure of geologic deciphering of space images of the Earth. In: “The study of natural environment from space,” issue 2, Geology and geomorphology, Moscow, All-Union Institute of Scientific and Technical Information, 1973, pp. 8—78, illus. Florensky, P. V., Lapinskaya, T. A, and Knyazev, V. 8., Some results of the petrographic study of the crystalline basement of the Volga-Urals petro- liferous region. Trudy Moskovskogo Instituta Neftekhimicheskoy i Gazovoy Promyshlennosti (Moscow Institute of Petrochemical and Gas In- dustry) Gostoptekhizdat, issue 26, 1969. Florensky, P. V., Deciphering of the deep structure of the Turan platform from space images in connec- tion with the prospecting for oil and gas fields. Izvestiya Vysshikh Uchebnykh Zavedenii, Geolo- giya i razvedka, 1973, no. 7, pp. 112-117, illus. PROCEEDINGS OF THE FIRST ANNUAL WILLIAM T. PECORA MEMORIAL SYMPOSIUM, OCTOBER 1975, SIOUX FALLS, SOUTH DAKOTA Geological Studies by Space Means in the U.S.S.R.‘ By V. G. Trifonov, V. I. Makarov, V. M. Panin, S. F. Scobelev, P. V. F lorensky, and B. P. Shorin-Konstantinov, Geological Institute of the Academy of Sciences of the U.S.S.R. Moscow, U.S.S.R. Wide spectrum of the geological tasks requires the using of various methods of research: from studies of the atomic, molecular, and crystal structure of min- erals and rocks to studies of the large structural forms and distribution of them in the planet. In the recent stage of the development of geology the solution of its many problems (first of all tectonic problems) is possi- ble only by studies and correlations of the geological objects and events in the planetary scale and taking into consideration of peculiarities of the Earth as a space body. Such planetological aspect of the geo- logical studies is provided (beside the traditional means and methods) with the new source of informa- tion on principle, that is the Earth‘s pictures from the satellites and orbital stations. These images are in visible and infrared parts of spectrum and are received by photographic, televisional and scannering means. Geological using of the space visible information that started about 10 years ago, now has achieved large successes in the USSR, USA and number of the other countries. We don’t want here to review the common progress in this branch of studies, but limit ourselves by observation of some works carried out in the Geological Institute of the Academy of Sciences of the USSR. First of all the space images started to be used for compiling and correction of geological, structural, and tectonic maps, for mapping and correlation of the large structural forms and zones. So, the map of the major faults and zones of deformations of southeast— ern Kazakhstan, Middle and Central Asia was com- piled with help of TV images with resolution 1—2 km 1Printed verbatim as received from the authors, except for minor changes in format and spelling. from meteorological satellites. The geological map of the central part of Tadjik depression (figs. 1 and 2) was compiled by deciphering of space photos from “Sojuz—9” with the higher resolution (100—200 m). The rather good quality of picture and integration of the brightnesses of the individual layers permitted to map objectively the lithological-stratigraphic units, folds and faults. The results of interpretation checked by the field observations. The comparison showed that de- ciphered map required essentially less of time and means than the earlier geological maps of the area in the same scale; but it didn‘t yield them in image of details, excelled them in exactness of image of faults and represented more objective generalization of the fold structure. The further development of the structural-geological trend of the space information using is connected with analysis of images in different spectral bands and its combinations. In the examples of images with resolu- tion about 100 m of Eastern Fergana and the north— western foreland of Pamir in Middle Asia we can see that in bands 0.5—0.7 mkm the Quaternary deposits and the more old rocks with rather high collector ability are deciphered better than all (figs. 3 and 4). In arid conditions of the Middle Asia the most evapora- tion occurs from these formations, because of it and peculiarities of vegetation they are seen as the more or less light places independently to the real color of the rocks. In near infrared part of spectrum the Prequaternary formations are deciphered better. The surface geo- logical structure is refiected with essential detailness in the images in band 0.8-1.1 mkm (mainly because of the reflection of various formations in the relief). In 355 356 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 1.—Large-sca|e space photo of Babadag-Karatau region of the central part of the Tadjik depression (spacecraft ”Sojuz—9,” cosmonauts A. G. Nikolaev and V. J. Sevastianov) band 0.7—0.8 mkm both the relief and the color of the rocks have the indicative role. The possibility of de- ciphering of details is less than in band 0.8—1.1. It permit to find the main structural elements, for ex- ample, the large tectonic lineaments (figs. 5 and 6) that are not seen in surface sometimes, but reflect the deep structure that we shall discuss below. Such linea- ments are deciphered better in images that are re- ceived by combination of various bands, for example by “subtraction” of bands from each other. It is especially productive to use the space images for studies of the geomorphology, the Neogene- Quaternary deposits and the modern structure of the Earth’s surface. In the orogenic areas there can be deciphered the genetic types and sometimes the age of relief; the units of the Quaternary cover (1, figs. 8, 9); and the modern folds and faults (figs. 7 and 8). It is possible to see not only large structural forms, re— flected by deformations of the modern deposits and the erosional forms, but very gently sloping uplifts GEOLOGICAL STUDIES BY SPACE MEANS IN THE U.S.S.R. floOO°O°'* . .. .._.-. . . . g ‘.\ ’ _@«_._'\\:;~"" " . \ , {'T‘; ‘_ \ . . t ... v “.1. ‘-»‘ . ‘. .. . '.". . ..:\ [23* [2:332 El‘fl3 m Lists LL46 2'31 m: is} -m -' ":32 23‘ :3 :‘1 :1; .. FIGURE 2.—Structural-geologic map of the central part of the Tadjik depression compiled using space photos. 1—Holocene and Upper-Pleistocene alluvium and proluvium; Z—Mid-Pleistocene loesses and Ioess-like deposits; 3—Lower- Pleistocene conglomerates (dashed lines are the boundaries of the geomorphological levels); 4—Mid-Miocene-Pliocene conglomerates and sandstones; S—Oligocene-Lower-Miocene sandstones, alevzolites, and conglomerates; 6—Eocene-Oligo- cene deposits mainly clays; 7—Campan-Bukhara clays, gypsum, and limestones (k-p); 8—Alb-Santon clays with limestones, sandstones, and gypsum (k); 9—Mesozoic and Cenozoic de- and limestones (k-p); 8—Alb-Santon clays with limestones by geological methods; 12—found by geomorphological meth- posits, undivided; 10—Paleozoic deposits. Faults: 11—found by geological methods; 12—found by geomorphological meth- ods; 13—masked by the Quaternary deposits and supposed. Geological boundaries: 14—mapped; 15—masked by the Quaternary deposits and inner boundaries in units; 16— supposed; 17—axes of the anticlines; and 18—axes of the synclines. among intermountain basins. The same uplifts are found also in space pictures of platform, where they reflected the masked structures of the plate cover that are in prospect for search of oil and gas (figs. 9 and 10). It is possible to receive the essential information 357 from space pictures on a modern volcanism. In Kam— chatka and Caucasus, the faults, the eruptive frac- tures and various types of the volcanoes are deciphered. In Landsat—1 images of Iceland, it can be seen not only volcanoes, volcanic chains, and faults of rift strike, but some hidden transverse elements of structure, not found before by land observations. The information from small-scale space images is uniform and independent on approachness of regions. It gives us the possibility to map and correlate the major neotectonic forms and zones and finally to com- pile the pattern of modern structure of the continents as a whole. It will permit to give answer such signifi- cant questions of geotectonics as the planetary regu— larity in localities and strikes of various modern struc- tural elements, and, at last, sources and mechanism of structural formation. The good expressiveness of modern structural forms, deposits, and relief, the minimum of the latest complications, and the general distribution made them the best subject for the solution of the tasks of the detail planetary correlation of the tectonic processes. The role of the space information for the solution of such tasks is not so large, but some utilizations can be presented. At first, the mapping of structural forms and fault zones and the determination of age of de- formed deposits are the elements of correlation them- selves. At second, the studies in Middle Asia showed that forms of relief of different ages have the specific peculiarities and indicative features permitting to map them on the large distance and by this way to corre- late the modern structural elements of the distant regions. New possibilities of the geological application of the space pictures that differ them from airphotos on prin- ciple are due to peculiarities of the generalization of image with the decrease of scale of the survey. The correlation of the results of deciphering of the space pictures with the geological, geomorphological, and geophysical data show that structural elements that one can see on them, are not only large, but located in deep horizons of the Earth’s crust and its basement. Geological manifestations of such zones of deep de- formations are often masked by the numerous surface deformations, so they can't be found by field re- searches and on airphotos. They are reflected better in the modern structure, but here the fragmentation of manifestations prevents to compile the common pic- ture that is seen in space images only. Let us observe that problem on the data of Eastern Caucasus. Space pictures of it were received from spacecrafts “Sojuz—9” and “Meteor” series. The ele- ments of “Caucasus" strike predominate in the geo- logical and modern surface structure of this region. 358 FIRST ANNUAL 71°46’50“? 41°16’16” 39°45'12" , 71°15’15”E PECORA MEMORIAL SYMPOSIUM 73’53’20'E 40°57’34"N 39°27’01"N 73°19'03"E FIGURE 3.—Scanner image of Eastern Fergana, Landsat—1, June 24, 1973, band 5. They are reflected on space photos with resolution 300—400 m, but the transverse lineaments are seen here beside them (fig. 11). The last correspond to the surface faults only partly, but often represent the boundaries of the different facies and thicknesses of deposits and the areas with different tectonic styles. Increasing of the intensivity of surface deformation, the Late Quaternary displacements and concentration of the air volcanoes occur along some lineaments. They correspond to the zones of high gradients of dipping, boundaries and bends of structural forms of the surface of the crystalline basement (where it is the deepest) and of the Conrad surface. The analysis of instrumental and microseismic data shows that the zones of abnormal going out of the seismic waves (fig. 12) and zones of high density of epicenters (mainly of shallow earthquakes) strike along the lineaments. The lineaments seem to reflect the zones of defor- mations on the depths 5—20 km. On the “Meteor" pictures with resolution about 1 km (fig. 13) the more part of those lineaments are seen badly absent. The lineament of “Caucasus" strike are seen here the best of all. They correspond to the long zone of southern slope of Big Caucasus where the area of strong and relatively deep (10—60 km) earth- quakes is located (fig. 14) according to the data of I. V. Ananjin. Along this zone the thickness of the Earth’s crust is essentially changed. Two lineamental zones of NE-SW strike are deciphered less clearly. They cor- respond to the areas of epicenters on depths to 20—30 km. GEOLOGICAL STUDIES BY SPACE MEANS IN THE U.S.S.R. 7 1 °46' 50”E 359 73°53’20"E 40°57’34”N 41°16’16”N . .. , _ . i <, “1*“ “cars“? ”a ‘ww-r’f‘ ~, ‘ . . «flirt "TH ' «IVAN l ...g‘- LL¢W+~L. to '1; e. IY'Q war ‘ 39°27'01"N 39°45'12”N 71°15'15”E 73°19'03”E FIGURE 4.—Geological interpretation of the image figure 3. Points are the various Quaternary deposits; lines dipped to right are the Lower-Cretaceous deposits; horizontal lines are Upper-Cretaceous deposits; crossing lines are the Paleozoic for- mations. Thus, the structural elements of the various depths are deciphered on the space pictures. The more the degree of generalization of image is, the deeper ele- ments of structure are reflected on it. The differences found in the structures of various depths permit some- times to assume the structural disharmony between the different depths. At the same time the tectody- namic analysis of the geological structural of surface and its comparison with the results of determination of the earthquake focus mechanisms and the directions of displacements in the Late-Quaternary faults, striked along the lineaments, give the arguments to suppose that the various horizons of the Earth’s crust are under the same stresses and differ only by the types of the rock deformations created by them. The analogous picture takes place in Tjan-Shan. On the space images of this mountain country there are two types of the modem structural forms: (1) various bands that correspond to ridges-uplifts and intermoun- tain basins and (2) the lineaments of the different or- ders. On space photos with the relatively high resolu- tion the ridges and basins that correspond to the individual foundation folds (anticlines and synclines), are seen. With the generalization of image there are observed the anticline and syncline zones and then only the up—warping and down-warping systems, that are the major elements of the modern structure. Ac- cording to the geological, geomorphological, and geo- physical data, the individual ridges-anticlines and basins-synclines and their zones are the deformations of the folded basement surface and reach only the Conrad surface; while the anticline and syncline sys- tems are fixed by the changes of depth of Mohorovicic surface. The longitudinal lineament corresponding to the zone of sharp change of basement depth, is deciphered on the space photos with resolution 300—400 m of Southwestern Tjan-Shan. On the TV-pictures with resolution 0.8-1.5 km this lineament is absent, but the long lineament of NE-SW strike is seen. The last cor- 360 FIRST ANNUAL 71°46’50"E 41°16'16”N 39°45'12"N 71°15’1 E PECORA MEMORIAL SYMPOSIUM 73°53’20”E ' , 40°57’34"N 9°27'01"N 73°19‘OG"E FIGURE 5.—Scanner image of Eastern Fergana, Landsat—1, June 24, 1973, band 6. respond to the zone of the essential change of the Earth’s crust thickness, to the boundary of high seismic activity area, and to zones of abnormal going out of the seismic waves. On the relatively large-scale space photos of Tadjik depression (resolution 100—150 m; “Sojuz—9” and Landsat—1) the elements of surface geological struc- ture with longitudinal folds are deciphered well. When the degree of generalization of image increases, the transverse lineaments are seen better and better. They reflect the disposition of the geophysical anomalies and correspond to the deformations of the crystalline basement surface and the deeper horizons of the Earth’s crust. The data under consideration show the possibility to advance with help of space pictures the problem of the structural and genetic relations of the different horizons of the lithosphere. Space pictures can’t claim to be the only source of information about deep struc- ture, but results of their deciphering permit to inter- pret the geophysical data more objectively. Of course, we can see on visible and infrared space pictures only the different elements of landscape, but the elements reflecting the structure of various deep horizons, are “in focus” on pictures with various degrees of generalization. Space pictures permit also a new interpretation of the correlation between the lithosphere and the atmos- phere. On the images of Tjan-Shan, Turan plate, Mon- golia, and northeastern part of the Russian platform there is observed the coincidence between the ele- ments of the cloud cover and the large disturbed zones GEOLOGICAL STUDIES BY SPACE MEANS IN THE U.S.S.R. 361 71°46‘50”E 41°16’16"N . 39°45'12”N 71°15'15”E 73°53'20”E 40°57’34”N »..v~’ ...... ...... ‘ 39°27’01”N 73°19’03”E FIGURE 6.—Geological interpretation of the image figure 5. Points are the various Quaternary deposits; lines dipped to right are the Lower-Cretaceous deposits; horizontal lines are Upper-Cretaceous deposits: crossing lines are the Paleozoic for- mations. of the Earth’s crust, the last reflected on the surface only indirectly sometimes. The relief can't define such localities of the clouds in these cases. The process of geological deciphering of the space pictures can be conditionally divided into three stages. The task of the preliminary stage is to represent the materials of the space survey in the most informative and convenient forms. The photographic filtering is very useful, for example “addition" or “subtraction” of the pictures in various spectral bands. The specific kind of filtering and of the objective analysis of pic- tures is its photometry in various scale. In the stage of deciphering (in basis of brightness and geometry of the images of the geological objects) the essential significance belong to the comparison of the pictures of various spectral bands and their com- binations, of the different degrees of generalization, and of the different season. So, the snow on the win- ter pictures masks many geological details, but em- phasizes some large linear elements of the structure. In the stage of interpretation the comparison of the results of deciphering with the geological, geomorpho- logical, and geophysical data is carried out. The inter- pretation is found on the studies of the indicative signs of the geological objects of the region under re- searches and of the similar ones. For studies of the indicative signs is useful the observation of the objects with different degrees of generalization (on land, from helicopters, aircrafts, on air and space pictures) is use- ful. As usual, the field check of the results of decipher- ing and interpretation is necessary. Let us consider the operations of each stage in the example of the studies of space pictures of the Lower Volga region (southeastern part of Russian platform). We have used as a basis the scanner color picture of the area with resolution about 100 m (fig. 15). Almost all the surface is the combination of the agricultural lands with the different brightnesses, so it does not permit to decipher geological elements, the larger, but poorly differed in brightness. It has been necessary to decrease the influence of brightness differences of the agricultural lands and to develop the features of pic- ture that reflect the geological structure. For this the contrast has been developed by the photographic 362 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM \ 42 02¢ 7V FIGURE 7.—Midd|e-sca|e photo of the lssik-Kul region (spacecraft "Sojuz—9,” cosmonauts A. C. Nikolaev and V. |. Sevastianov). means. The received picture has been photometried along the frequent net of profiles. The using of the most open split of the photometer and big step of interpolation have permitted to exclude the small fluc- tuations of the density and to develop the differences connected with the geological structure. It is the gen- eralization and the specific photometric filtering of the primary picture. The received map of the conditional brightness (fig. 16) has been compared with the maps of the geology the neotectonics and the basement surface structure, the gravimetric and magnitometric data. The compari- son has been carried out by the calculation of the co- efficients of correlation, represented on the special maps. The geological map is not correlated with the photometric one, but the magnitometric and partly gravimetric data are in accordance with the last. It has been shown by the correlation of the data that distri- bution of the brightnesses reflect the structure of the basement of this part of the plate. The application of the materials of the deep bore holes and the geophysi- cal data permit to compile (with using of the other space images of the area) the hypothetic structural- geological scheme of the basement (fig. 17). The mod- ern movements of the Earth’s crust correlate partly with structure of basement. All the operations under consideration (beside the compiling of the final scheme) are formalized and can be carried out by the computer. Thus, the last can be widely used for the solution of some tasks of decipher- ing, the signals of TV and scanner systems not been necessary to present in visible form. So, the geological deciphering and interpretation of space images can be used for the solution of the main matters of recent geology. The practical significance of the geological application is in the possibility to cheapen and to increase the quality of the geological survey, to plan the expensive geophysical researches better and to interpret their results more objective. Space pictures can be used for correction of the maps of seismic danger; for finding of structural forms and zones that are in prospect to search the underground waters, oil, gas, and number of the ore deposits; at last, for solution of the tasks of the engineer geology, GEOLOGICAL STUDIES BY SPACE MEANS IN THE U.S.S.R. 363 42°24 , N [1. 5. 6 47 1 -- . .- ' 2. a. _ :5 @309. FIGURE 8.—Ceologica|-geomorphological interpretation of the photo figure 7. l—Limneous and alluvium-proluvium plains of Late Pleistocene and Holocene; 2—fragments of the high accumu- lated plains of Mid-Pleistocene; 3—marginal and inner uplifts in basin, mainly Neogene deposits; 4—S—systems of modern uplifts built by the Paleozoic formations (4—zone of the young relief, steep deeply eroded slopes of uplifts; 5—zone of the fragments of the old relief of intervalley areas); 6—zones of intermountain narrow fault- basins; 7—faults; 8—figures of structural forms (1, 3, 4, 7—13, anticline forms; 2, 5, 6, syncline forms); and 9— area of detail observations. that is the election of the routes of communications, gation of the natural environment by space the places for reservoirs, dams, et cetera. means. Geology and geomorphology. Printing house VINITI, Moscow, 1973 (in Russian). BIBLIOGRAPHY Makarov, V. I., Scobelev, S. P., Trifonov, V. (3., Flor- Trifonov, V. G., Bysova, S. L., Vedeshin, L. A., Derev- ensky, P. V., Shchukin, Yu. K, Plutonic structure janko, O. 8., Ivanova, T. P., Kopp, M. L., Kurdin, of the Earth's crust on space images. Proceedings N. N., Makarov, V. I., Scobelev, S. «P., Florensky, of the Ninth International Symposium on Remote P. V., Problems of techniques of geological in— Sensing of Environment, vol. I. pp. 369—438. 15— vestigation of space images of the Earth. lnvesti- 19 April 1974. Ann Arbor, Michigan. 364 FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM FIGURE 9.——Large-sca|e photo of the eastern coast of Caspian Sea, Manghyshlak Peninsula, and Ustjurt (spacecraft ”Sojuz—12,” cosmonauts V. G. Lasarev and O. G. Makarov). GEOLOGICAL STUDIES BY SPACE MEANS IN THE U.S.S.R. 365 1 23,250,000 1’ ‘i‘ 0’ I; M #35 a?" FIGURE 10.——Structura| map of Manghyshlak and Ustjurt compiled using photos from "Sojuz—8” and "Sojuz—12.” There are shown the anomalies on space photos that correspond to local structures being in prospect for search of oil and gas. 1—lsolines of the surface of the Permian-Triassic deposits; 2—fault found by various methods; 3—faults found on space photos; 4—bore holes reached the Pre-Jurassic deposits; 5—photo-anomalies corresponding to local structures (their figures are on the map); 6—scarps of Ustjurt plateau; 7—boundaries of space photos from “Sojuz—8.’.’ Scale 1:3,250,000. FIRST ANNUAL PECORA MEMORIAL SYMPOSIUM 366 .Aonmw wci \>o:m:mm>om ._ .> Ucm 529:2 .0 .< BancoEmOQ :dINEOm: :Suwumqmv 3:230. «Ev. $26.. Ucm mammuzmu 58mm“ *0 90:3 vumam w_mUm.w_vE< IV 010(0H)_~, :4 (er—y) Si(.,_,,,A1,, where x is equal to (Al, Fe“) that substitutes di- rectly in octahedral position, and y is equal to A1 that is in coupled substitution in octahedral and tetrahedral position (Chidester, 1962, p. 46). Simi- larly, the formula for antigorite becomes 22 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVID'ERE MOUNTAIN AREA, VERMONT LL) c: in c”. N m EXPLANATION Shaded where exposed, no shading where covered Serpentinite and dunite Schistose altered amphibolite* - Massive altered amphibolite Coarse amphibolite* Contact, showing dip—Solid where exposed, dashed where inferred ,L Strike and dip of inclined schistosity L_I_L_|_LL Edge of quarry 99,800N Limit of exposure *Lines show trend of bedding or schistosity N [l] 1P 2P 3'0 4P 5P FEET l l l Geology by A. H. Chidester and 0 5 10 METERS J. C. Ratté. August 21, 1951 (Mg, F6”) FIGURE 4.—Geologic map of the contact between ultramafic rock and amphibolite, Eden quarry, Eden, Vt. V1 In practice, the formula of lizardite suffices as a ‘ 90 general formula for the serpentine group, and the = 9535‘?) Roman numerals designating the coordination posi- 1v l tion are dropped. Furthermore, it is not generally Si(4_y)A1y OIOKOH) mg. possible to determine precisely the role of the differ- 4 ‘ “ ent types of substitution on the basis of present (A1, Fe”) (my) 33: 5.65———y) 2 ULTRAMAFIC AND ASSOCIATED ROCKS 23 chemical analysis and sampling techniques, and it is sufficient to indicate simply that the trivalent ions in tetrahedral position are not necessarily equal to those in octahedral position. The simplified formula thus becomes [(Mg, Fe+2) (6—1‘) (A1; Fe+3)y] [Si(4—z)Alz] 010(0H)81 Where, theoretically, at is equal to 3/2 times the (Al, Fe+3) in direct substitution in octahedral position plus the Al in octahedral position that is in coupled substitution with Al in tetrahedral position; 11 is equal to the Al in octahedral position both in direct substitution and in coupled subsitution with Al in tetrahedral position; and z is equal to Al in tetra- hedral position in coupled substitution with Al in octahedral position. The precise formulas in the pre- ceding paragraph should make these relations clear. (See note.) NOTE—In addition to the compositional variations dis- cussed, it seems likely that the entrance of Al into six- and four-coordination in antigorite might cause structural modi- fications that would result in changes in the structural formula of antigorite: The replacement of Mg and Si by A1 would lead to better fit between the “brucite” layer and the Si205-layer, thus decreasing the curvature of the unit cell, and reducing the likelihood of “coincidence deficiency” (Kunze, 1956, p. 82) in the unit cell. In such a case, the structural formula of antigorite would change toward that of lizardite. Perhaps such a change would take place sud- denly, at a critical point with respect to the content of Al, in any particular unit cell. For macroscopic samples of antigorite, however, the compositions might be expected to vary statistically through a considerable part of the range between that of antigorite and that of lizardite. When the Al content is great enough to cause a good fit between the “brucite” layers and the Sigos layers, thereby reducing the curvature of the unit cell sufficiently to allow accommod- tion of an additional 6(0H) and 3 Mg per unit cell, probably the lizardite structure, rather than the antigorite structure, would be stable. The lizardite structure differs from the postulated structure of antigorite that contains a high con- tent of Al only in that the polar layers are not inverted at the midpoint of each supercell (81/2 unit cells of lizardite). Unfortunately, no general distinction based on optical properties can be made between varieties of serpentine that have been distinguished by X-ray, DTA, and electron microscope investigations. In a particular geologic environment, however, or per- haps only within the limits of a particular locality, a distinction on the basis of optical properties can be made between some of the varieties of serpentine. (Chidester, 1962, p. 46—48). Even then, it is gener- ally possible to distinguish only between chrysotile on the one hand and the platy varieties of serpentine (antigorite, lizardite, and six-layer orthoserpentine) on the other. And where two or more varieties are intimately intergrown, the distinction by means of optical properties may be uncertain or impossible. Thus, in the Belvidere Mountain area, chrysotile can generally be distinguished by its asbestiform habit and low indices relative to platy serpentine. The apparent optic axial angle in sections cut nor- mal to the fiber axis is 2V=0°; in oblique sections, the angle increases as obliquity increases. The ap- parent optic sign is (+); the sign of elongation, (+). Measured indices are n=1.532—1.545, N= 1550—1562; the measured value of N is probably near true y, which vibrates parallel to the fiber axes, but the value of n is intermediate between a and ,8. (See note.) NOTE.—It seems probable that, fundamentally, the optic sign of chrysotile is negative, the same as that of other varieties of serpentine. The apparent positive optic sign, and the apparent optic angle, which varies with the obliquity of the section with respect to the fiber axis, appear to result from the cylindrical structure of the lattice. In sections of orthochrysotile and clinochrysotile cut normal to the fiber axes, 'y vibrates parallel to the microscope tube, and all the transmitted light, vibrating in the plane of the section, ranges in index from a to ,3. Thus, the index observed is an average of the two; the mineral appears to be uniaxial and the optic sign, (+). In sections cut moderately oblique to the fiber axes, light vibrating parallel to the microscope tube is a nearly fixed large value of 7’. Light vibrating parallel to the long axis of the elliptical section ranges in index from a smaller 7’ to a’, so that the index observed is an average near [3. Light vibrating parallel to the short axis of the elliptical section ranges in index from a. to ,3, so that the index observed is a small value of a'. As a consequence, the observed interference figure is biaxial, 2V is moderate, and the optic sign is (+). As obliquity increases, the apparent optical axial angle increases until, when the fiber axes are parallel to the section, the interference figure is like that of a flash figure of an uniaxial positive mineral. The same line of argument would appear to hold for parachrysotile, but with this difference: If Bawy, the ap- parent optic angle, apparent optic sign, and sign of elonga- tion would be the same as for orthochrysotile and dine- chrysotile, but the apparent birefringence would be appreci- ably lower. If azfi, the relations of the apparent optic angle would be the same, but the optic sign and sign of elongation might be negative; the apparent birefringence would also be low. On the other hand, the platy varieties of serpen- tine (chiefly antigorite) have the following optical properties: 2V, small to moderate; optic sign, (—) ; sign of elongation, (+); birefringence colors, ab- normal blue; indices, a=1.560-1.565, Bmy=1.565— 1.573. These observations form the basis of data pre- sented in table 2, which relates approximately the structural parameters, habit, and optical properties of the serpentine minerals. 24 Many varietal names that designate chiefly types of serpentine defined on the basis of habit and as- sociation have been discredited; only antigorite, liz- ardite, six-layer o‘rthoserpentine, and the three types of chrysotile are now recognized as valid species. In this report, “picrolite” is retained, in addition to the species names listed above; it is used in a purely descriptive sense for nonasbestiform vein serpentine that is dense and of generally columnar or coarsely fibrous but locally massive habit. IGNEOUS ROCKS, SERPENTINITE, AND VEINS Ultramafic rocks that have more than about one- third of the essential primary minerals (olivine and pyroxene) remaining retain generally the appear- ance of the primary igneous rock and are classed in this report as dunite or peridotite. More highly ser— . pentinized rock is classed as serpentinite. The classi- : fication as dunite or peridotite is based on the origi- nal content of pyroxene. Rocks containing less than 5 percent pyroxene or recognizable pseudomorphs , after pyroxene are classed as dunite; those contain— . ing more than 5 percent, as peridotite. The transition from dunite to peridotite is com- . monly abrupt; layers that contain 5—10 percent py— roxene are in sharp contact with layers that contain ‘ little or no pyroxene. Less commonly, the transition 2 l l l l l l i l l I ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT is gradational over several centimeters. Both the dunite and peridotite grade into massive serpentinite. Normal to the layering, the transition is fairly sharp, the gradation from little or moderately serpentinized rock to highly serpentinized rock taking place within a fraction of a centimeter. Parallel to the layering, the transition is commonly very gradual, taking place over several meters. Chromitite forms sharply bounded layers in the dunite and massive serpentinite, and irregular pods in schistose serpentinite. Several types of veins crosscut layering in the dunite, peridotite, and mas- sive serpentinite; are parallel to the irregular schis- tosity of the schistose serpentinite; or are in transecting shear zones. Most of the veins are of ser— pentine, carbonate, and associated minerals, but magnetite also forms distinctive veins. Table 3 contains selected modes of ultramafic ig- neous rocks and their serpentinized derivatives. DUNITE AND PERIDOTITE Dunite crops out extensively on the southeastern slopes of Belvidere Mountain and was well exposed by stripping operations in the Lowell and C-area quarries. Areas predominantly of dunite are deline- ated on the large-scale maps of the Lowell quarry (fig. 5) and the C-area (fig. 7). Minor amounts of TABLE 2.—Unit-cell dimensions, habit, and optical properties of the structural varieties of serpentine minerals chrysotile . . . . S' _1 Ortho Clino Pam Antigorite Lizardlte orthdstrgfiiltine Unit-cell dimensions a _________________ 5.34 A 5.34 A 5.34 A 43 5 A 5. 31 A 5.32 A b _________________ 9.2 A 9.2 A 9.2 A 9. 2 A 9. 2 A 9.23 A c _________________ 14.63 A 14.65 A 14.63 A 7. 26 A 7. 31 A 43.59 A B _________________ __ 93° 16' __ 91° 23’ __ __ Optical properties 2V1 _______________ 0°-large 0°-large 0°—large Small to Small to ? _ _ moderate moderate(?) Optlc Slgn1 —__,_——— (+) (+) (+) or (—)" <—) — (—)? Slgn of elonga‘clon1 (+) (+) (+) or (—) " (+) (+) (+) a-mdex “' __________ 1532—1545 1532—1545 ? 1.560—1.565 ? [B-index ___________ ? ? ? ~-y ? ? 'y-index ___________ 1550—1562 1550—1562 ? 1565-1573 1 57 3 1.575(?) Optical orientation _ azc a:c a:c azc a:c fizb fizb [32b [32b sz ? 7:0. 7:0. 7:0, 72a P'yza Extinction ________ Parallel Parallel Parallel Parallel Pra ralell ? Abnormal bire- fringence colors _ None None None Blue Blue Habit Fibrous; a: Fibrous; a: Fibrous; b: Platy to colum— Platy to colum- Structureless to fiber axis fiber axis fiber axis nar; elon— nar; elon- platy; elon- gated parallel gated parallel gated parallel to b to a. to a. 1See text discussion on optic axial angle, optic sign, and sign of elongation, of chrysotile. 2Actually an (IX-index in the case of chrysotile. ULTRAMAFIC AND ASSOCIATED ROCKS 25 TABLE 3.—Selected modes of dunite and serpentinite [Tr, trace] Protolith, (based on A t content of pseudomor- n igo- _ _ . _ , . . Pyrox- Chro- Magne . Chi y- .. . Antho— Car- Gra Specimen No. phased pyloxene: Ollvlne . . rite or . Chloute Brucite . Sulfides Talc ~ dunite <5 percent ene mite tlte lizardite sotile phylllte bonate phlbe peridotite >5 percent) A. Dunite and massive serpentinite 105 _________________ Dumte _______________ 16 Tr -- 2 80 __ l .. 1 __ _- -_ -- 110 _________________ do ___________________ 33.3 __ Tr 3.8 59.3 __ 1.6 __ 2.0 __ __ -- —— AR—42 ______________ do ___________________ __ __ __ 7 83 __ 1 8 _- Tr 1 -- -— VAE—l ............. do ___________________ 90 __ Tr <1 10 <1 <1 _- __ -- -_ -- —— AR-13b ............ do ................... 63.0 __ __ 1.1 34.9 __ .7 .3 __ T1 _- __ -- AR—13f ____________ do .. ............ 76.4 _- __ 4.0 19.6 - __ __ __ _- __ -- -— AR—73 _____________ do _ ____________ __ _- _- 5.6 84.6 _ 4.1 5.7 __ __ -_ —_ - AV—15 _____________ do _________________ 1.3 __ __ 4.9 83.7 3.4 4.0 2.7 -- _- __ __ -— AV—29 _____________ eridotite ____________ 20 Tr __ 6 68 __ 1 _~ 5 _._ .. -_ -— AV—99 __ _Dunite _._ __ __ __ 5.1 78.2 _ 7.3 9.4 __ __ __ __ -_ AV—129 __ "Peridotite Tr Tr _- 10 ‘78 __ 2 Tr 10 _ Tr __ __ AV—248 _ _Dunite - . 85.9 _ Tr 1.0 12.7 __ Tr .4 __ _ -_ -_ —- A—BM-S'i—l ......... do ___________________ __ __ __ 5 76 __ __ __ __ __ __ __ 19 A—BM—S‘i—IZ _______ do ___________________ __ __ __ 5 95 __ __ _ _ _- __ __ -— A—BM-57—3 _________ do ___________________ 60 __ __ 2 37 __ 1 Tr __ _ __ -- -— B. Schistose serpentinite AV~74 _____________ Dunite _______________ __ _ __ 3.2 86.6 6.4 2.8 A __ __ .. .— AV—81 _____________ do ____________________ __ __ _- 4.0 80.2 13.2 1.3 1.2 _ __ 0.1 __ -— AV—209 _____________ do __________________ __ __ __ ‘l 91 1 _ __ _, 7 __ __ A—BM—RO ___________ do ____________________ __ __ 1 __ 99 __ __ ,_ , Tr __ Tr __ A—BM—53—20 ________ do ___________________ _ _ _ Tr 94 _ _ _ A f _ __ 1 5 A—BM—53—4‘3 ________ do ___________________ __ _ Tr 100 _- -4 __ __ __ ._ _- EQ-4-909 __________ do ____________________ _ A , _ 60 __ 2 __ __ __ 1 _ 37 CP—2—126 __________ do ___________________ _ f _ 7 90 _ Tr 3 __ __ _. __ __ peridotite are interlayered with the dunite but no- where form mappable units. Dunite is abundant in the ultramafic rocks on the steep slopes under and south of the abandoned aerial tramway line on the east side of Belvidere Mountain (fig. 3) but is not mapped separately. The dunite and peridotite are generally massive and range from grayish yellow green to dark green- ish gray. They are predominantly fine grained and break unevenly along minutely grainy surfaces char- acterized by cleavage flashes of olivine and tiny lus- trous silken threads and patches of serpentine. Oli- vine is the only essential mineral of the dunite, and some of the dunite is virtually free of pyroxene or its pseudomorphs; however, most of the dunite con- tained originally from a small fraction of 1 percent to as much as 5 percent pyroxene. The peridotite contains pyroxene as an essential constituent in addi- tion to olivine. The original pyroxene content of the peridotite ranged from 5 percent to about 10 per- cent. Chromite, magnetite, and sparse sulfides and sulfarsenides are common primary accessory min- erals of the dunite and peridotite. Serpentine and magnetite are ubiquitous secondary minerals, and brucite, chlorite, carbonate, and anthophyllite are common ones in both rocks. Both rocks vary widely in degree of serpentinization, but all contain more than 10 percent serpentine. Representative modes of dunite and peridotite are given in table 3A. On weathered surfaces long exposed to the air, dunite and peridotite characteristically range from buff or yellowish gray to moderate reddish brown, but weathered surfaces from which mantling vegeta- tion has been recently removed are commonly pale yellowish green to almost white. Layering is ubiquitous in the dunite and perido- tite. Most of the layers are 1—25 cm thick, and few are as thick as a meter. Some of the thinner layers consist of groups of uniform layers a few millimeters thick. Individual layers are distinguished by differ- ences in texture and color; the layering reflects chief- ly original differences in texture and content of accessory minerals but also marks differences in the pattern and degree of serpentinization. The distribu- tion of pyroxene pseudomorphs especially contrib- utes to the layering; the content of pyroxene is commonly significantly different in adjacent layers, and the alinement of the platy crystals parallel to the layering reinforces the layered effect. On weath— ered surfaces, differences in color and texture are accentuated by differences in weathered color of oli- vine, pyroxene, serpentine, and brucite, by the variously knotty or pitted weathered surface of py- roxenic rock, and by the increased contrast between silicate and metallic minerals. Much of the dunite and peridotite has a crude to pronounced foliation parallel to the layering. The foliation is due to the parallel alinement of platy minerals, particularly pyroxene, to the tendency of the rock to break along surfaces of contact between layers of contrasting composition, and, in a few specimens, to a discernible parallel alinement of crystallographic planes in the olivine grains. 26 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT CHROMITITE Concentrations of chromite (chromitite) are rare in the Belvidere Mountain area and were noted only in the main body of ultramafic rock. A layer of massive chromitite about 30 cm thick is exposed for a length of about 3 meters at 102,950 N., 90,940 E. (pl. 1) on the west peak of Belvidere Mountain. Sev- eral thin layers of chromitite 1—10 cm thick were exposed by stripping operations near 98,260 N., 97,160 E. (fig. 3 and pl. 1) in the C-area. These lay- ered bodies are conformable with the layering in the enclosing dunite and massive serpentinite. Where the layered dunite is folded, the chromitite veins are folded in conformity. A small irregular pod of mas— sive chromitite is exposed southwest of Corez Pond, at 92,700 N., 96,900 E. (pl. 1). The contacts are not exposed, and the pod is surrounded by talc-carbonate rock; so the structural relations are not known. Veins of pale-green coarsely foliated talc border and vein the chromitite pod. The tabular bodies of chromitite range in fabric from massive throughout to finely layered. The fine layering is as little as 1 mm thick and consists of alternate laminae predominantly of chromite and olivine or serpentine. The pod of chromitite is en— tirely massive. A sample (A—BM—29) of the 30-cm-thick layer of chromite from the west peak of Belvidere Mountain has a modal composition of: chromite, 66 percent; olivine and serpentine, 34 percent. The modal com- position of a sample (A—BM—53—116) from a tabular body 10 cm thick, in the C-area, is: chromite, 55 per- cent; magnetite (probable chromian), 13 percent; olivine and serpentine, 32 percent. SERPENTINITE Serpentinite has a variety of textural features and mineral associations, but can be divided into two textural end types: massive serpentinite and highly sheared and schistose serpentinite (hereafter re— ferred to simply as “schistose serpentinite”). Ser- pentinite intermediate in character is common. Massive serpentinite is well exposed near the west peak and on the southeast slopes of Belvidere Moun- tain, Where it is interlayered with dunite and perido- tite, and in the Eden, C-area, and Lowell quarries, where it is marginal to and interlayered with dunite and peridotite (pl. 1). Schistose serpentinite is well exposed in the three quarries, where it is principally adjacent to the contacts of the ultramafic bodies but also occurs in irregular zones that transect both dunite and massive serpentinite. Schistose serpen- tinite is also well exposed in the many small satellitic bodies around the main body of ultramafic rock. Moderately sheared serpentinite forms discrete masses in the Lowell and C-area quarries. It is map- ped generally with the schistose serpentinite, but plate 40 illustrates the variation in textural pattern of moderately sheared serpentinite. The general dis- tribution patterns of serpentinite are depicted in figure 3 and plates 3 and 4. Massive serpentinite is medium greenish or bluish gray to very dusky blue green or greenish black. It retains the gross textural features of the dunite or peridotite but breaks with a hackly rather than grainy fracture. The layering present in the dunite has been retained with little change except in the color. In several places, the massive serpentinite con- tains layers composed of diversely oriented flakes as much as 2 mm across; these layers are commonly not more than a few centimeters thick. Mineralogically, the massive serpentinite is like the partly serpentin- ized dunite and peridotite except for the relative pro- portions of the minerals. Relict olivine is commonly present, and remnants of pyroxene and anthophyllite are not uncommon, but in many places these miner- als are entirely altered to serpentine. Magnetite is more abundant than in the dunite and peridotite, particularly as very fine grains and dustlike parti- cles. Patches and veins of brucite and halos of chlo- rite around grains of chromite are common. Chro- mite, the larger grains of magnetite, and the sulfides and sulfarsenides are in the same relations as in the dunite and peridotite. Some of the massive serpen- tinite is slightly to highly graphitic, particularly in the flaky layers. Schistose serpentinite is similar in color to the massive serpentinite, but it retains almost none of the textural features of the igneous rock. It consists of irregular polyhedral units of unsheared serpentin- ite bounded by thin zones of highly sheared and schistose serpentinite. The unsheared polyhedral units range in size from tiny chips to masses several meters across; the schistose zones in which macro- scopic unsheared chips are virtually absent are com- monly a small fraction of a centimeter thick, but may be as thick as a few centimeters. The schistose serpentinite consists essentially of platy serpentin- ite; in most specimens, fibrous serpentine is absent from the groundmass, and in none does it constitute more than a few percent of it. Magnetite is the most abundant accessory mineral. Carbonate, brucite, chlorite, chromite, and sulfides and sulfarsenides are common but sparse accessories; talc is rare but local- ULTRAMAFIC AND ASSOCIATED ROCKS 27 1y abundant. Locally, graphite is sparse to very abundant in the schistose serpentinite, especially along the eastern contact of the Eden quarry body. Both the massive and schistose serpentinite weath- er to a very pale green, very pale buff, or white chalky surface. Commonly, smooth surfaces are marked by a reticulate pattern of finely etched lines. Remnants of olivine in the massive serpentinite and patches and flakes of brucite in both the massive and the schistose serpentinite weather brown to buff and show up against the white-weathered serpentine as distinctive specks and patches. SERPENTINE VEINS Serpentine veins in the ultramafic rocks have a wide variety of structural and textural relations and mineral associations, but they fall into two basic categories: asbestos veins and picrolite veins. As— bestos veins are of two habits: cross-fiber veins, in which the fiber axes are at a large angle to the vein walls, and slip-fiber veins, in which the fibers are nearly parallel to the vein walls. Picrolite veins, in which the serpentine mineral is dense rather than asbestiform, are usually coarsely fibrous or columnar but are locally massive. Some veins are composite and contain both picrolite and asbestos. Microscopic veins, which are observable only in thin section, or Whose textural features are discernible only under the microscope, can generally be classified in one or another of these categories, but some have features not displayed by macro-veins. In dunite and peridotite, serpentinized zones are conspicuous adjacent to both cross-fiber and picro- lite veins. In massive serpentinite, these zones are generally indistinct, and in schistose serpentinite none were observed along the rare tiny cross-fiber veins. Crosscutting relations between different kinds of serpentine veins are scarce, indicating that the veins are generally of the same age. In a few places, veins of slip-column picrolite crosscut veins of cross-fiber asbestos. Veins of cross-fiber asbestos were not seen to crosscut veins of picrolite, except in the limited sense that the two kinds of serpentine crosscut each other in composite veins, but such a relationship may exist. CROSS-FIBER ASBESTOS Veins of cross-fiber asbestos are the most conspic- uous serpentine veins and are the source of the most prized specimens of chrysotile asbestos, but they are quantitatively of relatively minor importance. Cross- fiber veins are composed chiefly of asbestiform fi- brous serpentine (chrysotile). The principal acces- sory mineral is magnetite concentrated sparsely to heavily along partings and at the margins of the veins, and locally intergrown with the fiber. Brucite, crudely fibrous in habit, is sparsely intergrown with the asbestos in some veins, and patches of calcite occupy irregular volumes in a few veins. On fresh surfaces of compact unopened fiber, the asbestos is lustrous grayish green, dusky yellowish green, or bronzy yellowish brown. Rarely, it shows faint and delicate variations in color in laminae parallel to the vein walls. Such “color lamination” is more common and more pronounced in some of the asbestos deposits in Quebec (Riordan, 1955, p. 67—81, fig. 4). When teased and separated, the fibers are silky and flexible and almost white. Weathered sur- faces are pale green, and weathered fiber generally rather stiff and brittle. Veins range in thickness from less than a milli- meter to as much as 5 cm; most are probably 0.5—5 mm thick, and veins thicker than 3 cm are rare. In tabular dimensions the veins range from a fraction of a centimeter to several tens of meters. The dis- tribution pattern varies greatly, from closely spaced subparallel veins of relatively uniform thickness (“ribbon veins”) to diversely oriented, widely spaced veins that vary greatly in thickness. In places, the veins form conjugate sets, but attitudes of the con- jugate set-s vary markedly from place to place; in other places no systematic orientation of the veins is discernible. Not uncommonly, some of the veins form conjugate sets, whereas others conform to no ap- parent system. In many places the spacing of the veins varies with their thickness: thin veins tend to be closely spaced, thicker ones, widely spaced. Cross-fiber veins show considerable diversity in the configuration of their walls, the relations of cen- tral partings or seams of magnetite, the distribution pattern of inclusions of serpentine, the orientation of the asbestos fibers with respect to the vein walls, and the relations between veins at their intersections. Figure 5 illustrates some of the configurations and relations shown by cross-fiber veins. The smaller veins, those a millimeter to several centimeters long, have mostly the form of thin doub- ly convex lenses. Larger veins are commonly tabular or sheetlike; they are of relatively uniform thickness throughout most of their extent, and wedge