F I ”17” GE 75 3? 3:?“n” “14:2: 15534.55 P0 \I. 9 IS A Stochastic Model for Predicting the Probability Distribution of the * Dissolved-Oxygen Deficit in Streams GEOLOGICAL SURVEY PROFESSIONAL PAPER 913 DOCUMENTS DEPARTMENT JUN 29 7976 LIBRARY 3 H Uri a u / ‘ mvsnsm OF CALIFORNIA // :‘1 w: ’3 C} .75 i; “at? w» MAY 24 1976 1155.1). A Stochastic Model fOr Predicting the Probability Distribution of the Dissolved-Oxygen Deficit in Streams By I. I. ESEN and R. E. RATHBUN j GEOLOGICAL SURVEY PROFESSIONAL PAPER 913 A description of the development and application of a stochastic model for predicting the probability distribution of the dissolved-oxygen deficit at points in a stream downstream from a waste source UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1976 l" ') “mu-y UNITED STATES DEPARTMENT OF THE INTERIOR THOMAS S. KLEPPE, Secretary GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress Cataloging in Publication Data Esen, I. I. A stochastic model for predicting the probability distribution of the dissolved-oxygen deficit in streams. (Geological Survey professional paper ; 913) Bibliography: p. Supt. of Docs. 110.: I 19.16913 1. Water—Dissolved oxygen—Mathematical models. 2. Biochemical oxygen demand—Mathematical models. 3. Random walk-s (Mathematics) 1. Rathbun, R. E., joint author. II. Title: A stochastic model for predicting the probability distribution . . . III. Series: United States. Geological Survey. Professional paper ; 913. TD737.E78 551.4’83 74—31208 For sale by the Superintendent of Documents, US. Government Printing Office Washington, D.C. 20402 Stock Number 024-001-02808-5 CONTENTS Page Symbols _________________________________________ iv Development of the probabilistic model—Continued Metric-English equivalents ________________________ viii Application of the technique and discussion Abstract _________________________________________ 1 of results—.Contmued Introduction 1 Hypothetical example ---------------------- """"""""""""""""""""" Sacramento River data _________--_________ Acknowledgments --------------------------------- 3 Estimation of the accuracy of the Monte Carlo Variations in the deoxygenation and reaeration procedure __________________________________ coefficients _________________________________ 4 Estimation of the variance of the oxygen-deficit Deoxygenation coefficient, K; ___________________ 4 distribution _________________________________ Reaeration coefficient, K: ______________________ 5 Extension of the model ___________________________ Correlation between deoxygenation and reaeration Monte Carlo computations _____________________ coefficients _________________________________ 6 Estimation of the mean and the variance of the Development of the probabilistic model ______________ 7 oxygen-deficit distribution ____________________ Solution of equations for longitudinal profiles Estimation of a stochastic critical time of traveL- of BOD and dissolved-oxygen concentration____ 7 Presentation and discussion of results __________ Random walk models __________________________ 8 Critical time of travel _____________________ Deoxygenation coefficient, K _______________ 8 Mean oxygen deficit _______________________ Reaeration coefficient, K; __________________ 9 Variance of the oxygen deficit ______________ Monte Carlo simulation technique for dissolved- Evaluation of the model ___________________________ oxygen deficit ______________________________ 9 Summary ________________________________________ Application of the technique and discussion of Literature cited ___________________________________ results _____________________________________ 11 Supplemental data ________________________________ ILLUSTRATIONS FIGURES 1—21. Graphs showing: 1. Distribution of the oxygen deficit estimated for the conditions of ‘case 1 _________________ 2. Distribution of the oxygen deficit estimated for the conditions of 'case 2 _________________ 3. Distribution of the oxygen deficit estimated for the conditions of case 3 ________________ 4. Mean oxygen deficit and 10 and 20 percentile limits as a function of time for the oxygen- deficit distributions estimated for the conditions of cases 1, 2, and 3 _________________ 5. Experimental dissolved-oxygen concentrations, mean dissolved-oxygen concentrations pre- dicted by the deterministic model (eq 39), and mean, 10 and 20 percentile limits of the dissolved-oxygen concentrations predicted by the stochastic model (eq 41); Sacra- mento River data _____________________________________________________________ 6. Distribution of the oxygen deficit as estimated from equation 41; Sacramento River data 7. Observed and predicted variance of the oxygen-deficit distributions as a function of time, Sacramento River data _________________________________________________________ 8. ‘P-parameter as a function of the reaeration and deoxygenation coefficients; time of travel of 1 day ______________________________________________________________________ 9. ‘I’-parameter as a function of the reaeration and deoxygenation coefficients; time of travel of 2 days ____________________________________________________________________ 10. W-parameter as a function of the reaeration and deoxygenation coefficients; time of travel of 3 days _______________________________________________________________________ 11. Critical time of travel (deterministic model) as a function of the deoxygenation and reaera- tion coefficients _______________________________________________________________ 12. Critical time of travel (stochastic model) as a function of the deoxygenation and reaeration coeflicients ____________________________________________________________________ 13. Percentage difference in deterministic and stochastic critical times of travel as a function of the ratio of the reaeration and deoxygenation coefficients __________________________ III Page 11 15 17 Page 13 13 14 15 16 17 18 21 27 28 IV 22. 23. 24. TABLE 1. 259°?“ Pass? 10. 11. 12. 13. Symbol a, _____________ B _____________ CONTENTS FIGURES 1—21. Graphs showing—Continued Page 14. Oxygen deficit as a function of the ratio of the reaeration and deoxygenation coefficients; deterministic and stochastic models ______________________________________________ 30 15. Oxygen deficit as a function of the ratio of the reaaration and deoxygenation coefficients; stochastic model and Taylor series approximation ________________________________ 30 16. Variance of the oxygen deficit as a function of the ratio of the reaeration and deoxygena— tion coefiicients; stochastic model and Taylor series approximation with the determi- nistic critical time of travel ____________________________________________________ 31 17. Variance of the oxygen deficit as a function of the ratio of the reaeration and deoxygenation coefficients; stochastic model and Taylor series approximation with the stochastic cri- tical time of travel ____________________________________________________________ 32 18. Variance of the oxygen deficit as a function of the ratio of the reaeration and deoxygenation coefficients; stochastic model with the deterministic and stochastic critical times of travel ________________________________________________________________________ 33 19. Variance of the oxygen deficit as a function of the ratio of the reaeration and deoxygena— tion coefficients; Taylor series approximation with the deterministic and stochastic criti- cal times of travel _____________________________________________________________ 34 20. Distribution of the variance among the terms making up the variance as a function of the ratio of the reaeration and deoxygenation coefficients _____________________________ 35 21. Variance of the oxygen deficit as a function of the deoxygenation and reaeration coefficients; stochastic model with the stochastic critical times of travel ________________________ 36 Flow chart for the computer program for the stochastic model for estimating the variance of the oxygen deficit __________________________________________________________________________ 48 Continuation of the flow chart for the computer program for the stochastic model for estimating the variance of the oxygen deficit ____________________________________________________________ 49 Completion of the flow chart for the computer program for the stochastic model for estimating the variance of the