out or lens out to knife edges at their margins. In some places, the wedging out is rather abrupt, in others, gently tapering. Some grossly tabular veins consist, AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA %/ WWW/«WW C 7 g / WW W B in \ l / 4% F D E ULTRAMAFIC AND ASSOCIATED ROCKS in whole or in part, of a series of en echelon gash veins. Most veins are nearly planar or only gently curved, but some are more irregular and depart con- siderably from planar form. Nearly all the veins have generally matching op- posite walls, and irregularities commonly correspond in minute detail. Inclusions of massive serpentine within cross-fiber veins are commonly seen to fit into cusps and irregular embayments in the adjacent wall. In a few places, matching fragments of frac- tured grains of chromite or of pseudomorphosed py— roxene appear on opposite walls of a vein. Almost without exception, matching irregularities, frag- mented crystals, and inclusions and embayments are at opposite ends ofbundles of fibers. Irregularities unequivocally indicative of replacement are rare, and the part for which origin by replacement is indi- cated constitutes only a small proportion of the total vein width. Many cross-fiber veins contain one or more part- ings of thin seams of magnetite—along which may be alined thin chips of serpentinite—that separate the vein into two or more layers of fibers. In others, single fibers extend the full width of the vein. On the basis of partings, it is a common practice to divide cross-fiber veins into single-fiber and double- or multiple-fiber veins (Cooke, 1937, p. 91—99). Many single-fiber veins have almost no magnetite associated with them, or only inconspicuous concen- trations at either or both walls. Others have mod- erately heavy concentrations of magnetite impreg- nating the wallrock bordering the veins. Double— and multiple-fiber veins generally have similar concen- trations of magnetite at either or both walls, in addi— tion to the central parting seams, but in some veins the margins are virtually free of magnetite. In some veins, the central parting or partings are fairly regular and occupy about the same relative position throughout the veins. In others, the part- ings are irregular and cut from near one side of the vein to near the other, or cut back and forth. In some veins, the partings are grossly smooth and con- tinuous; in others, they are strongly crenellated and discontinuous, the discontinuous segments capping bundles of fibers. In detail, even the smooth continu- ous partings are slightly crenellated. The partings are commonly of fairly uniform thickness, but many pinch and swell; some widen out to small irregular concentrations of magnetite—locally intergrown with the asbestos and of fibrous or columnar habit—- that may in places occupy nearly the whole thickness of the vein. The distribution pattern of chips of ser- 29 pentinite in the veins is similar to that of crenellated segments of magnetite partings. Single-fiber veins appear to be associated more commonly with massive serpentinite, and double- or multiple-fiber veins, with dunite, but each occurs in both types of rock. Veins of both types in dunite commonly have marginal concentrations of magne- tite at either or both walls, Whereas those in massive serpentinite have none, or only inconspicuous mar— ginal concentrations. In most cross-fiber veins, the asbestos is oriented so that its fiber axes are about normal to the vein walls. In few are fibers more oblique than 45°, but in some, the fibers approach the obliquity of slip-fiber. Even where the fibers display marked obliquity, evi- dence of shearing movement between fiber veins and vein wall is extremely rare or nonexistent. In both the single- and multiple—fiber veins, the fibers generally are straight and are uniformly oriented. In a rela- tively small proportion of the single-fiber veins, the fibers are flexed into open chevron-type folds along one or more planes parallel to the vein walls. In a few double-fiber veins, the fibers on opposite sides of a parting diverge in orientation by as much as 30°; commonly, such veins pass laterally into single-fiber veins that have a plane of flexure that is coplanar with the parting. The pattern of intersecting cross-fiber veins is al- most infinitely varied, but fundamentally the inter- sections are of two types, crosscutting and merging. All crosscutting veins are some variation of a basic X—form, whereas merging veins may have a Y, T, or X form. Most intersections are merging. Simple crosscutting intersections are rare, but complex in- tersections, which have both merging and crosscut- ting relations, are not uncommon. In simple crosscutting veins, one vein can be traced distinctly through another. The attitude of the fibers in each vein is virtually uniform at and on either side of the intersection. When crosscutting veins intersect obliquely, one or another of the veins is invariably offset forming an open Z-pattern. Which vein is offset depends solely on the geometric relations of the veins and the fiber orientation, rath- er than on age relations. Figure 5F, G, and H illus— trate some of the possible relations. In merging intersection, the fibers of the intersect— ing veins merge by gradual and systematic change in attitude, approximately radially about the edges of intersection of the vein walls (fig. 5 I-N). Where intersecting veins are about perpendicular to one an- other, the fibers commonly are radial to the four (or two, in T—shaped intersections) edges of intersection 30 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT of the vein walls (fig. 5 I and J). Where veins inter- sect in a small dihedral angle, the fibers commonly are radial only about the two (or one, in slant-T intersections) edges at the apices of the obtuse di- hedral angles of the walls (fig. 5 K and N). Com- monly, partings o-f magnetite and chips of serpen— tinite separate groups of fibers that have different radial foci. In predominantly single-fiber veins, such partings are generally thin and extend only for short distances from the junctions, or join the apices of opposite acute dihedral angles of the vein walls (fig. 5 J, L, and M). In double-fiber veins, the partings in the junction are commonly irregular and contain abundant magnetite, chips of serpentinite, and dis- ordered bundles of short asbestos fiber (fig. 5 I, K, and N). Complex intersections were observed only in mul- tiple-fiber veins and may be confined to them. Fea- tures characteristic of merging intersections pre- dominate. The possible relations are extremely varied, but commonly the layers of asbestos adjacent to the vein wall in the obtuse angles merge, and the layers adjacent to the vein wall in the acute angles partly or entirely crosscut each other and the merg- ing layers (fig. 5 N). For additional descriptions and illustrations of relations shown by asbestos veins, refer to Cooke (1937, p. 91-98) and Riordon (1955, p. 68—75). SLIP-FIBER ASBESTOS The term “slip-fiber asbestos” is applied to as- bestos of two distinctively different habits and modes of occurrence: (1) thin seams of short fibers dis- tributed along closely spaced irregular surfaces of schistosity in schistose serpentinite and along widely spaced minor shear surfaces in dunite and massive serpentinite, and (2) very long fibers in pockets along shear zones or faults that are fairly regular in attitude, tens or hundreds of meters in extent, and a centimeter to a meter wide. The proportion of slip- fiber asbestos in the schistose serpentinite is not ap- parent except upon close examination by an experi- enced observer, but such slip fiber constitutes most of the asbestos in the Belvidere Mountain area. The pockets of very long slip fiber in faults and shear zones are spectacular occurrences, but the amount of such material at Belvidere Mountain is negligible. Slip-fiber asbestos associated with schistose ser- pentinite is short, generally much less than a centi- meter. The fiber veins are thin, and most of the fibers appear to be nearly parallel to the vein walls. In detail, however, the fibers make a very small acute angle with the vein walls, and have a shingled—or, perhaps more accurately, a thatched—arrangement. Over small areas of a surface covered with slip fiber the fibers are essentially parallel to each other, but the orientation varies slightly from place to place. The overall pattern of such a surface forms an ir- regular mosaic of groups of fibers that have the same general orientation, but are slightly divergent from one another. The boundaries between divergent groups commonly coincide with irregularities in the schistosity surfaces. Most of the fiber is firmly at— tached to both walls; some is in sheared contact, but the shearing may have occurred during quarrying operations. In color, flexibility, and silkiness, the slip fiber in schistose serpentine, and also that in dunite and massive serpentinite, appears to be similar to the cross fiber, and the mineralogy does not differ sig- nificantly (see preceding section on “Cross-fiber asbestos”). Pockets of slip fiber in faults and shear zones con- tain fibers as much as 1 m long. All such slip fiber is intergrown with abundant fibrous calcite, which commonly is more abundant than chrysotile asbestos and is as delicately fibrous. The fibrous calcite is stiff and brittle, however, whereas the asbestos is flexible, soft, and silky. The fibers are parallel to the shear zones; no shingled or thatched arrangement is dis- cernible, and the fiber ends are unattached to either wall. PICROLITE Veins of picrolite have the same general distribu- tion and size range as veins of cross-fiber and slip- fiber asbestos, but are less abundant than either and are more variable in structure and mineral associa- tions. Nearly all the picrolite has a crudely fibrous or columnar structure, but some is megascopically structureless and massive. The picrolite ranges from pale olive through dark greenish yellow to dusky green; it is microcrystal- line, dense, and has a waxy luster. Coarsely fibrous to crudely columnar picrolite has a splintery frac- ture; dense and massive picrolite has a conchoidal fracture. The picrolite veins commonly contain small to large amounts of brucite. The brucite is commonly coarsely fibrous or flamboyant and is intergrown with the picrolite. Fibrous calcite and clots and dis- seminated grains of magnetite are abundant in a few veins. In most veins associated with dunite, peridotite, and massive serpentinite, the columnar structure is ULTRAMAFIC AND ASSOCIATED ROCKS 31 about perpendicular to the vein walls, as the fibers are in cross-fiber asbestos. In veins in schistose ser- pentinite, the columnar structure is nearly parallel to the vein walls, as the fibers are in slip-fiber veins. “Cross column” picrolite veins are mostly less than 2 cm thick, rarely as thick as 10 cm, and the columns are of comparable length. “Slip column” picrolite veins are seldom more than 5 cm thick, but the col- umns are commonly 15 cm to more than 30 cm long. The walls of cross-column picrolite veins common— ly have matching irregularities at opposite ends of columns, but irregular penetration of the walls by the vein minerals is rather common, though relative- ly minor. Some picrolite veins of the cross-column type are indistinctly laminated parallel to the vein. walls. The laminae reflect chiefly differences in color, but also slight differences in orientation of the coarse fibers and columns in successive laminae (Ri- ordan, 1955, p. 67—81, figs. 3 and 5). Few if any simple unlaminated veins contain partings of mag- netite, or significant amounts of serpentinite inclu- sions. On the other hand, compo-site veins that contain both cross-fiber asbestos and picrolite com- monly have partings associated with the cross-fiber asbestos, and laminated picrolite veins commonly contain thin discontinuous layers of magnetite paral- lel to the color lamination. COMPOSITE VEINS Most composite veins contain both cross-fiber as- bestos and cross-column picrolite, but some contain cross-fiber asbestos and slip-column picrolite; per- haps other composite types also exist. Commonly picrolite and asbestos form two or more distinct layers parallel to vein walls. The transition between layers is very sharp, and the boundary grossly reg- ular but minutely irregular in detail. In some com- posite veins, there is a lateral transition, gradational over a short distance, from a vein composed entirely of cross-fiber asbestos to a vein composed entirely of picrolite. A few composite veins consist of irregular intergrowths of cross-column picrolite, wispy bun- dles of wavy cross-fiber asbestos, and fibrous brucite. In the relations of the vein walls, mineralogy, and other features, the composite veins combine the features of the simple component veins. MICROSCOPIC VEINS Many of the microscopic veins represent in mini- ature one or another of all the categories of macro- veins, but Some have features apparently unlike those of any of the macro-veins. In one type, regular blades of platy serpentine, many times longer than they are wide, are alined parallel to the veins. Under low magnification they appear fibrous, but under very high magnification the apparently fibrous struc- ture is seen to be a perfect cleavage parallel to the length of the blades. Some of the veins of platy ser- pentine are composite with brucite veins. Typically, they consist of a central vein of brucite about 0.001 mm thick, bordered on either side by fibrous brucite veins about 0.005 mm thick, the fibers at a large angle to the vein; the brucite veins are succeeded outward by veins of platy serpentine about 0.01 mm thick. A second type of micro-vein consists, in thin section, of irregular elliptical particles of serpentine, probably spindle shaped, elongate parallel to the veins. In routine examination, the spindle-shaped particles appear to be structureless, but careful ex- amination under very high magnification indicates that they probably have a very fine fibrous structure about perpendicular to the spindle axes. A third type, observed in only a few thin sections, has an irregular fibrous structure at varied angles to the vein walls; the serpentine commonly is distinctly banded parallel to the walls and has a marked yel- lowish tinge. The anomalous appearance of this rare third type of vein is due to the orientation of the fiber axes at a relatively large angle to the plane of the thin section. Most of the micro-veins merge at their intersec- tions, but a few are crosscutting. In one section, yellowish veins of the third type, which were just described, are cut by micro-veins of colorless cross- fiber asbestos. Most of the micro-veins contain only serpentine, - but .some contain tiny grains and dustlike particles of magnetite, sparse flakes of brucite, and irregular patches of carbonate. SERPENTINIZED ZONES AT THE bIARGINS OF VEINS Marginal zones of alteration adjacent to serpen- tine veins are associated only with cross-fiber as- bestos veins and cross-column picrolite veins. Margi- nal alteration zones in dunite and peridotite are con- spicuous features, readily apparent even from a con- siderable distance (fig. 6). In massive serpentinite they are generally faint because of the decreased contrast in color; thus, the distinctness of the margi- nal zones varies directly with the proportion of un- altered olivine in the rock. The alteration zones consist essentially of serpen- tine, brucite, fine-grained or dusty particles of mag- netite, and tiny relict grains of olivine and chromite; variations in relative proportions, texture, and dis- 32 ASBE‘STOSFBEARING ULTRAMAFIC ROCKS 0F BE-LVIDERE MOUNTAIN AREA, VERMONT FIGURE 6.—Photograph of cross—fiber vein in dunite, showing marginal alteration zones. tribution impart a varied but distinct layering to the zones. The outer margins are generally marked by a thin layer that is 1 mm or less Wide; its color is grayish yellow green. Not uncommonly there are several such layers, of different shades of yellow green and increasingly less distinct inward toward the walls of the asbestos or p‘icrolite veins. The inner layers weather white, very pale green, or very pale buff; the outer layers, various shades of buff and light brown. Typically, immediately next to the vein the alter- ation zone consists of irregular shreds and chips of fibrous serpentine in a subordinate matrix of nearly isotropic serpentine of uncertain but probably platy habit. A few millimeters away from the vein scat- tered flakes and blades of platy serpentine appear. Outward toward the dunite, the proportion of fibrous serpentine and isotropic serpentine decreases, and the proportion of distinctly platy serpentine in- creases. Brucite occurs throughout the entire zone, commonly as disseminated flakes, and it tends to be more abundant in the outer part of the zone. Mag- netite occurs as scattered grains and dispersed dust- like particles; its distribution is commonly irregular, but in many marginal zones it is concentrated more heavily in a narrow zone immediately next to the vein, or between the asbestos and the vein wall. Very sparse relict grains of olivine occur chiefly in the outer parts of the alteration zones. Chromite is sparsely distributed as in the unaltered dunite. Some alteration zones consist almost entirely, ex- cept for abundant grains of magnetite, of diversely oriented shreds and flakes of fibrous serpentine and almost no isotropic or platy serpentine. Others con- tain many veins and disseminated flakes of brucite. Layering near the outer margins of some of the zones reflects chiefly different concentrations of brucite in the different layers. A few of the marginal zones are strikingly asymmetric in the distribution of minerals. In some of them, the marginal zone on one side of the vein shows the typical relations just described, whereas the zone on the opposite side of the vein consists almost entirely of brucite and only sparse bladed serpentine, fibrous serpentine, and magnetite. In others, the distribution of brucite is seemingly haphazard, and the relations of the ser- pentine minerals are irregular. The boundaries of marginal alteration zones are smooth and sharp in overall appearance. In detail, the transition from the altered zone to dunite is gradational over a distance of a fraction of a mil- limeter, and the boundary is minutely irregular. Where the alteration zone crosscuts layering in the dunite, tiny tongues from the alteration zone pene- trate as much as 1—2 mm along some of the layers. In some places, tiny fractures about normal to the marginal zone boundaries extend a few millimeters. into the dunite and serpentinite on either side of the boundary and commonly mark sawtoothed irregular- ities in the boundary. Geometrical relations of the marginal alteration zones are varied, but they conform in aggregate to a general pattern. Most of the marginal zones are roughly symmetrical with respect to the vein throughout most of its extent and are of about uni- form width along it, and the outer margins of the zones conform grossly with the vein walls. At vein intersections, the marginal zones are commonly somewhat thicker and more irregular. Not uncom- monly, the marginal zones thicken or thin locally in gently tapering fashion along a vein of uniform thickness; in some places the marginal zones main— tain a generally uniform thickness along a vein that varies appreciably in thickness. Along some veins, the marginal zone on one side is somewhat thicker than the zone on the other side, and not uncommonly the relation is reversed at different places along the vein. In a few places, veins as much as 5 mm thick have no marginal alteration zones, even in dunite, ULT‘RAMAFIC AND ASSOCIATED ROCKS 33 and some less than 1 mm thick have marginal zones several tens of times as thick. Though in a few places the ratio of vein thickness to total thickness of vein plus bordering alteration zones varies widely from the average, the ratio is fairly constant for the great majority of veins. For most of the cross-fiber asbestos vein in the Belvidere Mountain area, the ratio ranges from about 1 : 4 to about 1:10, and the average is probably near 1:6 or 1:7. No significant difference between veins in fresh dunite and in relatively completely serpentin- ized massive serpentinite was detected, but the data are insufficient to be conclusive. Few reliable obser- vations are available on picrolite veins, but the ratio appears to be somewhat larger, perhaps near 1 :3 or 1:4. These ratios are similar to those obtained by Cooke (1937, p. 103—110) in a careful study of as- bestos deposits in Quebec. AMPHIBOLE ASBESTOS A small mass of amphibole asbestos, consisting of fibers more than 30 cm long intergrown with fibrous calcite, was found in quarrying operations in a shear zone at or near the northeast contact of the Lowell quarry body; none was observed in place. The fiber is similar to masses of slip-fiber chrysotile, except that it is more brittle. Mineralogical examination by G. T. Faust (written commun., 1951) disclosed the asbestos to be a member of the tremolite-actinolite series. OTHER VEINS Less abundant and generally less conspicuous than the serpentine veins are veins of magnetite, car— bonate, brucite, and magnetite-brucite-carbonate. Of these, veins mainly of magnetite are the most abundant. Veins of magnetite are of two types: moderate to heavy concentrations of disseminated fine-grained magnetite in generally regular veinform masses in massive serpentine or dunite; and irregular pods and lenses of massive, coarse—grained chromian mag- netite in schistose serpentinite. The veinform masses of fine-grained magnetite generally are clearly con- trolled by layering or by crosscutting fractures, but most such magnetite is heavily disseminated in, and bears a replacement relation to, the adjoining ser- pentinite. In many places, seams of magnetite extend for considerable distances beyond the lateral limits of asbestos veins, along the same fractures. Most of the veins are fairly uniform and less than a centi- meter or so thick, but some widen out locally to form irregular masses several centimeters across which are broadly gradational into the serpentinite. Pods and lenses of massive chromian magnetite are very rare and no more than a few centimeters thick at Belvidere Mountain, but a mass more than 3 m thick and 30 m long is exposed about 3 km north— northeast of Troy village (Cady and others, 1963, p. 71, pl. 1). This pod is very irregular, and the mas- sive magnetite is everywhere in sheared contact with the enclosing schistose serpentinite. The magnetite rock in the pod consists of closely packed anhedral grains of magnetite 1—5 mm across in a matrix of bladed serpentinite. Ilmenite forms narrow rims around scattered grains of sphene that have cores of rutile and tiny veinlets along cleavage surfaces in magnetite. Calcite occurs as sparsely scattered grains and tiny veinlets. Magnetite constitutes 65- 75 percent of the rock; serpentine and carbonate, 35—20 percent; and ilmenite, rutile, and sphene, < 1—5 percent. Brucite and calcite and mixtures of one or both with magnetite commonly form scattered tiny vein- lets in the dunite, peridotite, and serpentinite. Com- monly the veinlets are fracture controlled. Brucite veins of this type associated with serpentine veins have been described under “Microscopic veins.” Some veins are of uniform width, have sharp boun- daries, and show little evidence of replacement. Some of the regular veins widen out locally to irregular volumes showing predominantly features character- istic of replacement origin. Many veins are entirely or predominantly irregular. Commonly, the veins having features characteristic of replacement con- tain narrow central fissurelike zones entirely of mag- netite. In others, the distribution of minerals is fairly uniform or entirely irregular. These veins range in composition from entirely calcite or brucite to various intermixtures of mag- netite, brucite, and calcite. In many, the textural re- lations of the three minerals are not diagnostic, but commonly the calcite replaces brucite. MINERALOGY AND PARAGENESIS Dunite, peridotite, and chromitite contain primary minerals, which are the earliest formed minerals in the decipherable history of the rock—that is, those that are not recognizably derived from preexisting minerals. These three rocks also contain secondary or metamorphic minerals, derived from the primary minerals or introduced into a host rock during the metamorphic history. Serpentinite is constituted en- tirely of secondary minerals or contains sparse relict primary minerals of the dunite and periodotite. Ser- 34 ASBESTOS-BEARING ULTRAMAFIC ROCKS 0F BELVIDERE MOUNTAIN AREA, VERMONT pentine veins and their marginal alteration zones, and brucite-calcite-magnetite veins are composed en- tirely of secondary minerals. Magnetite veins are largely secondary, though some of the veins parallel to layering in the dunite and peridotite, and some of the irregular pods may be entirely or largely primary. The primary minerals are olivine, pyroxene, chromite, magnetite (chiefly or entirely chromian), sulfides, and sulfarsenides. They vary little in com- position, habit, and association, or vary systemati- cally. The secondary minerals are varieties of ser- pentine, anthophyllite, brucite, chlorite, carbonate, and magnetite (nonchromian). They vary widely in textural relations and mineral association. The primary minerals are anhedral to subhedral grains that form a granular mosaic texture that is characteristic of many plutonic igneous rocks; the secondary minerals form irregular interlocking mosaics, felted aggregates, and diverse fibrous tex- tures that are characteristic of metamorphic min- erals; the secondary minerals also show character- istic zonal relations to one another and to primary minerals that are indicative of genetic sequence. The minerals in veins and marginal alteration zones commonly have a zonal distribution. Chromite grains and associated rimming minerals show distinctive zoning. A translucent core of chromite is surrounded by an opaque rim of chromite or chromian magnetite and magnetite, which in turn is rimmed irregularly by chlorite. In many places the chlorite rim is bordered by fresh olivine; more commonly, the zoned assemblage is surrounded by antigorite. In some such associations, zoning in the antigorite adjacent to the chlorite is indicated by a zonal pattern in the abnormal interference colors. In a few slides, grains of opaque chromite in massive serpentinite have par— tial rims of olivine tightly stuck to the chromite. Such grains have no rim of chlorite adjacent to the olivine, but where the olivine rim is absent the chromite is separated from the enclosing serpentine by a rim of chlorite. OLIVINE In the least serpentinized dunite and peridotite, olivine forms a mosaic of subequant grains pre- dominantly 0.5 to 2 mm across and of irregularly polygonal outline in thin section. A few crystals of olivine are much larger. One rock specimen con- tained a single crystal more than 3 cm long and 2 cm thick, which enclosed several small subhedral crys- tals of olivine of different crystallographic orienta- tion. All the olivine grains are highly fractured, but commonly many or most show uniform extinction. In every thin section, a few grains show undulatory extinction, and a few grains are twinned; in many thin sections, most of the grains show undulatory extinction, and many are twinned. The large single crystal shows pronounced undulatory extinction and locally a vague mosaic pattern of areas that extin- guish at slightly different positions. In even the freshest rocks, most of the olivine grains are slightly altered at their boundaries to bladed serpentine and are locally rimmed by tiny veins of chrysotile. Serpentine invades many grains along fractures and cleavage directions, particularly (010) cleavage. As serpentinization increases, the control by morphologic and crystallographic features is less evident. In massive serpentinite, olivine re- mains only as small, very irregular relict grains. Marginal zones of such relict grains are somewhat cloudy. Under very high magnification, these cloudy zones are seen to contain many closely spaced hair- line fractures, commonly only 0.002 mm apart, along which the olivine is serpentinized. Many of the tiny fractures have a six—rayed stellate pattern. The olivine shows several distinctive variations in habit. Inclusion-free grains of olivine are grayish yellow green (5GY 7/2)2 in hand specimen, and colorless in thin section. Most of the grains show only an irregular fracture, but many show good (010) and poor (001) and (021) cleavage. A few contain conspicuous planes of tiny inclusions, chiefly of unidentifiable nonopaque grains but also of mag- netite, along crystallographic directions (021), (001), and (101). In a somewhat similar manner, olivine is intergrown in a few places with skeletal crystals of magnetite so that ( 021) of the olivine is parallel to the dominant direction of skeletal growth of the magnetite. The result of this relation is that the olivine has apparently inclined extinction, and such intergrowths are easily mistaken for pyroxene grains in which magnetite is concentrated along cleavage directions. The olivine is generally uniform in optical prop- erties and shows no systematic variations within the ultramafic body. The optic axial angle, 2V, is about 85°—90°, the optic sign is (+), ,8=1.664—1.671, and the birefringence about 0038-0042. Although no analyses of olivine separated from the dunite were obtained, two analyses of the fresh- est available dunite (analyses 1 and 45, table 1; spe- cimen numbers AR—13f, AV-248, table 3A), com- paratively free of accessory primary minerals, yield reliable mineral formulas of olivine, based on cal- 2Color designations are based on the Rock-color chart of the National Research Council (Goddard, 1948). ULTRAMAFIC‘ AND ASSOCIATED ROCKS 35 culated modes. The calculated formulas for the olivine, corrected for contaminating spinellids, ser- pentine, and brucite, and reduced to ideal values, are: AR—13f: Mgi.96Fe+zo.o4Si1.ooO4.009 AV—248 2 Mg1.95Fe+20.OSSi] ,ooO4.oo~ The ratio of Fe+2/ Mg in the calculated formulas is 002—0025. The ratio of Few/Mg, based on the op- tical properties (AR—13f: 2V=88° (+ ) , ,8=1.66; AV—248: 2V=88° (+ ), (1:1.644, ,8=1.664, )1: 1.683) (Deer and others, 1962, v. 1, fig. 11) , is about 0.05 and 0.06. For AR—13f and AV-248, respectively, the ratio of total Fe/ Mg in the whole-rock analyses is 0.07 and 0.06, and of Fe+2/Mg, 0.036 and 0.046 (table 1, analyses 1 and 45). Inasmuch as an ap- preciable proportion of the Fe is contained in mag- netite and other spinellids and in serpentine, it is evident that the ratio of Fe+2/ Mg in the olivine must be appreciably lower than in the whole rock. Conse- quently, the calculated analyses are believed to indi- cate more reliably than the optical data the propor- tion of Fe in the olivine, though they may err slight- ly on the low side because of possibly slightly high estimates of the proportion of magnetite in the cal- culated modes. PYROXENE Though pyroxene originally constituted as much as 5 percent of the dunite and 5—10 percent of the peridotite, it is now largely altered to anthophyl- lite(?), antigorite, brucite, and magnetite, which are pseudomorphic after the pyroxene in gross out- line or retain features of the pyroxene cleavage. Un- altered pyroxene occurs only very sparsely as scat- tered relict cores of partly altered grains. The pseudomorphs after pyroxene are generally tabular and range in outline in thin section from irregularly elongate to elliptical or rectangular; most are 2—4 mm long and 1—2 mm wide. Some con- tain blebs and irregular stringers of olivine which appear to be relicts of inclusions in the pyroxene and of tiny veinlets of olivine that healed fractures in the pyroxene. Unaltered pyroxene is so sparse that it was not practicable to determine indices of refraction. Other optical properties, Where determinable, were uni- formly: 2V=60°, optic sign (+), extinction angle yAC=45°, nonpleochroic, and colorless. On the basis of these optical properties, the pyroxene is inferred to be diopside or augite that contains an appre- ciable content of Fe“. CHROMITE AND MAGNETITE Chromite, the sole essential constituent of the chromitite, is also a common constituent of the dunite, peridotite, and serpentinite but rarely con- stitutes as much as 1 percent of the rock. Magnetite, the sole essential constitutent of magnetite veins, is commonly a minor constituent of the chromitite and is a ubiquitous but generally minor constituent of the dunite, peridotite, and serpentinite; it is also present in most of the serpentine and other veins, commonly as a minor constituent but locally in pro- portions greater than 50 percent. In the chromitite layers and pods, chromite occurs as subequant to slightly elongate anhedral grains as much as 5 mm across, commonly in a closely packed granular mosaic. The chromite grains are commonly highly fractured and the fractures filled with platy serpentine. In some of the chromitite bodies, chro- mite is the sole spinellid; in others, the chromite grains are partially rimmed by, veined by, and inter- grown with varying but subordinate amounts of magnetite. The chromite varies appreciably in physical and optical properties. Hand specimens are black and have a dull metallic luster; the streak ranges from brown to brownish gray. The powdered chromite ranges, in reaction to a hand magnet, from non— magnetic to weakly magnetic. In polished specimens from the west peak of Belvidere Mountain and from the C-area quarry (A—BM—29 and A—BM—53—116; see section on “Chromitite” under “Igneous rocks, serpentinite, and veins” in the section on “Ultra- mafic and associated rocks”), the chromite has the following optical properties, as observed with an ore microscope: A—BM—29: gray in reflected light, having red in- ternal reflections under crossed nicols; isotropic, having black cross on white field under conoscopic observation using crossed nicols; dispersion of the reflection rotation (Cameron, 1961, p. 115) is DR, +7'>v; not scratched readily with a needle. A—BM—53—116: light gray in reflected light, no internal reflections detected, isotropic, black cross under conoscopic observation using crossed nicols; dispersion of the reflection rotation not detectable; not scratched readily with a needle. Magnetite in A—BM—53—116 is silver white in reflected light, isotropic, and does not show detectable dispersion of the reflection rotation. 36 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT In the magnetite veins, serpentine veins, and brucite-calcite—magnetite veins, magnetite forms irregular grains and aggregates 0.1—1 mm across; in addition, magnetite forms fibrous intergrowths with asbestos in many of the cross-fiber asbestos veins. In thin section, observed by reflected light, the mag- netite is light steel gray. In dunite, peridotite, and serpentinite, spinellids occur chiefly as grains that fall into three size ranges of distinctive habit, associations, and properties. The largest size range consists of euhedral to an- hedral grains 1—2 mm across, having grossly regular boundaries. Most have translucent cores that are dark reddish brown in transmitted light and brown- ish black in reflected light. Surrounding the trans- lucent core, and bordering fractures in it, is a shell of opaque spinellid of slightly higher reflectance. In massive serpentinite, and in moderately to exten- sively serpentinized dunite and peridotite, many of the grains contain a narrow outer rim composed of an aggregate of tiny grains that are steel gray in reflected light. The intermediate size group consists of anhedral opaque grains predominantly 0.1 mm across. These grains have generally an appearance in reflected light similiar to that of the inner opaque shells surrounding the translucent cores of the larger grains, but some are brighter and grayer. Many con- tain narrow rims of aggregates of tiny grains hav- ing steel-gray reflectance. The smallest size group consists of opaque dustlike particles and tiny grains and irregular clusters of grains generally less than 0.05 mm across. ‘ Other magnetite is varied in habit but uniformly steel gray in reflected light. Concentrations of mag- netite form tabular aggregates along relict cleavage directions in pseudomorphs after pyroxene and scaly aggregates of tiny grains 0.1—1 mm in diameter along schistosity surfaces in serpentinite. Skeletal crystals in the form of incomplete reticulate net- works of irregular laths and elongate particles of magnetite are locally intergrown with olivine. Though precise data on the variations in composi- tion of the spinellids are not available, several chemical analyses of chromite and magnetite, to- gether with observed variations in optical properties of the spinellids, provide general information on their composition. Chromite in the massive chromitite layers and pods ranges in composition from near that of the chromite end member of the spinel series to composi- tions considerably higher in iron and lower in chromium. A chemical analysis of massive chromite (sample A—BM—29, analysis 41, table 1) yields a cal- culated formula composition, corrected for serpen- tine and olivine, of (Fe + 20.49Mg0.43Mn0.05) (cr1.56Fe+ 20,20Al-025) 04 - The atomic ratio of Cr/Fe is about 2.3. A partial chemical analysis of material from a small chromite- magnetite vein (sample A—BM—53—116, analysis 42, table 1) indicates an average atomic ratio of Cr/ Fe of about 0.8. About one-fourth of the material was magnetite, so for the chromite mixture, the ratio of Cr/ Fe is near 1.1. The total range in composition of the chromite in the layers and pods is therefore at least as great as is indicated by these analyses, pos- sibly greater. Disseminated chromite in the dunite, peridotite, and serpentinite has a range in optical properties comparable with that of the analyzed specimens and probably has a similar range in com- position. The translucent reddish-brown grains prob- ably approach A—BM—29 in composition. The inner opaque rims and the larger opaque grains probably have an Cr/ Fe ratio as low as or lower than that of A—BM—53—116, and appreciably less Mg than A—— BM»29. All the magnetite appears to contain small amounts of chromium, aluminum, and magnesium, and lesser amounts of titanium and manganese. An analysis of magnetic concentrate from the tramp- iron magnet (sample VT—24, analysis 20, table 1), which consisted chiefly of magnetite but contained a small proportion of admixed chromite, is repre- sented by the following formula: (Fe+20,ssMgo.15) (Fe+31,85Alnflficr0fll) 04; an analysis of scaly aggregates of magnetite (sample A—BM—56—6, analysis 47, table 1) from schistosity surfaces in serpentinite is represented by the follow— ing formula: (Fe+20.$)0Mg0.()1) MnoplTiobi) (Fe+31.84CI‘0,00A10.09) O-l' SULFIDES AND SULFARSENIDES Sulfides and sulfarsenides occur sparsely in the dunite, peridotite, and serpentinite. They occur chiefly as isolated elongate blebs and irregular grains, but some are intergrown with magnetite. The sulfides are predominantly pale brassy yellow in reflected light. Most appear to be pyrite (Fesg), but some may be pyrrhotite (Fe1_,,S) . The sulfarsenides are silvery gray to grayish blue in reflected light. Specific identification of the min- erals was not made; elsewhere in Vermont, Clemmer and Cooke (1936, p. 12) identified gersdorffite in the ultramafic rocks. Probably, therefore, gers- ULTRAMAFIC AND ASSOCIATED ROCKS 37 dorflite, and perhaps arsenopyrite, are the principal sulfarsenides in the ultramafic rocks. AMPHIBOLE Two varieties of amphibole occur in the ultra- mafic rocks: anthophyllite as an alteration product of pyroxene, and amphibole asbestos in slip-fiber veins. Anthophyllite is common but sparse in the more pyroxenic dunite, peridotite, and massive ser- pentinite. In a few thin sections, many of the grains of anthophyllite surround relict cores of pyroxene; most grains, however, contain no relict pyroxene but preserve the form and cleavage of pyroxene grains which they have replaced. Many grains of antho- phyllite are slightly to extensively altered to serpen— tine. The anthophyllite is fibrous in habit. It is com- monly cloudy and composed of an aggregate of small grains, so that optical properties are difficult to determine. The color in thin section is pale clove brown. The optic axial angle, 2V, is about 90°; in- dices are about 1.625—1.632; and extinction is paral- lel to the fiber direction. Positive identification by optical methods is uncertain, but the mineral prob- ably is a magnesian anthophyllite. On the basis of Rabbitt’s (1948, fig. 4, p. 295) chart, it contains about 5 percent combined FeO+Fe20;,+Ti02+MnO, which corresponds to a formula composition in which the ratio of Mg/Fe+2 is on the order of 6. The amphibole asbestos was identified as tremo- lite—actinolite by George T. Faust (written commun., 1951). SERPENTIN E Serpentine constitutes an appreciable proportion of even the least serpentinized dunite and peridotite and is the principal constitutent of the massive ser- pentinite, the schistose serpentinite, and the serpen- tine veins. The habit and associations of the serpen- tine are many and varied. Fundamentally, individual particles are either fibrous or platy, but each kind has a wide variety of textural relations and habits. All the serpentine is moderate yellow green (5GY 6/2) to dusky yellow green (5GY 7/2) in hand specimen and colorless in thin section. Massive ma- terial commonly is darker because of abundant finely disseminated magnetite. Asbestiform serpentine is pale green to almost white when the fibers are sep- arated and flufi‘ed. On the basis of optical properties determined by the authors, and DTA and X-ray studies made by G. T. Faust (written commun., 1961), all or nearly all the platy serpentine in the dunite, peridotite, and serpentinite—that is, exclusive of serpentine veins and adjacent alteration zones in the dunite and mas- sive serpentinite—is identified as antigorite, and all the fibrous serpentine, as chrysotile. Some nearly isotropic serpentine apparently of platy habit (so- called serpophite) may be six-layer orthoserpentine. Some serpentine identified as antigorite appears crudely fibrous, but under highest magnification it proves to have a fundamentally platy habit; the ap- parently fibrous structure results from an excellent cleavage in very regular laths or flakes, or from the wispy habit. All the serpentine in the asbestos veins and the picrolite veins is chrysotile. The serpentine in the zones of marginal alteration bordering the serpentine veins is predominantly a mixture of lizardite and chrysotile, and, locally, minor anti- gorite. Platy serpentine is the most abundant type of ser— pentine mineral; it ranges in particle shape from bladed, lathlike, or flaky to subequant grains having very irregular boundaries, and in size from blades as much as 1—2 mm long to grains, flakes, and shreds as small as 0.005 mm in maximum dimension. In the relatively unserpentinized dunite and peridotite, large blades of platy serpentine commonly occur at and parallel to grain boundaries, and along cleavage and other crystallographic directions of the olivine. Where the olivine has a lattice-oriented fabric, the blades of platy serpentine are commonly parallel. Fine-grained platy serpentine commonly invades the olivine irregularly along grain boundaries and frac- tures. Platy serpentine that has replaced pyroxene and anthophyllite generally has a parallel or recti- linear arrangement of bladed particles. Large blades of platy serpentine commonly form diversely oriented, parallel, or reticulate patterns in a groundmass of fine-grained platy serpentine. In places, the coarse blades form flamboyant sheaves. Patches of platy serpentine pseudomorphic after pyroxene or anthophyllite have a distinctive parallel or reticulate pattern. In some samples, coarse blades of platy serpentine form a network that encloses sub- equant patches of fine-grained platy serpentine, or of very low birefringent serpentine that is mostly of platy habit, but that may be in part composed of diversely oriented submicroscopic fibers. The net— work of large blades has the same pattern as the grain boundaries of olivine in the dunite. Fibrous serpentine is generally markedly subordi- nate to platy serpentine but locally constitutes as much as 30—40 percent of the matrix serpentine— 38 ASBESTOS-BEARING ULTRAMAF‘IC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT that is, exclusive of macroscopic asbestos veins. In one thin section, all the matrix serpentine appeared to be fibrous. Much of the disseminated fibrous ser- pentine occurs in microscopic tabular to lenticular veinlets of cross-fiber asbestos, but in many thin sections an appreciable proportion forms wispy bundles and sheaves of parallel fiber not obviously veinform. All have a silky luster and parallel arrangement of the fibers. In many or most thin sections, the apparently nonveinform patches may result from the thin section just grazing the surface of tiny lenses of cross-fiber asbestos. That this is commonly the case is suggested by the observation that many of the isolated sheaves and bundles of fibrous serpentine give biaxial positive interference figures with small to moderate optic axial angles; such would be the result where the thin section is nearly parallel to the vein wall, at a large angle to the fiber axes. (See section on “Mineralogy of the serpentine group” under “Ultramafic and associated rocks”) However, in the samples very high in, or entirely of fibrous serpentine, the fibers cannot be veinform. Some patches of very low birefringent serpentine on which an interference figure cannot be obtained may be composed of diversely oriented, microscopically indistinguishable fibers, but most are of platy serpentine. Admixtures of fibrous serpentine modify the tex- tural patterns of platy serpentine in varying degree. The fibrous serpentine is distributed in tiny tabular veinlets that reflect the control of grain boundaries, fractures, and cleavage; in tiny lenticular veinlets; and in wispy sheaves and bundles. One thin section in which the matrix serpentine is entirely fibrous consists of an irregular interlocking mosaic of sub- equant wispy patches of parallel fibers. The appar- ent birefringence of the patches ranges from yellow to dark gray, depending upon the orientation of the fibers with respect to the thin section. Bundles whose fiber axes are at a large angle to the thin section have much the same appearance as very fine gran- ular antigorite. Fibrous serpentine in asbestos veins forms fine silky fibers that have uniformly parallel arrange- ment. In picrolite veins, it forms bundles and sheaves of slightly disarrayed and commonly less distinct fibers; the bundles and sheaves are nearly parallel and impart a coarsely fibrous or columnar structure. 2 In the massive serpentinite, the platy and the fibrous serpentine combine in an almost infinite variety of textures, but each is characterized by various combinations of a few basic textures. Small blades, laths, and elongate shreds generally form a felted mass of diversely oriented particles; not uncommonly, a few of the blades and laths form a rectilinear pattern in a felted matrix. Coarsely flaky particles of serpentine range from diversely oriented to subparallel, and in thin section present a variable, mosaic to schistose texture. Under crossed nicols, subequant granular particles of platy serpen- tine have a texture like that of galvanized metal. In particular specimens, either the felted texture or the “galvanized” texture may predominate almost to the exclusion of the other, or both may characterize the rock in varying degree. Textures characteristic of flaky particles commonly characterize an entire sam- ple, but some samples contain small admixtures of flaky serpentine dispersed throughout bladed or sub- equant particles. AN'mxnurr All the antigorite is generally similar in optical properties but varies appreciably within rather well—defined limits. Abnormal blue interference colors are characteristic. They range from very pale blue (53 8/2) to moderate blue (53 5/6) ; most are pale blue (near 53 6/2). Antigorite next to halos of chlorite that rims grains of chromite is deeper blue than the surrounding antigorite. The optic axial angle, 2V, ranges from 0° to 65°, but is predomi- nantly less than 20°. The small apparent optic axial angle may result largely from superposition of layers having random orientation about the c-axis. The optic sign is (——), and the sign of elongation (+). The [3 index of refraction ranges from 1.562 to 1.573 but is predominantly in the range 1.567—1.570. The a and y indices can seldom be measured reliably, but minimum a’ and maximum 7’ indices are seldom more than 0.003 smaller and larger than the ,8 index. The apparent birefringence ranges from 0.002 to 0.009 and is generally about 0.003-0.005. Analyses of pure antigorite were not obtained, but analyses were made of several samples of rock con- sisting principally or almost entirely of antigorite. By correcting for contaminating minerals whose compositions and proportions in the samples are fairly accurately known, the approximate composi- tions of the antigorite can be calculated. The calcu— lated formula compositions of the antigorite in these samples are tabulated below. ULTRAMAFIC AND ASSOCIATED ROCKS 39 Formula composition of antigorite calculated from chemical analyses of serpentim'te [Corrected for estimated contaminants of magnetite, chlorite. and carbonate] AV—81m (VT—4) , analysis 3, table 1 : (Mg5.6Fe+2o.o4Fe+so.1) (A10.ISi3.9) 09.!) (OH) 7.6) AR—73 (VT—6) , analysis 4, table 1: (Mg5.3Fe+20.02Fe+30.2) (A10.ZSi3.8) 09.9 (OH) 7.2 AV—127 (VT—8) , analysis 5, table 1 : (Mg5.oFe+20.4) (A10.1Si3.9) 09.8 (OH) 7.2 AV—209 (VT—16), analysis 12, table 1: (Mg5.2Fe+20.4) (AIOISi39) 010.0 (OH) 7.2 A—BM—53—12, analysis 21, table 1. (Mg4.sFe+20.3) '(A10.1Si3.9) 010.2 (OH) 6.3 A—BM—53—1, analysis 30, table 1 : (Mg5.2Fe+20,1Fe‘+30.04) (A10.1Si3.9) 09.7 (0H) 7.:: AV~127, analysis 55, table 1: (Mg5.oFe+20.4Fe+3o_05) (Alo.1Si3.9) 010.0 (OH) 7.2 CHRYSOTILE Structural varieties of chrysotile cannot be distin- guished from one another by optical properties. In the DTA and X-ray studies, clinochrysotile was spe- cifically identified in several samples; in other sam- ples containing chrysotile, the specific structural type of chrysotile was not identified. Clinochrysotile predominates greatly over orthochrysotile in world- wide occurrence. Orthochrysotile was not identified from deposits in Vermont, but specimens from Thet- ford, Quebec, contain as much as 7 percent ortho- chrysotile (Whittaker, 1956b, p. 862; Whittaker and Zussman, 1956, p. 114—115). Therefore, orthochry- sotile is probably present in small amounts at Belvidere Mountain, but subordinate to clinochry- sotile. All the chrysotile is colorless in thin section, and most shows normal first-order yellow birefringence colors. Some groundmass chrysotile in rock com— posed largely or entirely of chrysotile has low appar- ent birefringence that has a faint abnormal blue tinge. The anomalously low birefringence and ab- normal color may be caused by superposition of wispy layers of fibers having different orientations. Optical properties of the chrysotile are distinctive, but they vary appreciably in different specimens. The variable optic axial angle, depending upon the orientation of the thin section with respect to the fiber axes, and the positive optic sign, are distinctive of chrysotile (see section on “Mineralogy of the ser- pentine group” under “Ultramafic and associated rocks”). Another distinctive optical property is the low index relative to that of associated antigorite. The greater index of refraction, N, probably is gen- erally near true y; it ranges from 1.538 to 1.562, but is predominantly in the range 1545—1555 and is always less than the lowest index of associated anti- gorite. The smaller index of refraction, n, is of less fundamental significance, because it represents some average between a and [3. Measured values range from 1.518 to 1.555 and are predominantly in the range 1.538—1.548. The highest measured birefrin- gence was 0.020, but generally it is in the range 0.010—0.015. Extinction is parallel, and the sign of elongation (+) is parallel to the fiber axes. Two chemical analyses of chrysotile fiber were obtained, one of milled fiber (sample A—BM—57-1, analysis 50, table 1), and one of long silky fiber intergrown with fibrous calcite (sample AR—55, analysis 56, table 1). All rock particles and nearly all the magnetite were removed from the material in analysis 50 (A—BM—57—1) ; so the analysis is of vir- tually pure chrysotile. In analysis 56 (AR—55), the removal of all the magnetite from the material left a mixture of almost pure calcite (80 percent) and chrysotile (20 percent). A reliable formula com- position of the chrysotile was therefore obtained by subtracting from the analysis amounts Ca, Mg, and Mn equal to the content of 002. The calculated formula compositions are: A—BM—57—1: (Mgr;.02Fe+20.onFe+3o.13) (Si3.s7A10.12Fe+3o.m)09.80(OH) 9.77- AR—55: (Mgc_12Fe+20.o7Fe+3o.1o) (SixsoAlonoFe+30.02) 09,04 (OH) 9.15- Analyses of picrolite that is composed essentially of chrysotile yield the following calculated formula compositions : AR—52 (VT—15), analysis 11, table 1 : (Mgt;.oFe+20.ozFe+30,2A10.2) (Sili.tiAI(l.}Fe+30.1)011.0(OH) K."- AR—64, analysis 54, table 1 : (Mg5.3sFe+zn.olFel—xnpsAlom) (Sine-{Alana 010.0 (OH) 7. :~ AR—3x, analysis 10, table 1 : (Mgs..\nFe+20,01Fe+30.05(Si:x,94A10,02Fe+30.01) 09.8 (OH) 8.1- Analysis 10 agrees reasonably closely with the ideal formula; analysis 54, somewhat less so. Sam- ple AR—52, analysis 11, contained appreciable but undetermined amounts of brucite and magnesite, which accounts for the high content of Mg. Despite their limitations, the analyses do indicate a low con- 40 ASBESTOS-BEARING ULTRAMAFIC‘ ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT tent of Fe”, Fe”, and Al in the chrysotile in picro- lite veins. LIZARDITE Lizardite is generally indistinguishable from anti- gorite in habit and optical properties. None was dis- tinguished optically, but it was identified by X-ray and DTA studies in the zones of marginal alteration (George Faust, written commun., 1961). The platy serpentine in these zones of marginal alteration, which is the lizardite, has the general appearance and optical properties of antigorite. Analyzed samples (table 1, analyses 2 [AR—13f, VT—2] and 52 [A—BM—57—3]) of rock from alter- ation zones bordering chrysotile veins consisted al- most entirely of lizardite and minor amounts of chrysotile (identified on the basis of DTA and X-ray analysis), brucite, carbonate, and magnetite. Most of the magnetite was removed from sample AR—13f (VT—2) before analysis, leaving as significant con- taminants only brucite and magnesite, Sample A— BM—57—3 contained microscopically determined amounts of brucite, magnesite, and magnetite. Thus, for each analysis, the formula composition of the serpentine mixture could be determined with con— siderable accuracy. The calculated formula composi- tions, therefore, represent rather closely the compo- sition of the lizardite. They are: AR-13f (VT—2) : (Mg5.azFe+2o.02 Ni0.02Fe+30.10) (Si:z.ssAlo.osFe3+o.oc)09,71(AH)s.1c- A—B M—5 7 —3 : (Mg5.74Fe+20.01Nio.03Fe+3o,21) (Si3.79A10.16Fe+30.05) 010.04 (OH) 7.90- SIx-LAYER ORTHOSERPENTINE No six-layer orthoserpentine was positively iden- tified in the ultramafic rocks of the Belvidere Moun- tain area. The nearly isotropic serpentine locally abundant in the dunite and massive serpentine and tentatively identified as possibly six-layer orthoser- pentine has indices similar to those of antigorite, negligible birefringence, and no discernible struc- tural habit. No information is available on the chem- ical composition. CHLORITE Chlorite is a common accessory mineral of the ultramafic igneous rocks and serpentinite. It gener- ally constitutes less than 2 percent, and rarely as much as 5 percent of the rock. Most of the chlorite is associated with chromite and the larger grains of magnetite, forming complete or partial coronas which separate the grains from surrounding anti- gorite. In some of the schistose serpentinite, the chlorite occurs in isolated flakes or irregular patches of flakes adjacent to the spinellid grains, but not in distinctive coronas. In a few thin sections, chlorite is interlayered with antigorite in pseudomorphs after pyroxene. All the chlorite has similar optical properties: 2V =0°—20°, optic sign (+), sign of elongation (—), abnormal brown to pale-green and yellow-orange interference colors, fi=1.565-1.587, birefringence very low. On the basis of these optical properties, the‘. chlorite is inferred to range in formula composition from about (MngeHMAlw) (Si3,3Alo.7)Om(OH)2 to (MngeHMAlm) (SimAlLo) O10 (OH) 2. (See Chidester, 1962, p. 44—46, fig. 14.) BRUCITE Brucite is a common constituent of the ultramafic rocks and associated veins. It is generally a minor component of the slightly serpentinized dunite and peridotite and locally a major component of the ser- pentinite, in which it occurs as disseminated flakes and irregular aggregates of flakes. It is locally inter- grown sparsely with chrysotile in cross-fiber asbestos veins and is a major component of many veins of columnar picrolite. Brucite is an essential constit- uent of the marginal zones of alteration that border cross-fiber asbestos and cross-column picrolite veins and is commonly associated with magnetite and cal- cite in veinform aggregates. The brucite ranges in habit from flaky to fibrous. Observed optical properties are: uniaxial, optic sign (+), sign of elongation (—), w=1.566—1.573, bire- fringence~0.11. These optical properties indicate a composition near that of the end member, Mg( OH) 2, having at most only a small content of Fe+2. CARBONATES Magnesite, calcite, and dolomite are almost ubiq— uitous accessory constituents of the serpentinized ultramafic igneous rocks and of the several kinds of veins, and not uncommonly are major components. Generally only one species is present in a single hand specimen. Where more than one is present, the differ- ent species are generally distinguished by different habits or associations. ULTRAMAFIC AND ASSOCIATED ROCKS 41 Magnesite, the most abundant carbonate mineral in the ultramafic rocks, is generally sparse or absent in the least serpentinized dunite and peridotite but constitutes as much as 20 percent of massive and schistose serpentinite. The magnesite occurs in the massive serpentinite and partly serpentinized dunite principally as disseminated grains about 1—2 mm across, less commonly as aggregates of small grains. In the schistose serpentinite, it commonly forms small irregular masses and veins and locally is segre- gated in thin lenticles parallel to the schistosity. Measured indices range from 0):]..700 to w=1.705, indicating a composition near pure magnesite, MgCO:;. Calcite is associated principally with veins, in which it varies widely in abundance and habit; it occurs only sparsely in the serpentinite, in the form of dispersed grains and small lenticles. In cross-fiber asbestos veins, calcite is locally intergrown in fibrous or feathery form with the chrysotile or forms irreg- ular patches within the vein. In cross-column and massive picrolite veins, calcite forms irregular, com- monly flamboyant, intergrowths with the chrysotile. In masses of long slip-fiber chrysotile asbestos, finely fibrous calcite—megascopically distinguishable from the chrysotile asbestos only by its brittleness—com- monly forms as much as 50 percent or more of the asbestiform mass. Granular calcite is commonly as- sociated with brucite or calcite, or both, in small veins in the dunite, peridotite, and massive serpen- tinite. Calcite in all the associations shows only a small range in optical properties, from w=1.658 to a): 1.660. These properties indicate a composition near that of pure calcite, CaCO;.. Dolomite occurs sparsely in small veins in both the massive and the schistose serpentinite. Measured indices of w=1.684 indicate the composition of a somewhat magnesian dolomite. GRAPHITE Graphite occurs sparsely to abundantly in a small proportion of the serpentinite. It is confined to a few thin coarsely flaky layers in the massive serpentinite, to coarsely fibrous picrolite veins in such layers, and to relatively minor, but appreciable, volumes of schistose serpentinite. The graphite is disseminated sparsely in irregular cloudy patches in the flakes and shreds of serpentine and is concentrated moderately to heavily at the margins of flakes of serpentine and along slip cleavage surfaces and tiny shear zones in the schistose serpentinite. Thin sections of graphitic massive serpentinite contain 1—19 percent graphite; thin sections of schistose serpentinite, 5—37 percent. Specimens of massive serpentinite that were ana- lyzed chemically for graphite (analyses 25, 26, and 30, table 1) contained 1.4—14.4 percent. All the graphite is very fine grained and may be partly amorphous. X-ray powder photographs of graphitic serpentinite show only the strongest lines of graphite (Fred A. Hildebrand, written commun., Oct. 13, 1955; Nov. 9, 1955). Some of it, however, particularly that concentrated along grain bound- aries and in shear zones, shows flashes off crystal faces and is silvery gray in reflected light. STEATITE, TALC-CARBONATE ROCK, AND CARBONATE-QUARTZ ROCK Steatite, talc-carbonate rock, and carbonate-quartz rock are associated with the ultramafic bodies in dis- tinctive structural and petrologic relations. The three rock types are completely intergradational, but the compositional range and the association of rock types is related to structural setting. Talc-carbonate rock and steatite form concentric shells at the margins of ultramafic bodies, and quartz is very sparse or absent in such bodies. Talc-carbonate rock and carbonate-quartz rock are associated within ultramafic bodies as irregular or lenticular masses whose distribution suggests control by faults or shear zones. In the Eden quarry ultramafic body, the masses labeled “Outc” near grid line 94,000 E. (pl. 1) are predominantly of variably quartzose talc- carbonate rock in which no zonal distribution pat- tern is discernible. Near the south end of the Corez Pond body, carbonate—quartz rock and talc-carbonate rock show a zonal relation; talc-carbonate rock is marginal to a central mass of carbonate-quartz rock. Relations with nearby serpentinite are not exposed, but talc-carbonate rock appears everywhere to inter- vene between serpentinite and carbonate-quartz rock. Grains of chromite and magnetite are sparse relict igneous minerals in all three rock types. Table 4 contains selected modes of the end-member rocks and of some transitional types. STEATITE Steatite forms thin shells a few centimeters to a few meters thick at the outer margins of the small lenses of ultramafic rocks northeast of the Lowell quarry body, west of the Eden quarry body, and near the southeast corner of the Belvidere Mountain area (pl. 1). In the stripped area centered at 100,900 N., 97,200 E. (figs. 5, 11) , along the northeast contact of the Lowell quarry body, thin seams of steatite that 42 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT TABLE 4.—Selected modes of steatite, talc-carbonate rock, and carbonate-quartz rock [Tr, trace] Spe§i3men Talc Gag-11:21:33 Quartz Tremolite Chlorite Sphene Magnetite 3511:1355 Graphite Steatite A—BM—126 ____ 97 __ __ 3 __ __ Tr __ __ A—BM—53—3 ___ 99 __ __ __ ‘ 1 __ __ __ __ T2-6—652 _____ 70 __ __ 30 < 1 __ __ __ __ T2—8—1204 ____ 96 __ __ __ __ __ <1 __ 4 Talc-carbonate rock AV—212 _______ 88 10 __ __ __ __ Tr 2 __ EQ—1—175 _____ 60 40 __ __ __ __ __ Tr __ EQ—5—94 ______ 87 10 __ __ 2 1 __ __ __ T—32—568 _____ 60 40 __ __ Tr __ Tr __ __ T—32—664 _____ 60 40 -_ __ __ __ <1 __ __ Carbonate-quartz rock A—BM—108 ____ 45 45 9 __ __ __ 1 __ __ A—Ch—56—535c _ 1 70 29 _ _ Tr _ _ __ __ __ swell locally into pods as much as 2 m thick extend along bedding surfaces for tens of meters beyond the ends of serpentinite lenses. Steatite is relatively rare in the main ultramafic body. It is exposed at the contacts only near 102,000 N., 94,300 E., and 99,000 N., 92,000 E. (pl. 1). Sev- eral diamond-drill holes in the vicinity of these ex- posures, and at the south end of the Corez Pond body, penetrated as much as 1 m of steatite at the lower contact between the ultramafic body and the country rock. The steatite is uniformly fine grained, commonly schistose but locally massive, and medium gray to light greenish gray on fresh surfaces. Weathered surfaces are pale buff to light brownish buff. In some places, steatite in the outer part of the zone contains streaks and knots of minerals and subtle variations in color and texture that impart a faint layering similar in pattern to that in the adjacent country rock. The steatite consists of a fine-grained aggregate consisting of shreds and flakes of talc predominant- ly 0.01 to 0.1 mm long but locally as long as 0.3 mm. In the schistose steatite, narrow shear zones 0.001 to 0.01 mm wide, within which the particles are uni- formly alined, are separated by zones 0.005 to 0.1 mm wide, in which the particles range from diverse- ly oriented to uniformly parallel to the shear zones. The steatite is composed essentially of talc, but chlorite and dusty magnetite are sparse accessory constituents throughout the rock. Magnesite occurs in the transition zone between steatite and talc-car- bonate rock, and tremolite or abundant chlorite occurs in the zone transitional into tremolite rock or black- wall chlorite rock. Relict grains of the igneous min— erals chromite and magnetite are distinctive of the inner and major part of the steatite zone. Relict grains of sphene, apatite, and, rarely, other minerals of the country rock are distinctive of a narrow outer part of the steatite zone, particularly where relict layering is preserved. TALC-CARBONATE ROCK Talc-carbonate rock constitutes the central core, or forms an envelope from a few centimeters to sev— eral tens of meters thick about a central core of serpentinite, in the small ultramafic lenses northeast of the Lowell quarry body (pls. 3 and 4D), west of the Eden quarry body, and in the southeast part of the map area (pl. 1). Talc-carbonate rock also forms large irregular masses in the Eden quarry body and the Corez Pond body. The talc-carbonate rock consists of a fine-grained groundmass of talc throughout which are dispersed abundant single crystals and aggregates of mag- nesite. The porphyroblasts of magnesite range in size from less than 1 mm to as large as 3 cm and impart a faint mottling to the rock, which varies from medium gray to light greenish gray. Weathered surfaces are rusty reddish brown. The rock is pre- dominantly massive but locally has a crude spaced schistosity. Some of the massive talc-carbonate rock has faint layering resembling that of the dunite and massive serpentinite. The composition of the talc-carbonate rock varies widely, on a small scale throughout the rock body and on a large scale in transition zones into serpen- tinite, carbonate-quartz rock, and steatite. Outside the transition zones, representative samples contain about 60 percent talc and 40 percent magnesite. ULTRAMAFIC AND ASSOCIATED ROCKS 43 Relict igneous minerals and dusty magnetite com- monly constitute less than 1 percent of the rock, and chlorite is locally present in small amounts. CARBONATE-QUARTZ ROCK Carbonate-quartz rock forms an appreciable mass only at the south end of the Corez Pond body, near the center of a larger mass of talc-carbonate rock. Quartz occurs erratically in small to moderate amounts in the irregular bodies of talc-carbonate rock in the Eden quarry body and in the small pod of ultramafic rock near 91,500 N., 101,000 E. (pl. 1). The carbonate-quartz rock is medium to light gray or greenish gray and generally retains the gross ap- pearance of the serpentinite from which it was derived. Quartz forms a mosaic of anhedral grains 0.02 to 0.2 mm across, throughout which anhedral grains, rhombs, and aggregates of magnesite 0.1 to 2 mm across are irregularly interspersed. In the typical specimen, magnesite constitutes about 70 per- cent of the rock and quartz, 30 percent. Grains of chromite and magnetite occur as sparse relict igne- ous minerals. Small amounts of dusty magnetite and shreds of talc are common accessories. MINERALOGY AND PARAGENESIS Steatite, talc-carbonate rock, and carbonate-quartz rock are composed chiefly of talc, magnesite, and quartz that have characteristic felted and interlock- ing mosaic textures. Relict minerals of the preexist- ing rocks are volumetrically negligible but are diag- nostic of the metamorphic origin of the steatite, talc- carbonate rock, and carbonate-quartz rock. Granular magnetite, chromite, pyrite, and sulfarsenide are relict minerals inherited from the igneous rocks. Dustlike particles of magnetite, chlorite associated with magnetite grains, and serpentine are relict min- erals of the serpentinite. Chlorite and sphene in the outer part of the steatite zone are relicts of incom- plete alteration of the blackwall chlorite rock. Graphite is a relict mineral of the graphitic schist, amphibolite, and serpentinite. TALC Talc occurs in shreds and flakes generally less than 0.5 mm long. The optical properties are generally uniform. The optic axial angle, 2V, is small to mod- erate. The B-index of refraction varies through a small range, from about 1.588 to 1.591. Maximum interference colors indicate a birefringence of about 0.045. Chemical analyses of talc similar in optical prop- erties and geologic occurrence elsewhere in Vermont indicate a composition of about (Mg5.GFe+20.3A10.1) (Si7.9A10.1) 020 (OH) 4 (see Chidester, 1962, p. 79—80). The talc in the Belvi— dere Mountain area is inferred to have about the same composition. CARBONATE Only magnesite was noted in the talc-carbonate rock and carbonate-quartz rock, but calcite and dolo- mite in small amounts may possibly exist as relict minerals of the serpentinite. The magnesite occurs as anhedral to rhombic grains as much as 1 mm across and aggregates as large as 3 cm. The w-index of refraction ranges from 1.700 to 1.720. Analyzed carbonate minerals that have a simi— lar range in index from ultramafic rocks elsewhere in Vermont range in composition from about pure magnesite, MgCOg, to about (030.01Mg0.scFe+20.13) CO3. QUARTZ Quartz in the carbonate-quartz rock is uniformly fine grained and uniformly shows normal extinction. In rock containing little or no talc, inclusions are relatively rare except for magnesite grains enclosed in aggregates of quartz grains. In rock containing moderate to abundant talc, quartz grains and ag- gregates commonly enclose numerous shreds of talc as well. OTHER RIINERAIS Relict grains of chromite, magnetite, and sulfides and sulfarsenides differ little from those in the dunite, peridotite, and serpentinite (see section on “Chromite and magnetite” under “Mineralogy and paragenesis” in the section on “Igneous rocks, ser- pentinite, and veins”) and are inferred to have changed little or not at all in composition. Locally, grains of magnetite and chromite appear to have been replaced slightly by carbonate. Chlorite is a sparse relict of serpentinite or occurs in steatite transitional into blackwall. The optical properties and inferred composition are like those of chlorite in serpentinite (see section on “Chlorite” under “Mineralogy and paragenesis” in the section on “Igneous rocks, serpentinite, and veins”) or in the blackwall (see section on “Chlorite” under “Min-- eralogy and paragenesis” in the section on “Contact rocks”). 44 Graphite persists as a relict mineral only rarely in thin shear zones and along schistosity surfaces. Sphene is a rare relict mineral inherited from schist or amphibolite. CONTACT ROCKS Each of the three main associations of contact rocks—rodingite and serpentinite-chlorite rock, stea- tite and blackwall chlorite rock, and tremolite rock and chlorite rock—conforms generally to broad regu- lar patterns of interrelations and mineral assem- blages but varies considerably in detailed relations, in the proportions of major minerals, and in the identity and proportions of accessory minerals (see beginning discussion in section on “Ultramafic and associated rocks”). As the ultramafic bodies are generally conform- able with bedding schistosity in the enclosing coun— try rocks, the boundaries of the contact rocks also generally conform grossly with the bedding schis- tosity. Where the ultramafic bodies crosscut the bed- ding and where the configuration of the contact rocks is controlled by joints and shear zones, the boundaries of the contact rocks are grossly crosscut- ting; where the boundaries of a contact rock are irregular and the amplitude of the irregularity is appreciably greater than the thickness of bedding in the adjacent country rock, the contacts are crosscut- ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT ting in detail. At such places, wherever layering can be traced from one rock type to another, there is no discernible difference in thickness of layers in the different rock types, nor is there discontinuity in the bedding surfaces at the contacts between rock types. Table 5 contains selected representative modes of contact rocks. Table 6 contains modes of selected suites of specimens across contacts between ultra- mafic rocks and several kinds of country rock; it is intended to illustrate the variable and complex com- positional relations of contact rocks. THE RODINGITE AND SERPENTINE-CHLORITE ROCK ASSOCIATION Rocks of the rodingite and serpentine-chlorite rock association border serpentinite at contacts with the country rocks and at the margins of inclusions throughout almost the entire LOWell quarry body, in much of the Eden quarry body, and in probably a smaller proportion of the Corez Pond body. The as- sociation occurs both where the adjacent country rock is schist and gneiss, such as at the northeast margin of the Lowell quarry body, and where it is amphibolite, which borders much of the main ultra- mafic body elsewhere. The serpentine-chlorite rock zone is immediately adjacent to the serpentinite; the rodingite zone is adjacent to the unaltered country rock. TABLE 5.—Selected modes of contact rocks Carbon- . . Vesu- Epidote . Sulfide , Specimen DIOD- . Gar- , - Preh- Serpen- Chlo- Tremo- BIO- ate ADa- . Ilme- Magne- . _ Gia- No. side :1: net 0:02:52:- nite tine rite lite Talc tite 2:3; tite Rutile Sphene nite tite $35 phite A. Rodinglte TPT—Vt—IO ___ 29.2 -- 21.6 25.0 __ __ 19.6 __ __ __ ,_ __ __ 4.3 0.3 __ -_ __ __ 50 30 __ _ __ 20 __ __ ,_ T1' T1' __ __ __ __ __ 31.0 4.6 3.3 49.5 -1 __ 7 1 __ __ __ 0.3 __ __ 4.2 __ __ -_ -_ 10 4 '70 __ __ __ 15 __ __ __ __ __ __ ] __ __ __ -_ 40 T1 33 24 __ __ __ __ __ -_ __ __ __ 2 __ -_ 1 -- 65 __ 30 5 __ __ __ __ __ __ __ __ __ Ty __ __ Tr _- 89.4 __ 3.3 .3 _ _ 5.4 __ __ 1. __ __ __ Tr __ __ . __ 41.7 __ 10.4 32.4 2.0 _ __ __ ,- 8.2 5 1 __ __ .2 __ __ ._ ,_ 3.8 31.2 45.4 -_ __ __ 19.6 __ __ _ -_ __ __ __ __ __ __ _- 35 40 25 _ __ __ __ __ __ __ __ _- __ __ __ __ __ __ B. Serpentine-chlorite rock TPT—Vt—ll ____ __ __ __ __ __ 23.7 67.4 __ fl __ __ -- __ __ 8.9 __ -_ _, AR—20 ........ __ __ __ __ __ 51.1 45.5 __ __ __ __ __ __ __ 3.4 __ __ __ AV—5a ________ __ -_ __ __ ,_ 60 30 e- __ __ ,_ 2 __ _- 8 _- -- -— AV—90 ........ __ __ __ ._ __ 48.8 50.3 __ _- __ __ __ _- -. .9 __ _- ._ AV—238 _______ 4.0 __ __ __ 94.1 .3 __ __ __ __ __ __ __ __ 1.6 __ _. __ AV—241 _______ 1.1 __ __ __ 90.3 3.9 __ __ __ __ __ __ __ __ 4.7 __ .- __ C. Blackwall chlorite rock and tremolite-chlorite rock A—BM—82 ______ __ __ __ H __ __ 97 ,_ __ M -_ ,_ 1 -- _- __ __ 2 A—BM—109 _____ __ __ __ __ __ __ 97 __ ,1 __ __ __ 3 __ __ __ __ __ A—BM-53—41 __ __ __ __ 2 _z __ 88 __ __ __ __ __ -_ 9 __ __ .. _- EQ—1-740 _____ __ __ __ 1 __ ‘_ 89 __ __ __ __ T1 __ 10 __ __ -_ -- T—67—1004c ____ .._ __ __ <1 __ _.. 97 __ _ -_ ._ 1 __ 2 __ __ AV—206 _______ __ __ __ __ __ __ 73.7 7.4 ,_ 8.6 __ 4 __ 9.9 __ -_ -- __ A—BM—140 _____ __ __ __ __ __ ,_ 25 7o 5 __ __ __ __ __ __ __ __ -_ A—BM—53—57 __ __ __ __ __ __ __ 20 30 __ __ __ __ __ __ __ __ _z 50 AV—2o4 _______ __ __ __ __ __ -- 11.9 88.1 _- __ __ __ -_ __ __ _ __ __ AV—205 _______ __ -_ __ __ __ __ 6.3 93.6 __ -_ __ -_ __ ,z __ 0.1 AV—203 _______ __ _- -_ _- -- n 5.0 94.8 __ __ __ __ __ _- -_ __ .2 __ A—BM—130 ____ ,_ __ __ ,- __ __ 1 99 __ -- __ __ -_ __ __ __ __ __ ULTRAMAFIC AND ASSOCIATED ROCKS Both the rodingite and the serpentine-chlorite rock show variations related to structure or to lithology of the adjacent country rock or to both. The rodingite is uniformly about a meter thick where the adjacent country rock is schist or gneiss in conformable con- tact with the ultramafic body. Where the adjacent country rock at conformable contacts is amphibolite, the rodingite is commonly only a few centimeters or decimeters thick. Serpentine—chlorite rock, on the other hand, shows no significant variations in thick- ness related to lithology, ranging generally from 2 to 8 cm thick at conformable contacts. The rodingite is conspicuously thicker—in several places as much as 3 m—in both amphibolite and schist or gneiss where the contact of the ultramafite is strongly crosscutting and where septa and inclusions of country rock in “ the ultramafic body arehighly sheared and frac- tured. The serpentine-chlorite rock varies less markedly in thickness but is noticeably thicker at some structural irregularities, particularly where thin tongues of serpentinite penetrate along bedding planes of country rock at crosscutting contacts; such composite zones are as much as 1 m thick. Along joints that intersect the ultramafic contact, the ser- pentine-chlorite rock zone tapers from 5 cm thick to zero. In addition to such variations in thickness, the rodingite shows conspicuous variations in texture and grain size related to joints and shear zones. Contacts reflect the intrusive or replacement rela- tions of the associated rocks. The contact between serpentine-chlorite rock and serpentinite is generally regular and sharp but is commonly difficult to pin- point because of the similar appearances of the two rock types. In many places, the contact is a shear sur- face; elsewhere, the junction of the two rocks is tight. The contact between serpentine-chlorite rock and rodingite is very irregular and is gradational over a very short distance—commonly less than 1 mm (see fig. 7). The contact between rodingite and country rock is variable. In most places, the grada- tion takes place over a distance of 2 cm or less. In some places, particularly in the coarse amphibolite, the rodingite grades irregularly into unaltered coun— try rock over a distance of several centimeters to several decimeters; not uncommonly, such grada- tional zones contain an unusually large proportion of chlorite. In a few places, clear-cut zonal relations of con— tact rocks are not apparent. At the contact between serpentinite and coarse amphibolite at the west edge of the Eden quarry, no well-defined serpentine— chlorite rock and rodingite zones border the ser- pentinite. (fig. 4). Rather, a single zone, variable 45 FIGURE 7.——Photograph of contact between serpentine-chlorite rock and rodingite, northeast side of Lowell quarry. The rodingite, under the hammer, is pale buff, the serpentine- chlorite rock dark bluish green. in width, consists of variable proportions of diop- side, chlorite, and relict minerals of the amphibolite. Along the southwestern margin of the Lowell quarry body, serpentine-chlorite rock projects in narrow fingers beyond the rodingite zone along many of the numerous joints and small faults that intersect the contact of the ultramafic body; in a few places, the serpentine-chlorite rock is bordered at the outer margin by amphibolite, without an in- tervening zone of rodingite. Irregular veins of coarse chlorite, calcite, and magnetite are associated with the rodingite and serpentine-chlorite rock in a few places at the east- ern edge of the Eden quarry body and at the south- western end of the Lowell quarry body. Most such veins are in the rodingite and serpentine-chlorite rock but some penetrate into the serpentinite for a few centimeters. The veins appear to be largely fracture controlled, but in detail the boundaries are commonly slightly irregular and locally very ir- regular. RODINGITE Rodingite occurs in two distinctive textural varie— ties, each in a distinctive structural setting. That in structurally simple settings is uniformly fine grained and of dense flinty appearance. That in and im- mediately adjacent to highly fractured and sheared zones is coarser grained and more varied in ap- pearance. At the margins of the highly fractured zones, the coarse rodingite grades irregularly into 46 ASBESTOS-BEARING ULTRAMAFIC ROCKS 0F BELVIDERE MOUNTAIN AREA, VERMONT TABLE 6.—Modes of selected suites of specimens ITr, Distance (cm) into th(e ugtramafi: god); , — , m o e ‘ , _ Specimen No. contact? rock and Setlifign- (21?;- Mafige- Talc Tiieéno- (32::- country rock (+), measured normal to the contact A. Contact of the rodingite and serpentine-chlorite rock type in graphitic schist of the Hazens Notch AV—5a—1 _______ Serpentine-chlorite rock _________ + 6 60 _ _ _ 8 _ _ _ _ _ _ 30 AV—5a—2 _______ Rodingite ______________________ +7 _ __ _ _ _ __ _ _ _ _ _ _ _ 15 AV—5b _____________ do _________________________ +9 ___ ___ ___ ___ ___ 3 AV—5c _____________ do _________________________ +12 ___ ___ ___ ___ ___ ___ AV—5d _____________ do _________________________ +21 ___ ___ ___ _-_ ___ Tr AV—5e _____________ do _________________________ +28 ___ ___ ___ ___ ___ 3 «AV—6 ______________ do _________________________ +152 ___ ___ ___ ___ ___ ___ B. Contact of the rodingite and serpentine-chlorite rock type in graphitic schist of the Hazens Notch formation AV—131a _______ Rodingite ______________________ +9 ___ ___ ___ ___ ___ ___ AV—131b ___________ do _________________________ +91 ___ ___ ___ ___ ___ ___ AV—131c _______ Transition rodingite-schist ______ +167 ___ __ _ 3.7 _ _ _ _ __ _ _ _ AV—131d _______ Homblendic graphitic schist _____ +229 ___ ___ ___ ___ ___ 4.2 C Contact of rodingite and serpentine-chlorite rock type 1n coarse amphibolite of the Belvxdere Mountain A—BM—117a ____Roding1te ______________________ +3 ___ ___ Tr ___ ___ 8 A—BM—117b ________ do _________________________ +9 ___ ___ ___ ___ ___ 95 A— BM— 1176 ________ do _________________________ +24 ___ ___ __- ___ -__ 73 A—BM—117d ________ do _________________________ +30 ___ ___ ___ ___ ___ 22 A— BM— 118 _____ Amphibolite ____________________ +122 ___ ___ ___ ___ -_- 6 D. Contact of the rodingite and serpentine-chlorite rock type in fine amphibolite of the Belvidere Mountain [ Serpentine-chlorite rock containing irregular lenses AV—243 ________ Serpentinite ____________________ _1 93.8 _ _ _ 6.0 __ _ ___ ___ AV—242 ________ Rodingite ______________________ +6 ___ ___ -_- ___ ___ 19.6 AV—241 ________ Serpentine- chlorite rock _________ +15 90.3 ___ 4.7 ___ ___ 3.9 AV— 240 ________ Serpentinized rodingite __________ +30 88.3 _ -_ .9 __- _ __ .3 AV—239 ________ Rodingite ______________________ +40 ___ _ __ 7.2 __ _ __ _ 8.8 AV—238 ________ Serpentinized rodingite __________ +49 94.1 __- 1.6 ___ ___ ___ AV—237A ______ Rodingite ______________________ +63 _ _- _ __ 10 _ - _ _ __ 33 AV—237B __________ do _________________________ +64 ___ ___ Tr ___ ___ 24 AV—237C __________ do _________________________ +66 ___ ___ Tr ___ ___ 6 E. Contact intermediate between the steatite and blackwall type and the tremolite and chlorite rock type [Drill hole inclined about 70° to bedding schistosity. Last set of T2— 6— 584 ______ Serpentinite ____________________ _ 98 _ __ 2 ___ Tr ___ T2—6 6—652. 5 _____ Tremolitic steatite ______________ _ _ _ _ Tr _ _ - 70 30 TI‘ T2—6—654 ______ Tremolite- chlorite rock __________ Contact _ _ _ _ __ _ _ _ Tr 60 40 T2—6—655 ______ Tremolite rock __________________ _ ___ ___ _ __ ___ 98 2 T2—6—656 ______ Schist _________________________ _ ___ ___ Tr -__ ___ 8 T2—6—658 __________ do _________________________ _ ___ ___ ___ ___ ___ ___ T2—6—692 __________ do _________________________ _ ___ ___ 1 ___ ___ 13 F. Contact intermediate between the steatite and blackwall type and the tremolite and chlorite rock type [Drill hole inclined about 60" to bedding schistosity. Last set of T2-—8—1195 _____ Serpentinite ____________________ _ 83 _ __ 2 _ __ __ _ _ __ T2—8—1202 _____ Graphitic serpentinite ___________ _ 83 1 6 _ _ _ _ _ _ _ _ _ T2—8—1204 _____ Graphitic steatite _______________ _ _ _ _ _ __ Tr 96 _ _ _ _ _ _ T2—8—1205 _____ Graphitic tremolite rock _________ Contact _ _ _ _ _ _ _ _ _ s _ _ 98 _ _ _ T2—8—1206 _____ Graphitic schist ________________ _ ___ _ __ __ _ _ __ 2 ___ G. Contact of the tremolite and chlorite rock type in coarse amphibolite of the Belvidere AV—202 ________ Tremolite rock __________________ Contact ___ __s Tr ___ 100 Tr AV—203 ________ Tremolite—chlorite rock __________ +21 ___ M__ ___ ___ 94.8 5. 0 AV—204 ____________ do _________________________ +40 ___ ___ _-_ ___ 88.1 11 9 AV—205 ____________ do _________________________ +46 ___ ___ ___ ___ 93.6 6. 3 AV—206 ________ Chlorite- tremolite rock __________ +82 __h _-_ ___ ___ 7.4 73. 7 AV—207 ________ Amphibolite ____________________ +326 _-_ _u _-_ ___ ___ 8. 9 AV—208 ____________ do _________________________ +570 ___ ___ ___ ___ __s 7.9 H. Contact of the tremolite and chlorite rock type in coarse amphibolite of the A—BM—53—45 _ __Serpentinite ____________________ —122 96 ,__ 3 _ __ _ __ Tr A—BM—53—44 _______ do _________________________ —24 97 __ 3 ___ ___ ___ A—BM—53—43 _______ do _________________________ —3 100 ___ Tr ___ ___ ___ A—BM—53—42 ___Tremolite rock _________________ +18 ___ __- ___ 95 5 A—BM—53— 41 ___Chlorite rock ___________________ +46 ___ _" ___ ___ ___ 88 A— BM— 53—40 ___Amphibolite ____________________ +152 ___ ___ ___ ___ ___ 4 A—BM—53—39 _______ do _________________________ +244 ___ ___ ___ ___ ___ 2 1 Veins. ULTRAMAFIC AND ASSOCIATED ROCKS across the contacts of ultramafic bodies trace] 47 Car- bonate min- erals Epidote/ clinozoi- site Diop- side Preh- nite Vesuvia- nite . Horn- Gax net blende clase Flagio— Quartz Musco- vibe Bio- tite Apatite Sul— fide min- erals Sphene Gra- phite formation (Chc2) near the center of the northeast contact of the Lowell quarry body—Contact regular 10 4 70 55 ___ 10 40 Tr 33 40 ___ Tr ___ 53 5 -__ 30 ___ 2 1 ___ 60 2 ___ 25 ___ 65 _-_ ___ 3 ___ ___ _~_ ___ 30 ___ _-_ ___ 24 2 Tr MNNNr—t r—A (ChC2) near the center of the northeast contact of the Lowell quarry body—Contact very irregular 39.6 ___ 5.0 (1) ___ 53.6 __- ___ ___ ___ ___ ___ ___ __- 39.7 10.4 2.0 5.1 32.4 _ _ _ _ _ _ ___ 58.7 20.0 ___ Tr 6.1 2.8 1.2 14.1 16.7 36.0 8.2 Formation (€bc) at the upper contact of the Eden quarry body, west side of the Eden quarry 92 ___ ___ _ 3 ___ 10 ___ 17 ___ ___ 15 ___ 33 26 ___ ___ 34 36 22 Formation (be)) at the southeast end of the southwest contact of of rodingite grades sharply into massive rodingite at +60 cm] the Lowell quarry body 0.2 —§i2 ‘454 8 6 .0 ___ 0 0 0.5 ___ 3.5 4.0 ___ ___ 2 O ___' 20 46 1 38 ___ 24 ___ 4 4s ___ ___ ___ 2 42 7f} in graphitic schist of the Hazens Notch Formation (Chcg) beneath digits in specimen number indicates footage from collar of drill hole) the Eden quarry body, drill-hole T2-6 ___ ___ ___ Tr ___ ___ ___ ___ ___ ___ _-_ 53 ___ ___ ___ _~_ ___ 1 ___ 30 ___ ___ ___ __- 4 ___ 9 10 25 15 12 45 23 38 in graphitic schist of the Hazens Notch Formation (Chcg) beneath the Eden quarry body,drill-hole T2-8 digits in specimen number indicates footage from collar of drill holel __1 15 ___ ___ Tr ___ ___ 10 -_, 54 10 2O Mountain Formation (Cbc) near the southeast end of the northeast contact of the Lowell quarry body ___ ___ ___ ___ 8.6 ___ _-- __e ___ ___ ___ ___ ___ 3.2 76.5 ___ ___ __- ___ ___ ___ 6.4 70.4 8.2 48 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT the dense flinty rodingite, commonly within a few centimeters. FINE-GRAINIiD RODINGITIC The fine-grained, dense, flinty rodingite is light bluish gray, light greenish gray, or pale buff. Layer- ing, on the scale of that in the bordering schist, gneiss, or amphibolite, is typical of most of the rock and is marked by faint differences in color that are accentuated by weathering. Though generally rather uniform in megascopic appearance, the rock shows considerable variation in mineralogy and in micro- fabric. Significant variations in mineralogy are predomi- nantly on a scale larger than that which can be seen in a hand specimen, but some thin sections show significant variations within the area of the section, and a few show different assemblages of minerals in adjacent layers. The principal assemblages observed are: diopside-clinozoisite diopside-clinozoisite-garnet diopside—clinozoisite-garnet-chlorite diopside-clinozoisite—garnet-chlorite-vesuvianite diopside-clinozoisite-garnet-vesuvianite diopside—clinozoisite-garnet—prehnite diopside-clinozoisite-chlorite diop-side—clinozoisite-prehnite diopside-garnet-vesuvianite diopsite-garnet-vesuvianite-chlorite diopside-chlorite. (magnetite) clinozoisite-garnet-chlorite clinozoisite-chlorite garnet-chlorite-vesuvianite- (magnetite) In addition to the major minerals, sphene is an almost ubiquitous accessory mineral, and apatite, rutile, magnetite-ilmenite, and calcite are common accessories. Quartz, hornblende, sulfides, graphite, biotite, and plagioclase are common and locally abundant relict minerals of the country rock; micro- cline is very rare. The variations in mineralogy conform generally to two patterns. One type of variation is related to position in the rodingite zone and to local structural controls. Diopside and clinozoisite predominate at the inner margin, clinozoisite and chlorite at the outer margin; garnet and vesuvianite, though ir- regular in distribution and absent from much of the rodingite, are most abundant at the inner mar- gin of the zone and adjacent to coarse rodingite as- sociated with fractures and shear zones. The second type of variation is unrelated to position in the zone, but reflects original differences in composition of the country rock from which the rodingite was derived. This type of variation is exemplified by the different assemblages of minerals in different relict beds in a single specimen. Textures of the fine-grained rodingite may be grouped into only a few basic types, which in slight variations and different combinations contribute considerable textural diversity to the rock. Much of the rock is a fine-grained mixture of constituents resolvable only under high magnification. Some con- sists of an equigranular aggregate of subequant serrated grains that are readily resolvable under the microscope. A third type consists of small sub- equant grains of one or more mineral species, in- terstitial to which other mineral species form ir- regular patches, shreds, and wispy flakes. In a few places, groundmass material having one of the above basic textures contains conspicuously larger, diversely oriented, tabular crystals of diopside. Some of the rodingite derived from amphibolite consists of a fine-grained groundmass throughout which are scattered larger, rounded grains 0.1 to 0.5 mm across. Most of the rounded grains are zoned clino- zoisite; a few are diopside. All these textural varie- ties locally contain small veinform or irregular masses of distinctly coarser grained minerals, some of them flamboyant in habit. For the most part, layering in the rodingite is parallel to bedding, and where inequidimensional minerals are present, they are commonly alined sub- parallel to the layering. Not uncommonly, several of the basic textures are shown by different layers in a single specimen. In a few places where bedding is tightly folded, quartz is concentrated in the troughs and crests of tiny folds, and inequidimen— sional grains of dark minerals arranged tangential- ly to layering are concentrated along the limbs; the resulting layering is at a large angle to bedding. COARSE-GRAINED RODINGITE The coarse-grained rodingite is variable in color and texture and is generally massive. It ranges in color from light gray to various shades of green and brown, or is mottled in those shades. In texture, it ranges from medium fine-grained granular aggre- gates of anhedral and subhedral grains to very coarse aggregates of anhedral and subhedral crys- tals as much as 8 cm in maximum dimension, many of Which show excellent crystal forms. Some joint surfaces are covered with drusy aggregates of one or more minerals. In several places, large crystals form spectacular “pockets” from which interstitial ULTRAMAFIC AND ASSOCIATED ROCKS 49 calcite has been dissolved to give a vuggy appear- ance. Such pockets are the source of many prized mineral specimens. The coarse rodingite contains the same minerals and has a variety of mineral assemblages similar to those of the fine-grained rodingite, but the varia- tions are more irregular and on a larger scale, so that some masses of hand-specimen size consist almost entirely of only two or three minerals. In addition, calcite is an almost ubiquitous major con- stituent, and zoisite is locally abundant. The sur- faces of a few open joints are spotted with botry- oidal masses or radial aggregates of calcite and siderite. The principal assemblages observed in rock masses of hand specimen size are: diopside-vesuvianite-garnet-calcite diopside-vesuvianite-garnet-clinozoisite-calcite diopside—garnet-clinozoisite-calcite diopside-clinozoisite-calcite diopside-calcite garnet-clinozoisite-calcite garnet-calcite clinozoisite—zoisite. SERPENTINE-CHLORITE ROCK The serpentine-chlorite rock is fine grained, mas- sive, dark greenish gray to greenish black, and breaks with a hackly to conchoidal fracture. Though uniform in megascopic appearance, the serpentine- chlorite rock varies considerably in microscopic texture, and has a consistent pattern of variation in mineralogy. Serpentine and chlorite are the sole essential minerals of the rock, but magnetite and ilmenite are ubiquitous accessory minerals and are common- ly major components. The proportions of serpentine and chlorite are variable, and the variations are somewhat erratic; but, in general, serpentine pre- dominates over chlorite at the inner margin of the zone, next to the serpentinite, and decreases in abundance outward toward the rodingite zone. Near the contact with rodingite, chlorite commonly pre- dominates over serpentine. Relict minerals from rodingite are common accessories in the zone of transition from serpentine-chlorite rock to roding- ite, and the titanium minerals show consistent para- genetic relations. At sharp contacts, favorably situ- ated single grains consist of sphene on the roding- ite side and ilmenite on the serpentine-chlorite rock side. The textural pattern of the serpentine-chlorite rock ranges from a felted aggregate of interlocking flakes, shreds, and blades of chlorite and serpentine to a mosaic of irregularly polygonal and elliptical patches of nearly isotropic serpentine containing disseminated flakes and blades of chlorite and ser- pentine. In some places, the elliptical and polygonal patches of serpentine contain almost no chlorite, but the interstitial areas are predominantly chlorite. In the transition zone from rodingite, the interstitial areas of chlorite commonly contain remnant grains of diopside. Anhedral grains of magnetite and ilmen- ite are scattered throughout the rock. CHLORITE-CALCITE-MAGNETITE VEINS Chlorite-calcite-magnetite veins vary from tabu- lar to irregular in shape, from a millimeter to a decimeter thick, and from a decimeter to more than 3 m long. The contacts of the veins with the enclos- ing rocks are sharp but are commonly irregular in detail. Some of the veins are distinctly zoned, ch10- rite or chlorite and magnetite predominating at the margins of the veins and calcite predominating in the central part. In such zoned veins, the chlorite flakes and books at the outer margins are commonly alined normal to the contacts of the veins. Else- where, the distribution of the minerals is variable, and their orientation apparently random. The minerals of the veins are coarse. Chlorite and magnetite form spectacular euhedral to subhedral crystals. Calcite is mainly in anhedral aggregates interstitial to the chlorite and magnetite. Small knots of epidote are scattered sparsely and uneven- ly through the veins in a few places. THE STEATITE AND BLACKWALL CHLORITE ROCK ASSOCIATION Rocks of the steatite and blackwall chlorite rock association border all the ultramafic bodies where they contain talcose rocks at the margins. These sites include most of the small lenses of ultramafic rocks (see section on “Steatite, talc-carbonate rock, and carbonate-quartz rock” under “Ultramafic and associated rocks”), the central parts of the north- eastern and the western contacts of the Eden quarry body, and probably much of the southern end of the Corez Pond body. The zones are generally most con- spicuous and most fully developed in the schist and more obscure and narrower in the amphibolite. All the rocks of the association are intergradation- al, and they have interdependent distribution rela- tions. The steatite zone varies in width from a centi— meter to as much as several decimeters; the black- wall is typically 10 cm wide where the steatite zone is 30 cm or more but is appreciably thinner else- 5O ASBESTOS-BEARING ULTRAMAFIC ROCKS 0F BELVIDERE MOUNTAIN AREA, VERMONT where. A narrow zone of tremolite rock, commonly about 5 cm wide, intervenes between the steatite and blackwall in many places but is absent in the less mafic varieties of schist. Where the adjacent coun- try rock is albitic schist 0r gneiss, a zone as much as 8 cm wide at the outer margin of the blackwall con- tains a relatively high proportion of albite porphy— roblasts (the albite porphyroblast zone). The tremolite rock is grayish green and composed of interlaced to radial aggregates of tremolite needles as much as 3 cm long, intermixed with vari— ‘ able proportions of talc and chlorite. Sphene, epidote, and magnetite are sparse and rare accessories. Blackwall is typically dark greenish gray and fine grained. It consists essentially of chlorite but com- monly contains accessory sphene, ilmenite, rutile, and apatite; rarely, relict grains of quartz and albite occur. The blackwall commonly has poor to good schistosity. In thin section, the schistosity is seen to result from the partial to almost complete parallel alinement of the chlorite flakes (continuous schis— tosity), or from the presence of closely spaced tiny shear zones within which chlorite flakes are alined and between which the chlorite flakes range from alined to diversely oriented (spaced schistosity). In some places, both types of schistosity are present and are parallel. Albite porphyroblast rock is variegated in gray- ish green and light gray, the overall appearance de- pending chiefly on the proportion of albite porphy— roblasts. Except for its more visible content of a1- bite, the rock ranges in character and mineralogy from albitic schist to blackwall, with both of which it intergrades. Muscovite—albite-quartz-(chlorite) rock forms ir- regular masses near the western contact of the main ultramafic body at 99,800 N., 91,600 E. (pl. 1). The muscovite is compact and fine grained and is irregu- larly intermixed with mosaic aggregates of albite. Quartz and chlorite occur in moderate to small amounts. The rock replaces adjacent amphibolite, but its relations to nearby steatite and talc-car- bonate rock are hidden by a covered interval of about 1 m. This muscovite-albite rock may occupy the albite porphyroblast zone and may be simply an unusual and extraordinarily large variant mass. THE TREMOLITE ROCK AND CHLORITE ROCK ASSOCIATION Contact rocks of the tremolite and chlorite rock association occur in only a few places: along the northeast contact of the Lowell quarry body near 99,910 N., 98,270 E., and at the margins of the small serpentinite masses in the amphibolite tongue be- tween the Lowell quarry and Eden quarry bodies, near 100,500 N., 96,600 E. (pl. 3). In all places, the country rock is amphibolite. Tremolite rock is medium gray to pale green, fine grained, and massive. A very irregular and in- conspicuous spaced schistosity is locally discerni- ble. Needles of tremolite are rarely large enough to be distinguished readily without a hand lens. Trem- olite is the sole essential constituent; chlorite is common and is increasingly abundant away from the ultramafic body; magnetite and pyrite occur sparsely and erratically. Chlorite rock is dark greenish gray and fine grained. A crude spaced cleavage is commonly suf- ficently well developed that the rock breaks into ir— regular slabs. In addition to chlorite, tremolite is a common constituent; sphene, carbonate, apatite, and other relict minerals from the amphibolite oc- cur unevenly in the chlorite rock and are abundant in the transition into amphibolite. The tremolite and chlorite contact rocks border serpentinite without intervening zones of talc-car- bonate rock and steatite. Tremolite rock adjacent to the serpentinite is transitional by gradual increase in content of chlorite and decrease of tremolite into chlorite rock adjacent to the country rock. The con- tact between serpentinite and tremolite rock is sharp. The width of the tremolite rock and chlorite rock zones, and of the transitional zone between, ranges from a few centimeters to a few decimeters and varies with the size of the associated ultra- mafic body. The transition from chlorite rock to amphibolite is irregular and variable, but generally takes place within a few centimeters. MINERALOGY AND PARAGENESIS The minerals of the contact rocks differ widely