oxygen deficit ____________________________________________________________ 50 TABLES Page Dissolved-oxygen deficit from equation 4 and significant parameters of the stochastic oxygen-deficit distribution _____________________________________________________________________________ 12 Correlation coeificients between BOD and oxygen deficit _________________________________________ 14 Significant parameters of the oxygen-deficit distribution for the Sacramento River data ___________ 16 Variance of the oxygen-deficit distributions determined from the stochastic model and the Taylor series approximation (eq 65) __________________________________________________________________ 22 Critical time of travel for the deterministic and stochastic models and percentage difference ________ 26 Mean oxygen deficit for deterministic and stochastic models ______________________________________ 28 Mean oxygen deficit for stochastic model and Taylor series approximation; with Tc8 _____________ 29 Variance of the oxygen deficit from the stochastic model and the Taylor series approximation _______ 29 Distribution of the variance among the terms making up the variance estimated by the Taylor series approximation of the stochastic model _____________________________________________________ 35 Biochemical-oxygen-demand data for the Ohio River (from Kothandaraman, 1968) and results of the data analysis ___________________________________________________________________________ 43 Mean, variance, and coefficient of variation of the reaeration coefficient data of Churchill, Elmore, and Buckingham (1962) _____________________________________________________________________ 45 Correlation coefficient between the biochemical-oxygen-demand and the dissolved-oxygen concentration (from Moushegian and Krutchkofi', 1969) __________________________________________________ 46 Dissolved-oxygen concentration data for the Sacramento River (from Thayer and Krutchkofi', 1966)-- 47 SYMBOLS Definition Symbol Definition ______ Parameter equal to E—K—Ks, in re— BOD __________-____-Biochemical-oxygen-demand, in milli- ciprocal days. grams per litre. ______ Assumed deterministic part of the dis- solved-oxygen deficit, defined by C ___________________ Dissolved-oxygen concentration, in mil- equation 40, in milligrams per litre. ligrams per litre. CONTENTS V Symbol Definition C. __________________ Rate of addition of dissolved oxygen along the reach by all processes other than reaeration and photosyn- thesis, in milligrams per litre per day. C,,C;_1 ______________ Dissolved-oxygen concentration at the end of the time of travel T; and TH, respectively, in milligrams per litre. Co __________________ Dissolved-oxygen concentration at the upstream end of the reach, in milli- grams per litre. C. __________________ Dissolved-oxygen concentration at saturation, in milligrams per litre. 011(2) _______________ Coefficient of variation of the random variable Z. D __________________ Dissolved-oxygen deficit, or the differ- ence between the dissolved-oxygen concentration at saturation and the actual dissolved-oxygen concentra- tion. D3 __________________ Rate of removal of dissolved oxygen by the benthal layer on the stream bottom, in milligrams per litre per day. Dy. __________________ Dissolved-oxygen deficit for the kth iteration of the Monte Carlo simula- tion procedure, in milligrams per litre. Do __________________ Dissolved-oxygen deficit at the up- stream end of the reach, in milli- grams per litre. Dz __________________ Longitudinal dispersion coeflicient, in square feet per day. E __________________ Error in the fourth central moment of KlT, defined by equation 51. _______________ Expected value of the random variable Z. ________________ Function describing the variation with time I of the BOD at the upstream end of the reach, in milligrams per litre. _______________ Probability density function of the BOD, defined by equation 32. F(§) ................ Function describing the variation with time s“ of the dissolved-oxygen con— centration at the upstream end of the reach, in milligrams per litre. g ___________________ Function of the deoxygenation and re- aeration coefficients that gives the random part of the dissolved-oxygen deficit, defined by equation 54, in milligrams per litre. G1 __________________ Term giving the effect on the mean dissolved-oxygen deficit of random variations in the deoxygenation co- eificient, defined by equation 59, in milligrams per litre. G2 __________________ Term giving the effect on the mean dissolved-oxygen deficit of random variations in the reaeration coeffi- cient, defined by equation 60, in mil- ligrams per litre. E'(Z) f(§) J: fL(‘) Symbol Definition Ga __________________ Term giving the effect on the mean dissolved-oxygen deficit of correla- tion between the deoxygenation and reaeration coefficients, defined by equation 61, in milligrams per litre. G4 __________________ Term giving the effect on the variance of the dissolved-oxygen deficit of random variations in the deoxygena- tion coefficient, defined by equation 66, in (milligrams per litre) squared. G5 __________________ Term giving the effect on the variance of the dissolved-oxygen deficit of random variations in the reaeration coefiicient, defined by equation 67, in (milligrams per litre) squared. Ge __________________ Term giving the effect on the variance of the dissolved-oxygen deficit of correlation between the deoxygena- tion and reaeration coefficients, de- fined by equation 68, in (milligrams per litre) squared. h ___________________ Function of the deoxygenation coeffi- cient at the upstream end of the reach, the deoxygenation coefficient along the reach, the reaeration coefli- cient, and the BOD at the upstream end of the reach that gives the random part of the dissolved-oxygen deficit, defined by equation 72, in milligrams per litre. H __________________ Mean depth of flow, in feet. H1 __________________ Term giving the effect on the mean dissolved-oxygen deficit of random variations in the deoxygenation co- efficient along the reach, defined by equation 76, in milligrams per litre. H2 __________________ Term giving the effect on the mean dissolved-oxygen deficit of random variations in the reaeration coefii- cient, defined by equation 77, in milligrams per litre. H3 __________________ Term giving the effect on the mean dissolved-oxygen deficit of correla- tion between the deoxygenation co- efficient along the reach and the re- aeration coefficient, defined by equa- tion 78, in milligrams per litre. H4 __________________ Term giving the effect on the mean dissolved-oxygen deficit of random variations in the deoxygenation oo- eflicient at the upstream end of the reach, defined by equation 79, in milligrams per litre. H5 __________________ Term giving the effect on the mean dissolved-oxygen deficit of correla- tion between the BOD and the de- oxygenation coefficient at the up- stream end of the reach, defined by equation 80, in milligrams per litre. VI CONTENTS Symbol Definition Ho __________________ Term giving the effect on the variance of the dissolved-oxygen deficit of, random variations in the deoxygena- tion coefficient along the reach, de- fined by equation 84, in (milligrams per litre) squared. H7 __________________ Term giving the effect on the variance of the dissolved-oxygen deficit of random variations in the reaeration coefficient, defined by equation 85, in (milligrams per litre) squared. Hg __________________ Term giving the effect on the variance of the dissolved-oxygen deficit of correlation between the deoxygena- tion coefficient along the reach and the reaeration coefficient, defined by equation 86, in (milligrams per litre) squared. H9 __________________ Term giving the effect on the vari- ance of the dissolved-oxygen deficit of random variations in the BOD at the upstream end of the reach, defined by equation 87, in (milli- grams per litre) squared. H10 _________________ Term giving the effect on the variance of the dissolved-oxygen deficit of random variations in the deoxygena— tion coefficient at the upstream end of the reach, defined by equation 88, in (milligrams per litre) squared. H11 _-_- _______________ Term giving the effect on the variance of the dissolved-oxygen deficit of correlation between the BOD and the deoxygenation coefficient at the upstream end of the reach, defined by equation 89, in (milligrams per litre) squared. i ____________________ Denotes a summation index. I ___________________ Term giving the effect of the addition of BOD along the reach, defined by equation 27, in milligrams per litre. 