in grain size, textural relations, and variability of com- position, but all conform to a general pattern of paragenetic relations. Thus, minerals of the serpen- tine-chlorite rock replace minerals of the rodingite and locally those of the amphibolite; minerals of the rodingite replace those of the schist, gneiss, and amphibolite. Similarly, minerals of the steatite re— place minerals of the blackwall chlorite rock, and minerals of the blackwall replace those of the schist, gneiss, and amphibolite. Furthermore, in the roding- ite, some of the minerals have an orderly sequential relation: garnet and vesuvianite replace diopside and epidote, and chlorite commonly replaces all four of them. Garnet and vesuvianite commonly are euhedral in contact with calcite; vesuvianite is ULTRAMAFIC AND ASSOCIATED ROCKS euhedral in contact with garnet. All embay diopside irregularly, generally having scalloped boundaries convex toward the diopside, and all locally contain island remnants of diopside and of relict minerals of the country rock. Garnet, diopside, epidote, and vesuvianite commonly have ragged boundaries with chlorite, and in places form tiny island remnants in patches of chlorite. DIOPSIDE Diopside is an almost ubiquitous constituent of the rodingite, is generally a major constituent, and is commonly the most abundant mineral. In coarse rodingite, the diopside ranges in size and form from anhedral grains a few millimeters to several centi- meters across, to prismatic crystals as much as 3 cm long. Diopside in dense flinty rodingite is all fine grained and ranges in habit from subequant anhedral grains, wispy blades, and irregular poiki— loblastic grains to subequant subhedral grains and well-formed tabular crystals. The poikiloblastic grains enclose quartz, epidote, and other relict min- erals from the country rocks. In places, fine-grained diopside irregularly rims relict grains of coarse hornblende, and invades these grains along frac- tures and cleavages. Diopside in small veins and irregular coarser grained patches in fine rodingite is commonly bladed or flamboyant in habit. The diopside varies in color from moderate yel- low green to grayish olive. Euhedral crystals are generally short prisms terminated by basal pina- coids; {100}, {010}, and {111} faces are subdued. A few crystals are tabular, parallel to {010}. Prismatic cleavage is generally distinct. The diopside varies appreciably in optical prop- erties and inferred chemical composition. The optic axial angle, 2V, varies from 60° to 70°, the optic sign is (+), the extinction angle is nearly 45°, and the B-index of refraction varies from 1.671 to 1.705. These data indicate a range in composition, in terms of the diopside and hedenbergite end members, from He, to about H650, or from CaMgSigO.. to about ca (Mg(),5Fe+go,n) Sizou- CLINOZOISITE Clinozoisite is nearly as prevalent as diopside in the rodingite and not uncommonly is the most abun- dant mineral. In fine-grained rodingite, clinozoisite occurs as tiny anhedral subequant grains in an equigranular mosaic, as larger rounded grains in a fine-grained groundmass, and as subhedral crystals in patchy and veinform aggregates. In coarse 51 rodingite, clinozoisite forms striated bladed and an- hedral crystals as large as 3 cm across. The clinozoisite is dusky yellow green, light olive gray, or olive gray, and well-formed crystals are strongly striated. In thin section, it appears color— less and commonly has two good cleavages about at right angles. The fine-grained material is mostly un- twinned and shows uniform extinction. Coarse- grained material commonly is strongly twinned. Most of the large rounded grains set in a fine groundmass are zoned, the inner part of the grain commonly showing undulatory extinction and the outer part showing uniform extinction and slight- ly different optic orientation. The optical properties vary within a relatively small range. The optic axial angle, 2V, is large, and the optic sign varies from (+) to (—). Abnormal blue and lemon-yellow interference colors are dis- tinctive. The ,B-index of refraction ranges from 1.712 to 1.728 but is chiefly in the range 1.717—1.723. These properties indicate a range in composi- tion from close to that of iron-free epidote, Ca.Al,Al._.Si,,Om (OH) 2, to perhaps as much as Ca,A1., (Al‘Fe+3;) Sioogl (OI—1):) and a predominant composition of about calA-ll(AII,:KFe+:;U,T)SiciO'zl(011)2- ZOISITI‘Z Zoisite was noted only in coarse rodingite, where locally it forms coarse radial aggregates of bladed white crystals. The zoisite contains two optically distinct phases in intimate association. One phase has 2V=0°—50°; optic sign (+); dispersion, rv distinct; abnormal brown interference colors; indices a=1.701, B=1.701, y=1.706; optic orientation, a=y, b=a, c=,8. Zoisite having such properties is commonly correlated with ferrian or B-zoisite. The slight difference in optical properties suggests that the difference in composition between the two phases is also slight. The “B-zoisite” has only slightly higher indices than an analyzed sam- ple of zoisite listed by Deer, Howie, and Zussman (1962, V. 1, p. 188, table 32, analysis 3), in which the ratio Feti‘,/(A1+Fe+3) =0.033. The “a-zoisite” and “B-zoisite” from Belvidere Mountain are in- 52 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT ferred to have comparable Fe+i‘/ (A1+Fe+“) ratios, or formula compositions near Ca4Al4 (AlmFe‘L 30.2) SiGO“ (OH) 2. GARNET Garnet is a common and generally abundant con- stituent of both fine and coarse rodingite. It forms euhedral to subhedral crystals as much as 1 cm across in coarse rodingite, and small anhedral grains in fine rodingite. Garnet varies widely in color and optical properties in both types of rock and shows a variety of textural relations in fine rodingite. In fine rodingite, garnet forms massive aggregates of small grains and scattered clusters of grains. Patches of massive garnet have scalloped borders that are convex toward the diopside, and the garnet commonly encloses irregular shreds of diopside. Small veinlets of garnet or garnet and vesuvianite are common. Some such veinlets are complex; they consist of a central straight-sided vein about 0.05 mm wide of isotropic or nearly isotropic garnet, bordered on either side by narrower zones of bire- fringent garnet whose contact with diopside is very irregular—as if the central vein had formed by fis- sure filling, the outer zones by replacement. The garnet has a wide range of colors, chiefly shades of green, brown, and red. Most of the garnet is slightly to distinctly birefringent. The index ranges from 1.740 to- >1.789. In general, euhedral garnet in coarse rodingite and garnet in straight- sided veinlets are of low index, and garnet in the veinlets is isotropic; anhedral fine-grained garnet, whose relations suggest a replacement origin, is of relatively high index and is distinctly birefringent. However, there are many exceptions to these general relations; in particular, many of the large euhedral crystals of garnet are distinctly birefringent and of relatively high index. Microprobe analyses of garnets from rodingite, mostly coarse grained and euhedral to subhedral but representative of a wide range of colors and indices of refraction, indicate that the garnets are of two limited compositional ranges (grozand :alm :sp) :(1) 86—91:14—4:4—1:0—4, and (2) 45—66:55—32:0—2:0. Representative formula. compositions for each range are: (1) [CaMGManFeHMZ] [AngzFet‘O'mTiofll] Si,,..,0.., and (2) [Cangnm] [A10_,.,Fe+31_0,]Simon. Data are meager, but no correlation between compo- sition and physical and optical properties is apparent. VESUVIAN ITE Vesuvianite, though less common and abundant than diopside or garnet, is a fairly common and locally abundant constituent of the rodingite. It ranges in habit and grain size from anhedral fine grains to prismatic crystals as much as 3 cm in diameter and 10 cm long, some of them singly termi— nated by pyramids and basal pinacoids, and Belvi- dere Mountain has long been highly prized for vesuvianite by mineral collectors. Massive vesuvian- ite commonly consists of anhedral grains; isolated grains are commonly subhedral. As opposed to gar— net, vesuvianite generally is euhedral. The vesuvianite ranges in color chiefly from gray- ish yellow green to moderate yellow green. The opti- cal properties are generally uniform; the w-index is chiefly in the range 1.718 to 1.720. One specimen (AV—237) has an index 0):]..730. The optic axial angle range is moderate (40°—50°), the optic sign (— ) , the birefringence low, and interference colors of abnormal violet and yellowish brown are char- acteristic. On the basis of the light—green color (Deer and others, 1962, v. 1, p. 118) the vesuvianite is inferred to contain chiefly ferric, and little ferrous, iron. The indices are near those of a light-green vesuvianite listed by Deer and others (1962, v. 1, p. 116, table 21, analysis 5). Neither the exact mineral formula nor the correlation of optical properties with chemi- cal composition are adequately known, but on the basis of the data available, it appears that the aver- age composition of the vesuvianite, in terms of the most widely accepted ideal formula, is approximately cam (Mg1.0Fe+30.sFe+2o.2) A11[Si207]2[Si04] 5 (OHrF) 4- PREHN ITE Prehnite forms distinctive fine drusy aggregates on a few joint surfaces; it is rarely a sparse constitu- ent of fine-grained rodingite, where it forms an- hedral patches interstitial to diopside and clino-zois- ite. The optical properties are somewhat variable but show no consistent differences between drusy prehn- ite and that in the groundmass of fine-grained rodingite. The optic axial angle is 2V=70°—75°; the optic sign (+) ; indices, a=1.614—1.620, ,8=1.623- 1.634, y=1.642—1.643; extinction is parallel. The properties indicate a composition in which the ratio of Fe+3/(A1+Fe+3) is in the range 0.02-0.10. The average formula composition is inferred to be about caz (A1,1.9Fe+3o.1) Si3010 (OH) 2. ULTRAMAFIC AND ASSOCIATED ROCKS 53 CALCITE Calcite is a common and abundant constituent of coarse rodingite, where it forms anhedral grains interstitial to other minerals. It is a sparse constitu- ent of fine rodingite, of blackwall, and of tremolite rock. The m=index is uniformly 1.658 to 1.660, indi- cating a composition near that of pure calcite, CaCOg. MAGN ETlTE Magnetite is a relatively uncommon and generally sparse constituent of rodingite, blackwall chlorite rock, and tremolite rock; it is a ubiquitous and com— monly abundant constituent of serpentine-chlorite rock and of calcite-chlorite-magnetite veins associ- ated with rodingite and serpentine-chlorite rock. Magnetite in rodingite, blackwall chlorite rock, and tremolite rock is anhedral, subequant to irregu- lar, and has generally a distribution pattern similar to that of magnetite in adjacent country rock. Mag- netite in serpentine-chlorite rock is varied in habit, ranging in grain size from 0.015 to 1.5 mm and in form from anhedral to euhedral. Some thin sections contain both scattered large anhedral grains and abundant small euhedral to subhedral grains. In all sections, the proportion of magnetite in serpentine- chlorite rock is higher than in adjacent rodingite. Magnetite in the calcite-chlorite-magnetite veins forms crystals as much as 1 cm across which com- monly show octahedral and dodecahedral forms. No information is available on possible variations in composition of magnetite in the several contact rocks. An analysis of a sample from a calcite-chlo— rite-magnetite vein (table 1, analysis 46) gave a calculated formula composition of (Fe+2o.91Mgo.01Mno.mTide ivmn) (Fe+31,91A10.oncro.02) O4- CHLORITE Chlorite is the sole essential constituent of black- wall chlorite rock, an essential and major constituent of serpentine-chlorite rock and calcite-chlorite-mag- netite veins, and a commonly abundant constituent of rodingite and tremolite rock. The habit and tex- tural relations of the chlorite are varied. It occurs as blades, flakes, and shreds in blackwall, serpentine— chlorite rock, rodingite, and tremolite rock. It forms distinctive books of platy crystals in large calcite- chlorite-magnetite veins, as well as in varied tiny veinlets in the rodingite. In the large veins, these chlorite books are commonly hexagonal and are as large as 3 cm across. The chlorite is intimately inter— mixed, in apparent equilibrium relation, with ser- pentine in serpentine-chlorite rock; it invades and replaces garnet, vesuvianite, diopside, and clinozois- ite in rodingite; it replaces tremolite in tremolite rock; and it replaces relict minerals of the country rock in the blackwall. Chlorite in contact rocks is almost all of relatively low index, positive optic sign, negative elongation, and abnormal brown interference colors. It shows some minor but distinctive variations among differ- ent contact rocks. Chlorite in the calcite-chlorite-magnetite veins shows little variation in optical properties. An ana- lyzed sample (A—BM—56—4b, table 1, analysis 49) showed 2V=30°, optic sign (+), elongation (—), abnormal brownish-gray interference colors, arm/3: 1.576, 7:1.580. The formula composition calculated from the chemical analysis (Chidester, 1962, p. 132— 138) is (Mg4fiOFe+:0,lIMHOJIOJ) (A1().M:Fe+30,00C1'0.01Ti0.001) (Sill.0(iA10.9-l) 09.13(OHS.85F0.003) ' This composition fits rather closely with that based upon the optical correlations of Winchell and Winch- ell (1951, fig. 267; see Chidester, 1962, fig. 13, p.45). Measured indices of chlorite in blackwall range from ,8=1.594 to 3:1.613 and are predominantly near [3: 1.600. Chlorite of index B=1.600 has a com- position of about (Mg3.nFe+21.:;A11,2) Si2.TA1l.3) 010 (OH) 8- Observations elsewhere in Vermont (Chidester,. 1962, p. 66—69) suggest that the total range in com- position of chlorite in blackwall may be appreciably greater than is indicated by the relatively few opti- cal determinations made on blackwall chlorite at Belvidere Mountain, and that the variation is sys— tematic within the zone, the ratio of Mg/Mg+Fe+‘~’ increasing sharply near a contact with steatite and decreasing sharply near a contact with schist (Chi- dester, 1962, fig. 17, p. 67). Chlorite in serpentine—chlorite rock has a similar range in optical properties; variations appear not to be systematic with respect to position in the zone. Some of the chlorite is nearly isotropic, consists pre- dominantly of abnormal brown (optically positive) flakes intermixed with a relatively small proportion of abnormal blue (optically negative) flakes, and has indices of about ,8: 1.575. Such chlorite, on the basis of Winchell and Winchell’s data (Chidester, 1962, fig. 13, p. 45) is inferred to have a composition of about (MgS.UFe+:u.J;A10.T) (Si::.::A11LT)OW(OH) 3‘ 54 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE- MOUNTAIN AREA, VERMONT In index, much of the chlorite reaches a maximum of 3:162, and in birefringence reaches y—a=0.005. The corresponding limiting range in composition is inferred to be about (Mg2.6Fe+22.1A11.3) (Si27A11.3) 010 (OH) 8- Chemical analyses were obtained of three samples of serpentine—chlorite rock (TPT—Vt—11(VT—11) , AV— 78 (VT—13), and AR—110 (VT—20), analyses 7, 9, and 16, table 1). The distribution of minerals in ser- pentine-chlorite rock is erratic, so that accurate modes of the material analyzed are not available. It is estimated, however, that the material of analysis 7 consists predominantly of chlorite and serpentine in the proportion of about 25/75; that of analysis 9, about 50/50; and that of 16, about 60/40. Calculated formula values of the serpentine-chlorite mixture in each show R+3/Si ratios of 2/3, 1.4/3.3, and 1.3/3.4, and Mg/Mg+Fe+2 ratios of 0.88, 0.99, and 0.98. As the admixed serpentine has a much lower content of R+3 than the chlorite, these ratios set lower limits for the content of R+3 in the chlorite. Most of the chlorite must therefore have a composition near that indicated for chlorite of B—index= 1.62 (see above), a ratio of R+3/ Si of about 1. Chlorite in tremolite rock has about the same range in optical properties as that in blackwall but varies erratically. Measured indices of chlorite in rodingite range from 3:1.585 to ,8=1.621, corre- sponding to a range in composition of (Mg4.2Fe+20.7A11.1) (Si2.9A11.1)010(0H)8 t0 (Mg2.5Fe+22.2A11.3) (Si2.7Ali.3)Oio(OH)s~ Chlorite in analyzed samples of tremolite-chlorite rock (sample AV—205 (VT—17), analysis 13, table 1) and chlorite-tremolite rock (sample AV—206 (VT—18), analysis 14, table 1) is inferred to have a composition of (Mg3.6Fe+21.-'S) (Si2.9A11,1) 010 (OH) 8 on the basis of optical properties. Cross calculations , based on modal compositions were carried out on the two analyses (see Chidester, 1962, p. 132—138) ; these calculations yielded a calculated formula of (Mg3.5Fe+2i.2Mno.o2) (Fe+3o.1Al1.os) (Si2.5A11.15Tio.3Po.05) 010 (0H) 8.2) very close to that based on optical properties. Chlorite in the single and unusual occurrence of massive fine-grained chlorite in an irregular vein- like mass between serpentinite and rodingite is nearly isotropic, has a B=index of 1.632, and shows both abnormal paleapink and pale-blue interference colors. The composition is inferred to be about (Mg1.8Fe+22.sAli.4) (Si2.oA11.4) 010 (OH) 8- SERPENTINE Serpentine in the contact rocks occurs only in serpentine—chlorite rock, where it is everywhere a major component but varies in proportion from as little as 25 percent to as much as 95 percent. The serpentine has a variety of textural features. Most occurs as irregular flakes and blades, intimately intermixed with chlorite or segregated rather ir- regularly into layers predominantly of serpentine. Isotropic patches of macroscopically structureless serpentine appear to be pseudomorphic after garnet and vesuvianite. Rarely, tiny veinlets of fibrous ser- pentine occur next to the contact with the ultra- mafic body. The serpentine varies from colorless or very pale green to yellowish brown in thin section, and from virtually isotropic to distinctly abnormal blue or light gray under crossed nicols. Measured indices of refraction range from [3=1.548 to 18:1.573, and birefringence from nearly zero to 0.009. The sign of elongation is uniformly positive. The optic sign of all platy serpentine is negative; the apparent optic sign of fibrous serpentine is positive. X-ray and DTA studies by G. T. Faust (written commun., 1961) of two specimens of serpentine- chlorite rock indicate the presence of lizardite and chrysotile, in addition to chlorite. Chemical analyses of the serpentine-chlorite rock (TPT-Vt—ll (VT— 11), AV—78 (VT—13), and AR—110 (VT—20), anal- yses 7, 9, and 16, table 1) cannot yield accurate formula compositions for the serpentine minerals, but the average ratio of Mg/Mg+Fe+2 for the ser- pentine minerals must be greater than that of the mixture in each analysis (see preceding section on “Chlorite” under this section on “Mineralogy and paragenesis”). Thus the average ratio of Mg/Mg +Fe+2 must be near 0.99 or higher. No useful in- formation on the content of Al and Fe+3 in the serpentine minerals can be deduced. Though the content must be low (most Al and Fe+3 are required for the chlorite), limitations of the data allow for a range of variations considerably greater than pos- sible critical differences between difi'erent serpentine minerals. TRElVIOLITE ' With the exception of relict hornblende in roding- ite and blackwall chlorite rock, tremolite is the sole amphibole in contact rocks. Tremolite forms a thin zone between steatite and blackwall and in a few places occupies the position of the entire steatite zone. The tremolite varies from finely fibrous to ULTRAMAFIC AND ASSOCIATED ROCKS 55 blades as much as 1 mm wide and 5 mm long. The blades are commonly bent, broken, or folded; not uncommonly, laths of tremolite tail out into finely fibrous bundles of asbestiform tremolite. The optical properties vary only moderately: 2V=75°—85°, the optic sign (—), the extinction angle y/\C=15°—20°, and the y-index of refraction ranges from 1.625 to 1.636 but in most places is near 1.632. These optical properties correspond to a range in composition from pure tremolite, CazMg5Si8022(0H)2, to actinolite of formula com- position 032(Mg4tFe—i—got)Si8022(OH)2; mos-t is about ca2(Mg4.7Fe+20.3)Si8022(0H)2 (Foslie, 1945, fig. 1). A chemical analysis of tremolite-chlorite rock in the position of the steatite zone (sample AV—205 (VT—17), analysis 13, table 1), the tremolite of which has a y-index of 1.632, yields the following formula composition (after correcting for about 12 percent chlorite of ,8—index=1.599, and very small amounts pyrite and calcite) : ca1.9(Mg<1.9Fe+20.2)Si8.0022.0[00.4(OH)1.G]' ILMENITE, RUTILE, AND SPHENE All the contact rocks generally contain ilmenite, rutile, and sphene, singly or in combination, locally as major constituents. In the different contact rocks one or another of the three minerals commonly pre- dominates, not uncommonly to the almost complete exclusion of the others. In some of the contact as— sociations, the titanium minerals vary regularly be- tween rock types, and in some of the individual con- tact rocks they show regular variations with respect to position in the zone, or regular sequential rela- tions with one another. Ilmenite is a ubiquitous and commonly a major constituent of the serpentine-chlorite rock. It is rare in the rodingite and uncommon in the blackwall, in both of which it occurs only as relict grains in association with rutile or sphene, or both. The ilmen- ite ranges in habit from tiny laths to irregular grains. Rutile is relatively uncommon in the contact rocks. It is most abundant in the blackwall; there it forms prismatic crystals, or anhedral aggregates associ- ated with either ilmenite or sphene or both. Rutile occurs rarely in the rodingite, and always in associa- tion with ilmenite and sphene. Sphene is the most common and abundant mineral in the contact rocks. It is nearly ubiquitous in the blackwall and rodingite; it is common in tremolite rock, in steatite derived from schist or amphibolite, and in miscellaneous irregular types of contact rocks. .Only serpentine-chlorite rock is devoid of sphene. The sphene ranges in habit from irregular aggregates of tiny grains to diamond-shaped single crystals. The relations of ilmenite, rutile, and sphene are regular for each association but differ between as- sociations. In the rodingite, sphene generally occurs alone, but in some places, particularly in the outer parts of the zone, some of the sphene contains a core of ilmenite or a shell of ilmenite around a core of rutile. At the contact between rodingite and ser- pentine-chlorite rock, elongate grains that project across the contact consist of sphene on the rodingite side of the contact and ilmenite on the serpentine- chlorite rock side. In the blackwall, the overall re- lation among the three minerals is that of a core of ilmenite surrounded by a shell of rutile and an outer rim of sphene; any one or a pair of the three may be absent locally (Chidester, 1962, p. 69.) OTHER MINERALS Talc is a minor constituent of the transition zone between steatite and blackwall chlorite and is the essential major constituent of the steatite, the outer- most few millimeters of which belong to the con- tact-rock association. The properties are virtually the same as those of talc derived from ultramafic rocks. White mica, quartz, and albite occur principally in the outer margins of the rodingite and the black- wall zones, as relict minerals of the adjacent country rock. The three minerals are associated in the ir— regular masses of muscovite—albite-quartz rock which bear a replacement relation to the adjacent amphibolite near 99,800 N., 91,600 E. (see section on “The steatite and blackwall chlorite rock associa- tion” under “Contact rocks” in the section on “Ultramafic and associated rocks”). The mica oc- curs as compact fine—grained masses; the quartz and albite as irregular anhedral grains. The mica is identified as muscovite on the basis of X-ray pat- tern and optical properties of 2V=30°, optic sign (—), 7:1.599, y—a=0.034. Apatite, hornblende, biotite, sulfides, and graph- ite, all are relict minerals of the country rocks from which the contact rocks were derived. All are similar in habit and composition to their equivalents in the unaltered country rocks. 56 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT PETROGENESIS Igneous and metamorphic features and geologic relations of the ultramafic rocks attest to an igneous origin of the dunite, peridotite, and chromitite, to later alteration of much of the igneous rock to serpentinite, carbonate-quartz rock, serpentine and magnetite veins, talc-carbonate rock, and steatite, and to a regular sequential relation among them. Mineralogical and geological relations and textural features of the contact rocks attest to their origin by metasomatic replacement of the country rock and point to systematic genetic and chronologic rela- tions between the contact rock and the ultr‘amafic rocks. Direct evidence of relative ages based on field re- lations between the various ultramafic rocks and the contact rocks demonstrates only that the ser- pentinite is younger than the primary igneous min- erals, that the serpentine veins are younger than the consolidation of the ultramafic igneous rocks, and that the several talcose rocks are younger than the serpentinite. Inasmuch as contact rocks of the rodingite and serpe‘ntine-chlorite rock association, on the one hand, and those of the steatite and black— wall association or the tremolite and chlorite rock association, on the other, are mutually exclusive in occurrence, their field relations are ambiguous. Else- where in Vermont, steatitized ultramafic bodies con- tain talc pseudomorphic after asbestos (Chidester, 1962, p. 76—77), which demonstrates that the stea- tite is younger than the asbestos. Serpentine veins appear to be generally younger than the pervasive serpentinization because the veins, which are unsheared, were formed late in the tectonic history, whereas pervasive serpentiniza- tion was early. Pervasive serpentinization and the formation of serpentine veins may have overlapped in time to a significant extent. Vein formation took place in a relatively short interval late in the tee- tonic history of the area, whereas pervasive ser— pentinization probably took place over a much longer interval that began before final emplace- ment of the ultramafic bodies. Very probably, per- vasive serpentinization extended into the period of vein formation (see following section on “Serpen- tinite” under this section on “Petrogenesis”). IGNEOUS ROCKS An intrusive igneous origin for the dunite and peridodite is indicated by the locally crosscutting relations of the ultramafic body, the mosaic texture of olivine in dunite and peridotite and of chromite in chromitite, and the phenocrystic habit of the pyroxene and chromite. Layering of the igneous rocks is inferred to be, in part, at least, the result of crystal accumulation, but it may have been modi- fied and enhanced tectonically by fl-owage during emplacement of the mass. Some of the thinner layer- ing may be entirely of tectonic origin. Modification of the original igneous texture by cataclasis is in— dicated by the prevalence of undulatory extinction and glide twinning in part of the olivine throughout the dunite and peridotite, the vague mosaic extinc- tion in large crystals of olivine, and the parallel dimensional and crystallographic orientation of the olivine in some of the dunite. Olivine, pyroxene, and chromite formed early and more or less contemporaneously. The inclusion of olivine grains in some single crystals of pyroxene indicates that pyroxene was in part later than oli- vine. As crystallization proceeded, translucent reddish-brown chromite gave way to opaque chro- mian magnetite. The chromian magnetite formed as rims surrounding cores of translucent chromite, as dispersed discrete grains, as concentrations of small grains interstitial to olivine grains, as vein fillings in fractured chromite grains, and as thin discrete layers in dunite and peridotite. Pyrite and sulfarsenides formed at about the same time as magnetite, with which they are locally intergrown. Pyroxene is pseudomorphically replaced by antho- phyllite, which is in turn replaced by serpentine. Alteration of pyroxene to anthophyllite probably took place in an early postmagmatic stage, during cooling of dunite and peridotite. Other minerals in the igneous rocks are related to serpentinization, and will be discussed in the fol- lowing section. SERPENTINITE Many features demonstrate that serpentinite was formed by alteration of dunite and peridotite. For massive serpentinite, these include a gradational relation with dunite and peridodite, preservation of relief. layering, aggregate pseudomorphs after oli— vine, replacement of olivine by serpentinite, and the presence of relict olivine, pyroxene, chromite, and magnetite. For schistose serpentinite, the principal evidence is a gradational relation to massive serpen— tinite and a prevalence of relict chromite grains. Textural features of massive serpentinite indicate that serpentinization was effected without percepti- ble change in volume. There is no evidence in the schistose serpentinite bearing on volume change during serpentinization. Shear polyhedrons formed during tectonic transport are evidence of early ULTRAMAFIC AND ASSOCIATED ROCKS 57 serpentinization, perhaps beginning in the upper mantle. (See section on “Emplacement and struc- tural history.”) Antigorite, the predominant mineral in the ser- pentinite, formed chiefly by replacement of olivine and only in minor amounts by replacement of pyroxene or anthophyllite. Chrysotile, which is rela- tively minor in the groundmass serpentine, formed chiefly in tiny fractures in and at the margins of grains of olivine but locally appears to have replaced whole grains or groups of grains. Possible six-layer orthoserp‘entine, which occurs in the centers of mesh-structure serpentine, formed by replacement of olivine. Chromite and the larger grains of magnetite are relict igneous minerals. Magnetite in the form of dispersed fine dusty particles, scattered small grains, concentrations of tiny grains along relict cleavage surfaces of altered pyroxene, and aggregates of tiny grains that rim many of the chromite grains and the larger magnetite grains was formed during serpentinization. A small amount of chlorite was formed by the al- teration of pyroxene or anthophyllite to antigorite and chlorite during serpentinization, but most of the chlorite was formed by interaction between chromite or chromian magnetite and the surround- ing antigorite, during and immediately following serpentinization. During the reaction between the spinellids and antigorite, aluminum was lost from the spinellids, and entered the chlorite and locally the adjacent antigorite to form a thin halo of alumi- nian antigorite. The loss of aluminum from chro- mite and magnetite grains effected a change in com- position of the outer shell of the grains, but the re- sulting reaction zone cannot generally be distin- guished from the zones of opaque chromian magne- tite of igneous origin. Perhaps the thin opaque borders of fracture-s in translucent chromite grains are chiefly or entirely such reaction zones. Brucite in the groundmass of the serpentinite formed entirely by replacement of olivine, to which its relation is clearly shown in the partly serpen- tinized dunite. Magnesite, dispersed in the groundmass serpen- tine and as small lenticles, is a product of serpen- tinization of the olivine. Dispersed grains of calcite are probably all associated with late veins. Talc, which occurs only rarely in the schistose serpen- tinite and near the margins of the ultramafic bodies, is probably all later than the serpentinization. (See following section on “Steatite, talc-carbonate rock, and carbonate-quartz rock” under “Petrogenesis” in the section on “Ultramafic and associated rocks”) The origin of the graphite in serpentinite is un- certain. Concentration of graphite near the mar- gins of the ultramafic body probably occurred dur- ing serpentinization or steatitization, or both, and redistribution along schistosity and slip cleavage surfaces was probably accomplished mechanically during folding. Graphite was not seen in fresh dunite and peridotite. Trofimov (1940, p. 34—35) re- ported carbon-bearing peridotite in Siberia that con- tains traces to as much as 0.38 percent carbon; at Belvidere Mountain it is possible that such graph- ite—which would be very diflicult to detect micro- scopically in small amounts—was concentrated at the margins during serpentinization. Concentra- tions of graphite at the outer margin of the black- wall chlorite rock zone of steatitized ultramafic bodies elsewhere in Vermont have been interpreted as being formed by outward displacement from graphitic schist during the blackwall-steatite reac- tion (Chidester, 1962, p. 123, 127). Thayer (1966, p. 697—700) suggested that the reduction of carbon dioxide by hydrogen produced by decomposition of water during serpentinization may result in the formation of graphite. Hydrogen might also be formed in a similar manner during steatitization. In either case, reaction of such hydrogen with car- bon dioxide mobilized during regional metamor- phism would be expected at the margins of the ultra- mafic bodies. Such a reaction would explain well the distribution and apparent age relations of the graphite. SERPENTINE VEINS That irregularities in opposite walls of the ser- pentine veins match even down to fine details in- dicates that the veins have formed chiefly by frac- ture filling. Only locally has an appreciable propor- tion of a vein formed by replacement; elsewhere, replacement was negligible. In both the cross-fiber asbestos veins and the cross-column picrolite veins, the observation that matching irregularities in op- posite vein walls are at opposite ends of a group of fibers indicates that the axes of the fibers and col- umns mark the relative direction of movement of the walls during opening of the fissure. Though the relations are generally obscure for the so-called slip fiber in the schistose serpentinite, the orientation and length of the fibers probably also mark the di- rection and amount of movement on opposite walls of much of the slip~fiber veins. 58 ASBESTOS-BEARING ULTRAMAFIC ROCKS 0F BELVIDERE MOUNTAIN AREA, VERMONT Riordon (1955) concluded, for asbestos deposits in Quebec, that the delicate layering parallel to the walls in many of the picrolite and cross-fiber veins, and the interrelations of the veins, indicate that all the serpentine veins were injected into open fissures as “c-ollodial serpentine,” that the colloidal serpen- tine later recrystallized to picrolite, and that the picrolite subsequently recrystallized to cross-fiber asbestos. We propose, and favor, an alternative explana- tion for these features. We suggest, on the basis of the relations previ- ously described (in the section on “Serpentine veins” under “Igneous rocks, ser‘pentinite, and veins” in the section on “Ultramafic and associated rocks”) that the thin delicate layering in the veins records discrete increments in the opening of the fissures occupied by the veins. Differences in color of different layers probably reflect very slight dif- ferences in composition of successive layers; slight differences in orientation of the fibers in successive layers record slight differences in direction of rela- tive movement of the vein walls during successive increments of opening. We further suggest that When the rate of opening of the fissure equaled or only minutely exceeded the rate of influx of material into the fissure, the conditions led to an almost per- fect orderly growth of parallel tubular fibers of chrysotile. When the rate of opening exceeded the rate of addition of material to the fissure by a dis- crete but small amount, the fibers of chrysotile grew in a slightly disarrayed but subparallel fashion to produce coarsely fibrous to columnar picrolite. When the rate of opening of the fissure was so rapid that an appreciable void resulted, the chrysotile fibers formed in a sufficiently disarrayed manner so that the picrolite appears massive, without megascopical- ly discernible fibrous or columnar structure. The simple veins of chrysotile asbestos record a history in which the rate of opening of a fissure uniformly almost equaled the rate of growth of the fiber; the simple veins of picrolite record a history in which the rate of opening uniformly exceeded the rate of growth. The composite veins record a complex his- tory in which the rate of opening at times about equaled, and at times exceeded, the rate of growth of the vein material. Veins of unlayered asbestos and picrolite record a history in which the rate of opening was steady, or in which there was no sig- nificant change, in successive impulses, in the rela- tive direction of movement of the walls and in com- position of the vein material. Concentrations of magnetite along partings and at margins of serpentine veins probably consist chiefly of magnetite formed during fissure filling, but some may consist partly of original seams of magnetite along which fissures opened in the rock. In single-fiber veins, separation during opening of a fissure occurred repeatedly at the junction of fiber and magnetite seam or fiber and vein wall. In mul- tiple-fiber veins, separation occurred successively at opposite sides of a magnetite seam or at opposite vein walls. In veins with strongly crenellated part— ings, separation took place irregularly along the vein at one side or another of the partings in suc— cessive impulses of vein opening. Chips of serpen- tinite along partings mark sites where fractures extended locally into wallrock, rather than uniform- 1y along a junction of wallrock and magnetite seam or fiber. The relations of fibers at vein intersections indi- cate relative ages of fissures. Merging relations of the fiber indicate that intersecting fissures opened and vein material formed contemporaneously. Sim- ple crosscutting relations indicate that one fissure opened and its vein material formed entirely before the other fissure opened. Complex intersections in— dicate that fissures and vein materials are in part contemporaneous and in part sequential. Pockets of long slip fiber that are intergrown with abundant fibrous calcite in shear zones are difficult to explain. Perhaps they formed under early static conditions, by slow growth of patches of fiber that had previously been sheared into parallelism with the enclosing shear zones. The fibrous habit of much of the calcite and bru- cite, which occur in generally minor amounts in many of the serpentine veins, suggests that they formed largely contemporaneously with the serpen- tine, but calcite of granular habit appears locally to replace chrysotile asbestos. Marginal zones of alteration adjacent to serpen- tine veins are inferred to be contemporaneous with and genetically related to the veins. The zones were formed by replacement of olivine by chrysotile, lizardite, and brucite in dunite and massive serpen- tinite bordering the veins. Layering in marginal zones probably records successive outward waves of alteration corresponding to successive episodes of vein filling. The general relation that the width of the mar- ginal zone of alteration is proportional to the width of the adjacent vein suggests that the vein material (except water) was derived largely from immedi- ately adjacent wallrock. Many local departures from ULTRAMAFIC AND ASSOCIATED ROCKS 59 this general relation indicate that in places signifi- cant amounts of material moved along the veins. The relative amounts involved are illustrated in table 7, which compares the composition of equal volumes of cross-fiber chrysotile asbestos and of ser- pentinite from serpentinized zones bordering as- bestos veins with that of dunite, moderately ser- pentinized dunite, and massive serpentinite. This comparison is in terms of the modified standard cell described by Chidester (1962, p. 95—97). Table 7 shows that gains and losses in alteration zones are primarily dependent upon the degree of serpentinization of host rock. Minor differences in composition of host rocks account for other varia- tions, particularly of minor constituents. For veins in only slightly serpentinized dunite (table 7, sec- tion A), the principal changes are loss of Mg and Si from, and introduction of OH into, the border zones. For veins of average ratio of width of border zone to width of vein, the proportional loss of Si and Mg from the border zones considerably exceeds the proportional amount of those constituents in- troduced into chrysotile veins. The anomalous gain in Al and Fe+3 in the border zone, indicated in table 7, section A, results from the atypical character of the analyzed sample of dunite (AV—248). The sam- ple was selected for its freshness and freedom from pyroxene and magnetite and is therefore lower than average in its content of iron and aluminum. Sample AV—13f (table 7, section B) doubtless represents more closely the average content of aluminum and iron in the ultramafic rocks. These considerations indicate that during forma- tion of veins in relatively fresh dunite and perido— tite, the supply of constituents from host rock ad- jacent to vein walls was more than adequate to form the veins and that appreciable Si, Mg, Fe, Al, and Ca must have migrated outward beyond the domain of the veins. This migration could account for many magnetite veins in the ultramafic rocks and possibly for calcite-chlorite-magnetite veins in rodingite and serpentine-chlorite rock. In extensively serpentinized host rocks, the loss of Si and Mg (table 7, sections B and C) from the walls of the adjacent host rock is inadequate (for the average ratio of width of serpentinized border zone to width of asbestos vein) to supply the amounts in the veins. Clearly then, in most in- stances, those constituents of the serpentine veins cannot be exclusively, or even largely, of local deri- vation. They must have been supplied from more distant sources by the serpentinization process. This conclusion strongly suggests that pervasive serpen- tinization of the dunite and peridotite continued through the period of formation of serpentine veins. OTHER VEINS Thin veins and seams of magnetite are fracture controlled and are formed largely by replacement. The common association with asbestos veins and the invariable association with serpentinized zones tie them to formation of serpentine veins. Many vein- form masses parallel to layering may be partly of magmatic origin, but most or all were modified and enlarged during serpentinization. Because of their rarity and the lack of diagnostic features in the Belvidere Mountain area, irregular pods of chromian magnetite, such as the large pod exposed in the town (township) of Troy (fig. 1), are of uncertain mode of origin. They may be of mag- matic origin and modified by intense local shearing, or they may have formed by replacement along shear zones. Veins consisting of varied proportions of mag- netite, brucite, and calcite formed partly by frac- ture filling, partly by replacement. The association of many such veins with, and their local replace— ment relations to, serpentine veins indicates that they are genetically related to and were emplaced during and late in the stage of formation of serpen- tine veins (see preceding section on “Serpentine veins”). Though precise relations are unknown, amphibole asbestos probably formed late during serpentiniza— tion by reaction of material in the shear zone with magnesium-bearing solutions from the ultramafic body. STEATITE, TALC-CARBONATE ROCK, AND CARBONATE-QUARTZ ROCK Geologic relations of the rocks and paragenetic relations of the minerals demonstrate that talc-car- bonate rock and carbonate-quartz rock formed by replacement of ultramafic igneous rocks and ser- pentinite; steatite formed chiefly by replacement of serpentinite but partly by replacement of country rock. The absence of a serpentinized zone between some talc-carbonate rock and enclosing dunite in the Eden quarry body, and the zoned replacement rela- tion of talc-carbonate rock to serpentinite elsewhere (see Chidester, 1962), demonstrate that steatitiza- tion is later than and unrelated to serpentinization. Relict grains of chromite and chromian magnetite throughout talc-carbonate rock and carbonate-quartz rock indicate that the rocks were derived from ul- tramafic igneous rocks. The gradational relations 60 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT TABLE 7.—Chemical composition of equal volumes of cross-fiber vein chrysotile, marginal alteration zones, [Chemical composition is in terms of the content of the modified standard cell (MSC) (Chidester, 1962, p. 95, 132—137): gains and losses are deter- of the vein. The product is the proportional amount of each constituent gained or Width of Specimen N0. Zone zone Si Al Fe:H Fe“ Mg CB- A. Host rock, slightly AV—248 ____________ D unite ______________________ >1 m 41.6 0.19 1.4 3.5 76.3 0.09 A—BM~57—3 _________ Serpentinized border zone _____ 32 mm 34.1 1.5 2.7 .41 56.1 .02 A—BM—57—1 _________ Chrysotile vein _______________ 4 mm 32.6 1.0 1.2 .37 51.0 .20 Proportional gain or loss in the serpentinized border zone ____________ —240 +42 +41 6 —99 —646 —2 Proportional amount of con- stituent introduced into the , chrysotile vein ______________________ 130 4 5 2 204 ' 1 B. Host rock, moderately AR—13f (VT—1) _____ Dunite ______________________ >1 m 38.3 1.3 2.2 2.5 69.3 0.05 AR—13f (VT—2) _____ Serpentinized border zone _____ 28 mm 35.8 .83 1.6 .19 57.1 .00 AR—55 _____________ Chrysotile vein _______________ 4 mm 34.4 .8 1.0 .58 54.2 .00 Proportional gain or loss in the serpentinized border zone ____________ —70 —13 —17 —65 —342 —1 Proportional amount of con- stituent introduced into the chrysotile vein ______________________ 138 3 4 2 217 0 C Host rock highly serpentinized AR—73 (VT—6) ______ Dunite ______________________ >1 m 34.9 1.9 3.7 1.2 54.6 0.07 A—BM—57—3 _________ Serpentinized border zone _____ 80 mm 34.1 1.5 2.7 .41 56.1 .02 AR—55 _____________ Chrysotile vein _______________ 8 mm 34.4 .8 1.0 .58 54.2 .00 Proportional gain or loss in the serpentinized border zone ____________ —64 ——32 —80 —63 +120 —4 Proportional amount of con- stituent introduced into the chrysotile vein ______________________ 275 6 8 5 434 0 1The total cations per MSC is, by definition, equal to 100xF.-, between talc-carbonate rock and serpentinite and be- tween talc-carbonate rock and dunite and peridotite indicate that talc-carbonate rock replaced serpen- tinite, dunite, and peridotite directly. Relations of carbonate-quartz rock to dunite or peridotite and serpentinite are not directly ascertainable because of exposure conditions. However, the zonal distribu- tion and gradational relations of carbonate-quartz rock and talc-carbonate at the south end of the Corez Pond body, where carbonate-quartz rock forms a small central mass Within a larger mass of talc-carbonate rock, suggest that the carbonate- quartz rock and talc-carbonate sequentially replaced serpentinite or dunite, or both. The steatitization process is discussed in detail in an earlier report (Chidester, 1962) based on studies of extensive, well-exposed talc deposits elsewhere in Vermont. That report shows that the alteration of serpentinite to talc-carbonate rock is essentially iso- chemical with respect to the principal cations and involved chiefiyxthe introduction of CO2 and loss of (OH). Steatite formed as the result of metamorphic reaction between serpentinite and adjacent country rock, chiefly by outward migration of Mg and (OH) and is here rounded to one decimal place. Because of rounding procedures, the total tion by a number as large as 80), from the serpentinite and inward migration of Si from the schist. The two processes were essentially independent, though virtually contemporaneous. These conclusions are borne out in the Belvidere Mountain area, and similar analysis is not repeated here. However, the relations of talc-carbonate rock and carbonate-quartz rock to each other and to ser- pentinite and dunite provide additional information about the carbonatization of ultramafic bodies that warrants brief comment. Calculations based on the assumption that car- bonatization was isochemical with respect to the principal cations (see Chidester, 1962, p. 121—122) yield the approximate modal compositions for talc- carbonate rock and carbonate-quartz rock derived from dunite and serpentinite that are shown in table 8. Comparison of the chemical composition of modi- fied standard cells of rocks of the modal compositions shown in the table with that of dunite and serpentin- ite from which the talc-carbonate rock was derived indicates that only for the alteration of serpentinite to talc-carbonate rock can isochemical alteration take place without change in volume—that is, that the ULTRAMAFIC AND ASSOCIATED ROCKS 6} and host dunite and serpentinite; and proportional gains and losses of constituents during vein formation mined by multiplying the width of the zone or vein by the difference between the cell content of the host rock and the zone, or by the cell content lost per zone, or introduced into a vein; dashed line. not determined. Data f rom table 1] Na K Ti P Cr Ni Co Mn on 03%: , H F Cl 0 serpentinized dunite 0.00 0.00 0.00 0.00 0.21 0.24 0.02 0.14 0.10 123.7 23.9 ___ 0.01 178.2 .02 .01 .01 .00 .26 .23 .01 .08 .26 95.7 79.7 .00 .02 172.1 .02 .02 .01 .01 .12 .16 .01 .05 .53 87.2 73.8 .00 .03 158.4 +.6 +.3 +.3 0 +2 —.3 —.3 —2 +5 —896 +1786 0 +.3 —195 .1 .1 .04 .04 .5 .6 .04 .2 2 349 295 0 .1 634 serpentinized dunite 0.00 0.00 0.01 0.02 ___ ___ ___ 0.18 0.20 113.9 47.6 ___ 0.05 178.1 .00 .00 .00 .00 .13 .18 .01 .08 .30 96.2 81.1 .00 .03 174.1 .00 .00 .06 .00 .10 .10 .00 .00 .00 91.3 81.0 .00 .05 167.2 0 0 —.3 —.6 ___ _3 +3 —496 +938 ___ —-.6 _112 0 0 .3 0 .4 .4 0 0 0 365 324 0 .2 669 dunite (massive serpentinite) 0.00 0.00 0.01 0.03 ___ ___ ___ 0.08 0.18 96.7 72.3 ___ ___ 170.8 .02 .01 .01 .00 .26 23 .01 .08 .26 95.7 79.7 .00 .02 172.1 .00 .00 .06 .00 .10 .10 .00 .00 .00 91.3 81.0 .00 .05 167.2 +2 +1 0 _2 ___ ___ ___ 0 +6 —80 +592 ___ ___ +140 0 0 .5 0 .8 .8 0 0 0 730 648 0 .4 1338 does not necessarily equal the sum of the cations shown to the left of the total. In calculations of gains or losses from a zone (involving multiplica- the discrepancy may be fairly large. TABLE 8.