7' ___________________ Denotes a summation index. k ___________________ Deoxygenation coefficient along the reach or the rate constant for the biochemical oxidation of carbona- ceous material along the reach; as- sumed to be a function of the dis- tance downstream (x) or equiva- lently of time of travel (T), in re- ciprocal days. k4, km, la), kn ________ Deoxygenation coefficient along the reach after 2', i+1, j, and 71. steps, respectively, of the random walk process, in reciprocal days. It ___________________ Mean value of the deoxygenation co- efficient along the reach, in recipro- cal days. Symbol Definition K1 __________________ Deoxygenation coefficient or the. rate constant for the biochemical oxida- tion of carbonaceous material; as- sumed to be a function of the dis- tance downstream (9:), or equiva- lently of the time of travel (T), in reciprocal days. K1, _________________ Initial value of the deoxygenation co- efficient or the deoxygenation coeffi- cient at the upstream end of the reach, in reciprocal days. K1,, Kiwi, K1], K1,, __..Deoxygenation coefficient after i, i+1, j, and n steps, respectively of the random walk process; equivalently, K1,. is the deoxygenation coefficient at the end of time of travel T), in reciprocal days. K1 __________________ Mean value of the deoxygenation co- efficient, in reciprocal days. ______________ Mean value of the deoxygenation co- efficient at any time of travel T, in reciprocal days. K2 __________________ Reaeration coefficient or the rate con- stant for oxygen absorption from the atmosphere, in reciprocal days. K2,, K2, +1, Kai, K2,,-__-Reaeration coefficient after i, i+1, j, and or steps, respectively, of the random walk process; equivalently, K2]. is the reaeration coefficient at the end of time of travel T), in re- ciprocal days. K2 __________________ Mean value of the reaeration coefl'i- cient, in reciprocal days. K1(T) K2( T) ______________ Mean value of the reaeration coeffi- cient at any time of travel T, in re- ciprocal days. K220 ________________ Reaeration coefficient at 20° Celsius, in reciprocal days. K2,, _________________ Reaeration coefficient predicted by equation 15, in reciprocal days. Ks __________________ Rate constant for the removal of BOD by sedimentation and adsorption, in reciprocal days. L __________________ BOD or biochemical-oxygen-demand of carbonaceous material, in milli- grams per litre. L), LH ______________ BOD at the end of times of travel of T, and TH, respectively, in milli- grams per litre. L ___________________ Mean value of the BOD, in milli- grams per litre, defined by equation 33. La __________________ Rate of addition of BOD along the reach, in milligrams per litre per day. Lo __________________ BOD at the upstream end of the reach, in milligrams per litre. . L0,, ______ BOD at the upstream end of the reach at the end of 1, 2, . . . n steps of the random walk process, in milligrams per litre. L01, L02, . CONTENTS Symbol Definition Lo __________________ Mean value of the BOD at the up- stream end of the reach, in milli- grams per litre. m ___________________ Number of times that the Monte Carlo simulation procedure is repeated. n ___________________ Number of steps in the random walk process. 1) ___________________ Rate of production of dissolved oxygen by photosynthesis, in milligrams per litre per day. In or p1(AK1) ________ Probability that K1 takes a positive step in the random walk process. 112 or 102(AK2) ________ Probability that K2 takes a positive step in the random walk process. P[Z=y] _____________ Probability that the outcome of the event Z is y. q; or q; (AKI) ________ Probability that K1 takes a negative step in the random walk process. (12 or q2(AK2) ________ Probability that K2 takes a negative step in the random walk process. rp __________________ Rate of consumption of dissolved oxygen by plant respiration, in mil- ligrams per litre per day. r(Z1, Za) ____________ Correlation coefficient between the random variables Z1 and Zz. r(BOD, C) __________ Correlation coefficient between the BOD and the dissolved-oxygen con- centration. r(BOD, D) __________ Correlation coefficient between the BOD and the dissolved-oxygen deficit. le _________________ Set of n unifomly distributed random numbers used in the Monte Carlo simulation of random variations in the deoxygenation coefficient. R2]. _________________ Set of n uniformly distributed random numbers used in the Monte Carlo simulation of random variations in the reaeration coefficient. t ___________________ Time, in days. . . dz . T __________________ Time of travel or / —; subscripts .7 0 V and j—l indicate specific times of travel, in days. To __________________ Critical time of travel or the time at which the dissolved—oxygen deficit is maximum, in days. Tel) _________________ Critical time of travel for the deter- ministic model (eq 4), in days. Tea _________________ Critical time of travel estimated from the Taylor series approximation of the stochastic model (eq 74), in days. Var (Z) ____________ Variance of the random variable Z. V __________________ Mean velocity of flow, in feet per second or equivalent. at ___________________ Longitudinal position or distance downstream, in feet. X __________________ Amount of BOD removed or the amount of dissolved oxygen con- sumed, in milligrams per litre. VII Symbol Definition X1, XH _____________ Amounts of BOD removed at the end of times of travel T; and TH, re- spectively; in milligrams per litre. Z __________________ Random variable. Zm __________________ Mean value of a random sample of size m. a, ___________________ Drift coefficient for the deoxygenation coefficient, defined by equation 30. a2 ___________________ Drift coefficient for the reaeration co- efficient, defined by equation 37. I31 __________________ Variance of the total deoxygenation coefficient (K1+k), in (reciprocal days) squared. fl'l __________________ Variance of the deoxygenation coeffi— cient at- the upstream end of the reach, in (reciprocal days) squared. l3"1 _________________ Variance of the deoxygenation coeffi- cient along the reach, in (reciprocal days) squared. 132 __________________ Variance of the reaeration coefl‘icient, in (reciprocal days) squared. 3 ___________________ Confidence limit. A ___________________ Parameter of the stochastic model of Thayer and Krutchkofi' (1966) for estimating the distribution of the biochemical-oxygen-demand and the dissolved-oxygen deficit, in milli- grams per litre. AKl _________________ Step length of the random walk for the deoxygenation coefficient, com- puted from equation 31. AK2 _________________ Step length of the random walk for the reaeration coefficient, computed from equation 38. AT _________________ Incremental value of the time of travel, equal to T/n; in days. a ___________________ Expected deviation from the mean value of the random variable Z. n ___________________ Mean value. MU) ________________ Fourth moment of a binomial distribu- tion. m(K1T, N) _________ Fourth central moment of the limit- ing normal distribution of K1T, in reciprocal days to the fourth power. m(K1T, AT) __________ Fourth central moment of the scaled binomial distribution describing ET for finite AT, in reciprocal days to the fourth power. d, dx 5" ___________________ t— —or t—T, in days. a V 77' ___________________ The constant 3.14. U ___________________ Standard deviation. 0'2 ___________________ Variance. 