—Calculated modes of talc-carbonate rock and car- bonate-quartz rock formed by isochemical alteration of dunite and serpentinite, and consequent percentage increase in volume (:percentage loss of Mg plus Si in isovolumetric alteration) Carbonate- quartz rock Talc-carbonate rock From From From From . serpen- . a serpen- dunite tinite durum tinite Talc _________________________________ 50 60 _ _ _ _ Magnesite ____________________________ 50 40 70 65 Quartz _______________________________ _ _ _ _ 30 35 Percentage volume increase (:percentage loss of Mg plus Si) -_ 35 0 15 composition of the modified standard cell for each rock is the same with respect to the principal cations. For dunite, equal-volume alteration to talc-carbonate rock of the modal composition indicated in table 8 requires a loss, for both Si and Mg, of about 35 per- cent of the amount present in the parent rock; equal- volume alteration of dunite to carbonate-quartz rock requires a loss of about 45 percent. On the same basis, the alteration of serpentinite to carbonate- quartz rock requires a loss of about 15 percent for Si and Mg; the alteration to talc-carbonate rock, as stated above, is both isochemical and isovolumetric. Another way of stating the above relations is that isochemical alteration during carbonatization would require volume changes on the same order as those cited for changes in composition. The geologic and petrographic evidence available at Belvidere Moun- tain and elsewhere in Vermont (Chidester, 1962, p. 121; Jahns, 1967, p. 157) supports the conclusion that carbonatization was essentially isovolumetric. Furthermore, the irregular, abnormally wide, and mineralogically unusual zones of contact alteration observed in several places adjacent to talc-carbonate and carbonate—quartz rock masses in dunite suggest that appreciable material was displaced outward from the dunite during carbonatization. 0n the basis of the range in modal compositions observed, and on petrological considerations, it is inferred that the proportion of Mg to Si that was displaced varies con- siderably, but on the average is near their ratio in parent rock. Proportionately larger percentages of minor constituents, particularly A1 and Fe+3, were displaced; at the ultimate stage of carbonatization (carbonate-quartz rock), nearly all the Al and Fe+3 present in parent dunite or serpentinite was displaced. 62 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT These considerations suggest that, on the average, talc-carbonate rock and carbonate-quartz rock de- rived from serpentinite should differ characteris- tically from those rocks derived from dunite. Condi- tions of exposure at Belvidere Mountain do not permit quantitative evaluation of this conclusion, but the modes in table 4, section B and C lend support to it. The association of carbonate-quartz rock and quartzose talc-carbonate rock with faults or shear zones in the Eden quarry body, and the zonal distri- bution of carbonate-quartz rock and talc-carbonate rock with relation to serpentinite in the Corez Pond body, suggest that carbonate-quartz rock is localized near channels that provided free access of 00.2, and where the partial pressure of CO2 was consequently high. This relation, together with the petrologic con- siderations that have been discussed, indicate that carbonate-quartz rock is the end product of car- bonatization of dunite and serpentinite. The normal succession, where the partial pressure and supply of CO2 was sufficiently high, was the formation of talc- carbonate rock followed by the formation of car— bonate-quartz rock. Local variations in the supply or partial pressure of CO2 could account for local varia- tions in the paragenetic sequence of minerals. The presence of relict chromite in much of the steatite shows that the steatite was derived largely from ultramafic rock, but the presence of relict sphene in the outer part of the steatite zone indicates that the outer few millimeters or centimeters of steatite is derived from schist or amphibolite. The gradational relations of steatite and talc-carbonate rock in most places, and of steatite and serpentinite in only a few, indicate that in most places formation of steatite followed a preceding stage of formation of talc-carbonate rock, and only rarely was steatite formed directly by replacement of serpentinite. The invariable gradation outward of steatite into black- wall chlorite rock—in some places including an in- tervening narrow zone rich in tremolite—shows that steatitization of country rock followed an advance wave of chloritization and, locally, tremolitization. CONTACT ROCKS The diversity of contact rocks is related both to major differences in composition of the country rocks and to the several stages in the history of the ultramafic rocks. Inasmuch as the rodingite and serpentine-chlorite rock association exists within a few meters of the steatite and blackwall association, and both associa- tions are in the same kind of country rock (Chc, pl. 4D) , the two would seem to have formed under dif- ferent physical conditions and are inferred therefore to be of different ages. Because rodingite and ser- pentine-chlorite rock occur only adjacent to serpen- tinite, at contacts where talcose rocks are absent, this association seems to be genetically related to serpentinization or to some older stage. Blackwall chlorite rock, however, occurs only at contacts where the ultramafic bodies are marginally altered to tal- cose rocks. Therefore, this association seems to be genetically related to steatitization. (See Chidester, 1962, p. 89—93, for a detailed discussion of the age relations of serpentinite to steatite and blackwall.) It follows, from the conclusion that serpentinization is older than and unrelated to steatitization, that the rodingite and serpentine-chlorite rock association is older than and unrelated to the steatite and black- wall association. The association of rodingite and serpentine-chlorite rock with ultramafic bodies that contain large volumes of dunite suggests further that the rodingite and serpentine-chlorite rock may be entirely or partly related to the stage of emplace- ment of the ultramafic bodies. Neither the age of the tremolite rock and chlorite rock association relative to the other contact associa— tions nor its genetic relation to the ultramafic rocks can be determined with complete assurance on the basis of geologic relations. The absence of talcose rocks between tremolite rock and serpentinite would seem to rule out a genetic relation to talcose rocks. On the other hand, the similarity of the tremolite rock and chlorite rock association—except for the absence of the talcose rocks—to the steatite and blackwall association that contains a tremolite zone is striking. The most significant fact is that the tremolite rock and chlorite rock association is con- fined to amphibolite. This association is probably a genetic and temporal equivalent of the steatite and blackwall association, and probably forms, because of compositional limitations, where country rock bor- dering ultramafic rock is amphibolite. In each association, the zoned arrangement of the contact rocks peripheral to the ultramafic bodies, and the replacement relation the rock of a zone bears to that of the next zone outward, demonstrates a se- quential relation marked by successive mineralogical changes that proceeded outward from the ultramafic bodies. This sequential relation requires that the outer part of each zone be younger than the adjacent part of the next zone outward, but permits an entire zone to be younger than the outward adjacent zone. In the first case, the zones would be essentially con- temporaneous, and probably closely related genetical- ULTRAMAFIC AND ASSOCIATED ROCKS ly; in the second case, the zones might be of very different ages and perhaps unrelated in origin. The several rocks of the steatite and blackwall association (steatite, tremolite, blackwall chlorite rock, albite porphyroblast rock) are contemporane- ous and are genetically related. This conclusion is based upon the fact that nowhere do rocks of an inner zone cut out the rocks of an outer zone if the composition of the country rock is favorable for the development of all the zones. The conclusion is sup- ported by petrologic arguments which are essentially the same as those presented by Chidester (1962, p. 91—93) for deposits elsewhere in Vermont. The rocks of the tremolite rock and chlorite rock association are inferred also to be of contemporane- ous origin on the basis of their inferred similarity and genetic equivalency to the rocks of the steatite and blackwall association. Observed relations in the relatively scarce exposures of this association sup- port this conclusion. Though all are older than the steatitization, the rocks of the rodingite and serpentine-chlorite rock association show intersecting relations which indi- cate that, rather than being essentially contempora- neous, some of the rocks were separated in time of formation by a significant but probably small inter- val. The cutting out of rodingite by serpentine-chlo- rite rock at some places next to amphibolite, and the projection of serpentine-chlorite rock beyond the outer limits of rodingite zones along joints that in- tersect the zones, strongly suggest that the serpen- tine-chlorite rock is entirely younger than the rodingite. The intersection of rodingite by calcite- chlorite—magnetite veins shows that the veins are younger than the rodingite, but allows them to be the same age as serpentine-chlorite rock. The apparent absence of a serpentine-chlorite rock zone at the west side of the Eden quarry may be due to locally diffuse and inconspicuous development of the serpentine- chlorite rock zone. 0n the basis of the association of rodingite and serpentine-chlorite rock with serpen- tinite and dunite, and of the observed age relations of the rodingite and serpentine-chlorite rock, logical conclusions are that the rodingite may be related to the stage of emplacement of ultramafic rocks that were appreciably hotter than the surrounding coun- try rocks, and that the serpentine-chlorite rock and perhaps also the calcite-chlorite-magnetite veins are probably wholly related to serpentinization. Layering in contact rocks, which is observable commonly in rodingite and locally in serpentine- chlorite rock, blackwall, and steatite, is entirely in- herited from the protoliths, though considerably 63 modified by metasomatism. The distribution and kind of relict minerals, the similarity in appearance between layering in contact rocks and that in ad- jacent country rock, and the existence locally of layers that can be traced across one or more contact zones into adjacent country rock, all attest to the relict character of the layering. Where contacts are crosscutting and layers can be traced from one zone to another, absence of discontinuity or of systematic distortion of layers at contact between zones indi~ cates that replacement was on a volume-for-volume basis (see Chidester, 1962, p. 93—94) . Chemical changes that took place in the blackwall and steatite reaction are discussed in detail in an earlier report (Chidester, 1962, p. 122—124) and are not repeated here. Table 9 assembles analyses of suites of specimens of the rodingite and serpentine- chlorite rock association and of the tremolite rock and chlorite rock association. The analyses are pre- sented in terms of composition of equal volumes of rock (the modified standard cell) so that chemical changes can be deduced for reactions that took place on a volume-for-volume basis. These changes are con- sidered in the appropriate sections (“Rodingite,” “Serpentine-chlorite rock,” and “Rocks of the tremo- lite rock and chlorite rock association”) below. RODINGITE Rodingite formed by replacement and reconstitu- tion of country rock bordering ultramafic bodies. That the rodingite formed indiscriminately from schist, gneiss, and amphibolite is evident from its transitional relations to these rocks and from para- genetic relations of the minerals. The striking devel- opment of rodingite in schist and gneiss along the northeastern contact of the Lowell quarry body, and the relatively inconspicuous development in am- phibolite at a nearby contact, indicate that in com- parable structural settings the composition of the country rock exercises major control in the rodingite reaction. The paragenetic relations of the minerals dis- tinguish two stages in the formation of the rodingite. The first stage is characterized by alteration of mafic minerals to diopside, clinozoisite, and predominantly birefringent garnet; the second stage is character- ized by replacement of diopside, clinozoisite, and other minerals by isotropic garnet and vesuvianite. The two stages were probably not appreciably sepa- rated in time; the relations require only that stage one preceded stage two at any particular point. The distribution of the minerals indicates that the two ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT 64 .133 23 we $2 2.3 3 F505 2053 mg... «0 55m 0:» 72:6 32:38: go: wwov ~53 v5 .mm.~=cooo.a Mic—Sou «o ogaowm .woui :3:ch .28 3 vow—53 96: m; =5» {EXOOH 8 Edam .553ch \3 .mm AOmSC =8 93".:un ~55ch awn £535 133 93.; 9c| 93+ 9w| 3+ :.| «9+ 3+ 3| 3| 9N| 92+ 3| 9| 9”] 9fil ......... mTE> 3 3|?» .N 3+ Til 93+ ”3| 3+ 8. 8. 8. «9+ 92+ fifil 9| 9| 3+ 92+ ......... 51,5, 8 3&5 .H cc 53231 E om: awn Alv mwofi ho A+v END £09. E 35 9:: wov 8. S. :9 2. 3 my 9S «.2 3 NS 9:. Bzofiubétoio :5: Eu 3-.- 3:32nfia 2280 ...... ETE>V§NI>< .Oum—DQm—Afigu v—QO.~ 9E; 95 9:: «.N 3. 3: ad 8. 2. 9m 9% 9m 9 92 9S ".2313: 8 «5334‘ Ezofiobétoio ...... $TB>ZONI>< .u—uon 92: 9S 9": 2. mo. 8. S. 8. 3. 9S 9% 3“ v. 9” ".3 I: £5238,“ 3 «532% 323.3-quEw; ...... ASIE>E¢NI>< 92: 93 35 m: 85 86 35 85 cad 36 3m 9m 9— N.“ «.5 -- :2; 3:25: 3 upwuamua‘ .......... 3::anme ...... $TB>Z§1>< :omufiogmn “—2: STE—Au min :00.— oEoEub 05 mo 3qu .0 93! 93+ MEI :A 8+ «9| we. | «9| S.| wfil 92+ NNI 2| 2: 9mwl ......... 8&5 8 SJ?» .m 3+ N.w+ 3+ -- 8+ 8+ 2. + «9| 5| 2+ 3+ 9! 9+ 2+ 3.1 ......... $1.?» 8 malty .m «25+ «.mul 93+ -. 8+ 8.] 8| 3.1 3| 3+ 9+ 3| 3+ 5| 2+ ........ 3&5 8 max?» A ”mo :cmaauofiu E 092 hwg Alv me— ho AITV EmfiU x mm: 5: 98 5V NH. mm. m9 9“ 3 «6. M2. 3 m; 2: 2m ................ :92 E“2:50 ...... $23 052920 ...... annex/32¢!» .XOQH euro—:0 :3 .3 5:: 5V mu. 2. am. «9 E. 9: 9m 9m 9“ 9M: 98 $523“: Ea; Eu m ............... ow .......... ANNIH>VSTK< .XOOH wath—Jo 93: m: 2: 3V mu. cm. 3.. No. S. 93 N.“ 9“ km gm 9?. .2553»; 3 «53.23 ............ eguiuom ...... :mJECSTE‘ .v—Uoh w.m5 OE. owm «A mud 36 med cod cad HA mdv ES m4 9m wdm ....... mfiMEcoa 3 «53me wanEubcfinmgwm ...... SNIHNCSTKAV View 3 «59am: :omummoommw “—2: BEoEuéEEE—Bm can BEER: 2: «o mmoofl .m 3+ 3+ w.m+ «9| S.| 19+ S.| o c 3+ m3+ 3| 9| 3| 9m+ ......... SAC, 8 21.3, .w 9om| 95+ m.m~| 3+ co. + wo.| 5| o a. | 9mm! 9% .3| 9+ 9N| 93| .......... HTE> 8 SIS, .m 9m+ 9N+ 9+ 3+ 8| 3+ o “9| 3| 92+ MEI 9+ 9| :1 9| --- SJ?» 8 mfiummlsmé .V 3+ 2+ 3+ -- 5| “9+ 3+ «9! 9m| 92+ 9? 9N+ 3| 5+ 2| ilcTs> S wmummlsm|< .m 3+ 9m+ 3+ 8+ 8.... 29+ 5+ 3| 9m| 93+ 9 + 9~+ 9+ 9| 3| -- S|B> 8 37%|“:me .N ad... 95+ v.m+ 3.+ Ne.| mo.+ wail ov.| “6| v.NH+ ms IT 9m... 9| c.N| w.m| .......... SIE>3NHIH> A .MO EO—adaw&—§ n: Mums .Hwn Alv man: .5 A+V :50 8me «tom ham: 8. pm. 3. mm. vw. m3. QNH «.3 93 «in flow «.5 ...................... cc ................... oc ............ amfilmmIEmldw aw: Wm“ Wm: g. V mm. 3. w. E“. o.«. 92 v. 5 as ed 95 «.3 ...................... on ....... EEEEQEw $300 ......... waimmIEqu. aAmfi v.5 m2: 8. 3. ma. 3.. cm. NJ. N»: 5. NH as N; Nam: m.um ...................... ow ................... om. ............. olemmIEme 5w: 3 9o: 5. mm. 3. S. 2.. 9m 92 H. u 9w 9“ «.3 9mm ............... x2: .5550 ..... EzonEnEa mcrm ...... QTH>ET>< 9mm: 9mm 9m: we. Hm. 5. ea. 2. 2.. 9mm 9 m: 9.: 9H 3: 93. .......... SEEE: «a .6?an ............ mtmfiuomi STEESISJFF 9me 9.; 93 3. cm. 8. m9 8. co. mm. «.3 9m 9m 93 «.5 ....... 332:? 8 «532% ............... on ...... :TBEHTSIHAE .v—ooa Na: 9:. «.3 Ed wad m; 3.5 8.9 S; :3 98 99° 3 d: 95 - “23:8 ezmzbfi 3 £6326 moraiobnszmfiwm ....... $TE>§T>< 3:93:38“ 8 «53:3 :ofiumoomma in: Bio—aoéiuaumaom ER 8345.9. 2: mo mica .< 82 E o m @555 $0 :2 A E. M .2 do «2 4mm mi 2 «m 5369. 3ch 25 moom 52 5.58am Eon. HA wide Ea: mmmfidzmw .cwimfiagww «o: 6:: cw£mdfl .CMHINMH .mm d .Nme .hwuwwESOv .353: $3 chdwcflm @0519: 2: mo mar—3:00 0:» no mEhwu cm mm :omfimoafioo 33:5:an wmugwsvww 33:8 98 .83: \e 3533 3:3 .3 «6me .3 mficm @33383 ~32» :awtweafieu ~38§B~DI$ mamflb ULTRAMAFIC AND ASSOCIATED ROCKS 65 stages of alteration proceeded outward from the contacts of the ultramafic bodies oru-particularly in the case of the coarse rodingite—from fractures or fracture zones adjacent to the bodies. Minerals of coarse rodingite—garnet, vesuvianite, clinozoisite, zoisite, diopside, calcite, and locally, prehnite—formed largely by fracture filling, and the two stages distinguishable in the fine rodingite are not evident except locally at the margins of fissures. The fabric of fine rodingite is considerably differ- ent from that of the protolith. Alteration to diopside resulted generally in a reduction in grain size with respect to the protolith. Development of garnet and vesuvianite imparted a massive character to the rock. Clinozoisite, most relict minerals, and the titanium minerals retain generally the textural features and distribution pattern of the protolith. The first stage in the development of rodingite was marked by mineralogical changes and little or only moderate change in composition. Diopside formed almost solely by alteration of mafic minerals, chiefly and perhaps entirely hornblende; the rela- tion is shown by the incomplete alteration of horn- ' blende to diopside along cleavages and at margins of grains, at many places in the transition zone be- tween rodingite and country rock. Much, probably most, clinozoisite is a relict of the protoliths, but most was appreciably modified in composition (the content of iron reduced) during alteration. Only cores of the larger rounded grains appear to retain their original somewhat more ferriferous composition, The second stage was marked by greater composi- tional changes. Garnet and vesuvianite formed by replacement of a variety of minerals of contrasting composition—diopside, clinozoisite, albite, and other relict minerals of the country rock. Scalloped boun- daries of the garnet and vesuvianite, convex toward the rest of the minerals, and inclusion within them of diopside, clinozoisite, and other minerals, are evi— dence of the mode of origin. The abundance locally of island remnants of gar- net, diopside, epidote, and vesuvianite in areas of chlorite, and the persistence with which chlorite in- vades and embays both those minerals and the min- erals of the country rock, indicate the replacement origin of much of the chlorite. On the other hand, the occurrence in a few places of chlorite in contact with diopside, clinozoisite, garnet, and vesuvianite, but not embaying or invading them, suggests the com- patibility and contemporaneous origin of the min- erals concerned. There is no direct evidence of the age of chlorite that replaces the country rock min- erals, relative to chlorite that replaces the minerals of the rodingite. The simplest interpretation appears to be that the chlorite in apparent equilibrium with diopside and other minerals of the rodingite, and the chlorite that replaces the minerals of the country rock, are contemporaneous; chlorite that replaces the minerals of the rodingite is distinctly later. In the development of rodingite, the titanium min— erals appear to follow the progressive alteration of rutile to ilmenite to sphene, for sphene is greatly pre- dominant, rutile is present only as a core surrounded successively by shells of ilmenite and sphene, or il- menite forms the core alone. Some sphene, particu- larly euhedral grains, possibly originated in the alteration of slightly titaniferous hornblende to less titaniferous diopside. If so, the proportion is small, inasmuch as protolithic amphibolite contains about the same amount of titanium minerals as rodingite. Prehnite mostly fills tiny fractures in fine roding- ite. In rare instances where it is intermixed with other groundmass minerals, prehnite presumably formed by replacement, possibly of albite. Other minerals in rodingite are relicts of the protoliths. Increase in content of Ca is the dominant chemical change in the alteration of both amphibolite and schist to rodingite; the content of Na and K also decreased consistently (table 9, sections A and B). Other changes show no significant pattern. Infer- ences concerning changes in content of (OH) are unreliable because of widespread retrograde altera- tions (particularly chloritization) that affect both rodingite and amphibolite. Assemblages in the rodingite are of apparently higher metamorphic grade than those of the schist and amphibolite, but the narrowness of the rodingite zone indicates that factors leading to formation of rodingite were limited to a narrow zone at margins of the larger ultramafic bodies. The possible control— ling factors in the narrow zone include increased temperature, decreased partial pressure of water, and influx of Ca. Increased temperature in a narrow zone surrounding the ultramafic body may have re- sulted from the ultramafic rocks being hotter than surrounding country rock at the time of emplace- ment. A lowered partial pressure of water immedi— ately adjacent to the ultramafic body may be attributed to the desiccating effect of dunite in the larger bodies. Influx of Ca may have resulted from the outward migration of Ca from the ultramafic rocks as a result of serpentinization (see table 7, section A—D). 66 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT SERPENTINE-CHLORITE ROCK Serpentine-chlorite rock formed almost entirely by replacement of rodingite. Locally, along small faults that intersect the rodingite zone and where serpen- tine-chlorite rock transects the rodingite zone, ser— pentine-chlorite rock replaces amphibolite directly, but such places are relatively rare, and the amount of serpentine-chlorite rock so formed is small. These conclusions are evident from relations of the rocks, relict minerals, and preservation of a distinctive pat- tern of former grains and patches of garnet and vesuvianite at many places in the serpentine-chlorite rock. Almost the sole reaction in formation of serpen- tine-chlorite rock is wholesale conversion of the min- erals of rodingite—or, in rare instances, amphibolite —to a mixture of chlorite and serpentine. The only exception is a change involving the titanium miner- als sphene, rutile, and ilmenite. The predominance of chlorite in the outer part (next to rodingite) of the serpentine-chlorite rock zone and of serpentine in the inner part (next to the ultramafic body) indicates that at all stages during development of the zone chlorite predominated over serpentine near the interface with rodingite, and ser— pentine over chlorite near the interface with serpen- tinite. The initial stage at any particular point was thus marked by alteration of minerals of the roding- ite to chlorite and minor serpentine. As the zone widened, more and more of the chlorite must have been converted to serpentine, so that the end product was a zone composed at the inner margin predomi— nantly of serpentine and of only minor chlorite, and at the outer border, predominantly of chlorite and of only minor serpentine. During formation of serpentine-chlorite rock, both sphene and rutile altered to ilmenite, for both are absent, whereas ilmenite is present. The relation of single titanium-mineral grains at the contact be- tween rodingite and serpentine—chlorite rock (see section on “Serpentine chlorite rock” under “Contact rocks” in the section on “Ultramafic and associated rocks”), Which consist of sphene on the rodingite side and of ilmenite on the serpentine-chlorite side of the contact, further supports this conclusion. An increase in total content of titanium minerals in serpentine-chlorite rock as compared with rodingite requires a source outside the titanium minerals. Diopside, garnet, and vesuvianite are possible sources of both the required iron and titanium. Table 9, sections A and B, show that the dominant and only consistent chemical change was a several- fold increase in content of Mg. Other changes vary with composition of the country rock and reflect chiefly decreases in content of Al, Fe“, Ca, Na, and K imposed by the limiting compositional ranges of chlorite and serpentine. CHLORITE-CALCITE-MAGN ETITI‘I VElNS Veins of chlorite, calcite, and magnetite associated with rodingite and serpentine-chlorite rock are younger than rodingite, which they crosscut. They are also at least partly younger than serpentine- chlorite rock, for locally they crosscut it, but they are virtually contemporaneous with the serpentine- chlorite rock. The minerals of the veins formed largely by fissure filling, as is shown by the euhedral form of much of the chlorite and magnetite, the common orientation of columnar calcite and elongated books of chlorite normal to the vein walls, and the geometry of the veins—locally wedge form, generally smooth walls, matching irregularities in opposite walls (though rare) . The generally small scale of irregularities in- dicative of replacement implies that replacement was of minor importance in formation of the veins. These relations, together with petrologic consid- erations based on interrelations between serpentine veins, pervasively serpentinized dunite, rodingite, and serpentine-chlorite rock (see section on “Ser- pentine veins” under “Petrogenesis” in the section on “Ultramafic and associated rocks” and in the same section see “Serpentine-chlorite rock” under “Con- tact rocks”), suggest that chlorite-calcite-magnetite veins formed essentially contemporaneously with serpentine veins. These relations also suggest that the constituents were derived from excess cations (chiefly Mg, Fe, Ca, A1, Si) displaced from wallrock adjacent to the serpentine veins and from that part of the pervasive serpentinization of dunite that ac— companied formation of the serpentine veins. ROCKS ()I’ THE lefiATlTlfl ANI) BLA(IK\VALI. CHLORITE ROCK ASSOCIATION The geologic relations, relict minerals, and textur- al features indicate that blackwall and steatite are contemporaneous and that blackwall formed by re- placement of the schist bordering the serpentinite, whereas the steatite formed chiefly by replacement of serpentinite or talc-carbonate rock but partly by replacement of blackwall. Where the schist con- tained sufficient calcic and mafic minerals, a trem— olite zone formed between the steatite and black- STRUCTURE 67 wall; where it was albitic, an albite porphyro- blast rock zone formed at the outer margin of the blackwall zone. By analogy, it seems likely that a relatively high content of muscovite in the country rock may have led to heavy concentration of musco- vite at the outer part of the blackwall zone. The textural relations of the minerals indicate that albite porphyroblast rock has replaced the schist, blackwall has replaced the albite porphyroblast rock, and steatite has replaced the blackwall. Such rela- tions clearly indicate sequential development of the rocks. These relations and the contemporaneity of the zones indicate that the zones formed simultane- ously by outwardly migrating reactions, so that the outer part of each zone is younger than the inner part of the next zone outward. Tremolite rock in the blackwall-steatite associa- tion replaces both blackwall and steatite at either margin. It therefore widened at the expense of both and differs from other rocks in the association in that at its outer margin it is younger than the inner part of the blackwall and at its inner margin it. is younger than the outer part of the steatite. The zone as a whole, however, is the same age as the others. Chlorite, the sole essential constituent of the black- wall, formed by replacement of all the minerals of the schist. Albite formed in the adjacent albite por- phyroblast zone by enlargement of preexisting crys- tals of albite and by growth of new crystals; concen- trations of muscovite in the same relative position probably formed similarly. Tremolite grew by re- placement both of chlorite in the blackwall and of tale in the steatite. Talc in the steatite zone formed chiefly by replacement of serpentine, in some places directly but in most places with an intermediate stage of formation of talc-carbonate rock. Where tremolite rock is absent, talc also replaced chlorite in the blackwall zone for a short distance beyond the original contact of the ultramafic rock. In the blackwall, albite porphyroblast rock, and tremolite rock zones the titanium minerals under- went a progressive alteration from ilmenite—>ru- tile—>sphene, illustrated at-many places by cores of ilmenite surrounded successively by shells of rutile and sphene and by the great preponderance of sphene over ilmenite and rutile in general. The de- crease in content of sphene in the alteration of black- wall to steatite suggests that in the alteration, ap- preciable titanium was incorporated in talc. ROCKS OF THE TREMOLITE AND CHLORITE ROCK ASSOCIATION Petrogenic relations of the tremolite and chlorite rock association are somewhat doubtful, partly be- cause the rocks are unevenly distributed and er- ratically exposed. Chlorite rock replaces amphibolite bordering serpentinite, and tremolite replaces the chlorite rock. Tremolite may replace serpentinite, but evidence was not seen. The tremolite and chlorite rock association is inferred to be equivalent to the steatite and blackwall association in an amphibolite environment, and, by analogy, tremolite rock is in— ferred to be the same general age as chlorite rock. OTHER ASSOCIATIONS Irregular alteration zones in country rock adjacent to masses of talc-carbonate rock and carbonate- quartz rock in dunite superficially resemble the blackwall assemblage but are probably related to carbonatization rather than to steatite-blackwall a1- teration. Because exposure is inadequate, precise relations are not known. However, the general as- sociation of such alteration zones with talc-carbonate and carbonate-quartz masses in dunite, and petro- logic considerations as discussed before (in section on “Steatite, talc-carbonate rock, and carbonate- quartz rock” under “Petrogenesis” in section on “Ultramafic and associated rocks”) suggest that the alteration of country rock in these zones resulted from influx of material displaced from dunite when the dunite was altered to talc-carbonate rock and carbonate-quartz rock. The influx of constituents (principally Mg, Si, Al, and Fe) resulted in outward displacement of constituents of the schist, forming crudely zoned but irregularly distributed concentra- tions of talc, chlorite, sericite, and albite. STRUCTURE The structure of the Belvidere Mountain area is dominated by folds and by intrusive bodies of ultra- mafic rock, whose combined effect imparts a compli- cated sinuous pattern to the lithic units (pls. 1, 2, and 3). Faults play a comparatively inconspicuous role, being apparent chiefly in minor offsets in con- tacts of ultramafic bodies. In addition to major struc— tures, a wide variety of minor structural features—— layering, foliations, planar features, warped sur- faces, and lineations—impart distinctive character- istics to the rocks and bear consistent and signifi— cant relations to each other and to major structures. 68 ASBESTOS-BEARING ULTRAMAFIC ROCKS 0F BELVIDERE MOUNTAIN AREA, VERMONT MAJOR STRUCTURAL FEATURES On the basis of major structural trends (that is, those evinced by map patterns), the area is divided into three parts, each having characteristic rock as- sociations. In the northeastern part of the area, which consists chiefly of metamorphosed sedimen- tary and volcanic rocks, structural trends are pre- dominantly about east. In the southwestern part, which also consists chiefly of metamorphosed sedi- mentary and volcanic rocks, the trends are about north. In the zone between these two parts, which consists chiefly of intrusive ultramafic rocks, struc- tural features of the intrusive bodies are prominent, and structural trends are diverse (pl. 1). The bodies of ultramafic rock are grossly conform- able with the country rock formations and with in- tervening contact rocks. The configurations of the bodies vary in complexity with size. The smaller bodies show chiefly intrusive structures and are simple lenses. The large bodies show both intrusive structures and two sets of folds and are complexly warped. Two sets of major folds emerge from the pattern of diverse and contrasting structural trends. In the northeastern part of the area, the two sets of folds are markedly divergent, and east- to southeast- trending folds predominate greatly over southwest- trending ones. Southwestward, both sets warp around more and more into the south, and the two sets become less and less divergent, so that in the southwesternmost part of the area they are nearly parallel and trend about north. The character and style of the two sets of major folds (hereafter referred to simply as “southeast- trending” and “southwest-trending”) and the nature of the relation between the two sets are not known with certainty, because conditions of exposure are inadequate to determine precisely the map pattern of the rock units. Differences in style are not strik- ing, but the southeast-trending folds generally are compressed and attenuated, and the southwest-trend- ing ones are somewhat more open. Faults and shear zones in the area commonly do not intersect major folds, and so the relations of the one to the other cannot be determined unequivocally. Both faults and shear zones are unaffected by fold- ing, and both intersect and offset contacts of in- trusive bodies. ULTRAMAFIC INTRUSIVE BODIES Most of the major structural features of the ultra- mafic bodies, such as shape, layering, inclusions and septa, and schistose zones characterized by shear polyhedrons, are independent of the adjacent country rocks, but folds and shear zones are shared with the country rocks. In general, features of the ultramafic bodies are intersected by one or more sets of the folds and shear zones. SIZE AND SHAPE The ultramafic intrusive bodies are lenses for which the ratio of strike length to thickness appears to be roughly on the order of 5:1. Three—dimensional relations can be determined directly only for the main ultramafic body. In it, the intermediate axis is about parallel to the strike of the principal plane; the major axis is about downdip and roughly twice the length of the intermediate axis. The geometric relations of the other ultramafic bodies are inferred to be generally similar. The main ultramafic body is bordered on the southwest and northeast by southeast-trending syn- clines; between them, the now eroded upper contact of the body forms a broad anticlinal arch trending southeast about coincident with the major axis of the lens. Small sharp warps in the lens marked by in- folds of Belvidere Mountain Formation (€bf)—-one just southwest of the Lowell quarry, a second near 103,500 N., 93,000 E. on the northeast flank of Belvi— dere Mountain, and a third between the main peak and the west peak of Belvidere Mountain (pl. 1)— are produced by synclinal folds of the southeast— trending set. The folds that bring the main ultra- mafic body to the surface in an anticline in the vicin- ity of Corez Pond, and probably the broad syncline that warps the body below a cap of amphibolite at the crest of Belvidere Mountain, belong to the south- west-trending set. In overall attitude, the main body strikes about northeast and dips gently to moderate- ly southeast. CONTACT RELATIONS Intrusive contacts of the small ultramafic bodies are almost entirely conformable; those of the main ultramafic body are generally grossly conformable with the adjacent map units of country rock, but they are slightly to moderately crosscutting in many places. These contact relations are shown at various scales in figures 3 and 4, and plates 1—4. Both the strictly conformable and the crosscutting relations are best shown in the vicinity of the Lowell quarry, in the area of the tenuous connection between the Lowell quarry body and the Corez Pond body, and at several places along both the northeast and the STRUCTURE 69 southwest contacts. Abundant isolated inclusions and projecting septa of wallrock in the serpentinite fur- ther demonstrate the prevalence of slightly crosscut- ting relations in the Lowell quarry body. INCLUSIONS AND SEPTA Inclusions and septa of wallrock are exposed at many places in the Lowell quarry body and at a few places in the less well exposed Eden quarry body. A large projecting septum of coarse amphibolite sepa- rates two clawlike prongs of ultramafic rock at the northwest end of the Lowell quarry body near 101,000 N., 96,000 E. (pls. 1 and 3). Lenses of coarse amphibolite extend from this septum southeastward in the ultramafic body in an irregular and discon- tinuous pattern toward another septum that pro- jects from the northeast wall near the southeasern end of the quarry. Near 99,900 N., 98,300 E. (pl. 3, this septum becomes glued tight to the schist, but about 100 m farther southeast it is separated from the schist and is breached by a thin tongue of serpentinite that projects from the Corez Pond body and connects with the Lowell quarry body. A large slab of schist and gneiss projects conspicuously from the northeast contact near the center of the Lowell quarry (pl. 3). Several small septa of amphibolite are exposed near the southeast contact of the Eden quarry body (pl. 1). - Small pendulous or lenticular masses of amphibo- lite, completely isolated within the serpentinite or joined to wallrock by only a narrow neck are a dis- tinctive feature of the southwest contact of the Lowell quarry body between 100,000 N., 97,400 E., and 99,800 N., 97,600 E. (pl. 3). The contacts of these masses are conformable with bedding schistosity except in narrow necked-off parts of a mass and at faults, and are conformable with thin zones of ser— pentine-chlorite rock and rodingite that border the masses. Some of the masses may be nearly cigar- shaped in plan, but most or all are probably elon- gated vertically, so that their shape is roughly like that of an airplane Wing (pl. 43 and C). These fea- tures are discussed further in the section “Folds in ultramafic rocks.” LAYERIN G The overall pattern of layering in the ultramafic rocks is known only sketchily for most of the ultra- mafic bodies because of both insufficient outcrop and insufficient detailed study outside the active quarry areas. Scattered observations on the eastern slope of Belvidere Mountain and detailed mapping in the C- area (pl. 1) suggest that the layering there is very roughly conformable with the intrusive contacts. Layering in the dunite near the center of the Lowell quarry is markedly unconformable with the contacts of the body. In this large mass of dunite, the layer- ing has a mosaic pattern in which the units are ir- regular fault blocks a few tens to a few hundred feet across (pl. 3). SHEARING The outcrop pattern of shearing in the ultramafic bodies ranges from essentially linear tabular shear zones (fig. 3 and pl. 3) to very irregular schistose serpentinite masses characterized by shear poly- hedrons (pls. 3 and 4A). In places, tabular shear zones intersect and can be traced through irregular zones of schistose serpentinite. Elsewhere, the shear zones can be traced only a short distance into the ir- regular schistose areas, where they lose their identi- ty (fig. 3 and pl. 4A, B, and C). Some of the shear zones are about parallel to the northeast-trending set in the adjacent wallrock, and some—such as those in the dunite mass near the cen- ter of the Lowell quarry—have regularities of pat- tern throughout areas as much as a few hundred feet across. Many are diversely oriented, however, and no general pattern is discernible for the area as a whole. The irregular zones of sheared serpentinite are generally peripheral to the bodies. In many places, these general relations are obscured by irregularities. SOUTHEAST-TRENDING MAJOR FOLDS Several folds of the southeast-trending set are dis- tinguishable in the area. Most conspicuous is an east- trending syncline that parallels the eastern half of the northern border. Southwestward, successive syn- clines have increasingly southeast trends (compare the synclines northeast and southwest of the Eden quarry body). A syncline west of Scofield Ledges, near the 89,000 E. grid line (pl. 1), whose trough is marked by a long-narrow, north—trending belt of fine amphibolite, probably also belongs to this set of folds, but it cannot be distinguished with certainty from the southwest-trending set. These four narrow synclines are separated by appreciably wider anti- clinal tracts. Other, smaller folds in the southwest— ern part of the area may also belong to this set. All folds of the set tend to converge in plan toward the northwest corner of the area. Folds of the southeast-trending set are nearly iso— clinal; axial planes are about vertical. The gross map 70 ASBESTOS-BEARING ULTRAMAFIC ROCKS 0F BELVIDERE MOUNTAIN AREA, VERMONT pattern requires the fold axes to plunge gently to the east, southeast, or south. Except near the main ultramafic body, second-order features in the fold pattern are inferred from generally inadequate out- crop data. Most of those shown in figure 3 are large- ly interpretive, but none show mirror symmetry with respect to the axial planes of» the major folds. Rath- er, the pattern is the same in the region of the axial plane and in both limbs of the major fold. SOUTHWEST-TRENDING MAJOR FOLDS Folds of the southwest-trending set are subordi- nate to those of the other set. The principal ones are the synclinal and anticlinal folds at the southeast margin of the Eden quarry body, responsible for the complex map pattern in the vicinity west of Corez Pond (pl. 1) ; the folds in the contact between the Ottauquechee and Stowe along the southeastern mar- gin of the outcrop area of the Ottauquechee; the broad synclinal fold in the amphibolite that caps, Belvidere Mountain; and the folds that impart a zig- zag pattern to the northeastern contact of the Eden quarry body and to the nearby formations to the northeast. The folds are moderately open in style and have steep axial planes. The plunges of the axes are vari- able, depending largely upon relations to the south- east-trending folds and to the contacts of the main ultramafic body, but the predominant plunge is gent- ly to moderately south or southwest. Representation of the detailed configuration of the folds is to a con- siderable extent interpretive, though appreciably less so than in the case of the southeast-trending folds, but it is reasonably clear that the secondary fold pat- terns have mirror symmetry with respect to the axial planes of the major folds. FAULTS Observed faults, minor in their extent and struc- tural effect, are confined virtually to the vicinity of the ultramafic bodies, but such location may be large- ly due to conditions of exposure. A small fault at the northeast margin of the Eden quarry body just west of the Lowell quarry (pl. 3), exposed for a map dis- tance of about 100 m and inferred to extend along the contact for an additional 400 m, probably has a displacement of not more than a few tens of meters. The fault trends northwestward and dips moderate- ly to the southWest. Narrow shear zones a few tens of meters to more than a hundred meters long are abundant in the Vicinity of the ultramafic bodies (pl. 3). Most of them belong to a set that trends a little east of north and is nearly vertical. Others are diversely oriented, commonly irregular, and dip moderately to steeply. Though all types occur in both the ultramafic bodies and in the country rock, the set that trends east of north predominates in the country rock, the diverse— ly oriented ones predominate in the ultramafic bodies. The linear belts of talc-carbonate rock that trend northward within the Eden quarry body near the 94,000 E. mine coordinate may mark fault zones, but there is no direct evidence to so identify them. MINOR STRUCTURAL FEATURES Minor structural features (which are represented on maps by a structural symbol) in the area com- prise a wide variety of foliations, structural sur- faces, zones of distributed shearing, folds, and linea— tions. For these, the petrographic aspects have been described under the section on “Rocks.” In the pres- ent section, the geometric relations of the structural elements are described. The terminology is for the most part that com- monly used, but a statement of usage for some terms may be desirable (see Chidester, 1962, p. 17). Bed- ding schistosity is everywhere parallel to bedding, even in the noses of small isoclinal folds; it is gen— erally, perhaps always, a continuous schistosity, in which all platy minerals are uniformly parallel. Transverse or transecting schistosity transects bed- ding (and bedding schistosity) in suitable structural situations, such as in the limbs of folds that diverge sufficiently from the axial planes, and in the noses of isoclinal folds. Transverse schistosity is always a spaced schistosity which consists of discretely spaced surfaces of flexure, fracture, or shear that mark dis— continuities in the fabric of the rock. Slip cleavage is a transverse schistosity associated with clearly identifiable flexures or crinkles in an older schistosi- ty. Fracture cleavage consists of closely spaced frac— tures without significant alinement of minerals along them. Under field conditions, transverse schistosity is not always clearly and readily identifiable as eith- er slip cleavage or fracture cleavage; indeed, it may appear megascopically to be a continuous schistosity. Microscopically, however, it is always identifiable as a spaced schistosity. In much of the ultramafic rock, schistosity cannot be classified on the basis of intersecting relations and must be classified simply as continuous or as spaced schistosity. STRUCTURE The associations and intersecting relations of the minor structural features are systematic, and their spatial patterns are of differing complexity (homo- geneity and symmetry) at the scale of normal mega- scopic observations and for the field of the entire area. In general, the simpler the spatial pattern of a structural element, the larger the minimum domain for which the element is penetrative (for example, the wider the spacing of a regularly spaced schis- tosity) and the fewer the structural elements that intersect it. From these relations, it follows that the simpler the spatial pattern of a structural element, and the larger the minimum domain for which the element can be considered to be penetrative, the younger the element. On the basis of these relations, many of the classes of structural features can be divided into two or more sets of different age, such as older and younger folds and related structural elements, and several ages of schistosity. As a general rule, minor structures in the country rock cannot be related directly to those in the ultra- mafic rocks, including those that involve the con- tacts, because of the lack of suitable elements in common. Similarities in style and pattern, and other relations, permit indirect correlation. Because they are not directly related, however, the structural features in the two kinds of rocks are best treated separately. BEDDING Bedding commonly can be traced for as much as several tens of meters or the entire extent of an out- crop in quartzite, in highly quartzose schist and gneiss, and in much of the amphibolite. In the less quartzose thinly laminated schists, the bedding is similarly distinct and traceable in favorable struc- tural settings, but generally it is traceable for only a few centimeters or is virtually indistinguishable, particularly in the highly graphitic schist and in the chloritic schist characteristics of the Stowe. Indi- vidual bedding laminations appear to be parallel to larger bedding units in most places but diverge slightly where facies relations are evident, as in the area depicted on plate 4D. The attitude of bedding varies greatly, from vir- tually planar through simply folded to intricately contorted. Individual measurements of bedding atti- tudes throughout the area range virtually through 360° in strike and vary from horizontal to vertical in dip. In the northeastern part of the area, strikes deviate only moderately from east, and dips are 71 moderate southward. In the southwestern part of the area, strikes of bedding are predominantly a little either side of north; dips are steeply either side of vertical. The central wedge-shaped area has more diverse bedding patterns and no predominant trend. BEDDING SCHISTOSITY Bedding schistosity is best developed in rocks that have distinctive micaceous layers, but it has com- monly been obscured or obliterated by intersecting structural features. The spatial pattern of bedding schistosity is, of course, identical with that of bed- ding. LAYERING IN ULTRAMAFIC ROCKS Layers from a centimeter to as much as half a meter thick are conspicuous in most of the dunite and periodotite and in much of the massive serpen- tinite. The layering results from differences in color, texture, and mineralogy. The layering is nearly planar or only moderately warped in most places, but at several places in the C-area it is folded in simple patterns on a small scale. At the many small faults that offset the layer- ing by a few centimeters, the layering generally differs appreciably in attitude on either side of the fault. Layering is commonly transected at a high angle by zones of schistose serpentinite, but the re- lations are variable, and in places the boundaries of schistose zones are parallel to layering. Observed attitudes of layering are diverse and conform to no simple pattern. DIMENSIONAL ORIENTATION OF PRIMARY MINERALS IN DUNITE AND PERIDOTITE Most of the primary minerals in the ultramafic igneous rocks are about equidimensional; therefore, dimensional parallelism of minerals in these rocks is seldom noticeable. In places, p‘seudomorphs after pyroxene and rare relict grains of pyroxene are of tabular habit; such tabular grains are generally alined roughly parallel to the layering. Preferred crystallographic orientation of olivine in dunite is shown in a few thin sections by crude mass parallel extinction. The olivine is commonly alined with (010) cleavage parallel to preferentially serpentinized layers, which locally have a parallel continuous schistosity. The macroscopic relations of this foliation in the dunite and peridotite are not known, but these few microscopic observations sug- gest that the foliation may be parallel to layering. 72 ASBESTOS-BEARING ULTRAMAFIC ROCKS 0F BELVIDERE MOUNTAIN AREA, VERMONT SHEAR POLYHEDRONS Shear polyhedrons (Chidester and others, 1951, p. 7) consist ideally of closely packed polyhedral units of relatively unsheared massive serpentinite surrounded by generally thin zones of highly sheared schistose serpentinite. The polyhedral units com- monly present an irregular pattern, and the schis- tosity in the enclosing thin sheared zones conforms with the outlines of the units. In many places, the polyhedrons are distinctly triaxial, their principal planes are crudely alined with the zone of shearing, and the schistosity tails off at the ends of poly- hedrons tangentially to their principal planes. These relations are complicated and obscured in many places by the intersection of distributed shears re- lated to small faults. FOLDS Minor folds are prevalent in both the metamor- phosed sedimentary and volcanic rocks and the ultramafic rocks. In both, the folds can be grouped into several sets on the basis of style and interrela- tions, and, in both, the form surfaces (McIntyre and Weiss, 1957, p. 149; Weiss, 1959, p. 16) that define the folds can be divided into types that differ in kind, particularly in tectonic relations. FOLDS IN THE METAMORPHOSED SEDIMENTARY AND VOLCANIC ROCKS Minor folds in the metasedimentary and metavol- canic rocks may be classified principally into two groups on the basis of style and intersecting rela- tions: an older set of tightly compressed plastic folds, and a younger set of more open brittle-style folds which refold the older folds where the geo- metric relations are appropriate. Both these sets of folds range in amplitude from a few centimeters to as much as several meters and involve generally a considerable number of beds. A third type of fold is consistently on the order of a centimeter in ampli- tude and is shown only by the pattern of quartzose layers. These folds are plastic in style, and the limbs are tightly compressed. The distinctness of the fold form varies widely, so that in places the feature is distinguishable only as quartz rodding. In general, the axes of the older folds and of the folded quartz lenses are at a large angle to the axes of nearby younger folds. Some of the older folds are ‘ complex in form, and their structural elements vary markely in attitude. The younger folds, though also variable, are more regular and are uniformly simple in form. Form surfaces that define older minor folds and isoclinally folded quartz lenses are invariably bed- ding or bedding schistosity, whatever the scale of observation. Where older folds and folded quartz lenses are absent, bedding and bedding schistosity also serve as the form surfaces of the younger folds; where older folds are present, the form surfaces of the younger folds consist of transverse schistosity and axial surfaces of the older folds. Older folds.——Minor folds of the older set are sparse in the area. An excellent example of one, folded by a later fold, is exposed near 103,500 N., 94,300 E. (pl. 1). The limbs are isoclinal, the axial plane dips moderately south, and the fold axis plunges gently west overall, but all these elements are folded about a younger southwest-trending axis. A few other minor folds having isoclinal limbs and anomalous axial trends, but having no evident rela- tion to younger folds, are probably of the older set; the fold symbolized near 100,940 N., 96,580 E. is probably an example (pl. 3). The distinctive features of these folds appear to be a plastic style, virtually isoclinal limbs, nearly east-trending axes, and—de- pending upon position with relation to younger folds —locally warped and variable limbs, axial planes, and axes. Folded quartz lenses.—Isoclinally folded quartz lenses are relatively rare in the area and occur chiefly in the highly schistose rocks. The folded quartz layers are greatly thickened in the noses of the folds and thinned or pinched out on the limbs, so that the form is that of a flattened cylinder folded about axes parallel to that of the cylinder. In the more highly deformed lenses, the limbs are nearly obliterated, and the thickened noses coalesce so as to be nearly indistinguishable from quartz rods. The attitude of the rodding varies, but most rods are oriented about down the dip of bedding schis- tosity. Younger folds.-——Younger minor folds are abun- dant in the mevtasedimentary and metavolcan'ic rocks, but they vary in expression and abundance according to rock type. They range in style from moderately tight, straight limbed, sharp crested folds of generally small amplitude in the micaceous schist, to broad open warps in the more massive rocks such as the amphibolite. The younger folds vary appreciably in attitude, and the variations show a crude pattern with respect to areal position. Axial planes are predominantly steep; they range in strike from northeast or east in the northeastern part of the area, to north or north— east in the southwestern part. The plunge of axes STRUCTURE 73 ranges from steeply southward to gently north or south. The folds have dextral, neutral, and sinistral patterns, all associated with both northerly and southerly plunges, but dextral patterns associated with southerly plunges predominate. Folds near ultramafic bodies deviate from the general trend. FOLDS IN ULTRAMAFIC ROCKS Minor folds are common in the ultramafic rocks. The folds vary widely in form and orientation: from isoclinal or attenuated and plastic to open and brittle in style, and from steep to gentle and through a wide range of trends in the attitudes of axes and axial planes. They are defined by form surfaces of two types: (1) surfaces that bear no genetic relation to folding, such as layering in the ultramafic rocks and intrusive contacts; and (2) foliation that is demon- strably related to folding, such as schistosity and cleavage that is axial plane to folds in the contacts of the ultramafic bodies (fig. 3, pls. 1, 2, 3, and 4A, B, and C). The relation between the style and pattern of a fold and the kind of form surface that defines it is not simple. The sparse folds in layering range from nearly isoclinal to open and are variable in attitude and pattern. Folds in the contacts between serpen- tinite and amphibolite are mostly plastic in style, and the limbs are isoclinal or greatly attenuated. Some, however, are fairly open. Those in contacts between serpentinite and gneiss or schist are all open and tend toward a more brittle style. Most of the folds in the contacts have moderate to steep axes and a dextral fold pattern in plan, but a few are sinistral or neutral; near 101,000 N., 95,000 E. (pl. 1), a regular alternation of groups of dextral and sinistral small-scale folds that have gentle to mod- erate plunges results in a generally neutral pattern. Folds in the schistosity of the serpentinite are open and have fairly sharp troughs and crests. Some, such as those at 101,000 N., 95,000 E., are similar in style and attitude to nearby folds in the country rock. Others, such as those at 100,000 N., 97,430 E. (pl. 40), have attitudes that diverge greatly from those of folds in the country rock. TRAN SVERSE SCHISTOSITY Schistosity that transects or is transverse to bed- ding in the metamorphosed sedimentary and vol— canic rocks is varied in appearance and in relations to bedding and folding. Two classes are distin- guished on the basis of these relations: an older transverse schistosity, which bears an axial—plane relation to the older folds and which is virtually parallel to bedding, except in the noses of the older folds into which the schistosity abuts; and a younger transverse schistosity, which bears an axial-plane relation to the younger folds and is nearly every- where divergent from bedding. OLDER SCHISTOSITY The older transverse schistosity consists entirely of generally parallel shear surfaces so closely spaced that they are resolvable with difficulty or are only resolvable microscopically. These shear surfaces are parallel to bedding and bedding schistosity through— out most or the entire extent of observation, but wherever the relations to older folds are seen, the shear surfaces are axial plane to the folds and butt into them. Transverse schistosity generally parallel to bed- ding is seen chiefly in the gneiss and the more quartz- itic varieties of schist and is subordinate to contin— uous bedding schistosity. Within the field of most outcrops, the pattern of this older transverse schis- tosity is virtually identical with that of bedding schistosity; it is only in much larger fields that the two diverge significantly. Consequently, measure- ments of attitudes of bedding schistosity nearly always also serve to record attitudes of the older transverse schistosity; a separate symbol for the older schistosity is almost never recorded. YOUNGER SCHISTOSITY The younger transverse schistosity occurs in dis- tinctively different habits that correlate with lith- ology and that intergrade in conformance with changes in lithology. In granular rocks (quartzite and micaceous quartzite), the younger schistosity is a fracture cleavage; in quartzose mica schists, schistosity is a slip cleavage. In highly graphitic and phyllitic schists, the younger schistosity is generally classifiable as slip cleavage only in thin section. All types of younger transverse schistosity con- form to the same general spatial pattern and have about the same range of variation in attitude as do the younger folds. The schistosity is predominantly steep and varies significantly in strike only on a fairly large scale and in a simple manner. In the northeastern part of the area, it strikes chiefly northeast or east; in the southwestern part, it strikes north or northeast. Fracture cleavage—Fracture cleavage is char- acteristic of granular rocks such as quartzite and micaceous quartzite and is generally subordinate to bedding schistosity and to older transverse schis- tosity. The cleavage consists of parallel or subparal— 74 lel fractures spaced a few millimeters apart, along which differential movement is very small or indis— cernible. Micaceous minerals in the rock are not alined parallel to the fractures. Fracture cleavage is generally most conspicuous in troughs and crests of folds; commonly the cleavage surfaces are not strictly parallel but fan symmetrically on either side of the axial plane. Slip cleavage—Slip cleavage is a spaced schis- tosity that bears a regular relation to crinkles in an older schistosity and consists of discretely spaced surfaces of parting or incipient parting subparallel to the limbs of the crinkles and about parallel to their axial planes. It is distinctive of much of the schist and gneiss in the area. (See White, 1949, for a detailed discussion of slip cleavage.) It varies in expression from predominant over bedding schis— tosity to subordinate to it. The micaceous minerals of the rock are not generally alined parallel to the slip cleavage, though the shingled arrangement of micas in the attenuated limbs of crinkles commonly results in near parallelism. Where the attenuated limbs have actually been disrupted and sheared, some of the mica flakes are commonly parallel with- in the shear zones. Slip cleavage in graphitic and phyllitic schist is commonly not readily distinguished in the field. In such rocks, the transverse schistosity consists of very closely spaced shear zones, commonly of such fine texture that they cannot be detected without the use of a microscope. The transverse schistosity grades imperceptibly into slip cleavage. For the most part, the micaceous minerals are not parallel to the schistosity, but mica flakes are locally alined within individual shear zones; in some places such parallel flakes are abundant. Bedding and bedding schistosity in such rocks are discontinuous; segments can rarely be traced more than a few centimeters and in places form only shredded fragments. Where the noses of folds are discernible, the transverse schistosity is parallel to the axial planes of the folds. UNCLASSIFIED SCHISTOSITY When the relations of schistosity to bedding are unknown, the schistosity is here termed “unclassi- fied.” Commonly, the relations are unknown because bedding was originally poorly developed or has been largely obliterated during metamorphism at‘d defor- mation, or the conditions of exposure were poor. Probably most unclassified schistosity is equivalent to divergent schistosity in which bedding has been almost entirely obliterated. A small proportion may be equivalent to the older transverse schistosity. ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT Most of the structural attitudes recorded for unclas- sified schistosity are markely divergent from bed— ding and of the same pattern as the younger tran- secting schistosity. CLEAVAGE IN ULTRAMAFIC ROCKS Cleavage in the ultramafic rocks is confined chiefly to serpentinite. Locally, talc-carbonate rock, car- bonate-quartz rock, and steatite have similar cleav- age, but they are generally poorly exposed. Dunite and peridotite are chiefly massive and unfoliated. The cleavage is of three distinctive types: (1) paper-thin schistosity characteristic of the thin schistose zones that enclose the massive shear poly- hedrons; (2) subparallel irregular fractures or shear zones spaced as much as several centimeters apart; (3) sets of closely spaced fractures that in- tersect in a rhombic pattern, bounded by or grading into faults or shear zones of small displacement. In addition, relict-bedding schistosity is preserved in a few places in steatite derived from schist. The paper-thin schistosity characteristic of the thin zones surrounding shear polyhedrons is a fine- textured spaced schistosity. Between shear surfaces, the serpentine particles are common largely alined almost parallel to the shear zones, but rarely so well as to impart a good continuous schistosity. In the talcose and carbonatian rocks only the gross pattern of the shear zones remains, and the particles of talc are generally diversely oriented. The pattern of the schistosity conforms with that of the surfaces of the shear polyhedrons ; although very irregular in detail, the overall trend is in most places roughly parallel to the principal plane of the lenticular bodies. The widely spaced irregular parallel fractures and thin shear zones range in spacing from a fraction of a centimeter to several centimeters. This type of cleavage is distinctive of moderately sheared serpen— tinite containing shear polyhedrons and of many of the massive shear polyhedrons themselves. The atti- tude of this widely spaced cleavage is highly vari- able, and the cleavage is locally folded. At the south- western contact of the Lowell quarry body in the vicinity of 100,000 N., 97,420 E. (pl. 40), the cleavage is axial plane to bulbous tightly compressed folds. Closely spaced fractures that intersect in a rhom- bic pattern are of two principal types: ( 1) sharply bounded tabular zones a few centimeters to a meter or more thick that transect massive serpentinite and, locally, dunite or peridotite; and (2) branching distributive shear zones that transect schistose ser— pentinite characterized by shear polyhedrons and that grade into faults or tabular shear zones. STRUCTURE The relations of the rhombic cleavage are distinct where it forms sharply defined tabular zones. Both sets of intersecting fractures are steep, and they intersect in dihedral angles of 30° and 150° to form crude rhombic prisms generally less than a centi- meter on the rhombic edge. The prismatic axis is about down the dip, and the long diagonal of the rhombic section is parallel to the margins, of the zone. The strike of the fractures varies widely in different zones. Within zones of schistose serpentinite character- ized by shear polyhedrons, rhombic cleavage is gen- erally indistinct because of the interference of the irregular schistose zones that surround the shear polyhedrons. The relations between the three types of cleavage distinctive of the ultramafic rocks seem fairly simple, though not entirely beyond dispute. The schistosity surrounding shear polyhedrons inter- grades with spaced cleavage formed by parallel frac- tures and thin shear zones. The rhombic cleavage integrades with faults and shear zones of minor off- set and intersects both the schistosity that sur- rounds the shear polyhedrons and the widely spaced cleavage. Along the southwest margin of the Lowell quarry body (pl. 40), the widely spaced cleavage is axial plane to bulbous folds in the contact of the ultramafic body, and the minor faults and tabular zones of rhombic cleavage are axial plane to folds in the widely spaced cleavage. LINEATION S In addition to the axes of folds, significant linear features in the area include quartz rodding, crinkles, striations, and intersections of cleavage and bedding schistosity. Lineation resulting from mineral elonga— tion is virtually confined to the asbestos and picro- lite veins. This rather special kind of lineation is dis- cussed under both sections on “Serpentine veins.” Quartz rodding is at least partly, and probably entirely, equivalent to the isoclinally folded quartz lenses described in the section on “Folds in the metamorphosed sedimentary and volcanic rocks” under “Minor structural features” and is similar in attitude. The rods have the form of irregular ellipti- cal cylinders and are generally strongly striated parallel to the cylindrical axis. They are commonly about 1 cm by 3 cm in cross section and as much as 60 cm long. Lineation marking the intersection of cleavage and bedding or bedding schistosity takes several forms. Most commonly, for example, where the cleavage is slip cleavage or transecting schistosity, 75 the intersection is marked by a crinkle, which is gen- erally a millimeter or less and rarely as much as sev— eral millimeters in wavelength. Where fracture cleavage intersects bedding, the lineation is marked by a distinct fracture line. The strong striations dis- tinctive of quartz rods and isoclinally folded quartz lenses are the imprint of transecting schistosity on the quartz beds, and therefore are really lineations resulting from intersection of bedding and cleavage. In a few places where a conspicuous crinkle line- ation marks the intersection of slip cleavage and bedding schistosity, another set of crinkles is ori- ented at a large angle to the first set. Commonly, crinkles of the second set are smaller than those of the first. Intersecting relations between crinkles of the two sets are generally rather indistinct and am- biguous; where clearcut, the crinkles associated with the axial-plane slip cleavage are intersected and off— set by crinkles of the other set. Quartz rodding and striations on the quartz rods and folded quartz lenses have the same spatial pat- tern as the older fold axes. Lineation that marks the intersection of bedding schistosity and younger transecting schistosity (fracture cleavage, slip cleavage, or divergent schistosity) is strictly paral- lel to the axes of the younger folds. Crinkles that make a large angle with those associated with the transecting schistosity are about parallel to the older fold axes in most places; however, the inter— secting relations indicate that they are about the same age as the crinkles parallel to the younger folds and so are unrelated to the older folds. JOINTS Joints occur erratically in the area and are most conspicuous in the more massive varieties of both country rock and ultramafic rocks. Perhaps because of better conditions of exposure in the ultramafic rocks, joints seem more abundant there than in the country rock. Observations on joints in the country rock are scattered and confined almost entirely to the amphi- bolite; the joints show no discernible systematic pattern and no apparent significant relation to other structural features. In the gneiss at the northeast contact of the Lowell quarry body, a regular widely spaced set of joints was exposed by stripping oper- ations during quarrying; they are parallel to a faint slip cleavage which is axial plane to the younger folds (pl. 4D). These “joints” probably formed dur- ing quarrying operations by fracturing of the rock along planes of weakness determined by the slip cleavage. 76 ASBESTOS-BEARING ULTRAMAFIC ROCKS 0F BELVIDERE MOUNTAIN AREA, VERMONT Joints in the ultramafic rocks are virtually con- fined to the dunite, peridotite, and massive serpen- tinite. Their apparent absence in the talcose and carbonatian rocks probably results from conditions of exposure, for joints in talc-carbonate rock are locally conspicuous in ultramafic bodies elsewhere in Vermont (Chidester, 1962, p. 24). The joints range in individual configuration from nearly planar to very irregular; in their interrelations, they range from clearly recognizable systems of two or three sets, through clearly defined sets with no apparent systematic relation between sets, to apparently ran- domly oriented fractures. Even among conspicuous sets and systems of joints, no overall pattern is evi— dent. Particular spatial patterns prevail throughout areas several tens of meters in extent, but the rela- tions between such fields are varied. Throughout the main ultramafic body, many of the joints contain veins of cross-fiber chrysotile as- bestos. Commonly, all the joints of a system contain such cross-fiber veins, but in some places, joints of one set contain veins, whereas those of a second set, which is virtually complementary to the first, con- tain no veins. Where several sets of joints are clearly not members of a single system, it is common for one to contain cross-fiber veins and for the others to be barren. Joints in the massive serpentinite and dunite of the Eden quarry body in the Vicinity of the bodies of talcose rocks near 99,000 N., 92,000 E., and 100,- 000 N., 94,000 E. (pl. 1), are commonly the centers of irregular thin tabular zones of talc-carbonate rock. In many such places, the talc-carbonate alter- ation is confined to a single set of joints, and other sets of joints are barren of talc. Elsewhere in the main body, talc is almost entirely absent from all joints, except for a single occurrence a few centi- meters thick and a few meters in extent along a joint in the Lowell .quarry body near 99,850 N., 98,320 E. (pl. 3). F AULTS Observed faults in the area are virtually confined to the vicinity of the main ultramafic body, doubt- less largely as the result of better conditions of ex- posure in the quarry areas (fig. 3, and pls. 1—4). Nearly all of these faults are steep to moderate in dip, and range in displacement from only a few centimeters (so that distinguishing them from joints becomes largely academic) to a few deci- meters. They range in length from only a few deci- meters to a few hundred meters. In most, the zone of shearing is not more than 5 or 6 cm thick and is sharply bounded by rather smooth surfaces of un- sheared rock. Slickensides and gouges on the fault surfaces are mostly downdip. Most of the faults are entirely within the ultramafic bodies or intersect the contacts and extend only several tens or a hundred meters or so into the country rock; some, however, are confined to the country rock. Two predominant attitudes define two fairly apparent sets of faults, one trending northeast and one trending northwest. A few faults, most of which are very irregular in plan, are diverse in orientation and do not seem to fit either set. The fault at the northeast margin of the Eden quarry body, near 100,200 N., 96,600 E., (pls. 1 and 3), has been excellently exposed by stripping oper- ations throughout a length of about 100 m and ap- pears to be of greater magnitude than other faults in the area. A conspicuous gouge zone ranges from 15 to 25 cm in thickness, and the minimum displace- ment appears to be a few tens of meters (pl. 3, cross- section A—A’). The fault dips moderately southwest. Despite its greater magnitude, this fault is not greatly different from those of the northwest-trend- ing set of smaller faults, and it probably belongs to that set. Nearly all the faults that show evidence of relative movement sense are reverse faults, having only a small but commonly appreciable strike-slip com- ponent. In many of the faults the movement sense is indeterminate. Within the country rock zone, along all the faults that intersect the contacts of the Lowell quarry and Eden quarry bodies, the rock bordering the faults and the gouge and fragmented rock within the fault zones are altered for a few millimeters or centi- meters to calc-silicate minerals and to serpentine- chlorite rock (see section on “The rodingite and serpentine-chlorite rock association” under “Contact rocks” in section on “Ultramafic and associated rocks”) Within the ultramafic bodies, the faults are barren of calc—silicate minerals. Many are virtually barren of chrysotile asbestos, but some contain slip fiber asbestos and picrolite. TECTONIC AND PETROGENIC SYNTHESIS The Belvidere Mountain area records a history of sedimentation, volcanic activity, and emplacement of ultramafic rocks in an early Paleozoic eugeosyn— cline, followed by a complex history of folding and metamorphism which extended into the Devonian. Most of this history is imprinted in the rocks of the TECTONIC AND PETROGENIC SYNTHESIS 77 area and has been documented in appropriate sec- tions of this report, but establishment of some of the chronology and of the broader correlations de- pends upon regional studies which have been cited and summarized. The ultimate origins of the ultra- mafic rocks are largely inferential and encompass broad problems of petrology, global tectonics, and earth history, which are highly speculative and which allow for wide latitude in interpretation. Consequently, this aspect of the history has long been involved in intense controversy. (See Chidester, 1962, p. 87—88; Jahns, 1967, p. 155-156; Thayer, 1960, 1967, 1969, p. 511—515; Wyllie, 1967, 1969; Chidester and Cady, 1972.) ORIGIN OF THE ULTRAMAFIC ROCKS Two alternative modes of origin have generally been advocated for alpine-type ultramafic rocks: (1) accumulation by crystal fractionation of a complex (for example, basaltic) magma and (2) derivation from the upper mantle by one or more of several, commonly vaguely specified, processes. In either mode of origin, emplacement as a crystalline mass is generally recognized. Both models imply a close genetic relation between the ultramafic rocks and mafic volcanic rocks interbedded in a eugeosynclinal pile. In the first model (fractionation of a complex magma) the ultramafic rocks are regarded as crystal cumulates and the basalts are regarded as fluid fraction. In the simplest form of the second model (partial melting of upper mantle material), the ultramafic rocks are regarded as refractory resi- due and the basalts, as fluid filtrate. Many variations on this theme are possible. In earlier reports (Chi- dester, 1962, 1968; Cady and others, 1963; Cady 1969, p. 24-25), we have adhered to these conven- tional theories as alternatives for the origin of the ultramafic rocks of Vermont. Both theories, how- ever, have serious drawbacks to their acceptance, such as implied structural and petrologic features consequent upon a particular theory of origin, which do not fit observed field relations (for example, the absence of associated granitic rocks), and the need to invoke special conditions to explain the wide variety of relations shown by alpine-type ultramafic rocks. A theory of origin of ultramafic rocks based upon the concept of sea-floor spreading proposed by Hess (1962; Dietz, 1961) has been applied with consider- able success to full ophiolite complexes in such Widely separated areas as Papua-New Guinea (Davies, 1968), Oman (Reinhardt, 1969) , Greece (Moores, 1969), Cyprus (Moores and Vine, 1971), and California (Bailey and others, 1970). However, the application of the concept to the alpine-type ultramafic rocks of Vermont, which are exclusively intrusive into fairly continuous sequences of eugeo- synclinal sedimentary and volcanic rocks, requires modification of the model developed for the full ophiolite assemblages, which includes sheeted dia- base, pillow lava, and chert. We (Chidester and Cady, 1972) have recently presented such a model specifically to reconcile the mode of origin of the alpine-type ultramafic rocks of the Appalachians with the concepts of sea-floor spreading. The salient features of the model are summarized as follows: In late Precambrian time, mantle upwelling began along a sinuous belt that now marks the site of the Appalachian orogen. This sinuous belt was 300 to 800 km northwest of the line along which Africa and North America were later to separate by con- tinental drift, beginning in the Mesozoic. At this early stage, however, the two were parts of one continent, and the floor of the incipient Appalachian eugeosyncline was thick sialic crust. Outward move- ment of upper mantle rocks on either side of a de- veloping rift beneath the continental crust led to distention and necking of the continental crust and downwarping of the surface, thereby initiating the Appalachian eugeosyncline. Basaltic lava, formed by partial melting of material deeper in the upper mantle, erupted along the rift. As the subcontinen- tal rift continued to form, and as the accumulation of sediments in the eugeosyncline increased, basaltic lava erupted within the eugeosyncline as dikes and sills, surface flows, pyroclastic layers, and volcanic piles, which were continually reduced by erosion and incorporated as detrital volcanic beds into the sedi- mentary sequence. As a result of cooling and hydra- tion during upwelling, the peridotite of the upper mantle beneath the rift became partly serpentinized. As the upper mantle moved outward away from the rift beneath the continental crust, interactions at the surface between crust and mantle produced geanticlinal tracts and projections and irregular bulges of rocks of mantle composition into the crust. As activity continued along the rift zone and as the eugeosyncline formed, many masses of such material, varying widely in size, intruded the con- tinental crust and eugeosynclinal pile. Coarse layer- ing that reflects complex variations of primary min- erals in dunite and peridotite has the earmarks of magmatic crystallization and accumulation. Such layering is inferred to be a relict from the upper mantle. 78 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT EMPLACEMENT AND STRUCTURAL HISTORY The ultramafic rocks of the Belvidere Mountain area are representative of ultramafites in the north- ern Appalachian belt, which were emplaced at vari- ous times in the early and middle Paleozoic, during a long interval in which sets of early folds, plastic and similar in style and associated with an axial- plane schistosity, were successively formed (Cady, 1969, p. 44, 47, 108, 113). Some of the early folds are clearly Ordovician in age (Cady, 1969, p. 40—41; Stevens and others, 1969) ; some are dated as Devon- ian (Jahns, 1967, p. 138—141). The ultramafites were emplaced prior to a set of late folds, open and gen- erally parallel in style and having an axial-plane cleavage, that form the Green Mountain anticlinor- ium. The late folds are Middle and Late Devonian (Jahns, 1967, p. 141—144; Cady, 1969, p. 75—80). (see section on “Folds in ultramafic rocks”; see also Chidester, 1968, p. 348—350.) In neighboring Que- bec, ultramafic bodies are truncated by a Middle Ordovician unconformity (Cady, 1969, p. 23-24); emplacement of the ultramafites in the Belvidere Mountain area probably also took place in Early Ordovician. Plastic deformation of the rocks in the eugeosyn- clinal pile began early in the geosynclinal history and continued episodically throughout it. In this process, bodies of ultramafic rock were nipped off the upper mantle. Stresses were mainly subhorizon- tal and were triggered collectively, and in varying degree, by local uplift in the eugeosyncline (perhaps principally over the subcontinental rift), and by compaction-diagenesis-metamorphism in the wet eugeosynclinal sediments (Cady, 1969, p. 42, 143— 145.) These stresses produced tightly compressed plastic—style folds, and kneaded and squeezed previ- ously emplaced bodies of ultramafic rock along bed- ding planes and developing structural surfaces in the sedimentary and interbedded volcanic rocks. As the ultramafic rocks were transported to successively higher positions, they passed through newly de- posited and newly folded rocks. Serpentinization, already begun as pervasive but partial alteration of the dunite and peridotite when the upper mantle rocks were carried upward into a lower temperature regime, continued and was as- sisted by contact with wet sediments through which the bodies of ultramafic rock were kneaded upward. Transport was intermittent; episodes of movement were separated by periods of immobility. Sharing of the mass during transport subsequently facilitated the entry of water and consequent further serpen- tinization during periods of immobility. Such incre- mental serpentinization further increased the mo- bility of the mass in subsequent episodes of move- ment, during which dehydration of the serpentine under shear stress at high pressure and temperature (Raleigh and Paterson 1965, p. 3965; Reicker and Rooney 1966, p. 196), was perhaps also effective. A postulated sequence of events implies repeated epi- sodes of serpentinization, shearing, and dehydration in alternating stages of movement and immobility during tectonic transport, and increasingly exten- sive serpentinization accompanied by increased penetration of shearing. No unequivocal petro- graphic evidence was observed to confirm such a his- tory, but as Jahns (1967, p. 156) has stated, dis- seminated flakes of talc in massive serpentinite may be relicts of such dehydration. Serpentinization dur- ing the stage of tectonic transport produced chiefly highly sheared serpentinite at the margins of the body and in sheared zones that transect the central parts. As the process continued, pervasive serpentin— ization extended into the more massive zones, becom— ing complete in the outer parts of such zones and decreasing generally but irregularly inward. In the central parts of the larger bodies (such as the main body at Belividere Mountain), partial or only slight serpentinization resulted, and large parts of the olivine escaped alteration. In contrast to coarse layering (see section on “Origin of the ultramafic rocks” under “Tectonic and petrogenic synthesis”), fine layering in dunite, peridotite, and massive serpentinite reflects simple textural and mineralogical alternations and prob— ably was formed by flowage of the crystalline mass, particularly in the early stages of transport when entire masses were subjected to simple deforma- tional stresses. Complex shearing patterns, which formed in the serpentinite as it was kneaded through the country rock, led to the formation of shear polyhedrons in the marginal zones of the ultramafic bodies. Un- sheared units of serpentinite, ranging in size from small chips to masses several or many feet in diam- eter, were milled and rotated as they moved past one another, producing a matrix of finely comminuted and sheared serpentinite that forms shells around and fills interstices between polyhedrons. The mosaic pattern produced by diverse orientations of relict layering in adjacent larger polyhedral units results from rotation of the blocks relative to one another. In the central parts of the larger bodies of ultra— mafic rock, large masses of only slightly serpentin- ized dunite and peridotite were carried along rela- TECTONIC AND PETROGENIC SYNTHESIS 79 tively intact, mechanically buffered by yielding in the highly sheared marginal zones. After the emplacement of the ultramafic bodies at their present sites, the late set of folds, open in style and divergent from the early set in attitude, was im- posed on the country rock and the ultramafic bodies (Jahns, 1967, p. 141—144). Slight movement of the ultramafic rocks relative to the enclosing country rock may have occurred locally during the late epi- sode of folding. Such movement is suggested by the local absence of rodingite at the margins of the main Belvidere Mountain body. (See section on “Metamor- phic history” under “Tectonic and petrogenic syn- thesis”) Figure 8 illustrates the evolution of the structural pattern of the area and the emplacement of the ultramafic rocks. The three diagrams depict succes- sive stages in the geologic history, as if at each stage the rocks were exposed at the present surface. In favorable structural situations, such as at Belvidere Mountain, the late folding strongly warped tabular bodies of ultramafic rocks, bending layers and schistosity in conformity with late folds in the country rocks and producing a slip cleavage in folded schistose serpentinite identical in style with that in the adjacent country rock. Where intrusive contacts partly cut across early folds and partly conform to and were controlled by bedding surfaces warped by the early folds, the structural pattern of the contacts reflects part of the pattern of the early folds and the entire pattern of the late folds. Joint systems in the ultramafic rocks, which are best devel- oped in the relatively fresh massive dunite, are in- ferred to be related to the late folds. The pattern of joints, though locally consistent, varies irregularly from place to place in a body because of irregular- ities in the internal stress field induced by inhomo- geneities and irregularities in the ultramafic bodies. In unfavorable structural situations, such as in podlike bodies emplaced in simple homoclinal sec- tions, the effects of the late stage of folding are inconspicuous in the ultramafic bodies. Near the climax of the late period of folding, joints in the massive ultramafic rock and fracture surfaces in sheared serpentinite opened along zones under tensional stress. METAMORPHIC HISTORY Interpretation of the early metamorphic history of the ultramafic rocks depends to a considerable ex- tent upon inferences regarding their history before emplacement, which remains highly speculative. In particular, conclusions concerning two stages of the early history—origin of the primary ultramafic rocks and emplacement of the ultramafic bodies— are critical to interpretation of the metamorphic history. The following outline of events and their relations to one another is based upon the above dis- cussions of concepts of origin along a subcontinental rift and solid emplacement by tectonic transport. Serpentinization is inferred to have begun early in the stage of mantle upwelling along the rift be- neath the eugeosyncline, in accordance with the con- cepts developed by Hess (1962) for sea-floor spread- ing at an oceanic rift. Extensive further serpentin- ization occurred during tectonic transport of the ultramafic masses, as an integral part of the mech- anism of emplacement. Probably most of the ser— pentinization in each body was accomplished by the time the bodies were finally emplaced. After the emplacement of the ultramafic rock, initial metamorphic changes were confined to sys- tems not greatly larger than the ultramafic bodies (except for the sources of water). These changes in- cluded further pervasive serpentinization on a small scale, formation of serpentine and related veins in and adjacent to the ultramafic bodies, and the forma- tion of narrow contact metamorphic and meta- somatic aureoles adjacent to the larger ultramafic bodies. After formation of these features, meta- somatic changes related to regional metamorphism occurred. In part, these changes involved extensive migration of CO2 and resulted in the local alteration of dunite and serpentinite to talc, magnesite, quartz, and tremolite, in distinctive patterns of assemblages. Concomitantly, related alterations in the adjacent country rock produced distinctive contact-rock asso- ciations. The thermal history of a mass of upper mantle rock intruded into a eugeosynclinal pile would de— pend upon a complex interrelation among many vari- ables; primarily, however, the body’s temperature and its difference from ambient temperature would depend on its size and on the rate and distance of transport after detachment from the mantle. Initial- ly the body of rock would be at a higher temper- ature than that of the surrounding rocks. As it was kneaded through the pile, it would lose heat to its surrounding and gradually approach the ambient temperature. In terms of simple conductive loss, larger less far traveled bodies would be hotter than smaller farther traveled bodies at any particular time after detachment from the mantle. Though all the ultramafic rocks in the Belvidere Mountain area were emplaced at considerable depth and about simultaneously, thermal considerations 80 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT 1 |/\/ 1% (75122”— /_/\ A. AT AN INTERMEDIATE STAGE OF THE FIRST PERIOD OF FOLDING, PRIOR TO EMPIACEMENT OF THE ULTRAMAFIC ROCKS ING EVIPLACEMENT OF THE ULTRAMAFIC ROCKS FIGURE 8.—Isometric block diagrams illustrating critical stages in the structural history and emplacement of ultramafic rocks in the Bedvidere Mountain area. Rock units are shown at each stage as if they were exposed at the present land surface. A. At an intermediate stage of the first period of folding, prior to emplacement of the ultramafic rocks. B. At the end of the first period of folding, following emplacement of the ultramafic rocks. C. At the end of the second period of folding, and following metamorphism. Note the bodies of talc-carbonate and carbonate-quartz rock, absent in diagram B. Ch TECTONIC AND PETROGENIC SYNTHESIS C. AT THE END OF THE SECOND PERIOD OF FOLDING,AND FOLLOWING METAMORPHISM. NOTE THE BODIES OF TALC-CARBONATE [AND CARBONATE-QUARTZ ROCK, ABSENT IN DIAGRAMB EXPLANATION . TALC-CARBONATE ROCK AND CARBONATE-QUARTZ ROCK (ORDOVICIAN)1 DUNITE AND SERPENTWITE (ORDOVICIAN)1 OTTAUQUECHEE FORMATION (UPPER AND MIDDLE CAMBRIAN) BELVIDERE MOUNTAIN FORMATION (LOWER CAMBRIAN) HAZENS NOTCH FORMATION (CAMBRIAN(?)) ‘The Ordovician age designation refers to the age of emplacement of the Intrusive ultramafic rocks. not to the age of the parent igneous rocks or the metamorphic derivatives (see text subheading discussions under “Tectonic and petrogenic synthesis"). FIGURE 8.—Continued. 81 82 and the mechanism of emplacement advocated here ' suggest that the larger bodies differed significantly in temperature and mineralogy from the smaller- bodies at the time of emplacement. The larger bodies, such as the main ultramafic body at Belvidere Mountain, contained large masses of unaltered dunite in the central part of the body, and their temperatures were appreciably, though not greatly, above the temperature of the surrounding rocks. The smaller bodies were completely serpentinized when emp‘laced and were virtually at the temper- ature of the enclosing rocks. Under these circumstances, the temperature of the country rock at the margins of the large ultra- mafic bodies would be increased, Whereas that at the margins of the small bodies would not. Serpentiniza- tion of dunite would continue so long as water was available, producing such effects as partial desicca— tion and the addition of Ca and Mg (see section on “Rodingite” under “Contact rocks” in the section on “Petrogenesis” under “Ultramafic and associated rocks” and see tables 7 and 9). These conditions combined to produce narrow contact-metamorphic and metasomatic aureoles of rodingite around the larger ultramafic bodies that contained considerable dunite, but no discernible effects at the margins of the smaller completely serpentinized bodies. Limited serpentinization of dunite in the larger bodies extended into the period of late folding. As temperatures fell in the contact zones, serpentiniza- tion of dunite and consequent efllux of Mg led to replacement of rodingite by serpentine-chlorite rock in a narrow zone next to the contact of the ultramafic body. Near the climax of the late folding, the dunite and serpentinite opened up along joints, fractures, and shear surfaces under tensional stress. As such openings formed, they were filled with chrysotile to form cross- and slip-fiber asbestos veins and picro- lite veins. The character of the serpentine veins de- pended primarily upon local mechanical relations and less on physico-chemical variations. Because the stress field in the ultramafic bodies varied erratically owing to irregularities and inhomogeneities in the bodies, the direction and rate of opening along frac— tures and shear surfaces varied markedly from place to place and resulted in a complex distribution pat- tern of the various kinds of serpentine veins. Simul— taneously with the formation of the serpentine veins, constituents released during their formation formed in nearby fractures in the ultramafic bodies and con- tributed to the formation of chlorite—calcite-magne- tite veins in the bordering country rock. ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT During regional metamorphism, which culminated during and extended somewhat beyond the climax of the late stage of folding, dunite and serpentinite were locally altered to talc-carbonate rock and car— bonate-quartz rock. The distribution of these rocks at the margins of the ultramafic bodies, or in linear zones that appear to be controlled by faults or shear zones, indicates that the processes were controlled by access of CO2 into the ultramafic rocks. The general absence of such rocks adjacent to contacts where rodingite is present suggests that the rodingite acted as an effective barrier to the access of CO2, or that the structural conditions that led to preservation of the rodingite also prevented access of C02. The dis— tribution of carbonate-quartz rock with respect to talc-carbonate rock suggests that highly quartzose zones are centered along channelways that provided ready access of C02, and indicates that formation both of talc-carbonate rock and carbonate—quartz rock was a single process whose end product was carbonate—quartz rock. Where dunite and peridotite Were extensively altered to carbonate-quartz rock and talc-carbonate rock, expulsion of large amounts of Mg and Si led to extensive and irregular steati— tization, chloritization, and related metasomatic re- placement of the adjacent country rock. In the smaller ultramafic bodies that have no rod- ingite border, and in the larger bodies where roding- ite is locally absent, serpentinite and adjacent siliceous schist reacted to form steatite (chiefly by replacement of serpentinite) and blackwall chlorite rock (entirely by replacement of schist). Where the country rock bordering the serpentinite is amphibo— lite, tremolite was formed additionally, or in place of talc. Where rodingite borders serpentinite, it acted as an effective barrier to metamorphic interaction between serpentinite and schist—perhaps in part because, as an effective barrier to ingress of C02, it prevented the triggering action of CO2 metasomat- ism (Chidester, 1962, p. 120; Cady and others, 1963, p. 47; Cady, 1969, p. 155). CALCULATIONS Throughout this report, chemical changes in meta— morphic and metasomatic processes inferred to have taken place at constant volume are deduced by com— parison of the chemical content of equal volumes segregations of magnetite in the serpentine veins or (“modified standard 08115”) 0f I'OCkS (Chidester, 1962, p. 95). Chemical analyses in table 1 are pre— sented both in terms of Weight percentages and in terms of cation percentages, and cell factors (F0) CALCULATIONS 83 are included so that the content of the modified standard cell for each analysis can be computed readily by multiplying the cation percentage for each component by the cell factor. The method of deriving the modified standard cell, the methods of computation, and related prob- lems and methods of calculating mineral formulas from chemical analyses have been described in an earlier report (Chidester, 1962, p. 94—97, 129—205). In the present report, all serpentine formulas are calculated simply on the basis of (Si+1/2R+3) =4, because the methods of sampling and analysis do not warrant further refinements in calculation. (See text discussion in section “Mineralogy of the serpentine group” under “Ultramafic and associated rocks”; Bates, 1959.) In analyses of rocks consisting essentially of ser- pentine and minor brucite, the content of brucite is calculated from the relations: 1. NS+NB=C, and 2. 3/5N,=Si+1/2R+3, where Ns=equivalent molecular percentage of ser— pentine; N B=equivalent molecular percentage of brucite; C = sum of equivalent molecular percentages of serpentine and brucite; Si=cation percentage of silica; and R+3=sum of cation percentages of triva- lent elements (essentially Al+ Fe+3). Mineral formulas of amphibole and epidote were calculated from analyses of mineral separates, which contained small but significant amounts of impuri- ties, by solving simultaneous equations relating the two analyses, yielding calculated analyses of the pure minerals. The methods of calculation are described below. ANALYSES OF PURE AMPHIBOLE AND PURE EPIDOTE 1. Determine the mode of each analyzed sample (from thin section, slide mounts of sized grains of the material analyzed, or otherwise) and recalculate to the basis of amphibole+epidote= 100. Epidote sample (containing minor impurities) Amphibole sample (containing minor impurities) Amphibole M P Epidote N Q Where M+N= 100, and P+ Q = 100. 2. Recalculate each analysis of the impure material into equivalent molecular (= cation )percent. On the basis of the approximation that cation percentage= volume percentage, correct the analysis for impuri- ties of fixed composition (sphene, rutile, albite, and the like) by subtracting appropriate amounts of Ca, Ti, Si, Al, and so forth. The balance, recalculated to the basis of the sum of the cations= 100, represents, for each analysis, analyses of mixtures solely of amphibole and epidote. 3. From the modes recalculated to the basis of amphibole+epidote=100 (1, above), and the calcu- lated analyses of mixtures consisting solely of am- phibole and epidote (2, above), the following simul- taneous equations, valid for each constituent in the analysis, can be derived: (1) mCa+nCe=Cae (2) pCa+qcta=Ceu h l t M N P Q w ere m, n, p, q are equa 0-1—0—0, m’ m, 100. 0,, represents the cation percentage of each constitu- ent in the analysis of pure amphibole; (70 represents the cation percentage of each constituent in the analysis of pure epidote; 0.... represents the cation percentage of each constituent in the mixture of am— phibole and epidote; and C... represents the cation percentage of each constituent in the mixture of epi— dote and amphibole. 4. Solve (1) and (2) simultaneously: Multiple (1) by q: (3) qua + anP = qu” Multiply (2) by n: (4) pnC,,+an,=nCen. Subtract (4) from (3) and solve for C“: mel — pnC,1 = qC,1e — 1109,. q n (5) Ca = —‘_Cae '_ ‘——-Cea- qm — pn qm — pn Substitute (5) in (2) and solve for 0,: (6) Ce = _m—Cea _ Lone- qm — pn qm — pn 5. For each component, substitute appropriate values of m, n, p and q, and of Gen and Ca, in equa- tion (5) to calculate the analysis of pure amphibole, and in equation (6) to calculate the analysis of pure epidote. The resulting calculated analyses are in terms of cation percentages. 6. For example, in table 1, analyses 23 and 24, of concentrates of amphibole and epidote, respectively: m=0.9900 p=0.0152 n=0.0100 q=0.9848 Substituting these values in formulas (5) and (6), get: 84 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT TABLE 10.-—Calculatian of amphibole formula [Sample No. A—BM—534139a, Lab. No. 14338031. Analyses from table 1. 