'r, 'r' _________________ Dummy variables of integration. ‘1’ ___________________ Function of the deoxygenation and reaeration coefficients and the time of travel for estimating the variance of the dissolved-oxygen deficit, de- fined by equation 69. VIII CONTENTS METRIC-ENGLISH EQUIVALENTS Metric unit English equivalent Metric unit English equivalent Length Specific combinations—Continued millimetre (mm) = 0.03937 inch (in) litre per second (l/s) : .0353 cubic foot per second metre (m) : 3.28 feet (ft) cubic metre per second kilometre (km) .62 mile (mi) Area 10.76 square feet (ft?) square metre (m2) .386 square mile (mi?) square kilometre (km'—’) hectare (ha) 2.47 acres Volume cubic centimetre (cm3) 0.061 cubic inch (in3) 61.03 cubic inches 35.31 cubic feet (ft3) .00081 acre-foot (acre—ft) 10 7 acre-feet litre (1) cubic metre (m3) cubic metre cubic hectometre (hm3) ll II II II [I II II I! II (1: litre 23113 pints (pt) litre 1.06 quarts (qt) litre gallon (gal) .26 cubic metre .00026 million gallons (Mgal or 106 gal) cubic metre : 6.290 barrels (bbl) (1 bb1=42 gal) Weight gram (g) = 0.035 ounce, avoirdupois (oz avdp) gram 2 .0022 pound, avoirdupols (lb avdp) tonne (t) = 1.1 tons, short (2,000 lb) tonne : .98 ton, long (2,240 lb) Specific combinations kilogram per square centimetre (kg/cm?) : 0.96 atmosphere (atm) kilogram per square centimetre : .98 bar (0.9869 atm) cubic metre per second (ms/s) : 35.3 cubic feet per second (ftS/s) per square kilometre [(ma/s)/km2] : 01.47 cubic feet per second per square mile [(ft“/s)/mi2] metre per day (Ill/(l) : 3.28 feet per day (hydraulic conductivity) (ft/d) foot per mile (ft/mi) .9113 foot per second (ft/s) metre per kilometre (m/ m kilometre per hour (km/h) metre per second (m/s) feet per second metre squared per day (mfi/d) : 10.764 feet squared per day (ft2/d) (transmissivity) cubic metre per second (ma/s) : 22.826 million gallons per day (Mgalld) cubic metre per minute (mein) -2642 gallons per minute (gal/min) litre per second (l/s) _ 15.85 gallons per minute litre per second per metre [(l/s)/m] : 4.83 gallons per minute per foot [( al/min /ft] kilometre per hour g ) (km/h) = metre per second (m/s) 2 gram per cubic .62 mile per hour (mi/h) 2.237 miles per hour centimetre (g/cma) : 62.43 pounds per cubic foot (lb/ft“) gram per square centimetre (g/cmg) : 2.048 pounds per square foot (lb/ft?) gram per square centimetre = .0142 pound per square inch (lb/in”) Temperature degree Celsius (°C) 2 1.8 degrees Fahrenheit (°F) degrees Celsius (temperature) = [ (1.8 x “0) +32] degrees Fahrenheit A STOCHASTIC MODEL FOR PREDICTING THE PROBABILITY DISTRIBUTION OF THE DISSOLVED-OXYGEN DEFICIT IN STREAMS By I. I. ESEN and R. E. RATHBUN ABSTRACT A random walk model was developed for predicting the distribution of the biochemical-oxygen—demand for points downstream from a waste source for a stream system in which the deoxygenation coefficient is a normally distributed random variable. A Monte Carlo technique for simulating a random walk process was used for estimating the distribu- tion of the dissolved-oxygen deficit at downstream points in a stream in which both the deoxygenation and reaeration co- eflicients are normally distributed random variables. Equa- tions for approximating the mean oxygen deficit and the vari- ance of the oxygen deficit were developed by expanding the basic equation of the stochastic model in a Taylor series. The random walk model gave a lognormal distribution function for the biochemical-oxygen-demand. The frequency distributions of the oxygen deficit predicted by the stochastic model became flatter and skewed to the right as time of travel increased. The critical time of travel estimated from the stochastic model was always larger than the critical time of travel computed from the deterministic model; however, the percentage difference decreased as the ratio of the reaeration and deoxygenation coefficients decreased. The variance of the oxygen deficit at the critical time of travel was largest for small ratios of the reaeration and de- oxygenation coefficients and smallest for the large ratios. The variance showed the greatest dependence on the ratio at large values of the ratio and the smallest dependence at small values of the ratio. The variances of the oxygen deficit computed from the Taylor series approximation of the stochastic model were comparable to the variances obtained from the stochastic model for small times of travel; as the time of travel in- creased, the Taylor series approximation underestimated the variance. For computations at the critical time of travel, the variance estimated from the Taylor series approximation was less than that of the variance of the stochastic model over the entire range of conditions considered. The ability to predict the variances of the biochemical- oxygen-demand and the dissolved-oxygen deficit at points downstream from a waste source is extremely important, in view of the ever-increasing concern with the maintaining of water-quality standards. The stochastic model of this report is a valuable tool for predicting variances; however, further de— velopment and refinements of this and other models is still needed. INTRODUCTION The pollution of our streams and rivers caused by the discharge of excessive amounts of municipal and industrial wastes into them has been of increased concern in recent years. As a result of this concern, the Water Quality Act of 1965 was enacted. This law required all States to classify rivers and streams ac- cording to intended use and to adopt water-quality standards for each of the intended uses. Specific lim- its were required for 10 water-quality parameters for each of 9 designated water uses; these standards, as of 1969, for the different States were tabulated by the American Public Health Association (1969). Of the various water-quality parameters, the single parameter used most frequently to indicate the rela- tive state of pollution or health of a stream is the dissolved-oxygen concentration. The dissolved-oxy- gen concentration is the amount of free oxygen, that is, not chemically combined with other elements, available in the water for the respiration processes of the flora and fauna of the stream and for the oxi- dation of organic waste materials. Hence, knowledge of the response of a stream’s dissolved-oxygen con- centration to the addition of organic wastes is essen- tial to maintaining dissolved-oxygen concentrations that are adequate to support a desirable flora and fauna. The longitudinal profiles of the concentration of organic wastes, expressed in terms of BOD (bio- chemical-oxygen-demand), and of the dissolved-oxy- gen concentration, may be predicted from determin- istic models based on the principle of the conserva- tion of mass. The differential equations are 62L £+V§£= ,, —(K1+K3)L+L., (1) at ax arr" 2 pg 93? PC+K2(C.—C) at ax ax? ~K1L+C,,—DB+p—r,, (2) 1 2 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS where L = biochemical-oxygen-demand of carbona- ceous material; t =time; a; =distance in the longitudinal or direction of flow; V =mean flow velocity in the longitudinal direction; K, = deoxygenation coefficient or the rate con- stant for biochemical oxidation of carbonaceous material; K2 =reaeration coefficient or the rate con- stant for oxygen absorption from the atmosphere; K3 =rate constant for the removal of BOD by sedimentation and adsorption; , =longitudinal dispersion coefl‘icient; = dissolved-oxygen concentration; , = dissolved-oxygen concentration at satur- ation; a = rate of addition of dissolved oxygen along the reach by all processes other than reaeration and photosynthesis; =rate of addition of BOD along the reach; DB =rate of removal of dissolved oxygen by the benthal layer on the stream bottom; 17 =rate of production of dissolved oxygen by photosynthesis; and In, =rate of consumption of dissolved oxygen by plant respiration. Q QQU P The development of equations 1 and 2 assumes that: 1. The dissolved-oxygen concentration and BOD are uniformly distributed over each cross sec- tion so that the equations can be written in the one-dimensional form. 2. The processes described by the rate constants K1, K2, and K3 are first-order processes; that is, the rate of removal of BOD is proportional to the amount of BOD remaining, and the rate of reaeration is proportional to D, the dis- solved-oxygen deficit, which is the difference between the dissolved-oxygen concentration at saturation and the actual dissolved-oxygen concentration. 