0, looked for and not foundl Molecular (cation) present . . C Formula Impurities 1 C'a? C“: C“ (pliare Nos. $113321] (except epidote) (amxfiglsbde 100 C’" (from amphibole) (242—311 Sphene Rutile Albite epidote) )3 0’82 table 11’ 23.811335: SI 0 2 _____________________ 41.96 0.04 __ 0.03 41.89 41.97 36.97 42.02 6.66 Al 0 3/; ___________________ 13.40 __ __ .01 13.39 13.42 29.57 13.25 2.10 Fe 0 3/» ___________________ 2.52 __ __ _- 2.52 2.53 6.89 2.49 .39 Fe 0 ______________________ 9.08 __ __ __ 9.08 V 9.10 .74 9.19 1.46 Mg 0 _____________________ 16.24 __ __ __ 16.24 16.27 .71 16.43 2.60 Ca 0 ______________________ 9.64 .04 __ __ 9.60 9.62 24.05 9.47 1.50 Na 01/0 ___________________ 5.26 __ __ .01 5.25 5.26 .60 5.31 .84 K0 14- ____________________ .88 __ __ __ .88 .88 .10 .89 .14 H0 1/ ______________________ 7.86 __ __ __ 7.86 7.88 3.65 7.92 1.26 T1 0 3 _____________________ .58 04 03 __ .51 .51 .00 .51 .08 P0 ;/_ _____________________ .14 __ __ __ .14 .14 .14 .14 .02 _________________________ .06 __ __ __ .06 .06 .03 .06 .01 Cr 0 a] ____________________ .03 __ __ __ .03 .03 .08 .03 .005 Ni O ______________________ .03 __ __ __ .03 .03 .002 .03 .005 C0 0 ______________________ .01 __ __. __ .01 .01 0 .01 .002 Mn 0 _____________________ .18 __ __ _- .18 .18 .03 .18 .03 Cu O _____________________ .02 __ __ __ .02 .02 .005 .02 .003 Pb O ______________________ 0 __ __ __ 0 0 0 0 0 Zn 0 ______________________ 0 _ __ __ 0 0 0 0 0 Ga 0 3/ _____________________ .001 __ __ __ .001 .001 .002 .001 .0002 V0 2 ______________________ .02- __ __ __ .02 .02 .02 .02 .003 Sc 0 .»,/ ___________________ .010 __ __ __ .010 .010 .012 .010 .0016 YO 9. / :2 ____________________ O _ _ _ _ _ _ 0 O O 0 0 Yb 0 ./ ____________________ .0001 _ __ __ .0001 .0001 .0001 .0001 .00002 Zr 0 2 _____________________ .0005 _ -_ __ .0005 .0005 .002 .0005 .00008 Be 0 ______________________ 0 _- __ __ O O O 0 0 Sr 0 ______________________ 0 _- __ -- 0 0 0 0 0 Ba 0 ______________________ .0004 __ __ _- .0004 .0004 0 .0004 .00006 Total _______________ 107.92 12 03 .05 107.72 107.94 103.68 107.98 17.11 Less (HO1/g-i—F) ___. 7.92 __ __ __ 7.92 7.94 3.68 7.98 1.27 Cations _____________ 100.00 12 03 .05 99.80 100.00 100.00 100.00 15.84 1 Estimated mode of analyzed material: Amphibole ______ 98.80 Epidote _________ 1.00 Albite __________ .05 Sphene _________ .12 Rutile __________ .03 100.00 C... = 1.01030... — 0.010309“, Where: C. = 1.0156Cm — 0.01560“... Substituting appropriate values of Cno and 0..., con- stituent by constituent, in these equations, yields cal- culated analyses of pure amphibole and pure epidote. Thus, for SiO2 in the pure amphibole: SiO2 = 1.0103 x 41.97 — 0.0103 x 36.97 =42.402—0.381 =42.02. And so on (see tables 10 and 11). AMPHIBOLE FORMULA 1. Let the amphibole be represented by the follow- ing formula: + . (Na’K) .I'Caal:R+21R+21/+c+u (VI) u 2 10]” [Si._._._.P.A1...]W0..[o._.F.H.1.-.], IV and VI at the upper right of brackets desig- nate coordination numbers, (VI) in the for— mula represents vacancies in six-coordination position. 3 =tota1 Ca, t=total divalent ions, except Ca, u=R+3 in direct substitution for R+2 in six- coordination position, the excess charge being balanced by u/ 2 vacancies in the six-coordina- tion positions. Thus, u is equal to the total amount of R+2 minus the R +3 in coupled sub- stitution in six— and four-coordination posi- tions, minus the A11" balanced by Na and K, and minus the (R+:‘)"I balanced by deficiency of H. That is, u=R+3—2AIIV— (z+Na+K) , CALCULATIONS 85 TABLE 11.—Calculation of epidote formula [ Sample No. A—BM—53—139a, Lab No. 14330b. Analyses from table 1. 0, looked for and not found] Molecular (cation) present Impurities.1 Cien C”: Cue Ogre szrgula énistlesl‘zizli (except amphibole) (eglggte 10‘0 C’"e:- (from _ep(i%ote) (Si26) Sphene Rutile Albite amphibole) 2‘ C 9'“ table 10) :égllé’: gr: SI 0 2 ————————————————————— 36.90 0.36 __ 0.03 36.51 36.97 41.97 36.89 6400 A1 0 3/2 ——————————————————— 29.22 __ __ .01 29.21 29.57 13.42 29.82 4.85 Fe 0 3/2 ___________________ 6.81 __ __ __ 6.81 6.89 2.53 6.96 1.13 Fe O _____________________ .73 __ __ __ .73 .74 9.10 .61 .10 Mg 0 _____________________ .70 __ __ __ .70 .71 16.27 .46 .07 Ca 0 _____________________ 24.11 36 __ __ 23.75 24.05 9.62 24.27 3.95 Na 0 1/2 __________________ .60 __ _.. .01 .59 .60 5.26 2.53 0 K0 1/2 ____________________ .10 __ __ __ .10 .10 .88 2.09 0 H01/o ____________________ 3.61 __ __ __ 3.61 3.65 7.88 3.58 .58 T1 0 2 _____________________ .46 36 0.10 __ .00 .00 .51 .00 .00 PO 5/2 ____________________ .14 __ __ __ .14 .14 .14 .14 .02 ________________________ .03 __ __ __ .03 .03 .06 .03 .005 CI_‘ 0 3/2 ___________________ .08 __ __ __ .08 .08 .03 .08 .01 l 0 ______________________ .002 __ __ __ .002 002 .03 .002 .0003 CO 0 _____________________ 0 __ __ __ 0 0 .01 0 0 Mn 0 _____________________ .03 __ __ 1- .03 .03 .18 .03 .005 Cu 0 _____________________ .005 __ __ __ .005 005 .02 .005 .0008 Pb O _____________________ 0 __ __ __ 0 0 0 0 0 Zn 0 _____________________ 0 __ __ __ 0 0 0 0 0 Ga 0 «[4: ___________________ .002 __ __ __ .002 .002 .001 .002 .0003 V0 2 ______________________ .02 __ __ __ .02 .02 .02 .02 .003 Sc 0 3,. ___________________ .012 __ __ __ .012 .012 .010 .012 .002 Y0 3/2 ____________________ .004 __ __ __ .004 .004 0 .004 .0006 Yb 0 3p __________________ .0001 __ __ __ .0001 .0001 .0001 .0001 00002 Zr 0 .. ____________________ .002 _ __ __ .002 .002 .0005 .002 .0003 Be 0 _____________________ 0 __ __ __ 0 0 0 0 0 Sr 0 _____________________ .07 __ __ __ .07 .07 0 .07 .01 Ba 0 _____________________ 0 __ __ __ 0 0 0 0 0 Total _______________ 103.64 1.08 .10 .05 102.41 103.68 107.94 103.61 16.74 Less (H01/2+F) ____ 3.64 __ __ __ 3.64 3.68 7.94 3.61 .58 Cations _____________ 100.00 1.08 .10 .05 98.77 100.00 100.00 100.00 16.16 1 Estimated mode of analyzed material: Epidote _________ 97.27 ' .. 1.50 1.08 .10 .05 100.00 2 Dropped as contaminants. v = total F, substituting for OH, Al‘ I = 8 — Sl — P. w =total P, x=total (Na,K) , and AlIv balanced by (Na,K) , y=AlIV in coupled substitution for (R+3)VI, z=deficiency of H=2—F—H. (All values are calculated on the basis of O + OH + F = 24.) 2. From the analysis of pure amphibole, calculate the number of cations associated with each oxide on the basis of (O+OH+F) =24. 3. Ions are assigned to different positions in the structural formula as follows: Na, K, Ca, R“, Si, P, H, and F occur in only one position each in the structural formula, so the assignment of values is unambiguous. Each is assigned the total amount of that ion. 0 outside of the OH position is equal to 22. O in the OH position is equal to 2 — F. R+3y+z+u=R+3—Allv. Assignment of other values is evident from the structural formula and from the explanation of the symbols used in it (see 1, above). EPIDOTE FORMULA 1. The structural formula of epidote, in terms of the contents of the unit cell determined by Ito, Morimoto, and Sadanaga (1954), is [Ca.]"III [Al4 (Al,Fe+3) 2]V1[Si6]1"024 (OH) 2, Where the Roman numeral superscripts indicate the coordination position of the cations. The term (Al, Fe”)2 in six-coordination position indicates that Fe“ can substitute for Al only at particular sites in the lattice, and only to a maximum ratio of Fe+3/Al of 1 :2. Small amounts of R+3 may probably substi- tute for A1, and small amounts of R+2, for Ca. 86 2. The formula composition of epidote can be cal- culated from a chemical analysis in several ways—— for example, by setting 0 + OH equal to 26, or by set- ting Si equal to 6. As the determination of H2O+ ap— pears commonly to be of doubtful accuracy, and as Si is present very nearly or precisely in stoichio- metric proportions, it appears generally to be bet- ter and more convenient to calculate the formula on the basis of Si=6. 3. In table 11, the calculated analysis of pure epi- dote, column headed “0.. (pure epidote),” is derived by methods described above and is recalculated to the basis of Si=6 by multiplying each term by 6/3689. The subscript value for O in the formula (oxygen not combined with H) is determined by summing the oxygen associated with the cations in their oxide for- mulas and subtracting the equivalent of 1/2 H (to provide the additional 0 for OH determined as H20). The values for the different ions are then assigned to their proper place in the structural formula. REFERENCES CITED Albee, A. L., 1957, Bedrock geology of the Hyde Park quad- rangle, Vermont: U.S. Geol. Survey Geol. Quad. Map GQ—102. 1962, Relationships between the mineral association, chemical association and physical properties of the chlorite series: Am. Mineralogist, v. 47, nos. 7—8, p. 851—870. Aruja, Endel, 1945, An X-ray study of the crystal structure of antigorite: Mineralog. Mag., v. 27, no. 188-, p. 65—74. Bailey, E. H., Blake, M. 0., Jr., and Jones, D. L., 1970, On-land Mesozoic oceanic crust in California Coast Ranges: U.S. Geol. Survey Prof. Paper 700-C, p. C70—C81. Bain, G. W., 1932, Chrysolite asbestos; II, Chrysotile solu- tions: Econ. Geology, v. 27, no. 3, p. 281—296. ————1936, Serpentinization of Vermont ultrabasics: Geol. Soc. America Bull., v. 47, no. 12, p. 1961—1979. 1942, Vermont talc and asbestos deposits, in New- house, W. H., ed., Ore deposits as related to structural features: Princeton, NJ, Princeton Univ. Press, p. 255—258. Bates, T. F., 1959, Morphology and crystal chemistry of 1:1 layer lattice silicates: Am. Mineralogist, V. 44, nos. 1—2, p. 78—114. Bowen, N. L., and Tuttle, O. F., 1949, The system MgO— SiOg—H20: Geol. Soc. America Bull., v. 60, no. 3, p. 439—460. Bowles, Oliver, 1955, The asbestos industry [revised]: U.S. Bur. Mines Bull. 552, 122 p. Cady, W. M., 1956, Bedrock geology of the Montpelier quad- rangle, Vermont: U.S. Geol. Survey Geol. Quad. Map GQ—79. 1960, Stratigraphic and geotectonic relationships in northern Vermont and southern Quebec: Geol. Soc. America Bull., v. 71, no. 5, p. 531—576. ASBESTOS—BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT ——-—1967, Geosynclinal setting of the Appalachian Moun- tains in southeastern Quebec and northwestern New England, in Clark, T. H., ed., Appalachian tectonics: Royal Soc. Canada Spec. Pub. 10, p. 57—68. 1968, The lateral transition from the miogeosynclinal to the eugeosynclinal zone in northwestern New England and adajcent Quebec, in Zen, E-an and others, eds., Studies of Appalachian geology—northern and mari- time: New York, Interscience Publishers, p. 151—161. 1969, Regional tectonic synthesis of northwestern New England and adjacent Quebec: Geol. Soc. America Mem. 120, 181 p. Cady, W. M., Albee, A. L., and Chidester, A. H., 1963, Bed- rock geology and asbestos deposits of the upper Mis- sisquoi Valley and vicinity, Vermont: U.S. Geol. Survey Bull. 1122—B, 78 p. Cady, W. M., Albee, A. L., and Murphy, J. F., 1962, Geologic map of the Lincoln Mountain quadrangle, Vermont— Bedrock geology: U.S. Geol. Survey Geol. Quad. Map GQ—164. Cameron, E. N., 1961, Ore micrOScopy: New York, John Wiley and Sons, 293 p. Chidester, A. H., 1953, Geology of the talc deposits, Sterling Pond area, Stowe, Vermont: U.S. Geol. Survey Mineral Inv. Field Studies Map MF—ll. ——1962, Petrology and geochemistry of selected talc- bearing ultramafic rocks and adjacent country rocks in north-central Vermont: U.S. Geol. Survey Prof. Paper 345, 207 p. 1968, Evolution of the ultramafic complexes of north- western New England, in Zen, E-an and others, eds., Studies of Appalachian geology—northern and mari- time: New York, Interscience Publishers, p. 343—354. Chidester, A. H., Billings, M. P., and Cady, W. M., 1951, Talc investigations in Vermont, preliminary report: U.S. Geol. Survey Circ. 95, 33 p. Chidester, A. H., and Cady, W. M., 1972, Origin and em— placement of alpine-type ultramafic rocks: Nature: Phys. Sci., V. 240, no. 98, p. 27—31. , Chidester, A. H., Stewart, G. W., and Morris, D. C., 1952a, Geologic map of the Barnes Hill talc prospect, Water— bury, Vermont: U.S. Geol. Survey Mineral Inv. Field Studies Map MF—7. 1952b, Geologic map of the Rousseau talc prospect, Cambridge, Vermont: U.S. Geol. Survey Mineral Inv. Field Studies Map MF—8. Clemmer, J. B., and Cooke, S. R. B., 1936, Flotation of Vermont talc-magnesite ores: U.S. Bur. Mines Rept. Inv. 3314, 12 p. Cooke, H. C., 1937, Thetford, Disraeli, and eastern half of Warwick map-areas, Quebec with chapters on the Beau- ceville, St. Francis, and Lake Aylmer series, by J. H. Clark: Canada Geol. Survey Mem. 211, Pub. 2440, 160 p. Davies, H. L., 1968, Papuan ultramafic belt: Internat. Geol. Cong., 23d, Prague, 1968, Rept. Proc., sec 1, p. 209—220. Deer, W. A., Howie, R. A., and Zussman, Jack, 1962—1963, Rock-forming minerals, v. 1—5: New York, John Wiley and Sons, 5v. (v. 1, 3, 5, 1962; v. 2, 4, 1963). Dietz, R. S., 1961 Continent and ocean basin evolution by spreading of the sea floor: Nature, v. 190, no. 4779, p. 854—857. Doll, C. G., Cady, W. M., Thompson, J. B., Jr., and Billings, M. P., 1961, Centennial geologic map of Vermont: [Montpelier] Vermont Geol. Survey, scale 1:250,000. REFERENCES CITED 87 Efremov, N. E., 1939, Classification of the minerals of the serpentine group: Akad. Nauk SSSR Doklady, new ser., v. 22, no. 7, p. 432—435. Eskola, Pentti, 1932, On the principles of metamorphic dif— ferentiation: Finland Comm. Géol. Bull. 97, p. 68—77. Faust, G. T., and Fahey, J. J., 1962, The serpentine-group minerals: U.S. Geol. Survey Prof. Paper 384—A, 92 p. Faust, G. T., and Nagy, B. S., 1967, Solution studies of chrysotile, lizardite and antigorite: U.S. Geol. Survey Prof. Paper 384—B, p. 93—105. Foslie, Steinar, 1945, Hastingsites, and amphiboles from the epidote-amphibolite facies: Norsk. Geol. Tidsskr., v. 25, p. 74—98. Gillery, F. H., 1959, The X-ray study of synthetic Mg-Al serpentines and chlorites: Am. Mineralogist, v. 44, nos. 1-2, p. 143—152. - Goddard, E. N., and others, 1948, Rock-color chart: Wash- ington, D.C., Natl. Research Council, 6 p. (republished by Geol. Soc. America, 1951, reprinted 1970). Gruner, J. W., 1937, Notes on the structure of serpentines: Am. Mineralogist, v. 22, no. 2, p. 97—103. Hess, H. H., 1962, History of ocean basins, in Engel, A. E. J., and others, eds., Petrologic studies—A volume in honor of A. F. Buddington: New York, Geol. Soc. America, p. 599—620. Hey, M. H., 1954, A new review of the chlorites: Mineralog. Mag., v. 30, no. 224, p. 277—292. Hitchcock, C. H., 1861, The Azoic rocks, in Hitchcock, Ed- ward, Hitchcock, Edward, Jr., Hager, A. D., and Hitch- cock, C. H., Report on the geology of Vermont; descrip- tive, theoretical, economical and scenographical: Clare- mont, N.H., Claremont Mfg. Co., v. 1, p. 452—558. Hitchcock, Edward, Hitchcock, Edward, Jr., Hager, A. D., and Hitchcock, C. H., 1861; Report on the geology of Vermont; descriptive, theoretical, economical, and sceno- graphical: Claremont Mfg. Co., Claremont, N.H., v. 1, 558 p.; v. 2, 982 p. Ito, T., Morimoto, N., and Sadanaga, R., 1954, On the struc- ture of epidote: Acta Cryst., v. 7, pt. 1, p. 53—59. Jahns, R. H., 1967, Serpentinites of the Roxbury district, Vermont, in Wyllie, P. J., ed., Ultramafic and related rocks: New York, John Wiley and Sons, p. 137—160. Keith, S. B., and Bain, G. W., 1932, Chrysotile absestos; I, Chrysotile veins: Econ. Geology, v. 27, no. 2, p. 169—188. Kemp, J. F., 1901a, Notes on the occurrence of asbestos in Lamoille and Orleans Counties, Vermont: U.S. Geol. Survey Mineral Resources U.S. 1900, p. 862—866. 1901b, A new asbestos region in northern Vermont [abs]: Science, new ser., v. 14, p. 773—774; Am. Geolo- gist, v. 28, p. 330; New York Acad. Sci. Annals, v. 14, p. 140—141 (1902). Kourimsky, Jiri, and Filéakova, E., 1954, Vyzkum hadcfi z dobsinskych hald Research of the serpentine from Dob- sina: Prague Narodni Mus. Sbornik, v. IOB, no. 6, Mineralogia, no. 2, p. 3—13. Kourimsky, Jiri, and Satava, V., 1954, A contribution to the question of the determination of minerals of serpentine group: Prague Narodni Mus. Sbornik, v. IOB, no. 4, p. 1—19. Kunze, Gunther, 1956, Die gewellte Struktur des Antigorits, I: Zeitschr. Kristallographie, v. 108, no. 1 and 2, p. 82—107. 1958, Die gewellte Strukture des Antigorits, II: Zeitschr. Kristallographie, v. 110, no. 4, p. 282—320. McIntyre, D. B., and Weiss, L. E., 1957, Construction to scale of block diagrams in orthographic projection: Geologists’ Assoc., London, Pr0c., v. 67 (1956), pt. 1—2, p. 142—155. Marsters, V. F., 1904, A preliminary report on a portion of the serpentine belt of Lamoille and Orleans Counties: Vermont State Geologist Rept. 4, p. 86—102. 1905, Petrography of the amphibolite, serpentine, and associated asbestos deposits of Belvidere Mountain, Ver- mont: Geol. Soc. America Bull., v. 16, p. 419—446. Maser, Morton, Rice, R. V., and Klug, H. P., 1960, Chrysotile morphology: Am. Mineralogist, v. 45, nos. 5—6, p. 680— 688. Moores, E. M., 1969, Petrology and structure of the Vourinos ophiolitic complex of northern Greece: Geol. Soc. Amer- ica Spec. Paper 118, 74 p. Moores, E. M., and Vine, F. J., 1971, The Troodos massif, Cyprus and other ophiolites as oceanic crust, evaluation and implications, in A discussion on the petrology of igneous and metamorphic rocks from the ocean floor: Royal Soc., London, Philos. Trans, ser. A, v. 268, no. 1192, p. 443—466. Morgan, W. J., 1968, Rises, trenches, great faults, and crustal blocks: Jour. Geophys. Research, v. 73, no. 6, p. 1959— 1982. Nagy, B. S., and Faust, G. T., 1956, Serpentines—natural mixtures of chrysotile and antigorite: Am. Mineralogist, v. 41, nos. 11—12, p. 817—838. Olsen, E. J., 1961, Six-layer ortho-hexagonal serpentine from the Labrador trough: Am. Mineralogist, v. 46, nos. 3—4, p. 434—438. Page, N. J., 1968, Chemical differences among the serpentine “polymorphs”: Am. Mineralogist, v. 53, nos. 1—2, p. 201—215. Perry, E. L., 1927, Summary report on the geology of Plymouth and Bridgewater, Vermont: Vermont State Geologist Rept. 15 (1925—26), p. 160—162. 1929, The geology of Bridgewater and Plymouth Townships, Vermont: Vermont State Geologist Rept. 16, (1927—28), p. 1—64. Rabbitt, J. C., 1948, A new study of the anthophyllite series: Am. Mineralogist, v. 33, nos. 5—6, p. 263—323. Raleigh, C. B., and Paterson, M. S., 1965, Experimental de- formation of serpentine and its tectonic implications: Jour. Geophys. Research, v. 70, no. 16, 3965—3985. Ramberg, Hans, 1945, Petrological significance: of sub-solidus phase transitions in mixed crystals: Norsk Geol. Tidsskr., V. 24, p. 42—74. 1952, The origin of metamorphic and metasomatic rocks: Chicago, Univ. Chicago Press, 317 p. Reicker, R. E., and Rooney, T. P., 1966, Weakening of dunite by serpentine dehydration: Science, v. 152, no. 3719, p. 196—198. Reinhardt, B. M., 1969, On the genesis and emplacement of ophiolites in the Oman Mountains geosyncline: Sch- weizer. Mineralog. u. Petrog. Mitt., v. 49, no. 1, 30 p. Riordan, P. H., 1955, The genesis of asbestos in ultrabasic rocks: Econ. Geology, v. 50, no. 1, p. 67—81. 88 ASBESTOS-BEARING ULTRAMAFIC ROCKS OF BELVIDERE MOUNTAIN AREA, VERMONT Roy, D. M., and Roy, Rustum, 1954, An experimental study of the formation and properties of synthetic serpentines and related layer silicate minerals: Am. Mineralogist, V. 39, nos. 11—12, p. 957—975. 1955, Synthesis and stability of minerals in the system MgO-AlQOa—SiOngO: Am. Mineralogist, v. 40, nos. 3—4, p. 147—178. Selfridge, G. C., J r., 1936, An X—ray and optical investigation of the serpentine minerals: Am. Mineralogist, V. 21, no. 8, p. 463—503. Smith, C. H., 1958, Bay of Islands igneous complex, western Newfoundland: Canada Geol. Survey Mem. 290, 132 p. Stevens, R. K., Church, W. R., St. Julien, Pierre, 1969, Age of ultramafic rocks in the northwestern Appalachians [abs.]: Geol. Soc. America, Abs. with Programs, [v. 1], pt. 7, p. 215—216. Sykes, L. R., 1967, Mechanism of earthquakes and nature of faulting on the mid-oceanic ridges: Jour. Geophys. Re- search, v. 72, no. 8, p. 2131—2153. Thayer, T. P., 1960, Some critical differences between alpine- type and stratiform peridotite—gabbro complexes: Inter- nat. Geol. Cong., let, Copenhagen, 1960, Rept., pt. 13, p. 247—259. 1966, Serpentinization considered as a constant-volume metasomatic process: Am. Mineralogist, v. 51, nos. 5—6, p. 685—710. 1967, Chemical and structural relations of ultramafic and feldspathic rocks in alpine intrusive complexes, in Wyllie, P. J., ed., Ultramafic and related rocks: New York, John Wiley and Sons, p. 222—239. 1969, Alpine-type sensu strictu (ophiolitic) peridotites —Refractory residues from partial melting or igneous sediments?, A contribution to the discussion of the paper “The origin of ultramafic and ultrabasic rocks” by P. J. Wyllie in Deep-seated foundations of geological phenom- ena—Upper Mantle Proj. Sci. Rept. 24: Tectonophysics, v. 7, no. 5—6, p. 511—516. Trofimov, V. S., 1940, [The carbon-bearing peridotites of the region of the village of Kalgachikha, Archangel dis- trict]: Akad. Sci. Nauk S.S.S.R., Sér. Géol., no. 6, p. 20—35. (In Russian, English summ.) Vermont State Geologist, 1898—[1947?], Reports, 1898—1947: Burlington, Vt., v. 1—25. Weiss, L. E., 1958, Structural analysis of the basement system at Turoka, Kenya: Overseas Geology and Mineral . 1 Zussman, Jack, Brindley, G. W., and Comer, J. J., 1957, Resources, v. 7, no. 1, p. 3—35; no. 2, p. 123—153. (Re- printed in 1959 in one volume: London, H. M. Sta- tionery Office, 65 p.) White, W. S., 1949, Cleavage in east—central Vermont: Am. Geophys. Union Trans, v. 30, no. 4, p. 587—594. Whittaker, E. J. W., 1951, An orthorhombic variety of chry- sotile: Acta Cryst., v. 4, pt. 2, p. 187—188. ———1952, The unit cell of chrysotile: Acta Cryst., v. 5, pt. 1, p. 143—144. 1953, The structure of chrysotile: Acta Cryst., v. 6, pt. 8—9, p. 747—748. 1956a, The structure of chrysotile; II, Clino-Chryso- tile: Acta Cryst., v. 9, pt. 11, p. 855—862. 1956b, The structure of chrysotile; III; Ortho-Chry- sotile: Acta Cryst., v. 9, pt. 11, p. 862—865. 1956c, The structure of chrysotile; IV, Para-Chry- sotile: Acta Cryst., v. 9, pt. 11, p. 865—867. 1957, The structure of chrysotile; V, Diffuse reflexions and fibre texture: Acta Cryst., v. 10, pt. 3, p. 149-156. Whittaker, E. J. W., and Zussman, Jack, 1956, The char- acterization of serpentine minerals by X-ray diffraction: Mineralog. Mag, v. 31, no. 233, p. 107—126. 1958, The characterization of serpentine minerals: Am. Mineralogist, v. 43, nos. 9—10, p. 917—920. Wilson, J. T., 1965, Transform faults, oceanic ridges, and magnetic anomalies southwest of Vancouver Island: Science, v. 150, no. 3695, p. 482—485. Winchell, A. N., and Winchell, Horace, 1951, Description of minerals, Pt. 2 of Elements of optical mineralogy—an introduction to microscopic petrography [4th ed.]: New York, John Wiley and Sons, 551 p. Winchell, Horace, 1958, The composition and physical prop- erties of garnet: Am. Mineralogist, v. 43, nos. 5—6, p. 595—600. Wyllie, P. J., ed., 1967, Ultramafic and related rocks: New York John Wiley and Sons, 464 p. 1969, The origin of ultramafic and ultrabasic rocks, in Deep-seated foundations of geological phenomena— Upper Mantle Proj. Sci. Rept. 24: Tectonophysics, v. 7, nos. 5—6, p. 437—455. Yoder, H. 8., Jr., 1952, The MgO—A1201r—Si0r—H20 system and the related metamorphic facies: Am. Jour. Sci., Bowen Volume, pt. 2, p. 569—627. Zussman, Jack, 1954, Investigation of the crystal structure of antigorite: Mineralog. Mag., v. 30, no. 227, p. 498—512. Zussman, Jack, and Brindley, G. W., 1957, Serpentines with 6-layer ortho-hexagonal cells: Am. Mineralogist, v. 42, nos. 9—10, p. 666—670. Electron diffraction studies of serpentine minerals: Am. Mineralogist, v. 42, nos. 3—4, 133—153. A Page Actinolite, in muscovite-quartz- chlorite schist of Belvidere Mountain Formation __ Aibite, detrital origin ____________ in albite porphyroblast zone _- 67 in amphibolite of Belvidere Mountain Formation _9, 10, 11 in contact-alteration zone ___- 19 in Hazens Notch rocks _______ S in metamorphosed sedimentary and volcanic rocks__12, 14, 15 in muscovite-albite-quartz- (chlorite) rock ______ 50 in schist of Belvidere Mountain Formation __ 10 in schist of Stowe Formation __ 12 relict grains in blackwall _____ 50, 55 replaced by garnet ______ 65 replaced by vesuvianite ________ 65 porphyroblast rock zone ______ 50,67 Allanite, in metamorphosed sedimentary and volcanic rocks ___- 15, 16, 17 Alteration, gains and losses _______ 59 isochemical __________________ 60 isovolumetric ________________ 61 Amphibole, in amphibolite ________ 14 in Hazens Notch rocks ________ 8 in metamorphosed sedimentary and volcanic rocks __12, 14, 17 in schist of Stowe Formation _ 12 Amphibole asbestos __________ 33, 37, 59 Amphibolite _______ 7, 8, 9, 10, 11, 18, 45, 48, 49, 50, 55, 62, 63, 65, 67 Anthophyllite ___________________ 34 fibrous _________________ 37 in dunite and peridotite ______ 25 pyroxene altered to anthophyllite ____ 35, 37, 56 relict in serpentinite _ _ 26 Anticlinal arch __________ _ 68 Anticlines _____ _ 8 Antigorite ___ _ 38 aluminian ___________________ 57 crystal structure modifications _ 23 habit ____________________ 19, 23, 37 in zones of marginal alteration bordering serpentine veins _______________ 37 interlayered with chlorite in pseudomorphs after pyroxene ____________ 40 predominant mineral in serpentinite _________ 57 replacement of olivine by antigorite ___________ 57 replacement of pyroxene or anthophyllite by antigorite ________ 35, 40, 57 INDEX [Italic page numbers indicate major reference-1 Antigorite—Continued Page rimming mineral of chromite or magnetite grains _____ 34, 40 Apatite, in amphibolite of Belvidere Mountain Formation __________ 10, 11 in blackwall—chlorlte rock _____ 50 in metamorphosed sedimentary and volcanic rock -_.._ 16 in rodingite _________________ 48 in schist of Belvidere Mountain Formation _ 10 in schist of Stowe Formation _ 12 relict mineral from country rocks _______________ 55 relict mineral in steatite ______ 42 Arsenopyrite ____________________ 36 Asbestos, amphibole __________ 38, 37, 59 replaced by talc ____________ 56 Asbestos veins, brucite intergrowth with chrysotile ______ 18 calcite intergrown with chrysotile ___________ 41 composition 37 cross-fibre veins _____ 27, 38, 57, 76, 82 geometric relations ______ 29 magnetite intergrowth with asbestos ____________ 36, 59 slip-fiber veins _____ 27, 30, 57, 76, 82 calcite intergrowths with chrysotile ___________ 18, 41 Augite __________________________ 35 B Bedding _________________________ 71 crosscut __________ 44,, 45, 48, 56, 68 folded ______________________ 48 lineations resulting from intersection of bedding and cleavage ________ 75 original differences in composition _________ 7 relict _______________________ 16 in albitic gneiss of Hazens Notch Formation _____ 8 in amphibolite, schist. greenstone of Belvidere Mountain Formation __ 7, 9 in Ottauquechee Formation __________ 11 in rodingite _____________ 48 in schist of Hazens Notch Formation __________ 7. 8 transverse to schistosity ______ 8 Biotite, in amphibolite of Belvidere Mountain Formation _ 9, 10, 11 in country rock ______________ 48 in Hazens Notch rocks _______ 8 in metmorphosed sedimentary and volcanic rocks ________ 12, 13, 14, 15 Biotite—Continued Page in schist of Belvidere Mountain Formation __________ 10 in schist of Stowe Formation _ 12 Blackwall chlorite rock, albite porphyroblast rock zone ________________ 67 association with ultramafic rocks where marginally altered to talcose rocks _______________ 62 Brittle-style folds ________________ 72 Brucite _________________________ 34, 40 “brucite" layer in crystal structure ___________ 23 fibrous ___________________ 18, 30, 31 in dunite and peridotite ______ 25 in microscopic veins __________ 31 in picrolite veins ____________ 30,40 in serpentinite ____________ 26, 27, 57 in serpentinized zones at margins of veins __________ 31, 32, 40 intergrowth with chrysotile _ 18, 27, 40 olivine replaced by brucite ___- 57 pyroxene altered to brucite ___- 35 Brucite-calcite-magnetite veins _________ 33. 34, 35, 36 C C-area quarry ___________________ 24 Calcite, associated with brucite in veinform aggregates _ 40 columnar ___________________ 66 fibrous _________ 18, 30, 33, 39, 41, 58 in amphiboiite of Belvidere Mountain Formation _ 11 in asbestos veins ____________ 27,41 in hlackwall rock 53 in metamorphosed sedimentary and volcanic rocks ___- 16,17 in phyllite, schist, gneiss, and quartzite ____________ in picrolite veins _ in rodingite _____ in serpentine veins in serpentinite ______________ 41, 57 in serpentinized ultramaflc igneous rocks “-1 40. 41, 57 .. n steatite. talc-carbonate rock. and carbonate-quartz rock ________________ 43 in tremoiite rock _____________ 53 intergrowth with chrysotile ___ 41 replaces brucite _____________ 33 Calcite veins, in ultramnfic and associated rocks _____ 41 with chlorite, and magnetite_-_45, 159, 53, 59,63, 66 with magnetite and brucite __ 33, 36, 59 Calcite veinlets, in dunite, peridotite, and serpentinite _________ 33, 41 89 90 Page Carbonate-quartz rock _____ 41, 43, 56 82 as tabular, irregular, or lenticular masses in main ultramafic body ______ 18, 41 bordered by talc—carbonate rock ________________ 41 derivation from ultramafic igneous rocks and serpentinite _________ 59 metamorphic reaction with dunite, serpentinite, and country rock ___- 60,61 relation to dunite, peridotite, and serpentinite _____ 60 relation to faults and shear zones _______________ 62 Carbonates ______________________ 34 in amphibolite of Belvidere Mountain Formation _ 10,11 in chlorite rock _____________ 50 in dunite and peridodite __ __ 25 in Hazens Notch rocks ________ 8 in igneous rocks, serpentinite, and veins ___- 26, 33,34,40 in metamorphosed sedimentary and volcanic rocks __- 16 in schist of Belvidere Mountain Formation __________ 10 in schist of Stowe Formation __ 12 in serpentinite _______________ 26 in steatite, talc-carbonate rock, and carbonate-quartz rock ________________ 43 in tremolite and chlorite rock association __________ 50 Carbonatization ____________ 60, 61, 62, 67 Chemical analyses of rocks _______ 7 Chlorite ________________________ 34, 40 at contact between serpentinite and amphibolite ______ 45 formed by interaction of magnetite and antigorite ___________ 57 in amphibolite of Belvidere Mountain Formationfi, 10, 11 in blackwail rock ________ 50, 53, 67 in chlorite rock ______________ 50 in dunite and peridotite ______ 25 in Hazens Notch rocks ________ 8 in metamorphosed sedimentary and volcanic rocks _ 12, 18,15 in rodingite _________________ 48, 53 in schist of Bclvidere Mountain Formation __________ 10 in schist of Stowe Formation , 12 in serpentine-chlorite rock _ 49, 53, 66 in serpentinite _______________ 26 in steatitc, talc-carbonate, and carbonate-quartz rocks_42. 43 in tremolite rock _____________ 50, 53 pyroxene altered to chlorite ___ 40, 57 rclict mineral in steatite from blackwall chlorite rock _ 43 calcite. and magnetite veins in rodingite and serpentine-chlorite rock ___- 45, 4.0, 53. 59, 63, 66 Chloritization ________________ 62, 65, 82 Chromite _______________________ 85 as layers in chromitite ___ _ 26 associated with chlorite __ _ 40 composition ___________ - 36 in asbestos veins _____________ 29 in dunite and peridotite -_ 25, 26, 56 in serpentinite __________ 26, 56, 57 in serpentinized zones at margins of veins _____ 31, 32 INDEX Chromite—Continued Page in steatite, talc-carbonate rock, and carbonate-quartz rock ______ 41,42, 43, 59, 62 intergrowth with magnetite ___ 35 primary mineral _____________ 34 replaced by carbonate ________ 43 zoning of grains ______ 34 Chromite-magnetite vein __________ 36 Chromitite ____________ 24, 26, 33, 36, 56 Chromitite veins _________________ 26 Chrysotiie ______________________ 39 Chrysotile, asbestos - 18, 27, 30, 37, 41, 59 habit _____________________ 19, 23, 37 in picrolite veins _____________ 37, 40 in serpentinechlorite rock _____ 54 in veins __________________ 34, 58, 59 replacement of olivine by chrysotile ___________ 34, 57 subvarieties ________ _ 19. 24, 39 Cleavage, axial-plane _____________ 78 fracture ____________________ 73 definition __ 70 in ultramafic rocks ___________ 71, slip __________________ 16,17, 73, 7‘4 definition _______________ 70 in rocks of Belvidere Mountain Formation __ 9 in schist of Hazens Notch Formation __________ 8 spaced, chlorite rock _________ 50 Clinochrysotile __________________ 39 Clinozoisite, in gneiss ____________ 13 in rodingite ________ 48, 49, 51, 63, 65 relict mineral _______________ 65 replaced by chlorite in rodingite _ 53, 65 replaced by garnet ___________ 65 replaced by vesuvianite _ 65 Conchoidal fracture Contact relations ________________ Contact rocks ___- Corez Pond body __________ 18. 41, 42, 43, 44, 60, 62, 68 Country rock (schist and amphibolite) ___ 7, 19, 44, 45, 50, 62 Crystals, bent __________________ 10 Crystals or grains, amorphous _____ 16 elastic ______________________ 17 cryptocrystalline _____________ 16 embay engulf fractured ________________ 10, 34, 56 healed fractures __________ 18, 15. 35 inclined extinction ___________ 34 invasion of _______ 13.14, 15. .‘H. 51 mortar structure ____________ 11 overgrowths _________________ 12 polysynthetic twinning ________ 12 serrated ____________________ 48 spindle-shaped _______________ 31 sutured _____________________ 12 strained ____________________ 12 undulatory extinction __11 34. 51, 56 uniform extinction ___________ 51 zonal pattern _______________ 34 1) Dimensional orientation of primary minerals in dunite and pcridotite _______ 71 Diopside. at contact between serpentinite and anipliibolite __________ 45 enclosed in rodingitc by garnet _ 52, 65 hornblende replaced by diopside _ 65 in dunite and peridotite ______ 35 in rodingite 1-“ ~18, 49. 50, 5], 63, 65 replaced by chlorite in rodiugite ____________ 53, 65 Diopside—Continued Page replaced by garnet ___________ 65 replaced by vesuvianite __ 65 Displacement along faults ________ 76 Dolomite. in metamorphosed sedimentary and volcanic rocks ______ 16 in serpentiuite ______________ 41 in serpentinized ultramafic igneous rocks _______ 40 DTA studies, chrysotile group _____ 39 lizardite ____________________ 40 serpentine ___- 37 Dunite _____________ 18, 24, 26, 32, 33, 35, 37, 40, 41, 42, 43 association with multiple‘fiber asbestos veins _______ 29 magnetite veins ______________ 34 marginal zones of alteration adjacent to serpentine veins _______________ 31, 32 picrolite veins 30 relation to association of rodingite and serpentine-chlorite rock 62 replaced by talc-carbonate rock _____________ 60, 61, 62 serpentine veins _ ___ 59 spinellids ____________________ 36 E Eden quarry body __ 18. 41,42. 43, 44, 45. 50, 59, 62, G3 Epidote. in amphibolite of Belvidere Mountain Formation _ 9, 10, 11 in Hazens Notch rocks _______ 8 in metamorphosed sedimentary and volcanic rocks _ 12,15,17 in rodingitc _________________ 50, 51 in schist of Belvidere Mountain Formation __________ 10 in schist of Stowe Formation _ 12 in tremolite rock _____________ 50 Epidote-amphibolitc facies ________ 9 Eugeosyncline ___________________ 76 F Facies. epidote-amphibolite ________ 14 Faults ___________________ 62, 68, 7'0, 76 control of distribution of rocks _ 41,82 Fluorapatite ____________________ 16 Folds ___________________________ 72. 78 in metamorphosed sedimentary and volcanic rocks _ 9. 68. 69. 7'0, '72 in ultramatic rocks ___________ ’73 Foliation. in ainphibolite of Belvidere Mountain Formation _ 10 in schist of Hazens Notch Formation __________ 8 Formations. Beividere Mountain Formation __________ 7, 8 Hazens Notch Formation ..... 7 Ottauquechee Formation ______ 7, 11 Stowe Formation ______ --__ 7,11 Formulas. amphibole ______ ___- 14,15 amphibole. calculations - __1_ 83, 8!, antigoritc ___________________ 22, 39 brucite. calculation __________ 83 chlorite in blackwall rock _____ 53 chlorite in calcite-chlorite- magnetite veins ______ 53 chlorite in chlorite-tremolite rock ________________ 54 chlorite in metamorphosed sedimentary and volcanic rocks _______ 14 chlorite, in rodingite _________ 54 Formulas—Continued Page chlorite in serpentine‘chlorite rocks _______________ 53, 54 chlorite in tremolite _________ 54 chlorite in ultramafic rocks and serpentinite _________ 4O chromite ____________________ 36 chrysotile ___________________ 21, 39 clinozoisite __________________ 15, 51 diopside ____________________ 51 epidote _____________________ 15 calculations _____________ 83, 85 garnet _____________________ 15, 52 lizardite ____________________ 40 magnesite ___________________ 43 magnetite in calcite—chlorite- magnetite veins ______ 53 olivine _____________________ 35 picrolite ____________________ 39 pistacite ____________________ 15 prehnite ____________________ 52 serpentine, calculation ________ 83 talc ________________________ 43 tremolite ____________________ 15, 55 vesuvianite __________________ 52 zoisite ______________________ 52 Fracture cleavage ________________ 73 Fractures _______________________ 82 G Garnet, alteration product ________ 65 in amphibolite of Belvidere Mountain Formation _ 10,11 in Hazens Notch rocks ________ 8 in metamorphosed sedimentary and volcanic rocks _ 12, 14, 15, 17, 18 in rodingite ___- 48, 49, 50, 52, 63, 65 in schist of Belvidere Mountain Formation __________ 10 in serpentine-chlorite rock ___- 66 microprobe analysis __________ 52 replaced by chlorite in rodingite ____________ 53, 65 replaced by serpentine ________ 54 Gersdorflite ______________________ 36 Glide twinning __________________ 56 Gouges _________________________ 76 Graphite, displacement of graphitic schist during blackwall- steatite reaction _____ 57 in amphibolite of Belvidere Mountain Formation _ 10,11 in country rock ______________ 48,55 in gneiss ___________________ 13 in Hazens Notch rocks _______ 8 in metamorphosed sedimentary and volcanic rocks ___ 16 in phyllite and schist of Ottauquechee Formation __________ 11 in serpentinite ___________ 27, 41, 57 in steatite, talc-carbonate, and carbonate— 4- __ —— __ _.. __ __ 1 1 1 1 1 1 3 4 __ _ _ __ i ' _- 10 Analyst 1 ____________ _- 1 1 1 _- 1 -- -- 1 1 __ 4- -_ -- -- -_ -_ -_ 1 2 2 2 2 _- -- 2 2 2 2 2 2 2 2 2 2 2 2 4 3 1' i ‘i "i q "i 3 3.3.3351... Cation” percent (based on atomic weights of 1955) Si ___________________ 33 6 37.24 39.59 ‘.1 40.70 40.3 29.30 50.1 32.3 39.48 35.78 39.0 49.3 31.5 41.2 32.1 49.5 57.0 53.3 3.73 41.58 50.34 41.96 36.90 -- -- 37.73 40.83 46.76 43.3 46.1 46.3 47,9 42 8 48 57 50 19 33 62 — - A1 ------------------ 1.1 .86 1.33 .0 1.08 15.8 17.14 18.4 11.5 .16 4.88 1.3 2.7 15.4 15.9 10.1 18.1 17.1 20.5 1.80 1.51 8.59 13.40 29.22 ,_ __ 7.06 1740 15.75 157 17.1 15.0 15.7 1531 I 1114 :76 :15 fig; —— i362 317:1; 3:502 3:11-33 £34235 40.3% 40.3% 6.51;: 811 F8 ,. ________________ 1.9 1.52 .85 .8 .57 1.6 2.43 1.8 1.7 .89 1.97 1.0 .4 .8 1.0 1.6 2.4 2.2 1.4 57.10 .25 1.31 2.52 6.81 __ -- 2.39 2.96 1.51 3.2 2.7 2.1 2.0 3.5 . .57 .49 1.11 63.80 " '64 1'32 4'34 2'83 270 '57 '59 .19 A Fe ________________ 2.2 .20 .57 .2 4,50 90 6.03 4.3 .39 _ .09 . .16 '41 3.0 8.9 10.4 .78 2.6 3.1 5.4 27.29 3.64 3.02 9.08 .78 -_ _ 1.63 8.77 6.84 6.4 5.2 6.4 5.4 6.7 . 3.66 4.57 2.81 30.60 33.04 1:10 '42 1'55 ‘43 4'51 '46 4'56 '11 i366 Mg _________________ 60-8 59-35 57-25 .5 52432 11-7 43-32 0-5 52-5_ 0890 04.64 025 33-1 32.8 20.0 51.9 7.4 5.1 5.0 7.28 51.59 27.58 16.24 .70 _- -_ 50.17 1329 113,8 14.7 9.9 125 11.8 15.4 . 2698 2583 61.67 .80 .07 4810 58:46 55:28 5864 47-08 57-28 52.59 10.54 Me Ca —————————————————— '04 -00 06 ~07 -06 20‘02 ~2( 9‘§ 30 00 1-10 -81 11-2 5'7 7-2 1-1 18.8 13-7 5-4 ~20 .12 10.84 9.64 24.11 -_ _, .00 10.86 11_79 12.1 135 10.1 10.8 11.5 ‘ 12.48 13.52 .07 .11 49.88 .63 23 07 '024 ‘03 -22 0 ‘04 40-«97 C g Na __________________ .00 .00 __ .00 .00 .31 .00 8.1 .00 00 .00 .00 .04 .02 3.0 .00 .09 .25 3.2 .00 .51 2.63 5.26 .60 _ _ W .00 3.92 3.74 3.1 4.3 5'3 5‘3 3'5 _ 24 .77 .00 11 '02 -02 -02 .00 .00 .00 .00 a K ------------------- .00 .00 -- .00 .00 .00 .00 .36 .00 00 .00 .00 00 .00 .12 .00 .02 .60 4.7 .00 .10 .20 .88 .10 -- __ .00 .73 .18 52 .09 .10 '11 .30 _ .11 .44 .00 —— '04 -02 '04 -01 .01 .00 .00 . Na T1 ------------------ .01 .00 -- .01 .01 .82 .70 .63 .58 .01 .06 .01 .01 2.0 .87 .65 .63 .55 .71 .01 .08 .01 .59 .46 ,_ __ .01 .83 .71 54 .7 .7 53 _7 . .03 05 00 3 25 7] -01 '007 .00 .007 .72 .01 .00 .00 K. P ___________________ .02 .00 _— ~03 .01 .15 -10 13 -14 -00 -01 -01 .01 33 .09 .17 .18 .18 .26 .00 .09 .11 .14 .14 __ -- .00 .01 _07 .09 .12 .10 ‘10 '13 . .04 __' :00 ' :02 :008 '00 -00 ~07 .08 '00 8(1) 121 CT —————————————————— -- '13 13 '22 -7 '04 -— -- 02 ‘01 -- -- —— —— —— -— ~— —- 1-22 -26 ‘11 -02 -02 __ -4 .21 .07 .02 .05 .03 .02 .02 .02 . _20 _10 .17 _55 “5 _11 .14 :29 :27 :05 '03 -16 -02 Cr 11‘ —————————————————— —- ~19 15 '21 -- '05 -- -- 11 '08 -— -- -- —— —— —— —7 ~— ~82 ~21 -09 -03 .002 __ __ .15 .02 .007 .02 .009 .006 .007 .007 . .07 .08 .19 _11 0 .02 .18 .29 _24 .06 '07 '24 ~02 101 CO —————————————————— —- -0062 004 '0091 -- ‘013 -- -- 004 '0057 -- -- ~« —— ~— —— —— ~- -077 -005 4004 -01 0 __ __ .006 .007 .003 .004 .003 .002 .003 .004 . _002 .002 .02 .04 0 _01 .0057 _083 0075 01 '0056 -0063 -0020 0 Mn —————————————————— 16 -08 05 -08 -13 ~18 32 21 ~30 02 ‘02 -06 08 15 24 30 21 17 13 .45 .05 .16 .18 .03 _, __ .12 .23 .17 .16 .16 .23 .16 J9 ' .09 .20 .11 .37 007 .05 .06 .09 ~08 -46 ~05 -13 -11 Mo Cu __________________ -- 0051 01 ~011 —— 0052 —— —— 0006 ~0011 —— —— —~ —— —— —— —— —— ~0037 0006 ~0007 -02 .005 _- 4- .004 .02 .003 .003 .003 .007 .008 .006 . .003 .003 .0005 .03 0 .005 .0023 00008 :00057 :0058 100067 0040 '0015 C111 Eb —————————————————— —— -— -- -- *7 ‘r -_ " " ‘— 7' -- -- -- —- —— ‘-— -- -0 g 3 8 8 ~— -- 8 8 0 0 0 0 .0006 0 0 0 0 O 0 __ __ __ _ ' Pb n —————————————————— —— —— —— —— ~— —— —~ -— —— —— fl -— —— —— —- —— —- ~- 4- -4 0 0 0 0 0 0 o 0 0 0 0 " " 7’ “ Ga __________________ __ <.0006 <.0006 <.0006 -- 0006 -— —— <-0004 <-0004 —— __ __ __ __ __ __ -_ 0 0 0 .001 .002 0 002 002 002 002 002 002 002 0 0 0 0 0 ’6 '6 " " " “ '7 ‘7 Z“ _- __ . . . . . . . 0 0 V ——————————————————— -- 002 ~002 '002 -— .8068 -- —- <‘00(1) '3373 -- —- —— —— —— —— -— —— 0141 ~002 -002 -02 -02 __ —_ -001 .02 .02 .02 .01 .010 .02 .02 . .003 .01 0 .1!" 0 .001 .0040 .0017 .0040 0.018 0 0017 0 0023 8 8a SC —————————————————— —— .002 ~002 <‘0006 -- ' 5 —- -- <‘00 6 ‘ 62 -- -— —- —— —— ~— -- -— 0 0 0 .010 1112 __ __ 0 .012 .009 .009 .007 .007 .007 _007 0 0 0 0 0 .0006 .0011 0011 002 0081 ' 0' 0 S Y _________ 1 _________ __ <.0005 <.0005 <-0005 -_ .002 —— —— <-0005 <-0005 -_ —— —— —— __ __ __ -1 0 0 0 0 .004 __ __ 0 .003 .003 .002 .003 .002 .002 003 0 0 0 .0001 0 0 0 0' 0' '0023 “ 0 0 Y0 Yb —————————————————— —— <‘0001 <40001 <-0001 -_ ‘0001 4- -* <‘0001 <'0001 *— —— -- —~ ~— -~ —- -~ 0 0 0 .0001 .0001 __ 1_ 0 .0002 .0001 .0001 .0002 .0001 .0001 0002 0 0 0 __ 0 0 0 0 0 '00012 —— 0 0 Yb Zr —————————————————— —— <001 <.001 <-001 ~- 0046 —- —— <-001 <.001 —— -— —— —_ -— -- -- -- 0 0 0 .0005 .002 __ __ 0 .001 .002 .0005 .002 .002 .002 502 0 0 0 __ 0 0 0 0 0 0- __ 0 0 Z Be ------------------ -- 0 0 0 —‘ 0 “‘ " 0 0 ” -- —- -- -- -- -- -- 0 0 0 0 0 __ __ 0 0 0 0 0 0 0 0 0 .0011 .0011 _ 0 0 0 0 0 0 -- 0 0 Br 3, __________________ -- 0 0 0 __ 0 -- __ 0 0 -_ -- -- __ _- -- _- _- .0003 0 0 0 .07 _- _- -- .004 .01 .01 .04 .01 .01 .005 0 0 0 -- .0005 _ 0 0 0 0 “‘ 0 0003 sf Ea __________________ __ 0 0 0 _- 0 __ 4_ 0 0 __ __ _ __ __ __ __ -- .0001 0 0 .0004 0 -- __ (g 38 .0004 .0004 .0007 .0004 .0004 .0004 0 0 0 0 0 0 _: .0001 0 0 00004 H 0002 0 0' Ba ___________________ -_ -_ __ __ _- -4. —_ —— 4— —— —— —— —— -— —— —— —— 44 —— __ — _- __ __ __ . __ -- __ __ __ -_ -- 253 581 298 -- ' ' 04+ __________________ .18 .31 .19 .19 .02 .07 .17 .06 .15 .32 1.27 1.18 .13 2.4 <.07 1.3 <07 <07 <07 -- -- -- __ _- __ .12 .01 .01 <06 <07 <05 .06 <05 05 _ 08 1.. -- -- -- -- ~- -— —— —— —— C . T0191 _________ 100.0 100.00 100.00 100.00 100.00 100.0 100.00 100.0 100.0 100.00 100.00 100.00 100.0 100.0 100. 100.0 100.0 100.0 100.0 100.00 100.00 100.00 100.00 100.00 __ __ 100.00 100.00 100.00 100.0 100.0 100. 10<0. 100. 106.0 166.00 106.00 100.00 166.00 163.33 100263 1003311) 1003015) 10060 10066 10036 10036 16666 04 Total Associated elements H ___________________ 418 8431 7723 748 7501 199 7613 55 777 8115 7853 712 190 602 328 791 130 50 195 856 6675 713 786 361 _- -- 6976 1763 1581 108 114 171 138 140 874 1263 1028 1450 8857 8463 5 2 -- g1 ------------------- -- 04 .30 -_ 02 -- -_ _- -- -_ -- _- -_ -- _- _- -_ _- __ -- -_ '33 .23 .06 .03 _- _- '33 .09 .03 .03 .00 .00 .03 .09 .63 .00 __ j :03 :00 7 383 83:33 ”'33 ”~33 75-3; 15-33 g __________________ . . . -_ __ __ -- _- -- __ -_ __ -_ __ __ __ __ __ __ . -_ __ -_ _- __ . -- __ __ _1 __ __ __ __ __ .01 ' ' ' ‘ s ___________________ -- __ _- -- -- 159 1‘822 15§:3 1§§7 "6 _- -- .02 .02 .69 .03 .04 .04 .90 -- -_ -_ -_ __ -- __ 1 .59 __ -_ __ -- __ __ __ ,_ __ __ __ ‘ '" '03 '02 '02 '02 ‘00 -02 -01 g1 0,“ _________________ 156.3 181.00 179.55 176.6 179.22 19 7 . . . 18 .93 179.85 177.0 160.4 174.6 164.7 179.8 167.1 169.9 170.6 138.09 176.26 155.06 151.54 157.10 __ __ 67.76 158.37 162.69 157.0 160.4 162.1 161.8 158.0 14799 153.65 144.05 13505 ~- __ " -_ “ " “ “ ‘— 0 notin 011950 062- 1141 9607 10194 1014 10417 1399 10175 1587 1007 9914 9878 1084 1411 1096 1324 981 1541 1640 1511 12953 10951 14793 14868 15349 __ - 9776 14072 14686 1462 1490 1450 1480 1440 13925 14102 12461 13455 15000 13525 19572 10151 19202 10132 1021: 10323 1576: 8‘“ ' 011 d F35 ----------------- 11394 “00620 _- 09672 -- 11663 160.9300 11091 09635 -- -_ 09547 11191 10134 10793 09599 11120 10667 09599 13455 09409 11343 11414 12234 __ - 10109 11570 11178 10649 10981 10853 10732 14321 11838 11632 12373 11264 03863 03725 09784 03574 09773 05633 09691 11663 F.93t1“ an 002 1 R, rapid analysis (colorimetric methods). S, standard analysis (conventional methods). 2 Mineral abbreviations: ab, albite; amphib, amphibole; calc, calcite; chl, chlorite; mt, magnetite; qtz, quartz; ser, sericite. “Reported as R209, which includes any or all of the follow- ing: A1305, total Fe as Fe-205, T109, P905, MnO. 4 Total Fe calculated as FeO. 5 Because of the unfavorable ratio of MgO to CaO, as much as 0.1 percent of CaO may have been missed. " Percent P, calculated as elemental phosphorus. 7 “Less 0” is a correction for oxygen reported in oxides but actually displaced by S (assumed to be in pyrite), F, and Cl, as follows: O for SE73 W8, 0 for F 4 W41, O for Clglé Wm, where W9, W1», and Wm Stand for weight percentage of S, C], and F, respectively. Fe in pyrite is assumed to have been reported as Fe405. ‘Analysts are keyed to chemical analyses by the following numbers: 1. mews ©9745” 10. 11. 12. 13. 14. 15. Rapid Rock Analysis Project, B. W. Brannock, project leader. Chlorine by S. M. Berthold. M. Seerveld, analyst. Faye H. Neuerberg, analyst. M. K. Carron, analyst, Alkalies by B. W. Brannock. Fluorine by S. M. Berthold, Mineral separations by W. F. Ottenbridge. Laura E. Reichen, analyst. E. J. Tomasi and Faye H. Neuerburg, analysts. Paul L. D. Elmore, Katrine E. White and Samuel D. Botts, analysts. Fluorine by L. E. Reichen and S. M. Berthold. Fluorine and free carbon by L. E. Reichen and S. M. Berthold. S. M. Berthold, analyst. Joseph I. Dinnin, analyst. M. Balazs, analyst. M. K. Carron, analyst. 9Value near limit of sensitivity; reported upon re-exami- nation of plates by Harry Bastron. “’ Other elements determined spectrographically are keyed as follows: 1. Looked for but not found (IWS—652): Ag, Au, Hg, Rh, Pd, Ce, Nd, Sm, Os, Ir, Pt, Mo, W, Re, Ge, As,Sb,BL(M,TLIn,La,Hf,Th,Nb,Ta,U,LL P. B. 2. Looked for but not found (IWS—699): Ag, Au, Hg, Rh, Pd, Ir, Pt, M0, W, Re, Ge, Sn, As, Sb, Bi, Te, T1, In, La, Ce, Hf, Th, Nb, Ta, U, P, B. 3. Looked for but not found (IWS—664): Ag, Au, Hg, Rh, Pd, Ce, Ir, Pt, Mo, W, Re, Ge, Sn, As, Sb, Bi, Tl, In, La, Hf, Th, Nb, Ta, P, B. 4. Looked for but not found (IWS~737): Ag, Au, Hg, Os, Ir, Pt, Mo, W, Ge, Sn, As, Sb, Bi, Cd, Tl, La, Nb, Ta, U, P, B. LlAnalysts are keyed to spectrographic analyses by following numbers: 1. Raymond G. Havens, analyst. 2. Harry Bastrom, analyst. 3. Harry J. Rose, analyst. 4. Janet D. Fletcher, analyst. Ru, Sn, Cs, Ru, 0d, Ru, Cd, Pd, Th, th e 1” Derived from weight percent value reported for “R203 [, which] contains any or all of the following: A1206, total Fe as Fe204, T109, P905, MnO” as follows: From 112032008, the value Mn:0.01 was subtracted; the remaining R500 (20.07) was assumed to be all FeO, calculated as Fe206. 1“ Calculated, based upon ideal value. ”Otztotal oxygen, including 0 associated with metallic ions and O in OH and C02. “Fezcell factor. Cation percentages, when multiplied by the cell factor, yield values that denote the contents of equal volumes of rock (the modified standard cell) ; the values are in terms of atoms per cell. ‘“ Based on an assumed Sp.Gr. of 2.62. UNITED STATES DEPARTMENT OF INTERIOR GEOLOGICAL SURVEY TABLE 1.——Ckemical analyses, spectrographic analyses, calculated cation percentages, and cell factors (F..) of some rocks and m FDashed leaders (I) means not determined. Zero (0) means looked for spectrographically but not detected] inerals associated with nltramafic rocks in northern Vermont PROFESSIONAL PAPER 1016 TABLE 1 . . e g 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 , 4::2$2$;fi:,. i 2 2 § 2 g s R R s s R R R R R R R R S S S S S S S S S S S S S R R R R R R s s s s S S S 4: g g :‘ g 2. g :1 ”i :4 _4_4‘_§§44:fi_:_:§g7—e—e—ee—e——~——e—~——e~—~e- A ‘ ‘ Sequence Sequence Sequence - Sequence Magnetite Graphitic G‘raphitic Graphitic . - - . . . . . . Magnetite . Calcite Chl ‘t . . _ S S “H Serpentine Serpentine . Serpentme- . . . from tramp- Flaky . . . _ . , . , _ Coarse Fine Coarse Fine Fine Albitic C . Graphltlc ' . . Graphitlc Massxve Chromite Magnetite Magnetite fro hl- Magnetlte 01” e Chrysotile . ' Serpentlnite . h ‘ ' . . . - - . - MaSSIVE - . - - - _ - ~ L G h t . . Amphibole Amph1bole Ep dote Serpentmite Serpentmite fine . fine . fine Serpentimte . . . . . . . . . . fl 02.1156. fi Glaphltl-C . . _ - _ . , m c f from chl- from cm. a b t s t t b (1 ~ - - _ Ch t1! ‘2. m M 4.22:. 2:332:22“ eieeeiite 2323332“ .. “974444,, ‘23:: 2.2:: 4:: h etgeggei ‘ me... .4342... .4424... eeeieieiite teeteeiitee eeeieeeiite eeieeieeiie eeeeeiteei ”4.221.. eeieiiitemie 4432.444... eeieieiitee 222223“ 322.223. “°2.33“‘ “32.233 00:23:33.“ 2323 .3 3222 3:30.000 00000 . A000 1 vein are k k h'b l't roe — roe ’ r00 [ ( ‘ . (q .. ‘ , . 1 1 e _ _ _ . roe , ' Laboratory or 5341089 E2131 E2132 5341094 132133 531:)f098 10:34 2mgig—i100 9 534101 E2135 E2136 53—1104 53—1105 53—1106 53—1107 53—1108 53—1109 53—1110 53—1111 E1947A 143378 143379 1433809 143380b 143381 143382 143383 143384 143385 1:13.36 $26337 1356868 143389 143390 143391 143392 143393 143394 1433959 143395b 148091 148092 148093 148094 148095 0:000? 149992 i300: i300]? 1:222: 152565 1:220:59 15256 mmgmwn Serial No. _ _ _ _ _ _ _ _ _ _ _ _ fl _ _ _ _ g ‘H _ _ _ 0636 E1941 E . 7 152568 152569 152570 _ . F. _ _ _ _ 1e . I I I V— 0 AVI2 3 AR— ARI110 Aft—110 AR—lll A—BM— A BM— A BM A BM A BM A BM A BM A BM A BM A BM A BM A BM A BM— AeBM— A—BM_ A—BM— A—BM— AIBMI AIBM_ _ _ _ _ _ _ _ _ 1942 E1943 E1944 E1945 E1946 E19 e ------------------ 70.04017 0.033500 (V'r‘m (VT—23) wee» ee 4 0040.0 e 0000 egg 4g Ag 105—6130 4,44 45,44 4‘00 .57.... .5... .54., A5,... Ag _ I I '0 - 64 127 55 Chemical analysis (weight percent) - . 