3. Only the carbonaceous demand of the waste is significant. If the nitrogenous demand is im- portant, an additional term must be added to equations 1 and 2. If D,, K3, La, Ca, and DB have negligible effects on the BOD and dissolved-oxygen profiles, and if steady-state and uniform-flow conditions exist, then equations 1 and 2 reduce to the classical equations of Streeter and Phelps (1925) . The solutions under these conditions are L=Lo exp (—K1T) (3) and KL, D—C’.,—C—K2_K1 [exp ( KlT) -exp (—K2T)]+Do exp (—K2T) (4) where Lo =BOD of the carbonaceous material at the upstream end of the reach; Do =dissolved-oxygen deficit at the upstream end of the reach; and T =time of travel from the upstream end of the reach to the point of interest at longitudinal position x, or ' dot ac T=o 7—7 (5) The various other modifications of equations 1 and 2 and the various types of analytical and nu- merical solutions of these equations that have ap- peared in the literature were reviewed by Bennett and Rathbun (1972). The deterministic equations for the longitudinal distributions of the BOD and dissolved-oxygen con- centration, that is, equations 1 and 2 or similar equations with constant coefficients, give one value for the dissolved-oxygen concentration at each downstream point for a specific set of conditions. Similarly, most water-quality standards state that the dissolved-oxygen concentration must not drop below one specific concentration. However, because of the presence of random components in natural processes, there is a nonzero probability that the dissolved-oxygen concentration will fall below the concentration predicted by the deterministic equa- tions, and a dissolved-oxygen concentration that is on the average sufl‘icient to assure healthy fish does not prevent fish kills. Hence, an interest has de- veloped in determining the probability distributions of the BOD and dissolved-oxygen concentration at downstream points in the reach. Previous studies have described three methods for the estimation of the probability distributions of the BOD and dissolved-oxygen concentration. The first of these methods, developed by Loucks and Lynn (1966), predicts the probability that a dissolved-oxygen concentration less than some spe- cific concentration will exist for some specified time period at a. point downstream from a waste dis- charge. A Markov model was used and a first-order transition probability matrix for the low-flow sea- INTRODUCTION 3 son was formed by making use of the available streamflow data. The term in the jth. row and kth column of the transition matrix of a first-order, discrete-time, discrete-outcome Markov process gives the probability that a process which is in state 7' at time t will be in state k at time t+ 1. To each discrete average daily streamflow, one set of pa- rameters such as K1, K2, K3, temperature, Lo, and Do were assigned. Then, the critical or maximum dissolved-oxygen value for each set of parameters was assigned a probability in the transition matrix. Loucks and Lynn (1966) also give methods for de- termining the probabilities of violating a particular stream standard for two or more consecutive days and of violating a standard when sewage flow de- pends on streamflow. The second of these methods considers the prob- lem essentially as a stochastic birth and death proc- ess. Thayer and Krutchkofl" (1966) assumed that the BOD and dissolved-oxygen concentration changed by a small amount A in a short time interval owing to additions of BOD and dissolved oxygen along the river, benthal demand, sedimentation of solids to the river bottom, deoxygenation, and reaeration. In their model, a change of size A in the concentration was considered to constitute a change of one state, and it was assumed that a change of more than one state in time AT had a probability of 0(AT). The probability of a change of one state was assumed to be proportional to AT. Thayer and Krutchkofl" (1966) determined the mean, variance, and probability distribution of BOD and the dissolved-oxygen deficit for a number of different stream and waste conditions and showed that the predicted mean value of the BOD and the oxygen deficit is the same as those computed by deterministic equations for the same stream and waste conditions. The variances of the BOD and the deficit, at any time of travel T, were found to be linear functions of A, and a numerical value for A could be determined by experimentally determining the variances at some point downstream from the point of addition of the waste to the stream. The third of these methods is the Monte Carlo method proposed by Kothandaraman (1968). He considered the deoxygenation coefl‘icient, K1, and the reaeration coefficient, K2, to be random variables and showed them to be normally distributed. Then, he randomly selected values of K1 and K2 from normal distributions with known means and variances and computed the oxygen deficit from the Streeter- Phelps equations for a preselected number of flow times. In this manner, he was able to estimate the probability distribution of the oxygen deficit at downstream points. Thus, the problem of predicting the response of the dissolved-oxygen concentration of a stream to an organic waste load can be approached in either of two ways: deterministic or probabilistic. In the de- terministic approach, the dissolved-oxygen concen- tration is predicted by solving two coupled differen- tial equations (equations 1 and 2 or modifications of these) with appropriate assumptions and boundary conditions. On the other hand, the random nature of those factors such as turbulence, mean velocity, depth of flow, and type and concentration of waste that determine the coefficients of these equations and random variations in the type and concentration of the input waste loads suggest that a probabilistic approach should be more appropriate to the prob- lem. However, the probabilistic models developed thus far are either too restrictive in the assump- tions necessary or require extensive field data at some point upstream to predict the probability dis- tributions of the BOD and oxygen deficit at down- stream points. The purpose of this report is to: 1. Present a discussion of the random variations to be expected in the deoxygenation coefiicient, K1, and the reaeration coefficient, K2, and pos- sible correlation between these parameters. 2. Describe the development of a random walk model for estimating the probability distribu- tion of the BOD when K1 is a random variable. 3. Describe the use of a Monte Carlo technique for estimating the probability distribution of the dissolved-oxygen deficit when K1 and K2 are random variables. 4. Describe the application of these techniques to the estimation of the variance and the per- centile limits of the dissolved-oxygen deficit at downstream points. 5. Describe the extension of the model so that the input BOD could be considered as a random variable, with particular emphasis on the critical time of travel when the maximum oxy- gen deficit occurs. ACKNOWLEDGMENTS This report is a modification of the Ph. D. disser- tation presented to Colorado State University by the senior author (Esen, 1971). The major profes- sor was Dr. E. V. Richardson, and committee mem- bers were Dr. J. Gessler, Dr. J. P. Bennett, and Dr. R. B. Kelman. This study was a part of the U.S. Geological Sur- vey research program on reaeration in open-channel flow. VARIATIONS IN THE DEOXYGENATION AND REAERATION COEFFICIENTS The processes of deoxygenation and reaeration are generally considered to be first order; that is, the rate of deoxygenation is directly proportional to the amount of BOD remaining to be oxidized, and the rate of reaeration is directly proportional to the dissolved-oxygen deficit remaining to be satisfied. Therefore, deoxygenation may be described by the equation dL __= _ K, d t L (6) and reaeration by the equation dD ———= —K D d t 2 (7) The deoxygenation coefficient depends on the type of waste and the reaeration coeflicient on the hydraulic conditions in the channel. Because variations in the type of waste and in the hydraulic conditions are in general random, there is ample basis for consider- ing K1 and K2 as random variables. DEOXYGENATION COEFFICIENT, K1 Possible explanations for variations in the deoxy- genation coefl‘icient, K1 were discussed in detail by Kothandaraman (1968). He pointed out that be- cause the characteristics of municipal wastes vary considerably with time, the rate parameter, K1, also varies considerably. This parameter characterizes biological processes which depend on the response of living organisms to their environment and hence, these processes do not have the uniformity of a chemical reaction. He concluded that because most of the contributing factors were random in nature, the variations in K1 could also be considered random and hence treated in a probabilistic manner. Kothandaraman (1968) determined K1 values and ultimate or total first-stage BOD values of carbona- ceous material for the Ohio River data collected and published by the U.S. Public Health Service (1960). He applied several statistical tests to 83 average values of K1 which were determined by the least- square procedure of Reed and Theriault (1931), and he accepted at the 5 percent significance level the hypothesis that the K1 values were normally dis- tributed with a mean of 0.173 days—1 and a variance of 0.0044 days-‘2. MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS We analyzed the Ohio River data to determine the variation of K1 with distance downstream or travel time through the reach. The BOD versus time data given by Kothandaraman (1968) were used in this analysis, and these data are presented in table 10 in “Supplemental Data.” Also given in table 10 are the mean values of the deoxygenation coefl‘icient (K1) and the total first-stage or carbonaceous BOD (Lo) determined by Kothandaraman (1968) . If equation 6 is integrated and the result rear- ranged, then it can be shown that X=Lo[1-eXp (—K1T)] (8) whereX =Lo—L, the amount of BOD removed in travel time T (also the amount of dis- solved oxygen consumed up to time T) ; Lo =total first stage or carbonaceous BOD; and L =BOD at time T. It follows that K1= —-%ln[1—X/Lo] (9) The value of K1 for each increment of travel time was computed from 1 x-—ac-_ K..= ____z 1_#_‘] 10 J Tj—Tj_1 n[ Lj—l ( ) where Lj=Lj_1 BXD [_K1j(Tj—Tj—1)] (11) and j = 1, 2, 3, 4, and 5. The index 1' indexes the value of the variable at the end of the travel time T,-, and L0 was determined by the Reed and Theriault (1931) procedure. Values of K1 obtained from equa- tion 10 for the Ohio River data are given in table 10 in “Supplemental Data.” Travel times larger than 5 days were not considered because biological proc- esses become less predictable at large times. The variances of the K1 values along the reach were com- puted and these variances are presented also in table 10. Inspection of the deoxygenation coefficient values given in table 10 shows that K1 in general decreases with increase in T, or equivalently, distance down- stream. A possible explanation for this is that the more easily degraded material is oxidized first in the stream. No attempt was made in the present study to develop a relation between K1 and T; how- ever, the theoretical developments to be presented were generalized to take this variation into consideration. VARIATIONS IN THE DEOXYGENATION AND REAERATION COEFFICIENTS 5 It was also observed that the variance of the initial values of K1 at T=0 was considerably larger than the variance of K1 along the reach for a spe- cific initial value of K11. A Chi-square test of the initial values of K1 showed that, at the 5 percent significance level, the data were compatible with the assumption that they were normally distributed with a mean of 0.205 days”, and a variance of 0.00538 days—2. Two extreme values of K1, com- puted from observations numbered 37 and 80, were neglected in the analysis. The coefficient of varia- tion of the initial K1 values was 0.357. The average variance of the K1 values along the reach was 0.0015 days—2. It was assumed that the K1 values along the reach were normally distributed with the above mentioned variance, but this hypothesis could not be tested, because for T<5 days, each series of ob- servations contained only four data points. The coeflicient of variation, C”, of the mean values of the deoxygenation coeffiicent, K1, for the Ohio River data was 0.066/0.173 or 0.38. In the present study, no attempt was made to substantiate the pres- ence of the same K1 value in all streams; however, as a first approximation, the coefficient of variation of K1 is assumed to be 0.,(K1) =0.35 (12) Thus, the assumption is that the dispersion or spread of the K1 values about the mean is essentially the same as in the Ohio River data. Although the data in table 10 indicate that there is a difference be- tween the variance of the initial values of K1 and the variance of the K1 values along the river reach, a coefficient of variation of 0.35 will in general be used to estimate both of these variances. The differences obtained in the probability distribution of the dis- solved-oxygen deficit by considering different values of C1,(K1) for the initial values of K1 and for the values of K1 along the reach, and by assuming the same 0,, (K1) for both variances, are discussed in the examples presented later. The probabilistic model to be described in the next section is capable of considering both variations in the initial value of K1 and variations in K1 along the river reach. Another possibility is that the initial K1 values are deterministic, but the variance of K1 is very large during the first few hours. No data were available to test this supposition. REAERATION COEFFICIENT, K2 In general, the reaeration coefficient, K2, of a stream may be considered as a property of the flow in the channel. As such, the variation with time of K2 may be defined as the sum of a constant term, a periodic component, and a random component (Ma- talas, 1971). The constant term may under some cir- cumstances itself be a function of time or distance downstream. The periodic component results from seasonal changes or perhaps diurnal changes, for example, in the quantity of flow. The random com- ponent has no simple physical explanation but is the resultant of a very large number of physical causes; for practical purposes, this random component may be considered as the inherent characteristic of tur- bulent flow (Matalas, 1971). Because the intensity of turbulence at a point in a stream has an approxi- mately normal distribution (Batchelor, 1959), there is in turn a basis for considering K2 as a normally- distributed random variable. The K, data of Churchill, Elmore, and Bucking- ham (1962) for streams of the Tennessee Valley are generally considered to be the best available data for natural streams. The reaeration coefficient was com- puted from K,j=__1_.-m[ 08—0] ] (13) Tj_Ti-1 Cs_Cj——1 =saturation concentration of dis— solved oxygen; 0 =dissolved-oxygen concentration; 3', j—I =value of variable at the end of the time of travel Ti, TF1; and 0,—00 =dissolved-oxygen deficit, where C, is the dissolved-oxygen concentra- tion at the upstream end of the reach, assumed to be known. where 0, Several equations were developed by multiple re- gression analysis, and each of these equations was tested by comparing computed K2 values with the geometric means of the experimental K2 values for each reach studied. The equation recommended for use by Churchill, Elmore, and Buckingham (1962) was K220 = 5.026V0.969H—1.673 (14) =mean velocity of flow in feet per second; H =mean depth of flow in feet; and K =reaeration coefficient in days—1 at 20°C. For V in metres per second and H in metres, the constant term in equation 14 is 2.178. Equation 14 had a correlation coefl‘icient of 0.822. Kothandaraman (1968) conducted a regression analysis of the Churchill, Elmore, and Buckingham where V 220 6 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS (1962) data using arithmetic mean values for each ‘ reach and obtained l K220=5.827V°-924H-1-7°5 (15) 1 with a correlation coefficient of 0.917 (for V in ‘ metres per second and H in metres, the constant i term in equation 15 is 2.304). He then analyzed the distribution of the percentage errors, with the per- centage error defined as (Kg—K2,) (100) /K2,, where K2 is the arithmetic mean value, and K2,, is the value predicted by equation 15. He found that the per- centage errors were approximately normally dis- tributed with a mean of zero and a standard devia- tion of 0.368. His analysis assumes that the value of K220 given by equation 15 is correct and that devia- tions from it are probabilistic in nature; it also as- sumes that once a process starts with a certain value of K2, the rate of reaeration is constant for the reach. In the present study, we assumed that the devia— tions of measured values of K2 for a specific reach of stream from the mean of all the measured K2 values for that reach were normally distributed. This as- sumption is in contrast with that of Kothandaraman (1968) in that he assumed that deviations of the means of the experimental K2 values from K2 values computed from equation 15 were normally dis- tributed. The arithmetic means, variances, and the coefficients of variation estimated from the K2 data of Churchill, Elmore, and Buckingham (1962) are presented in table 11 in “Supplemental Data.” The mean value of the coefficients of variation for all streams studied was 0.307. Therefore, for our study, we have assumed that the coefficient of variation of the reaeration coeflicient is 0.3, or 01) (K2) = 0.3 (16) CORRELATION BETWEEN DEOXYGENATION AND REAERATION COEFFICIENTS Increased turbulence and mixing in a stream re- sult in a more rapid rate of reaeration and hence a larger K2 and also a larger K1 as a result of in- creased bacterial degradation of wastes; conversely, lower levels of turbulence result in smaller K1 and K2 values. Similarly, an increase in temperature in- creases both K1 and K2. Therefore, a positive corre- lation between K1 and K2 is expected. Although the interdependence between BOD and dissolved-oxygen deficit, D, is known, little effort has been directed toward determining the numerical value of the correlation coefficient for BOD and D. There has, however, been one theoretical approach to the problem by Moushegian and Krutchkoff (1969) who used the stochastic model of Thayer and Krutchkoff (1966) to estimate the correlation coeffi- cient. The correlation coefficient was defined as Correlation coefficient _ covariance [(BOD variance) (D variance)]1/2 The correlation study was run with an initial BOD input (Lo) of 12.4 mg/l (milligrams per litre), a saturation concentration (0,) of 10.4 mg/l, an ini— tial dissolved-oxygen concentration (00) of 5 mg/l, and a value of 0.1 mg/l for the A-parameter of Thayer and Krutchkof‘f (1966). Different values of K1, K2, and T were used, and the resultant correla- tion coefficients between BOD and dissolved-oxygen concentration obtained by Moushegian and Krutch- koff (1969) are presented in table 12 in “Supple- mental Data.” The correlation coefficient is positive at all times which implies that the correlation coeffi- cient between BOD and D is negative. Moushegian and Krutchkoff (1969) found that in- terpretation of the results of the correlation coeffi- cient study was difficult, but in general it was ob- served that the correlation coefficient between BOD and dissolved-oxygen concentration decreased as time of travel increased and the difference between K2 and K1 increased. On the other hand, as the time of travel becomes small, correlation between BOD and D approaches —1. This may be seen as follows: for an initial dissolved-oxygen deficit of zero and small travel times, the exponentials in equations 3 and 4 may be approximated by their Taylor series expansions. Hence L~LO (1 —K1AT) (17) (18) and D~L0K1AT (19) where AT=an incremental value of the time of travel. The Moushegian and Krutchkoff (1969) study yielded no information on the possible correlation between K1 and K2. Therefore, in the present study, the correlation coefficient between ,K¥-and K4 was assumed to have the value 'r(K1, K2) =0.5 (20) This correlation coefficient was applied to the K1 and K2 values at the same instant of time, although the presence of a small lag period is possible. The effect of different correlation coefficient values on the oxygen-deficit distribution will be discussed in the examples to be presented. DEVELOPMENT OF THE PROBABILISTIC MODEL 7 DEVELOPMENT OF THE PROBABILISTIC MODEL SOLUTION OF EQUATIONS FOR LONGITUDINAL PROFILES OF BOD AND DISSOLVED- OXYGEN CONCENTRATION The longitudinal profiles of the BOD and dis- solved-oxygen concentration downstream from a source of organic biodegradable waste are described by equations 1 and 2, respectively. The assumptions inherent in these equations were discussed previous- ly. With the additional assumptions of (1) the ef- fect of longitudinal dispersion on the BOD and dis- solved-oxygen concentration profiles is negligible relative to other factors; and (2) the effect of photo- synthesis and respiration is negligible; equations 1 and 2 reduce, respectively, to L L 6—+ a—=—(K,+K3)L+L, (21) at 896 and ?—C—+V£=K20,—K1L+Ca—DB—K20. (22) at 896 These equations, as well as equations 1 and 2, as- sume that the water temperature, and hence the saturation concentration of dissolved oxygen, is constant. The effect of longitudinal dispersion on the BOD and dissolved-oxygen concentration profiles for steady-state and uniform flow conditions was dis- cussed by Dobbins (1964) who concluded that the effect was negligible for the largest value of D, known at that time. A much larger value was found later by Yotsukura, Fischer, and Sayre (1970), but the mean flow velocity was also large; and according to Dobbins’ analysis, the effect of dispersion would still be negligible. In estuaries where dispersion be- comes large and velocities small, the effect of longi- tudinal dispersion cannot be neglected. The effect of photosynthesis may be extremely important in cer- tain situations; the literature and procedures for treating photosynthesis were discussed by Bennett and Rathbun (1972). The water temperature may increase in the downstream direction as a result of both natural and manmade causes; this problem was, considered by Liebman and Lynn (1966). Thus, longitudinal dispersion, photosynthesis, and changes in water temperature may be important in certain situations; for the present study, however, they were neglected. Following Li (1962), equations 21 and 22 were solved for a stream system with the following characteristics : 1. Hydraulic conditions may vary with distance downstream but are steady at each cross section. 2. The BOD and dissolved-oxygen concentration at the cross section at which the waste is added to the stream may be functions of time; in- herent in the assumption of one-dimensionali- ty is the requirement that the distance neces- sary for complete lateral and vertical mixing of the BOD be small relative to the distance downstream to the cross section(s) of interest. 3. The rate coefficients K1, K2, and K3 are functions of the time of travel or distance downstream. 4. The distributed source and sink terms, that is, 0,, La, and D B, vary with distance downstream but are steady at each cross section. Time of travel, T, and distance downstream, 90, may be interchanged through the relation 3” dx T= [0 7 Details of the solution have been presented previous- ly (Esen, 1971), and only the results will be pre- sented here. For the longitudinal profile of the BOD, the equa- tion obtained was (23) 1. + / (K1+K3)d'r' U rm + foTLae 1' —- / (Kl+K3)dT 0 L=e d1- (24) where z is given by ” dx €=‘t—fo 7—; =function describing the variation with time g of the BOD at the upstream end of the reach; and 1-, 1" = dummy variables of integration. (25) f(€) Comparison of equation 24 with equation 3, which describes the BOD profile for the approach of Streeter and Phelps (1925) , shows the differences in the results. If the BOD at the upstream end of the reach is independent of time, K3 and L, are zero, and K1 is independent of time of travel in the reach, then equation 24 reduces to equation 3. 8 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS For the longitudinal profile of the dissolved-oxy- gen deficit, the equation obtained was /T sz'r’ —/;(2dr 0 T D=C,—C=e- [Cs—FQH/o (DB—0m dr [1(K2—K1—Ka)d'r' [7(K2—K1—K3)d7" T o T 0 mo] Kle d.