40.6 30.32 53.5 34.9 42.02 38.11 41.7 53.7 32.1 42.6 34.4 50.9 58.2 54.6 3.02 44.30 55.54 44.86 38.18 I I __ __ I 42.02 42.60 48.93 45.7 48.4 48.9 51.0 45.0 46.6 55.25 55.66 5.1 3.1 __ __ _ e - , F 31100 “““““““““““ 3% 39132 4%; 3% 42.03 13.5 15.05 16.7 10.6 .14 4.41 1.2 2.5 13.3 13.9 9.2 15.8 14.8 17-8 1-24 1.37 3-36 12.15 25.65 I I I I I 1.00 15.40 14.88 14.0 15.2 14.3 14.2 13.5 4.8 1.10 .72 5.2 __ __ __ 40E: 2:32 (21:?) —— i300? 39088 38.76 31.83 32.92 43.62 42.90 7.66 8104 F50 ““““““““““““““““ 29 2:29 1223 5.4 .93 2.2 3.35 2.6 2.4 1.26 2.79 1-4 -5 1~1 1-4 2-3 3-3 3-0 1-9 61-47 ~36 1-92 3-58 9-37 —— —— I I —— 3-54 4.11 2.10 4.4 3.7 2.9 2.9 4.9 2.9 .86 _72 6.4 __ __ __ 1.77 67.58 6544 3 08 '89 1'85 :41): 1.39 10.77 .69 .88 .15 A140... Feb """"""""""" 3:0 .26 .48 1.6 5.78 10.9 7.46 5.5 .50 .12 .21 5-2 3.9 10.9 12-8 1.0 32 33 5-6 26-44 4-64 3-98 11.61 -91 I I I I —— 2.17 10.94 8.55 8.1 6.6 8.1 6.9 8.4 6.3 4.98 6.06 145 e 43_0 e 803 4 83.0 4_02 29.16 28:90 - 1-38 .53 1.96 3.353: 3.71 .82 .83 .30 F8203 M 0 """""""""""" 47.5 42.35 41.21 40.6 36.90 7.9 30.13 3.9 38.2 42.07 39.05 37-6 24-2 22-4 13.9 37-3 5-1 3-5 3-4 396 36-88 20-42 11-65 -49 I I I I I 375-47 9.30 7.99 10.4 7.0 8.9 8.4 10.9 22.0 20.59 1922 13.1 __ __ ~_ 49.54 .43 1 30 06 33-88 41.40 39.25 . 5.58 .60 5.75 _16 FeO CEO ““““““““““ 04 00 0 00 .07 0.00 18.8 .26 9.8 .35 “.00 1.09 .81 11.4 5-4 6-9 1-1 18-1 13-0 5-2 -00 -12 11-16 9-62 23.28 I I I I I .00 10.57 11.51 11.9 13.2 10.0 10.7 11.3 8.1 1325 14.00 __ __ 00 - 56-28 -62 .2 415.79 32.69 41.48 3723 8.30 MgO Na 0 """""""""" :00 .‘00 :00 .00 .00 .19 .00 4.8 .00 .00 .00 .00 .02 .01 1.6 .00 -05 .13 1.7 ~00 28 1-50 2-90 .32 I I I I I .00 2.11 2.02 1.7 2.3 2.9 2.9 1.9 .11 _14 .44 __ __ : I :00 -- —— - .06 .03 .00 .00 .03 0.00 :00 44,87 CaO 8.6 ’0'"_:::0:::::: .00 .00 .00 .00 .00 .00 .00 .30 .00 .00 .00 .00 .00 .00 .10 .00 .02 .48 3.8 '00 .33 .53 .33 '00 __ __ __ __ __ .9g '02 _5g .51 _07 _08 .09 .25 .06 .10 ‘38 __ __ __ __ .00 :: :: __ .03 .0: '00 .3: .3? '00 '00 '00 iiii) HIo-OIIII- _______ I .57 .24 I -09 I 20 —— —— ~89 -47 —— -- —- ~- -- -- -- 3 0 1'04 10-66 1‘18 1'2 - —— —— —~ ~— —— - - - - I I I I I .27 .20} 3 0 .10 __ __ I -70 1-00 -19 -43 .52 .07 . . 2 ~ I 0 .3 13.44 12.42 12.0 11.82 3.0 11.81 .88 12.6 12.95 12.54 11.4 3.1 9.2 5.0 12.7 2.0 1.8 . . . . . 6 .56 I I __ __ I 11.64 2.76 2.48 1.7 1.8 2.7 2.2 2.2 4.2 1.49 2.10 - —— I I 3. . . . . . . .06 .26 H40 r1048. “““““““““““““ 7.01 .00 .00 .02 .01 1.1 .96 .89 .84 .01 .08 -02 .02 2-7 1.2 .93 -86 -75 -96 ~01 -11 ~01 ~83 -63 I I I I I .01 1.15 .99 .76 1.0 1.0 .89 1.0 .08 .04 07 .21 ~_ 2.1 2 1 .33 132 —— 10 —— 1340): 1330i 11.3% 13%: 11.99 12.29 11.88 2.70 1:400 C(gf ““““““““““““ .15 .24 .15 .15 .02 .05 .13 .05 .12 .25 .99 .92 .10 148 <305 1.0 <<-05 <305 <<-05 __ —- —— -- -_ __ -- __ _~ -_ .10 .01 .01 <305 <305 <305 <305 <;05 <305 __ __ _2 __ __ __ .07 —— -35 -47 >12 .21 .99 .01 .00 .01 710, pg“. """""""""""" .03 .00 .00 .04 .01 .18 .13 .16 .18 .00 .01 .01 ~01 .40 -11 22 .22 ~22 ~31 ~00 -11 ~15 ~17 .17 I I I __ I .00 .01 .08 .11 .15 .13 .12 .16 .06 '06 __ 52.0 __ __ __ 00 -~ —- —— -03 .01 .00 . 0 .14 .04 .03 35.54 co2 60_LI:::::::::::::::: _- .00 .00 -- __ __ _- -_ __ e— ee ee -_ —— —— —— —— —— —— —~ -gg ~08 -02 -01 -- -- -_ —— —— '88 .03 01 .01 .00 .00 .01 .03 .01 .01 .00 __ ~_ __ :fi ~_‘ :: :: :: :01 .00 -00 '00 .33 .5? '83 '08 iio gl —————————————————— __003 __-02 __-Ol :: :_- _-: :: :: :: :j :j j: “.01 “.01 "38 “02 ".02 “.02 I 49 I I I I I I I I I 335 I I I j: j: j: j: :j :j : g1 :: “ : _ '01 -- -~ -~ -- 4‘2 -01 -01 ~01 -00 -01 :01 CI §e:6f::_:::_ ........ I :57 .23 I gt; I g; I —— 99 ~93 e— —— I —~ —— -_ ._ -- 1'33 ‘32 -- -- e— -‘ -- -- -- -- 33 ~- -_ —— —— —e —— -- -- —— —— 47-1 34.3 _ —— .33 .67 2.00 I: i: “.19 “.39 “.36 “.04 “.04 “.22 ".03 30 i0 --------------- ‘7 ‘ ‘ “ ' “ ' I I ' ‘9 ‘ 0 I I I I I I I I ' ' I I I I I I I I ' I “ “ ‘7 “ " -- -— —— —— —— I —— . I I I __ .24 .38 .32 .08 09 32 03 .~ “ . 22 .39 .26 .38 .0 .0 .08 10 .18 .29 .38 26 .20 .16 .43 .06 .21 28 .04 I I _ I I .15 .28 21 .20 20 .29 .20 .24 .20 . 2 - - - N10 g/InO ________________ .22 -10 __-OG —_ 10 n 16 —_ —_ ~h _4 __ a _H O __ 1_ __ __ __ __ __ __ __ __ __ __ :_ 14.4 11.2 2_9 5_9 4.3 1.42 __ __ __ __ __ __ __ 6.9 1&2 .00; :: :: —— :‘ '15 '35 24 '01 ‘06 ~08 ~11 10 .56 .07 .16 .15 M110 ‘iI800{_:::::I:I:: 10f. 10007 9985 100 9999 99. 10030 99. 10. 9991 9991 100 100.00 100.00 100.14 101.01 100.01 100.01 100.18 9975 9900 99:3 99%? 99:3 __ __ __ _- __ 100:: 993: 9933 100.00 100.00 100.00 100.00 100. 102. 9958 10049 968 __ I: _: 10006 10095 10058 :: 0022 0091 10034 10026 10015 0993 10029 10017 C Sum Less 0 ....... .01 .00 .00 I I I e— —— —— e— I —— - 0 - - 0 - . -‘ 0 - - - —~ —— —— —— —— - e - - . . . . .01 .00 .00 .00 <7 W 7_ __ .00 __ __ __ .00 .00 ‘00 '00 -00 -00 -01 .00 T _____ 101. 100.07 99.85 100. 99.99 99. 100.30 99. 10. 99.91 99.91 100. 100. 100. 100. 101. 100. 100. 100. 99.75 99.60 99.65 99.61 99.69 I I I I I 100.37 99.90 99.94 100. 100. 100. 100. 100. 102. 99.58 100.49 96.8 I I I 100.06 100.95 100.5 - ' ‘ - - . Less 0 AnalysigtllIIiIII 1, 2 3 ( 3 1 3 1 3 1 1 3 3 1 1 1 1 1 1 1 1 4 5,2,6 5,6 5.6 5.6 7 7 7 7 7 8,2 8 8 9,10 9,10 9,10 9,10 9,10 9,11 12,6 12,6 13,6 13,6 13,6 13,6 14 15 15 8 12 D10222 2,9691 122134 12,026 142115 49233 1.9228 1.9%” Analygt‘ltal Sp'Gr'Eigfifif _______ 289 ,_ _, 264 -- 332 __ 300 262 __ -- 261 298 288 304 262 314 304 292 483 256 299 310 344 -_ __ __ __ __ 265 323 341 293 304 300 295 313 292 302 307 390 409 458 462 301 502 481 273 245 238 270 262 274 260 268 289 Sp'Gr'Epwfler 5.19 5.21 ‘ ' ' ' " ' ' ‘ ra‘ns Spectrographic analysis (weight percemt) C __________________ I 0.0060 0013 I 0-013 I 00057 —— I 09007 0.0013 —— —— I I I I I e— 00034 0.0007 0.0009 0.02 0.005 I I I I I 0.005 0.02 0.004 0.003 0.003 0.01 0.003 0,006 0.007 0.004 0.003 0.002 __ __ __ 0.0009 . 2 4 .I. I I_ P10 __________________ __ __ __ I I I I __ I __ __ __ __ I __ __ __ __ I __ 0 0 0 0 __ __ __ __ I 0 0 0 0 0 0 .03 0 0 0 0 0 __ __ __ 0 EO 0'02 (0 0-006 0-0026 000009 000064 0.0065 0.00077 0.0046 0.0019 Cu Zn __________________ __ I I __ I —— I __ I I __ —— __ I —— I I I —— 0 0 0 0 0 __ I __ I I 0 0 0 0 0 0 0 0 0 0 0 .5 __ __ __ 0 u “ 0 —- -- -- —— —— I I I Pb Mn _________________ I I I __ I —— I I I I I —- —— —e e— —— —— —- e— —— 061 335 604 -1 -03 —— I __ I I .1 .30 .2 .1 .04 .07 .2 .2 .1 .09 .3 1 __ __ __ '1 __ :‘ -- -- —- —— —— —— I I 04H C0 __________________ I .0066 0038 I .0096 __ -013 I -_ .0040 .0050 __ __ __ __ __ __ __ —— - - - .01 0 I __ I __ _- .007 ’7 003 .004 .003 .002 .003 .004 .002 _002 _002 _2 __ __ __ .02 .03 :05 ‘6 -- -~ _ —~ —v —— __ __ __ n Ni __________________ I .17 .11 I .26 I .056 I I .080 .074 I I I I I I I I .69 .2 .1 .03 .002 I I I I I .1 .02 008 .02 .009 .006 .01 .01 .07 .08 .09 .1 __ __ __ .4 .09 .3 0 g; 410250 30134 E376 (050 3:38 .0368 .0024 09 Ga __________________ I <.0005 <.0005 I <.0005 I .0005 I I <.0005 <.0005 I I I I I I I I 0 0 0 .001 .003 I I I I I 0 .002 .002 .002 .002 .002 .002 002 0 0 0 0 __ __ __ 0 __ __ 0 0- 0' 0- 0. 0. 0. .1 .018 N1 Cr —————————————————— —— I I —e —— e— e— -— I -— e— I —— -— —— —— —— —— —— —— 0096 '002 602 433 -07 e —— I —— I ~30 '83 8% .05 .03 .02 .02 .02 .2 .2 .1 __ __ __ __ .2 h v 0 X 0 0 8a V 4444444444444444444 I .002 002 I 002 I 0023 —— —— <-001 -0066 I I —— I —— —— —— e— - . - - 2 -02 __ I __ I I .‘ 1 . . .02 .01 .009 .02 .02 .002 .003 .01 _04 __ __ __ 0 .1 .008 0 .001 h 0 -- —~ —— __ __ __ r Sc .................. _- .001 .002 __ .<.0005 __ .0046 _- _- <30005 .0049 __ __ -_ __ __ __ __ -_ 0 20001 0 .008 .009 _ __ _- __ -- 10001 .009 .007 .007 .006 .006 .006 _006 0 0 0 _003 __ __ __ 0 __ _‘ 0 .0005 fogii '0000 -38§4 .8534 0.0013 0.0020 3 'v Y ................... I <.001 <.001 I <.001 I .003 I I <.001 <.001 I I I I I I I I 0 0 0 0 .006 I I I I I 0 .005 .005 .004 .005 .004 .004 .005 0 0 0 0 __ __ __ _001 __ __ 0 0 0 0- 0- .0035 0 0 0 Se Yb __________________ R <.0002 <_0002 __ <_0002 __ .0003 __ __ <.0002 <.0002 __ __ __ __ __ __ I I 0 0 0 .0003 .0004 __ I __ I I 0 .0005 0004 .0004 .0005 .0004 .0004 .0005 0 0 0 0 __ __ __ 0 __ __ 0 0 0 0 0 ‘00030 0 0 Y Ti __________________ I I __ I __ I I —— I __ _ — I __ I I I —— —— —— -006 -01 .4 .3 I I I __ I .006 .8 7 .4 .5 .4 .7 .8 .02 009 _01 _1 __ __ __ 0 u u 0 u - 0 Yb Zr __________________ __ <.002 <.002 I <.002 I .0071 I I <.002 <.002 I I I I I I I I 0 0 0 .001 .003 I I I I I 0 .002 004 .001 .004 .003 .003 .003 0 0 0 .002 __ __ __ 0 h h 0 0 -(-) _0 '0 '0 _0 _0 I T1 Be __________________ __ 0 0 -- 0 -- 0 _- __ 0 0 __ __ -_ __ __ __ __ __ 0 0 0 0 __ _- -_ __ -_ 0 0 0 0 o 0 0 0 0 .0002 .0002 0 __ __ __ 0 -_ _- o 0 0 0 0 0 0 0 0 Zr S, __________________ __ 0 0 g, 0 _ 0 _- _- 0 0 __ __ -_ _- __ __ __ __ .0004 0 0 0 .1 __ __ __ -- -- 0 .006 .02 .02 .06 .02 .02 .007 0 0 0 0 __ __ __ 0 -_ -- .001 __ 0 0 0 0 0 0 0 39 B. __________________ _- 0 0 __ 0 -- 0 _- -— 0 0 __ _- _- __ __ —_ __ —— .0002 0 0 .001 0 -_ __ __ -_ __ 0 .001 .001 .002 .001 .02 .02 .007 0 0 0 0 __ __ __ 0 0 0 0 _' .0003 0 0 0001 0004 0 0.0004 Er Ca __________________ W <01 .044 __ 04:, -- __ I I <01 I I I I I I I _- I 11 3 "i ‘1 '1 I I __ I I 1 1 1 1 1 1.001 1.001 3 :1 0 ”i .05 I __ I .06 .06 .06 I I I .052 .017 ‘ '16 026 C: Others” _____________ __ __ I I I I __ I __ I __ __ __ __ __ __ __ __ I __ __ __ __ __ _» __ __ __ __ __ 1 2 __ __ __ 3 __ n 4 u __ __ I- . __ AnalySt ‘1 ———————————— —— 1 1 —- 1 e— 1 —e —— 1 1 -— —e —— —— —— —— —— —e 1 2 2 2 2 I I I I I 2 2 2 2 2 2 2 2 2 2 2 2 __ __ __ 2 3 3 4 3 1 1 —1~ ”i ‘1 ‘1 _1_ 22:32:11 Cation” percent (based on atomic weights of 1955) Si ___________________ 33.6 37.24 39.59 36.1 40.70 40.3 29.30 50.1 32.3 89.48 35.78 39.0 49.3 31.5 41.2 32.1 49.5 57.0 53.3 3.73 41.58 50.34 41.96 36.90 I __ I __ I 37.75 40.83 46.76 43.3 46.1 46.3 47.9 42.8 341 48.57 5019 5.6 __ __ __ 3362 027 0. ‘ _ “‘ 41 __________________ 1.1 .86 1.33 2.0 1.03 15.8 17.14 18.4 11.5 .16 4.88 1.3 2.7 15.4 15.9 10.1 18.1 17.1 20.5 1.80 1.51 3.59 13.40 29.22 __ I -_ I -_ 1.06 17.40 16.71,- 15.7 17.1 46.0 15.7 15.1 41 1.14 .76 6'7 __ __ __ .15 1.8.0? 38% —— i302 31711: 3:350“: gig 13:31? 40.42 40.66 6.53 3, Fe 00 ________________ 1.9 1.62 .86 3.8 .67 1.6 2.43 1.8 1.7 .89 1.97 1.0 .4 .8 1.0 1.6 2.4 2.2 1.4 57.10 .25 1.31 2.52 6.81 I __ I __ I 2.89 2.96 1.51 3.2 2.7 2.1 24, 3,5 1.6 '57 '49 5.3 __ __ __ 1.11 63.80 61.05 : :64 1-32 434 2-83 2.70 :5 .98 .15 A1” Fee ________________ 22 .20 .37 12 460 90 603 43 .39 .09 - 16 _41 30 89 104 .78 26 31 54 2729 364 302 908 .73 -_ _- __ _- __ 163 877 684 64 52 64 54 Gg 39 Sgfi 457 133 __ __ __ 231 3060 2937 .{04 110 -42 135 :3 451 .4: 59 .19 Fe: Mg _________________ 60.8 59.35 57.25 56.5 52.32 11.7 43.39 5.5 52.6 08.90 04.64 02.9 33.1 32.8 20.0 51.9 7.4 5.1 5.0 7.28 51.59 27.58 16.24 .70 __ I __ __ I 50.17 13.29 11.38 14.7 9.9 125 11.8 15,4 24.0 26.98 25.83 21.4 __ __ __ 61.67 .80 2.40 .07 48.10 5846 55-28 5864 47-08 57.28 74.56 .11 Fe + la __________________ .04 .00 .06 .07 .06 20-0 .27 9.8 ~35 .00 1.10 .81 11.2 5.7 7.2 1.1 18.8 13-7 5.4 .20 .12 10.84 9.64 24.11 __ __ __ __ __ .00 10.86 11.79 12.1 13,5 10.1 10.8 11.5 6.3 12.48 13.52 .08 u __ __ .07 .11 .11 49.88 63 ‘23 -07 ~024 -03 .22 52.59 10.54 Mg Na __________________ .00 .00 I 00 .00 .37 .00 8.7 .00 .00 .00 .00 .04 .02 3.0 .00 .09 .25 3.2 .00 .51 2.63 5.26 .60 I I __ I __ .00 3.92 3.74 3.1 4.3 5.3 5.3 3.5 .16 24 .77 __ __ __ __ .00 __ > _ __ :11 -02 -02 -02 .00 .00 .04 40.97 Ca K: ___________________ .00 .00 __ .00 .00 .00 .00 .36 .00 .00 .00 .00 .00 .00 .12 .00 .02 .60 4.7 .00 .10 .20 .88 .10 _- _- __ __ __ .00 .73 .18 .62 _09 .10 _11 .30 _06 .11 _44 __ __ __ __ .00 __ __ _ .04 .02 ~04 .01 .01 .00 .00 .00 pqa Ti __________________ .01 .00 I .01 .01 .82 .70 .63 .58 .01 .06 .01 .01 2.0 .87 .65 .63 .55 .71 .01 .08 .01 .58 .46 __ I __ __ __ .01 .83 .71 .54 .7 .7 4,3 ,7 .04 .03 _05 J7 __ __ __ '00 1.25 .09 0 .01 :007 -00 -007 .72 .01 33 .00 K. P ___________________ .02 .00 I .03 .01 .15 .10 .13 .14 .00 .01 .01 .01 .33 .09 17 18 .18 .26 .00 .09 .11 .14 .14 __ I I __ I .00 .01 .07 .09 .12 .10 .10 '13 .04 .04 __ 4.3 __ __ __ '00 __ __ H .02 .008 :00 -00 .07 .08 .00 '01 T. CT —————————————————— -— '13 ‘13 -— '22 7- ‘0‘} -- n '02 ‘01 7‘ -- -- -- -— ~~ -- -- 1‘22 ‘26 11 -02 .02 —— —— —— I I 2} .07 .02 .05 .03 .02 .02 .02 .17 _2() '10 40.8 __ __ __ '17 .66 1.96 0 _11 ‘14 29 :27 '05 ‘03 ‘16 ~03 5 Ni —————————————————— —— ~19 ~15 —— ~21 ~— ~09 -- -- ‘11 '08 , ~~ —- —— —— —— —— —— -- ~82 21 -09 003 .002 I I I I I .10 .02 .007 .02 .009 .006 .007 .007 .05 .07 .08 _11 __ __ __ 19 _11 .38 0 '02 .18 29 ‘24 '06 ~07 .24 .02 N1: Co __________________ —- -0062 -004 —— -0091 —— .013 —— —— ~004 -005‘ —~ —— __ —— -— __ —- —— -077 905 -004 -01 0 __ __ __ __ _- .006 .007 .003 .004 .003 .002 .003 .004 .002 .002 .002 _22 __ ~_ __ .02 .04 .06 0 .01 .0057 .083 0075 ~01 ~0056 .0063 .0020 1 Mn __________________ .16 08 05 08 '03 .18 820052 .21 .30 .8006 .8011 06 08 15 24 30 .21 .17 .13 .3337 .3006 .5807 '03 4635 I I I I I 634 -(2)3 -17 .16 .16 .23 .16 .19 .12 .09 .20 1.20 I __ __ .11 .37 .25 .007 .05 .06 09 '08 '46 '05 '13 '11 £3,121 Cu —————————————————— —— 9051 -01 - 11 —— - —— —- - - —- —— —— —— ~— —— —- —— - - - . - . —— —— __ I I . - 2 .003 .003 .003 .007 .003 .006 .005 .003 .003 .003 __ __ __ .0005 .03 .03 0 .006 .0023 ‘00008 '00 - ~ . . Pb __________________ _I __ __ —_ —_ e— -— -—— —— —— —— ~~ —— ——~ I- __ __ __ —— __ 0 0 0 0 _I __ __ __ _g 0 O 0 0 0 0 .0006 0 0 0 0 0 __ __ __ 0 _H __ 0 —_ H I_ ~_ 057 -0058 000067 .0040 .0015 $3 Z“ —————————————————— —— —— —- —- -- H —— n “ u " “ -- —- -- -- d- -- -- 0 0 0 0 0 —— —— —— I I 0 0 0 0 0 0 0 0 0 0 0 .50 __ __ __ 0 __ __ 0 __ __ __ —“ - —— —— Z Ga __________________ __ <-0006 <-0006 __ <-0006 e— -0006 e— -— <~0004 <-0004 —— —— I I I I __ —— 0 0 0 .001 .002 I I __ __ I 0 .002 .002 .002 .002 002 .002 _002 0 0 0 0 __ __ __ 0 __ __ 0 0 0 0 —(—) _0 7) -6 _0 G11 V ___________________ h .002 .002 __ .002 __ .026 I I <.001 .0073 I I I I I I I I -0141 .002 .002 .02 .02 I I I I I .001 .02 .02 .02 .01 .010 .02 .02 .002 .003 .01 .06 __ __ __ 0 .15 .012 0 .001 .0040 .0017 0040 018 0017 0023 0 Va Sc —————————————————— —— ~002 ~002 -- <30006 —— -0058 —— —- <30006 ~0062 e— —— —— —— —— —— __ ~— 0 0 0 .010 .012 -_ __ -_ __ __ 0 .012 .009 .009 .007 .007 .007 .007 0 0 0 _005 __ __ __ 0 __ 0 .0006 .0011 0011 0002 -0081 - 0. 0 Y ___________________ __ <.0005 <.0005 I <-0005 —— 002 —— —— <~0005 <~0005 —~ —— —— —— —— —— -- 4— 0 0 0 0 .004 __ __ __ __ 0 .003 .003 .002 .003 .002 002 .003 0 0 0 0 __ __ _ .0004 __ __ 0 0 0 O- 0- .0023 __ 0 Sc Yb __________________ __ <.0001 <.0001 __ <.0001 _- .0001 —- —— <00001 <00001 I —— -— —— -— I. —— —_ 0 0 0 .0001 .0001 _ __ __ __ __ 0 .0002 .0001 .0001 .0002 .0001 .0001 .0002 0 0 0 0 __ __ __ __ __ __ 0 0 0 0 0 -00012 __ 0 0 Y Zr __________________ __ <3001 <3001 __ <3001 _- .0046 __ —_ <3001 <3001 —— —— __ __ __ -- _- 0 0 0 .0005 .002 _ __ __ __ __ 0 .001 .002 .0005 .002 .002 .002 .002 0 0 0 .001 __ __ __ __ __ __ 0 0 0 0 0 0. -_ 0 0 :6 Be —————————————————— —— 0 0 —— 0 —- 0 -- -- 0 0 —- -— -- -- -- ~— -- -— 0 0 0 0 0 —— —— —— I I 0 0 0 0 0 0 0 0 0 .0011 .0011 0 __ __ __ __ __ __ 0 0 0 0 0 0 —— 0 0 Br SI‘ __________________ __ 0 0 __ 0 __ O __ __ 0 0 __ __ II -I _. __ I- I- .0003 0 0 0 .07 __ __ __ I. II e—— .004.01.01.04.01.01,005 0 0 0 0 __ __ __ __ __ __ _0005 __ 0 0 O 0 —_ 0 0003 Se Ba —————————————————— I 0 0 —— 0 —— 0 —— —— 0 0 —— e— — —— — -- -- —~ 0001 0 0 0004 0 I I I I I 238 .0004 .0004 0007 .0004 .0004 .0004 0 0 0 0 0 __ __ __ 0 0 0 0 __ .0001 0 0 00004 —— 0002 0 0. B: c ____________________ _ __ I __ I I I I I I I —— I _ I — —- —— I e— I I I I I I I I - —- I I I I I I 25.3 5.81 2.98 I __ __ __ I __ __ __ __ _ ‘ ' C‘+ __________________ .18 .31 .19 .19 .02 .07 .17 .06 -15 ~32 1-27 1-18 -13 2-4 <.07 1-3 <-07 -07 <.07 I I I I I I I __ __ .12 .01 .01 <06 <07 <06 <06 <06 <05 __ __ .3 __ __ __ .08 __ __ ,4 50.00 .46 - 61 —— 15 —— 27 -— 19 I 05 I 04 1135 84+ Total _________ 100.0 100.00 100.00 100.00 100.00 100.0 100.00 100.0 100.0 100.00 100.00 100.00 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.00 100.00 100.00 100.00 100.00 I __ __ __ __ 100.00 100.00 100.00 100. 100.0 100. 100. 100.0 100.0 100.00 100.00 100.0 I I I 100.00 100.00 100.00 100.00 100.00 100300 10000 10000 10000 10000 10000 10000 Total Associated elements H ___________________ 41.8 84.31 77.23 74.8 75.01 19.9 76.13 5.5 77.7 81.15 78.53 71.2 19.0 60.2 32.3 79.1 13.0 5.9 19.5 8.56 66.75 7.13 7.86 3.61 I I __ I __ 69.76 17.63 15.81 10.8 11.4 17.1 13.8 14.0 205 8.74 12.63 21.9 __ __ __ 19.28 .450 4.41.24 __ 88.57 8463 75 60 83 2 F V F ................... I .00 I I __ I I I I I I I I I I I I I I I '33 .23 .06 .03 I I I I __ .33 .09 .03 .03 .00 .00 .03 .09 .02 4,3 .00 __ __ __ __ __ __ __ __ .03 .00 :00 :03 1733 70-35 75(1): 153(5) g Cl __________________ .04 .03 .02 I I I I I I I I I I I I I I I I I . I I I I I I I I . I I __ __ __ __ __ __ __ __ __ __ __ __ .01 __ __ __ __ . . - - . . S —————————————————— n u u ‘7 ‘7 *‘ "‘ n n __ 9 " __ ‘02 ‘02 '69 ‘03 '04 '34 0'90 7‘- 09 1’00 2 155 0 15I54 157710 “ 7‘ “ _‘ "' 167.32 15g 37 10B 69 15,? 0 _0 —— —— —— __ __ _I .2 __ __ __ __ __ __ __ __ 03 02 '02 '02 '00 '02 '01 gl 0“ _________________ 1563 18100 17955 1766 17922 1599 17822 1598 1787 180.3 17985 1770 1604 1746 1647 1798 1671 169. 17.6 138. .6 .6 . . __ __ _- __ __ . . . . 16.4 1621 1613 1580 1219 14739 15365 1423 __ __ __ 14:05 13535 13:38 15000 18474 18160 17744 —— —— __ _- __ Otnot in 0H and Co. _ 114.1 96.07 101.94 101.4 104.17 139.9 101.75 153.7 100.7 99.14 98.78 103.4 141.1 109.6 132.4 98.1 154.1 164.0 151.1 129.53 109.51 147.93 143.68 153.49 I __ I __ 97.76 140.72 146.86 146.2 149.0 145.0 148.0 144.0 101.4 139.25 141.02 119.8 I __ I 124.61 134.55 133.14 50.00 95.25 95:75 10154 100102 gig; i030: i033: 10000 8:“ ' H .‘5 _________________ 1.1394 “0.9620 I 0.9672 I 1.1663 “0.9300 1.1091 09635 I I 0.9547 1.1191 1.0134 1.0793 0.9599 1.1120 1.0667 0.9599 1.3455 0.9409 1.1343 1.1414 1.2234 I I I I _ 1.0109 1.1570 1.1178 1.0649 1.0981 1,0853 1.0732 1.1321 1.3343 1183;; 1.1632 12006 __ __ __ 1.2373 _ __ - - - . not In 0 and 002 1.1264 0.8863 0.8725 0.9784 0.9574 0.9773 0.9633 0.9691 1.1663 F. ‘5 1 R, rapid analysis (colorimetric methods). S, standard analysis (conventional methods). 2 Mineral abbreviations: ab, albite; amphib, amphibole; calc, calcite; chl, chlorite; mt, magnetite; qtz, quartz; ser, sericite. “Reported as R404, which includes any or all of the follow— ing: A1405, total Fe as F8203, T104, P205, MnO. 4 Total Fe calculated as FeO. Because of the unfavorable ratio of MgO to CaO, as much as 0.1 percent of CaO may have been missed. 6 Percent P, calculated as elemental phosphorus. “‘Less 0” is a correction for oxygen reported in oxides but actually displaced by S (assumed to be in pyrite), F, and Cl, as follows: 0 for 8313/8 W4, 0 for Fg.4 We, 0 for Clgl/a Wm, where Ws, Wet, and Wm stand for Weight percentage of S, Cl, and F, respectively. Fe in pyrite is assumed to have been reported as F8202. “Analysts are keyed to chemical analyses by the following num 1. mews #0 ©W99 1 1 12. 13. 14. 15. bers: Rapid Rock Analysis Project, B. W. Brannock, project leader. Chlorine by S. M. Berthold. M. Seerveld, analyst. Faye H. Neuerberg, analyst. M. K. Carron, analyst, Alkalies by B. W. Brannock. Fluorine by S. M. Berthold. Mineral separations by W. F. Ottenbridge. Laura E. Reichen, analyst. E. J. Tomasi and Faye H. Neuerburg, analysts. Paul L. D. Elmore, Katrine E. White and Samuel D. Botts, analysts. ‘“ Other elements determined spectrographically are keyed as follows: 1. Looked for but not found (IWS~652): Ag, Au, Hg, Ru, Rh, Pd, Ce, Nd, Sm, Os, Ir, Pt, Mo. W, Re, Ge, Sn, As, Sb, Bi, Cd, T1, In, La, Hf, Th, Nb, Ta, U, Li, Cs, P B 2. Looked for but not found (IWS—699): Ag, Au, Hg, Ru, Rh, Pd, Ir, Pt, Mo, W, Re, Ge, Sn, As, Sb, Bi, Te, 0d, T1, In, La, Ce, Hf, Th, Nb, Ta, U, P, B. 3 Looked for but not found (IWS—664): Ag, Au, Hg, Ru, Rh, Pd, Ce, Ir, Pt, Mo, W, Re, Ge, Sn, As, Sb, Bi, Cd, Tl, In, La. Hf, Th, Nb, Ta, P, B. 4. Looked for but not found (IWSJ737): Ag, Au, Hg, Pd, Os, Ir, Pt, Mo, W, Ge, Sn, As, Sb, Bi, Cd, Tl, La, Th, 1" Derived from weight percent value reported for “R204 [, which] contains any or all of the following: A1203, total Fe as Fe206, T104, P205, MnO” as follows: From R203=0.08, the value Mn:0.01 was subtracted; the remaining R203 (20.07) was assumed to be all FeO, calculated as F9203. ‘3 Calculated, based upon ideal value. “Oeztotal oxygen, including 0 associated with metallic ions and O in OH and 002. ‘fiFczcell factor. Cation percentages, when multiplied by the cell factor, yield values that denote the contents of equal volumes of rock (the modified standard cell); the values are in terms of atoms per cell. 1“ Based on an assumed Sp.Gr. of 2.62. Fluorine by L. E. Reichen and S. M. Berthold. Fluorine and free carbon by L. E. Reichen and S. M. Berthold. S. M. Berthold. analyst. Joseph I. Dinnin, analyst. Nb, Ta, U, P, B. “Analysts are keyed to spectrographic analyses by the following numbers: 1. Raymond G. Havens. analyst. 2. Harry Bastrom, analyst. 3. Harry J. Rose, analyst. M. Balazs, analyst. M. K. Carron, analyst. 9Value near limit of sensitivity; reported upon re—exami— nation of plates by Harry Bastron. 4. Janet D. Fletcher, analyst UNITED STATES DEPARTMENT OF GEOLOGICAL SURVEY 6 THE INTERIOR Base from US. Geological Survey to“? Unpublished Jay Peak special map, l:10,000, 1953 “owl Wm (“B HIHU N in H1 APPROXIMATE MEAN DECUNATION, 1978 GEOLOGIC MAP OF THE BELVIDERE MOUNTAIN AREA, EDEN AND LOWELL, LAMOILLE SCALE 1:12 000 V2 lMlLE .5 1 KILOMETER y___( CONTOUR INTERVAL 20 FEET DATUM is MEAN SEA LEVEL l—fl (Outc) quartzite (Cha) ' K INTEIORiGEULOGlCAL SURVEY, RESTON, VA._19787G75268 V Geology by A. H. ChidesterC. A. Ratte’, and J. C. Flatte’, 1952—53. Geology of the Lowell quary area is based on plate 3. This map incorporates results of invetigation by the mine exploration staff of the Industrial Products Diviion, GAF Corporation (formerly Vermont Asbestos Mines Division otthe Ruberoid Company), as well as restudy by Chidester of areas striped after 1953 and until June 1960 CORRELATION OF MAP UNITS Camels Hump Group ORDOVICIAN Lower Ordovician Upper and Middle Cambrian CAMBRIAN Lower Cambrian CAMBRIAN(?) DESCRIPTION OF MAP UNITS RANK IN 4 a 44° CHAMPLAIN ,—. —— \—-——— NEW ' _..—-— lam ‘CHITTEN ULTRAMAFIC IGNEOUS ROCKS AND DERIVATIVES (ORDOVICIAN)1 serpentine (antigorite and (or) lizardite) and small varied proportions of chrysotile; and schistose serpentinite (Ous). In extensive covered areas, unit composed typically of about 40 percent magnesite and 60 percent talc; steatite is composed essentially of talc. Talc-carbonate rock is intergradational with steatite and also with carbonate—quartz rock (Ouc). Steatite commonly has a thin selvage of tremolite; where appreciably less than 0.3 m thick, a steatite body commonly consists chiefly or entirely of tremolite and minor chlorite - CARBONATE-QUARTZ ROCK—Composed essentially of 70 percent -—+ magnesite and 30 percent quartz; intergradational with talc—carbonate rock STOWE FORMATION (LOWER ORDOVICIAN)—Grayish—green quartz— sericite—chlorite schist characterized by numerous lenticles of granular white quartz parallel to schistosity BELVIDERE MOUNTAIN FORMATION (LOWER CAMBRIAN) FINE AMPHIBOLITE—Greenish—gray to medium—bluish—gray, distinctly bedded fine amphibolite (inside the hornblende isograd) and actinolitic and calcareous greenstone (outside the hornblende isograd). Individual crystals are rarely discernible to the naked eye €bs MUSCOVITE—QUARTZ—CHLORITE SCHIST—Silvery—green, contains abundant coarse spangles of white mica and scattered large nodules of magnetite commonly 10 mm and locally 20 mm across COARSE AMPHIBOLITE—Dark—greenishgray or greenish—black, distinctly BZOA bedded; contains minor amounts of fine amphibolite. Hornblende crystals are predominantly 5—10 mm long and range to as much as 25mm HAZENS NOTCH FORMATION (CAMBRIAN (?))—Within the formation in the northem part of the area, early minor folds diverge markedly in trend from 73° __.._.._.-—i—__—___~ l~ PROFESSIONAL PAPER 1016 PLATE 1 fi-Fi Contact—Solid where accurately located, dashed where approx— 70 . . . . imate. Arrow shows direction and amount of clip '''''' * Contact——Location based upon magnetometer survey T— Fault or shear zone—Solid where accurately located, dashed where 5” approximate. Arrow shows direction and amount of clip —-l——> Anticline—Showing approximate trace of axial plane and direction of plunge of axis. Inferred from stratigraphic pattern —'(——> Syncline—Showing approximate trace of axial plane and direction of plunge of axis. Inferred from stratigraphic pattern PLANAR FEATURES it. Strike and dip of inclined axial plane of minor fold Strike and dip of bedding i Inclined Vertical EB Horizontal Strike and dip of schistosity parallel to bedding ULTRAMAFIC ROCKS, UNDIFFERENTIATED——Chiefly: dunite, composed 78 . . . . . . . _ . 4‘ Inclined essentially of oliv1ne but generally consrderably serpentinized, masswe serpentinite, which intergrades with dunite, composed chiefly of bladed 1 Vertical Strike and dip of schistosity divergent from bedding, in nonbedded rocks, or where relations are indeterminate probably includes some bodies of talc—carbonate rock and steatite (Outc) i I l' d and carbonate—quartz rock (Ouc) “C me Ous SCHISTOSE SERPENTINITE—Matrix composed chiefly or entirely of bladed + Vertical serpentine (antigorite and (or) lizardite); chrysotile confined almost entirely Strike and dip of slip cleavage to shear surfaces; little or no relict olivine 63 . 4— Incllned Outc TALC—CARBONATE ROCK AND STEATITE—Talc—carbonate rock is , ————» ‘9‘ Vertical LINEAR FEATURES Generally combined with one of the above planar symbols for features in bedded rocks Bearing and plunge of minor fold axis or lineation Fold axis or crinkle related to the longitudinal folds of the Green Mountain anticlinorium a Folds transverse to and warped by the longitudinal folds "M W I Pattern, in plan, of folded or crinkled bedding Strike and dip of layering in ultramafic rocks 50 OTTAUQUECHEE FORMATION (UPPER AND MIDDLE CAMBRIAN)— 4* Indined Graphitic sericite—quartz phyllite and associated thin beds of dark—gray —H— Vertical quartzite, light—green quartz—sericite-chlorite phyllite, light—buff sericite :fi: Horizontal phyllite, quartz—pebble and feldspar-granule conglomerate, and sericitic _1__l__l_. Hornblende isograd—Boundary between greenschist and epidote— amphibolite facies. Hachures are on side of epidote—amphibolite facies (higher metamorphic grade) Diamond-drill hole—Cross locates collar, tick locates bottom of hole; value in degrees gives inclination from the horizontal 68° Inclined #‘33 Vertical Triangulation point in mine survey EQ~5 D J Geologic section lines in fence diagram, plate 2. Location of section lines was influenced largely by the position of critical drill holes R late folds, and patterns of refolded folds are clearly exposed in several ‘94? , , _ , , outcrops. Major folds are inferred to display the same patterns. Outcrops are Mme dmlmg‘gnd lmes’ marked by Stakes on the ground too sparse to document the relation, but the pattern is illustrated by (Q diagrammatic crinkles and warps in favorably situated contacts of map units (6’0 GRAPHITIC QUARTZ—MUSCOVITE—CHLORITE-ALBITE SCHIST—Chcl is oldest. Contains interbeds of fine amphibolite and actinolitic greenstone QUARTZ—ALBITE—MUSCOVITE—CHLORITE GNEISS—Greenish—gray QUARTZ—MUSCOVITE—CHLORITE SCHIST—Greenish—gray t0 gray- ish—olive—green {7 Ch UNDIFFERENTIATED SCHIST AND GNEISS ‘The Ordovician age designation refers to the age of emplacement ofthe intrusive ultramafic rocks, not to the age of the parent igneous rocks or the metamorphic derivatives (see text subheading discussions under “Tectonic and petrogenic synthesis"). 72° - “§-- —--‘________— K \ NorthTroy QUADRANGLE ' l XORLEANS 0 Lowell i, 1 Area of , ”I this report Wk HYDE PARK ‘ ~ - QUADRANGLE 0 Eden M’HS gALEDONIA \ \ /\ U ,4 45‘ _——-"\\ \ ’\ $0 (J $0 G E s is! ' N 65 s 0., N e g / I/ ll, \‘ 80 *1? ,«w I N D s 0 R\~ It \ ~I\ \ [Q $ a 7 \\ A {L} m ZOMILES 10 20 30 KILOMETERS AND ORLEANS COUNTIES, VERMONT UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY DESCRIPTION OF MAP UNITS ULTRAMAFIC IGNEOUS ROCKS AND DERIVATIVES (ORDOVICIAN)1 ULTRAMAFIC ROCKS, UNDIFFERENTIATED—Chiefly: dunite composed essentially of olivine but generally considerably serpentinized; massive serpentinite, which intergrades with dunite, composed chiefly of bladed serpentine (antigorite and (or) lizardite) and small varied proportions of chrysotile; and schistose serpentinite. Includes some talc—carbonate rock, carbonate—quartz rock, and steatite (Outc), particularly at the margins of small pods of ultramafic rock Outc TALC—CARBONATE ROCK AND STEATITE, MINOR CARBON- ATE—QUARTZ ROCK—Talc-carbonate rock, composed of about 40 percent magnesite and 60 percent talc, intergrades with steatite, composed essentially of talc, and carbonate—quartz rock, composed of about 70 percent magnesite and 30 percent quartz STOWE FORMATION (LOWER ORDOVICIAN)—Grayish—green quartz- sericite—chlorite schist characterized by numerous lenticles of granular white quartz parallel to schistosity OTTAUQUECHEE FORMATION (UPPER AND MIDDLE CAM~ BRIAN)—Graphitic sericite—quartz phyllite and associated thin beds of dark—gray quartzite, light-green quartz-sericite—chlorite phyllite; light—buff sericite phyllite, quartz—pebble and feldspar-granule conglomerate, and sericite quartz BELVIDERE MOUNTAIN FORMATION (LOWER CAMBRIAN) FINE AMPHIBOLITE AND GREENSTONE—Greenish—gray to medium— bluish—gray, distinctly bedded fine amphibolite (inside the hornblende isograd) and actinoliti- and calcareous greenstone (outside the hornblende isograd) €bs MUSCOVITE—QUARTZ—CHLORITE SCHIST—Silvery—green, contains abundant coarse spangles of white mica and scattered large nodules of magnetite commonly 10 mm and locally 20 mm across COARSE AMPHIBOLITE—Dark-greenish-gray or greenish—black, distinctly bedded; contains minor amounts of fine amphibolite. Hornblende crystals are predominantly 5—10 mm long and range to as much as 25 mm HAZENS NOTCH FORMATION (CAMBRIAN(?)) GRAPHITIC QUARTZ-MUSCOVITE—CHLORITE—ALBITE SCI-HST QUARTZ—ALBITE—MUSCOVITE—CHLORITE GNEISS—Greenish-gray QUARTZ—MUSCOVITE-CHLORITE—SCHIST—Greenish—gray to gray— #_ ish—olive—green Ch UNDIFFERENTIATED SCHlST AND GNEISS Contact—Dashed where extrapolated above ground surface Fault—Dashed where extrapolated above ground surface CORRELATION OF MAP UNITS Lower ORDOVICIAN Ordovician Upper and Middle Cambrian Lower Cambrian CAMBRIAN CAMBRIAN (?) ‘The Ordovician age designation refers to the age of emplacement ofthe intrusive ultramafic rocks, not to the age of the parent igneous rocks or the metamorphic derivatives (see text subheading discussions under “Tectonic and petrogenic synthesis”). 0 7.33:3. 7‘ $7777. A\ '\ ,, .3777: I 73777:: 7.71177 7 77.1 ,s 77.777 .77.: 75““ 7 57$. ,7 7,...7.,,,.77. “7.77 “i . . 77,... L , 2,, . $77: 7 7777' 7,777 I. .1 .m. .7 at: 777. 5 7 7 “7,773,117: , “@777 ,i 7‘ sin. .7 . ”757777“ f 7.77 :7, .7757 7;: 7. at“ .7777 77, 7‘77. . M7,...777 W774. with; Vim? .. 77 .7 ,.,,. ,‘77.77,7',,7,,,t .7 77 @7777 ‘. ,i 777 :27 ELL—e s37 3 1‘7 ‘.‘.; ,7, ’ 77:7 7.7,. il s .7 L as .1ng 7. :1 579s \ “OE E 327:: 2.1%, 7 s. .7. . .77 ‘ 7‘7 7.7.7 {’77 7.777 . ~ 5 W751“, - 7‘7 .. ,‘7777777777, 7739177775“ ‘ 5.7.7 7 77.7,...‘7‘ " ”i" w W 7 'i 777 .7, , 7. ., Ch 743‘ 7.“ 777717.. 754,7, 7;). 7.1.7.7.. 7.77.77.77.77 77.7.7 7 iii. ‘ it in It 77:77:75“, ' ‘7 $57,, ““7 777.777, (is, 1,. it 111;: :7. 3.7». 2.3% €5.53; .7: .3, 7%: M4: ,7. . S5553: ‘ «Essa.» 1 3‘3? 3: , 77 77.7,? ,7, 7. . 7.777... 777.777,; 7 77,777,, 7 77,7». 7 ’7 7777 a 5. s73... .. : s. sax-{77$ «Ki: . ,, 7.7.77 "7 {77777777, 7,. ‘ 77.7777‘77777/7‘77777777 7,7 777777777 i7‘..7«;7 77. 9,7 7,77 s 7. 7 77 . ‘77, ,7 7 7.7. , 73.7.: 7 «”772: .77 777.,- ,.:.,7777. :7. .7 .7. 7.77.7.7. .7 . 7:777. 777 5‘.“ .7 7. 77. 77,}: 77777. .7 77, 7,7775,- 7'.‘ '1 I.“ L 7 " 7.77,, ,.7.s,7, . ,7717777' , 7,, .. , . ' - . . 7 7. 7 7 . 7 . .7772“: ‘ ,,. ., ,, . 7. . r If, 7,, . fl, 7‘7 7, 77,. ,R 7'77 w‘ L :3; igis‘“ ‘~ X s3? ~7‘ ,\ 7 .i. 77 ,7 . ”7, 7 777777777. 7777777777777 T. 5‘ < ‘A 5:31. 7 77771777777777: ’7 $92: x333. 7 .77 ’77 ”.777 77‘77‘7‘7: . 77, 77,77 777...:7‘“ 7777.,777 ,3 1.7.7,,7,,._,7,.,7 , ., , , 7.57.7277. - "7.77.7.7: 7 . , .,.,...,., 7.7.7 ., 77 .“ 7 7 ‘77..‘77.:7:77‘: 7 7777.773".- 77 r .-..s;-v.7';:7‘,,7:t, s ,7... .77 7.77.7777... a. .7777 7, ., 3.77.77. 7,,.7,.L..77 7;..7 . 77 mfihkfii“, ‘7‘ ‘ilé‘awiAAV-t“ .7, 7777777777., V \ 7,.c 77.77 “ 7.7.7717“; 7 “A 7 \ WW‘WWA .7 7 ,.»r ilxx‘ii”\,7lf 't“.:7‘..\i» 1““;“37'fl7‘fi, "7.77.7 abstinwfi 7.77 7.7 7 7 7 7:“ .flsggsilflwis 7« - .7 .7 my?» 17 Mi 7 3.7.7.777. ' ,7?s%7hm,i 77.77791“. ., W777. - 7.777337. .7797 t ,7: 77‘7““‘7‘77 77.7.7777. 77.7 7 .‘w‘allfi N” 7 . . 7 7 7777.77 L:- H; .4. 5:777 «WM-:1?- “ i “Mm: 7 ‘7 is “i l» ”I ' .7‘7 ., .7 .7 .477‘77: 3,3,, 7,7. 7“ 7». 777M- 7 WWW“ “ 7‘ £5377 7‘“ ,7... 3.7.7.7.,» , 7 7 7 .77 .MH 1.4 7:77:77: 7 7.7.7 77:777.: t‘fii’ehe Q‘Itliilill 7 “:9.“ 777:7 ,7. ‘iitil‘if‘la’i‘iaifia 7737,7777 77777.77 77.7,... .7777, 7 777,773.,7;771737,7,,71.13,7 ,77... .... . . -. $77,303 7:777‘7‘77‘7‘ 7,. . ‘77. . ‘ 7‘ 7. _ 7‘377.7.-‘»77‘:.777 7.7%: 7‘37: 7.7777 77.77 717E,.,|,,, ' ’ 7 ,7 isw, .7 ‘77:: .‘.7 2‘ i . , ~7 777777.777. , 7.. - , ~ ,7 -‘ 1:77.77 77777:“. . fl, .7 . .77,,,fi3,77, . - ,,.7*-..7.7.7.-.7.77 .. 7,, ,7 , 7,7,7,- .. 7. .. 7 7 7,77,77,77,, 77, «$677,? ,7"? .77 , . , , .7 7.7.7 . . ,,.,, , _ ~1777i,‘ 77, 77.7%777‘77777 77 7 ‘ ;. p7” :7 ’ 7 ‘ ' "L7" [7,3,], ,7 7 7 7,771.77. ’ .7777, 3‘ is; $72 is F. 7.. 777777.777» 7 , .7, 7777.77.77.77‘777‘7: 7 , 7.....77rii7..7 ‘ . 7,3,7, ..,,,7,7,7‘ . .7 $7,777,; , .. "77‘ i. 7 77777.17 7 77.777.77.77 7.7717777 177‘ 757777271515. 3? -..7- Wars 7777.. 77 . 7 7. ‘73". 17777 :3 1,. b 17‘. n. 7'7. 7 «771.777.7777 7774 7.7.777 . 177,777.77”. W?” 777%8777? .77777‘7’77'7‘77’W77‘7 .77 77777777» . ..,,,7,.,7f,,. ., , ,, 77: ,1, £511: , 5:; ~ gs =si 3 77.7 7777777717777 .7, . , 7.11%., 77.77.77 . 77777.. ’ W71"? .,.7,7.i. '7 ‘ ,.,,7 w ,7 ,1 777.77’ ,7.,77777.777y ,. :7 77,777,. 7 7 M, 7,7“ 77...,77, 7 77777777777 “37777” 7 ,7. .7 s .3: is; g? 7.777777. 774.77%?“ 1 .7 77,7 . 77‘, 77 , 7.7 7"”(7'7‘ 777 13:15 .57, 7 $2752 2:: ‘7 a ‘3‘ 7 "7 m. '7‘”? 7/ 7.777715%? 77 is ii 53 a, 7... 77 77 7-. eggs 5’? i‘ v as; :3 7777‘ a“ s : {ti 7...» $5 3i. p ,7 ‘1 ‘1‘. 7 i‘ 77.77, 231% . 57.22777 sagas s ‘ ,2— ;iiz‘ t ‘$\ #5" 77771777.» ,, .. 7 7 9’7»...77"7 . W77 7 “ 7 W7”? 77? ,, 77.77 ‘77: ,, . . ”£737,st ,. 473537 H773“ ,’ '7 .777, 17“" u 7777:7777 7. «777777757777 7 77 7777.77.77 ~77.~7 7.,ti,,q;7,,, , '7 p71,, 57., L 3%" {a 31:57: ear 1 WWW‘T‘ 757,777.. y,, . 377 7,575,. is. .3. 15‘ 3% gr. 7}. 27:7 ,,777 ’c r77 ’77.:7 77.7 7.7. .7 ’ ‘7 7.7.27.7 .i1 77.7: 1% .7 5 £1 ,4 7.77.7777 ”77.177777 7’7'7’7 was ‘L‘3r727‘ '2‘1‘7/7‘7; 777 777.17% 7 ’t‘léi " ”l 35 7.. 7 , 77.7 7:."77777 ”7777 7“»? 2.“?7‘7f‘i71i77 77777777731747.7757?“ 7,, r.. is .7. .., 1‘ 1 ‘11} s ‘7’7 .‘Wipr . . “7'77" 7 7... 7.77.7.7. 73.7.7. 777,. 7777777777 , . . .. . _ 77 , Kw’d . ,y , 77.7. , . 7 . .7. .. 7.771,,7’7172‘7 777,7, . 7. . 7 . 7 7 ~i , 4.7. 1. , . i. 7, 7 7»: 77.7., ‘77' , 77,7 .7... ,7. 7 , , . . ,.. .77777777777777777. 7,7,7, 7 «i. . 7 7 ,7 7 7‘7 777. 77777777 7 7 . .77. 7 ,, iii/’17: "72/2: "7x77“ 7 ‘Z‘r . :77.” .. 7 7, .7 .7777. 77.: , '. W777. ,,. 7 . 7. 7, ,0 _ 7,, 7777.77,.7777777277777777” 7' ,7 . . 7 . . ,_ g n, 7.777777777777777737777 . 7 , , . 77 ,7 WW”, SEE. .7 .77 7.7.7. ,..7., ,77,7. 7.7 7 7 a 7.. 777,727? ‘ 7 .. W77 7 i777:77(774777.777i 777.777.71.777: 7, ..7. 7.7 777 77.777 7.77”“? H’ .7 7. 'r: I?" 7 JJJM'H .i‘. r( 7 77 .r' ’77 ‘ 19774., D ‘ H7177 ’i 77’? 7717‘777‘71177777 (7 W7 . “Ali... “ « ”7,73,; ,7 . . i». 7 7777,77 , 7 7477417777., 777.: > ”“7757, 1‘7; 7777,71, 7 77 7/7 7 MW 57777 MM 1‘ 7 . . 7.7.47.7 . .. . .7 7, 77.7 7777 7,7,77,77,77 , M77777,“ 7 .77‘7'777777777‘77 7 .77/7777777775,,77 7 7 .. 777-7777773777! .777, 7.7777777 :7 7.77:: u; .7 477.117?” . ,7 7 . ,7 ,y 7000 7,, 77;. ‘ ,7}: ”.7’ 7.1,: 7.77 “i‘ 7", 7 . , ,7: , . PROFESSIONAL PAPER 1016 PLATE 2 / 050.00 4/ Compiled by A, H. Chidester, 1960. Based on Plate 1 [SOMETRIC FENCE DIAGRAM OF THE BELVIDERE MOUNTAIN AREA, EDEN AND LOWELL, LAMOILLE LAND ORLEANS COUNTIES, VERMONT ,_ UNITED STATES DEPARTMENT OF THE INTERIOR ' PROFESSIONAL PAPER 1016 PLATE 3 ‘ GEOLOGICAL SURVEY LLJ 3 ;;J LL} Lu ‘ Lax Lu“ 9 ‘ g; c: . ‘~ _ ‘ :3 Cr :1 v , . C7. s : CORRELATION OF MAP UNITS . 3 , ; Osd QUATERNARY I I Unconformity I , 7 7 . 7 ; _ 01,200 N ORDOVICIAN I Lower I Camels Hump Cambrian CAMBRIAN Gm" CAMBRIAN(?) DESCRIPTION OF MAP UNITS I I . Qsd SURFICIAL DEPOSITS (QUATERNARY)—Shown only on structure sections ULTRAMAFIC IGNEOUS ROCKS AND DERIVATIVES1 (ORDOVICIANV ‘ .I I I ULTRAMAFIC ROCKS, UNDIFFERENTIATED—Chiefly dunite (Oud), I I . I massive serpentinite (Ouds), and schistose serpentinite (Ous); may include a few small masses of talc-carbonate rock and steatite (Outc) DUNITEl—Composed essentially of olivine, but all of unit has been partially serpentinized; grades into massive serpentinite (Ouds) MASSIVE SERPENTINITE3—Composed chiefly of bladed serpentine (antigorite and (or) lizardite); chrysotile sparse to moderately abundant in > matrix, locally prominent as cross—fiber veins. Relict grains of olivine persist M- I _ I I but range widely in abundance. The rock retains the textural appearance of TM ' f j 3 dunite I SCHISTOSE SERPENTINITE3—Matrix composed chiefly or entirely of bladed serpentine (antigorite and (or) lizardite); chrysotile confined chiefly to shear surfaces; little or no relict olivine 1 _ _ , TALC-CARBONATE ROCK AND STEATITE3—Talc—carbonate rock is I I/ . I I I composed typically of about 40 percent magnesite and 60 percent talc, and it I is intergradational with steatite, composed essentially of talc. Steatite - commonly has a thin selvage of tremolite; where appreciably less than 0.3 m I I I , thick, a steatite body commonly consists chiefly or entirely of tremolite and 3 minor chlorite i -- ~ , , , , , , - ~ , , ~ , . , , , , ,7 ISO/I30.“ N BELVIDERE MOUNTAIN FORMATIONl (LOWER CAMBRIAN) FINE AMPI-IIBOLITE—Greenish-gray to medium-bluish-gray, distinctly bedded. Individual crystals of hornblende are rarely discernible to the naked eye COARSE AMPHIBOLITE—Dark-greenish-gray or greenish—black, distinctly bedded; contains minor amounts of fine amphibolite. Hornblende crystals are predominantly 5—10 mm long and range to as much as 25 mm '“WTIITTTT' \ ”300,600 N HAZENS NOTCl-I FORMATION (CAMBRIAN (?)) GRAPHITIC QUARTZ—MUSCOVITE-CHLORITE-ALBITE SCHIST—In pl. 1, which includes other carbonaceous units of the Hazens Notch, this unit forms a part of a unit designated as {3th QUARTZ-ALBITE-MUSCOVITE—CHLORITE GNEISS—Greenish—gray 10040001” , I I ,, I , _ - , ~ 3 , ’ , 7 ‘ a 3 3 _ r 3 . I 50 Contact—Solid where exposed, dashed where readily inferred. Arrow shows direction and amount of dip (vertical, inclined) -------- Indefinite or gradational contact 3 -------------------- Limit Of exposure—Within quarry, also serves as boundary between “'00 “00 N .1 3 undifferentiated ultramafic rocks (Oul (ultramafic rocks masked by I "I . 100,200 ‘I quarry rubble) and the several varieties of ultramafic rock 3 Quarry rubble—Shown only on structure sections I Fault or shear zone—Solid where exposed, clashed where readily inferred. 40 75 Barbed arrow shows direction and amount of dip (vertical, inclined); diamond-headed arrow shows bearing and plunge of slickensides 3 3 ill Strike and clip of axial plane of minor fold I i PLANAR FEATURES I > I 3 3 I fl Strike and clip of bedding 1 00 >30 N ,,,,, I I 1 . ‘ ' " ” ‘ ' ”3‘ < , , — UU’IIIIO N Strike and dip of schistosity parallel to bedding i _A5_0 Inclined > I ' i i 3 - i + Vertical ? : ' i s > ,r , Strike and dip of schistosity divergent from bedding, in nonbedded rock, or 23 . _ I where relations are indeterminate I I V ;I J Inclined i > I M I l l l / 1 ; + Vertical I; I; 3 fl Strike and clip of slip cleavage , .. . . . I I I . I 95‘ 0 . N , .. ' ' ,, , ,, ' ~ I" 09"" LINEAR FEATURES Generally combined with one of the above planar symbols for features in 0 100 200 300 400 500 FEET bedded mks llrirliiiii-Ill II II | 1 || 0 100 150 METERS 18—59 ——30 Bearing and plunge of minor fold axis or lineation CONTOUR INTERVAL 10 FEET ARBTTRARV lMiNE) DATUM rs mm run run " 7' 3 42/ I e; E If” Pattern, in plan, of folded or crinkled bedding ABOVE MEAN SEA LEVEL 3 i 3, Strike and dip of layering in ultramafic rocks I ‘5 44—45 Inclined I»: I WWWEWLL 777777 ‘l'l— Vertical I g 3 Diamond-drill hole on map—Cross locates collar, tick locates bottom of hole; I TI ‘ i 13—54 value in degrees gives inclination from the horizontal. _ _ 3 :3 _ 42° Inclined APPROXIMATE MEAN l , l DECLINATION, 1970 3 3 3 +GP—5 Vertical r , i , i Diamond-drill hole on cross sections—Circle indicates the collar, double circle I I GP—3 the piercing point, and tick the bottom of the hole. 3 .3 3 T Within 10 feet of plane of section 9.0 r it .. I I .. . ; """" 7—56 : 9: Projected more than 10 feet onto plane of section I : l - , _ , _ , , , , , y _8—53 I 3 I “awn-we,» , 3 ' > . ' , p _ 3 , , , , , , , ‘ ‘ , I ‘ I _ I. ' I 3 I 14E; Inclined to and intersecting plane Of section 315 db Triangulation point in mine survey 3 II I I l l l\I II Foot of dump or stockpile MAR—”AL... ”new.” III)?! I} l l l l | l I l Top of quarry wall or edge of bench (904/ é Mine drilling grid lines, marked by stakes on the ground V ‘Darker color where exposed, lighter color where covered 2The Ordovician age designation refers to the age of emplacement of the intrusive ultramalic rocks, not to the age of the parent igneous rocks or the metamorphic derivatives (see text subheading discussions under “Tectonic and petrogenic synthesis“) 3Within the ultramafic body, rock covered by surficial deposits or by quarry rubble is shown on the geologic map as 3 3 3 III undifferentiated ultramafic rock (Ou). On structure sections, where drill-hole dab: and extrapolation of surface 3 3 3 g . 3f exposurepermit,identifiablevarieties ofultramaficrock(Oud,0uds,0us)aredistinguished beneathsuchcovered 3 3 areas. . I. 3. ... w? w ,“iI “I Compiled by A. H. Chidester, 1960. Compiled hiefly from 1:600-scale geologic planetable map by A. H. Chidester, W. M. Cady, A. L. Albee, P. L. Weis, and G. M. Koch, July— September 1951; revised in part by Chidester and J. C. Ratté in June 1952, by Chidester and Cady in October 1957, and by Chidester and 0. Fl. Nichols in October 1957 and October 1958. The portion west of 96,000 E. is based chiefly on a 1:3,600-scale planetable map B by Chidester and C. A. Ratté, June— A . FEET C 2—52 ugust 1953 I400 ‘ Conveyor FEET 1* 1300 / iii 0 d A A METERS s D’ 3 3 400 — - , . ~ GP-8 w1300 , , ,3 . . , , I, . .. 3%: 05d .13—54 34—53 ' ' . ‘ I “I , I" i 1200 Conveyor Qsd FEET It . FEET i 3400 # 1200 05d — 1100 73° 72° — 1100 450 ----nr--—-rr'—i"—'“—"—‘ “--~r--—--~--—--—---— Chg ' ’l-l-l / JAY PEAK \ uNorth Troy METERS Thin zones of QUADRANGLE V_ __ 1100 Ch METERS , , , , . , , -3 , , , steatite and L ' 1300 300 9 _ _ , , , , , b' k n 1000 . Ous —1000 300— . , w ‘ -. _ * f- ? ‘ comma at — I F R A N K I. I N l , , XORLEANS _ 1200 w 1000 _ 0Lowell 300 — i 900 900 Area of this report HYDE PARK QUADRANGLE v — 900 _ 1100 _ 300 800 g g Thin zones of Steatite and g blackwall at contacts v 1000 — 800 i 700 L 700 200 __ _ /\§ALEDONIA \ I \ e 900 — 700 H 500 _ 600 \I 200 — ’/ 800 — 500 O 500 3 ’ — 500 u-l Z L 700 —~ 500 ~ 400 * 40“ —’——\ ,7 EM 100 — _ — 500 — 400 - 300 — 300 g9 44° R A N G E Os 0‘51 Ch 100 _ c _ 500 — 300 — 200 — 200 Chg ‘/ _ 400 — 200 — 100 — 100 A \ , 0 10 20 MlLES Chg __ , . ’ " I 3 / 0 10 20 30 KILOMETERS 300 100 0 0 0 I 0 INTERIOR—GEOLOGICAL SURVEV, HESTON, VA —197B—G75268 Ticks left of vertical line in meters. Ticks right of vertical line in feet. GEOLOGIC MAP AND STRUCTURE SECTIONS OF LOWELL QUARRY AND VICINITY, LOWELL, ORLEANS COUNTY, VERMONT UNITED STATES DEPARTMENT OF THE INTERIOR PROFESSIONAL PAPER 1016 GEOLOGICAL SURVEY 7 PLATE 4 \. \l GQ be.” 100 FEET 10 20 METERS CONTOUR INTERVAL 5 FEET AHBlTRAHV [MINE] DATUM Is ABOUT 155 FEET ABOVE MEAN SEA LEVEL CORRELATION OF MAP UNITS 10 FEET ‘ I. g } ORDOVICIAN ZMETEBS ,, , ' . y A I Lower, >CAMBRIAN Cambrian } CAMBRIANl?) DESCRIPTION OF MAP UNITS ULTRAMAFIC IGNEOUS ROCKS AND DERIVATIVES1 (ORDOVICIAN)2 ULTRAMAFIC ROCKS, UNDIFFERENTIATED—Chiefly: dunite (Oud) composed essentially of olivine but generally considerably serpentinized; massive serpentinite (Ouds), which intergrades with dunite, composed chiefly of bladed serpentine (antigorite and (or) lizardite) and small variable proportions of chrysotile; and schistose serpentinite (Ous) DUNITE—Composed essentially of olivine, but all of unit has been partially serpentinized; grades into massive serpentinite (Ouds) MASSIVE SERPENTINITE—Composed chiefly of bladed serpentine (antigo— rite and (or) lizardite); chrysotile sparse to moderately abundant in matrix, locally prominent as cross—fiber veins. Relict grains of olivine persist but range widely in abundance. The rock retains the textural appearance of dunfie SCHISTOSE SERPENTINITE—Matrix composed chiefly or entirely of bladed serpentine (antigorite and (or) lizardite); chrysotile confined almost entirely ‘ ‘. , . . ~ ' . » . . _ a . . , . , , , to shear surfaces; little or no relict olivine. In plate 4 C lines give generalized it . , I ' ' , . . A- I . .. . . 5 . , I ' I , ‘ 7 _ l ' ‘ 3 pattern of schistosity, including zones of intersecting schistosities bounded by I ' ' ' ' ‘ Geology by A. H. Chidester and A. L. Albee, , , shear surfaces September 1951 A , TALC-CARBONATE ROCK AND STEATITE—Talc-carbonate rock is com— posed typically of about 40 percent magnesite and 60 percent talc, and it is intergradational with steatite, composed essentially of talc. Steatite com— C. CENTRAL PART OF THE SOUTHWEST CONTACT monly has a thin selvage of tremolite; where appreciably less than 0.3 m thick, a Steatite body commonly consists chiefly or entirely of tremolite and minor chlorite 100 150 200 FEET _ , ‘ , , 1 _ _ g -_ .‘i .w , , ‘_ ‘ , , , , , » - , , , . . . . _ , Geology find topography by G‘ M. Koch, , , BELVIDERE MOUNTAIN FORMATIONl (LOWER CAMiaRIAN) . . _ _ . i y . ~, , 2 g; ). _ . ' , . ‘ ‘ « I, P. L. Wars, A. L. Albee, W. M. Cody, and g, _ RODINGITE (LIME-SILICATE)—A1teration zone of dropSide—vesuVianite— 1 [I l l y l ' ‘ 4: . T i ‘ f ,, . , If ‘ ', I 3 ‘T _ ‘ Geology and topography by W. M. Cady, .; - I A- H- Chidester, JUIsteptember 1951 epidote—gamet rock adjacent to the ultramafic pluton. In plate 4C lines give 20 30 4O 50 METERS ,~ , , , J “ . ' , > , A. H. Chidester, A. L. Albee, generalized pattern of bedding mggyigyg 'DNATUEQVIQQUEEET FEET . - - ' . 2’9; Wag-Vila [in- :1} tK§§hCthiglevs-tgfptember , _ g FINE AMPHIBOLITE—Greenish-gray to medium—bluish-gray, distinctly bed- ABOVE MEAN SEA LEVEL and J C. Ratté, June 1952 ded. Indivrdual crystals of hornblende rarely discemible to the naked eye. In ‘ plate 43 and C lines show generalized pattern of bedding GRAPHITIC FINE AMPHIBOLITE COARSE AMPHIBOLITE—Dark—greensih-gray or greenish-black, distinctly bedded. Crystals of hornblende are commonly 5—10 mm long and range to as much as 25 mm HAZENS NOTCH FORMATION (CAMBRIAN (?)) RODINGITE (LHVIE—SILICATE)—Alteration zone of diopside—vesuvianite— epidote-gamet rock BLACKWALL CHLORITE ROCK GRAPHITIC QUARTZ-MUSCOVITE—CHLORITE-ALBITE SCHIST GREENISH-GRAY QUARTZ-ALBITE—MUSCOVITE—CHLORITE GNEISS MAFIC INTERBEDS RICH IN CHLORITE AND EPIDOTE were“ 35%“ : , I, g; , :3 , > f it: ,, \ _ g fig, .4 , A??? ~ , ‘ .f . ’ , ‘ ,‘ ~ _ . , ' , g 50 Contact—Solid where exposed, dashed where inferred. Arrow shows " ’ °" R? L A ‘ 50 FEET , .' ‘y ', ‘ , g 3’ 3 y_: ‘ g . direction and amount of dip (vertical, inclined) ”Age $3,, if N ,; ~ A“ , ,,,,g,; , . . “ 'y 1 ~ " 1; I, _- ******* Indefinite or gradational contact I I 10 METERS ‘ - _ . . . _ . . _ CONTWE INATENITYSAIABEUTEEEFEET ‘ _ . . u I 1 " : , ‘ 7 _ Limit of exposure—~Within quarry; also serves as boundary between ARBITHARYABUVEl'aEAfi SEALEVEL _ ‘ , . , , 3 I, , “ ,2 , r, _ r V, undifferentiated ultramafic rocks (Ou) (ultramafic rocks masked by quarry rubble) and the several varieties of ultramafic rock N a; exée is A”: . «52% QE’AWA‘“ Megiaw g; \ W, _ . i _ I ‘ _‘ . ‘ I 50 47 Fault or shear zone—Solid where exposed, dashed where readily inferred. we . W- _, , . ‘9 . Wm ,_ ?“ -i‘ » . » . ». . ' M- :1. . A? . - ,- - -w, , w . . . . . . ,M g g z . .. , .. -. . _, ., , , _ _, - 4 g . , , “ism w, w . , 1,. ,~ . . . - - » er. r. . _ . ‘ g _ . ~ . Barbed arrow shows direction and amount of dip (vertical, inclined); . . , diamond—headed arrow shows bearing and plunge of slrckensrdes wee“ Anticline—Showing trace of axial plane and plunge of axis 335???? 35%,. e» , 3 Syncline—Showing trace of axial plane and plunge of axis Rig . a, we; ., Strike and dip of axial plane of minor fold ”\“v‘gcéégwe M, ‘31:? Strike of vertical axial plane of minor fold and plunge of axis PLANAR FEATURES Strike and dip of schistosity parallel to bedding Inclined : . , ', , ‘ . w , ,0 x _ » W? _ ‘ . , 3 , , ; I, g . °‘)’1°‘ak > We >42: . i Wm“ ' u L i \ \ , , ’ , ‘ _ . vv . a . ,- , 7 Strike and dip of schistosity divergent from bedding, in nonbedded rock, or where relations are indeterminate Strike and dip of slip cleavage LINEAR FEATURES {gr 3 Generally combined with one of the above planar symbols for features in bedded rocks ewmmggye i if» —->30 Bearing and plunge of minor fold axis or lineations Pattern, in plan, of folded or crinkled bedding Strike and dip of joints Inclined —9— Vertical 75 —=I—— Trace of joint—Showing dip 5f] 4’7 Trace of mappable horizon in amphibolite—Arrow shows direction and amount of clip 60 _r_r_ Trace of layering in ultramafic rock—Double ticks show direction and amount of Clip -—r—r-r~r— Top of quarry wall 100 l , I I l ' . , » .‘ , E” '+ , ‘ . m1 ,1 w * . ‘Darker color where exposed, lighter color where covered 7:“ . . 1, A 1 . 20 30 40 M ETERS ‘ T - r' ‘ ' A” .. ' _ ‘ .; _ 2The Ordovician age designation refers to the age of emplacement of the intrusive ultramafic rocks, not to the age of ‘ ' A. ) . 1 ”19”» , , the parent igneous rocks or the metamorphic derivatives (see text subheading discussions under “Tectonic and CONTOUR INTERVAL 5 FEET , A z; ,1 , ii, ,: , pmgemc Synthesis..)_ AHBITRARY lMINE) DATUM IS ABOUT 155 FEET ' '3 ti ‘ » ABOVE MEAN SEA LEVEL Ar.“ ~"A 7" Geology and topography by D. Fl. Nichols and A. H. Chidester, October 1957 and October 1958, modified in part from 1:600-scale planetable map by Chidester, W. M. Cady, A. L. Albee, P. L. Weis, and G. M. Koch, July—September 1951 Geology and topography by D. R. Nichols and A. H. Chidester, October 1958, area south of INTERIOR—GEOLOGICAL SURVEY, HESTON, VA—197SHS75258 CENTRAL PART OF THE NORTHEAST CONTACT 100.750N based'arge'vupontiBOO-Sca'eplanetab'emapbv ChidesterrW-M-Cadvr E. SOUTHWEST SIDE OF THE NORTHWEST PART OF THE LOWELL QUARRY A. L. Albee, P. L. W813 and G. M. Koch, July—September 1951 GEOLOGIC MAPS SHOWING CONTACT RELATIONS AND STRUCTURAL FEATURES OF THE LOWELL QUARRY ULTRAMAFIC BODY, LOWELL, ORLEANS COUNTY, VERMONT