+f K116 d7 (26) where [T (K1+K3)dT' T I=j; Lae d-r; and (27) F (C) =function describing the variation with time g of the dissolved-oxygen concentration at the upstream end of the reach. In equation 26, the term C,—F (g) is the dissolved- oxygen deficit at the upstream end of the reach; the first integral term is the net effect of the addition of dissolved oxygen along the reach by all processes other than reaeration and photosynthesis and the removal of dissolved oxygen by the benthal layer on the stream bottom; the second integral term is the BOD added at the upstream end of the reach; and the third integral term is the BOD added along the reach. Comparison of equation 26 with equation 4, which describes the longitudinal profile of the dissolved- oxygen deficit for the approach of Streeter and Phelps (1925), shows the differences in the results. If the dissolved-oxygen concentration at the up- stream end of the reach is independent of time, La, DB, Cu, and K3 are zero, and K1 and K2 are inde- pendent of time of travel in the reach, then equation 26 reduces to equation 4. Equations 24 and 26 describe the longitudinal pro- files of the BOD and dissolved-oxygen deficit, respec- tively, for a stream system in which the rate coeffi- cients K1, K2, and K3 are unknown functions of the time of travel. Application of these equations to the estimation of the probability distributions of the BOD and the dissolved-oxygen deficit for a stream system in which K1 and K2 are normally distributed random variables is described in the following sections. RANDOM WALK MODELS There is considerable justification, as discussed previously, for considering the deoxygenation coeffi- cient, K1, and the reaeration coefficient, K2, as ran- dom variables. Therefore, it is possible to consider T T that the values of the integrals f0 Kldr and [0 sz-r are attained as a result of a simple random walk. DEOXYGENATION COEFFICIENT, K1 Assume that in a time interval of AT the quantity KIAT can take only two possible values, K1 (T) AT +/_\.K1 and K1(T) AT—AKl. To these two values the following probabilities are assigned: P[K.(T>AT=IT0 AK1 T (30) lim (AK1)2_ 5.3% AT “BIT ‘31) where 011 and ,81 are finite. Then the probability dis- tribution of K1 is found to be no_rmal (Bailey, 1964; Feller, 1968) with a mean of Kl—alfll and a vari- ance of [31- For the limiting variance to remain finite, (AK1)2/AT must be of the order of unity, and for the mean to remain finite, p,—q1 must be of the order of AKl. The equations of this section satisfy these conditions. These concepts were applied to equation 24 to esti- mate the probability distribution of the BOD for a stream system in which K1 is a random variable. In this analysis, the following assumptions were made: 1. The BOD added along the reach is much smaller than the BOD added at the upstream end of the reach, that is K Ladr<«f(§); —[)T(K1+Ks)d1 and hence the variance of the term I e is negligible, where I is defined by equation 27; and 2. The rate constant for the removal of BOD by sedimentation and adsorption, K3, is deter- ministic throughout the river reach under con- sideration, that is, independent of flow time. These assumptions are valid for many situations observed in natural streams; in fact, the effects of La and K3 were completely neglected by several in- DEVELOPMENT OF THE PROBABILISTIC MODEL 9 vestigators (Streeter and Phelps, 1925; Kothan- daraman, 1968). Considering K3 as a random vari- able does not introduce any complications into the analysis if K3 is considered as normally distributed with known mean and variance. For appreciably large values of La, the probability distribution of the BOD cannot be found analytically, but Monte Carlo methods can be efficiently used. Details of the development of the probability dis- tribution of the BOD have been presented previously (Esen, 1971). It was found that the BOD was dis- tributed according to the lognormal distribution, or 1 1 fL(') —[\—/27r_—_Tzfil][[l‘_l exp (*foT (K+K3)d1)]] {ln[L—I exp (—/0T (1?.+K3)df)] ‘fo (K+Ks)dr+Ta1181—l" NW exp _ 2sz for T _ L>I exp (—f0 (K1+K3)dr) and fL(-)=0forLéIexp (—/0T (image) (32) The mean value of L for any time of travel T was found by integrating the product of L and the den- sity function between the limits — 00 and + 00. The result was L=exp[ln f(§) -—foT (K1+K3)d1+TlX1,31+ 1122181] +1 exp of: (Z+K.)dr) (33) Similarly the variance of L at any time of travel T was found by integrating the product of (L —L) 2 and the density function between the limits — 00 and + co , and the result was var (L)=exp[2[ln f(c) —f0T (K+K3)d1+Ta1/31] +21%] —exp[2[ln f(§) - [OT (K.+K3)dr + T111181] + T231] (34) The probability density function of L given by equation 32 reduces to that of Kothandaraman (1970), if K3 and L, are 0 and K1 is considered to be independent of the time of travel. However, one of the advantages of the random walk model from which equation 32 was developed is that it can con- sider K1 as well as the mean and variance of K1 as a function of time of travel. Thus in the most general case, ,8, and 011 are functions of the time of travel. If the variance of K1 is considered as constant through- out the river reach of interest, ,81 is constant, and if it is further considered that the mean value of K1 is also constant, then 021 can be taken as zero. The mean and variance of K1 are measurable quantities, and numerical values can be assigned to K1, a1, and ,81. REAERATION COEFFICIENT, K2 Proceeding exactly analogously as for the deoxy- genation coefficient, we assume that in a time inter- val of AT the quantity KZAT can take only two possi- ble values, K2(T)AT+AK2 and K2(T)AT—AK2. To these two values, the following probabilities are assigned: P[K2(T)AT=K2(T)AT+AK2]=p2(AK2) (35) P[K2(T)AT=K2(T)AT—AK2]=QZ(AK2) (36) where K2(T) is the mean value of K2 at time of travel T and p2(AK2)+q2(AK2)=1. Let us further assume that lim [q2(AK2) ‘p2(AK2)] =2 (37) AK2-90 AKZ T lim (151(2): rise. TT ‘3” (38’ where 012 and 182 are finite. Then the probability dis- tribution of K2 is found to be normal (Bailey, 1964; Feller, 1968) with a mean of Kz—azfiz and a vari- ance of ,82. These equations also satisfy the condi- tions previously given for the limiting variance and the mean to remain finite. Because of mathematical complexities, it was not possible to apply this random walk model for the re- aeration coefiicient to equation 26 and obtain ana- lytically the distribution function of the dissolved- oxygen deficit. Therefore, the distribution of the oxygen deficit was estimated by using a Monte Carlo simulation technique. MONTE CARLO SIMULATION TECHNIQUE FOR DISSOLVED-OXYGEN DEFICIT The Monte Carlo simulation technique is a method by which a complex system with random components is numerically operated by random numbers chosen in such a manner that they simulate the physical behavior of these components. In the general sense, each random component of the system is represented by a numerical value randomly chosen from some probability distribution. 10 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS Monte Carlo simulation essentially involves three steps: 1. Select representative probability distributions which will simulate the physical behavior of the random components of the system. 2. Generate numerical values of the random vari- ables from the probability distributions. 3. Use variance-reducing techniques to accelerate the computations. A random walk model is well suited to simulation by Monte Carlo methods because each random com- ponent (the increments added to K1 or K2 in the present study) is allowed to take only two possible discrete values. This requires the simplest of random variable generation techniques, the generation of uniformly distributed random variables. The meth- ods for the generation of uniformly distributed ran- dom variables are discussed in various books (such as Hammersley and Handscomb, 1964), and almost all digital computers have a built-in subroutine avail- able for this purpose. The determination of the probability distribution of the dissolved-oxygen deficit is based on equation 26 which may be written 1' -—/ K2d'r o D=C,,-—C=B+f(§)e 7 / (K2—K1—K3)d-r’ 0 K16 d1- (39 ) where 7. K2117" o _];TK2dT T B=e c..—I~"(c)+/o (DB—Cm d? T [)T(K2—K1—K3)d'y' +f K116 d7 (40) Additional assumptions are necessary besides those made in the development of the model for the proba- bility distribution of the BOD. These are: 1. the initial dissolved-oxygen deficit, CS—F (1:) , is small and the variance of the term [Cs—F (0] _ A K2d7' e is small; and 2. the difference between the benthal demand term, DB, and the term for the addition of dissolved oxygen along the reach, Ca, is small, and the 1' —/ [(2017 o variance of the term e T LTKMT’ ]; (DB-Ca)e d1. is small. With these assumptions and approximating the integrals in equation 39 by finite sums, it follows that n —\K1 C,—C—B=f ex — KziAT) ————1— (C) p( {E1 [K21_K11_K3 K15(K25+1_K3)_(K2i_K3)K (K,,—K,,—K,) (K2,+1—K, n—l + 2 i=1 1i+1 "K3) i+1 i Z (sz—K1j_K3) AT 1 exp 3': K m. 1n +m exp 21(K2FK15'K3MT ] .7: (41) Note that for nAT=T, K11=K12= . . . =K1n, K3=0, and K21=K22= . . . =K2n, equation 41 reduces to the classical oxygen sag equation of Streeter and Phelps (1925) (recall eq 4). If K3 should be a normally distributed random variable like K1 and K2, then Monte Carlo simulation requires three sets of uniformly distributed random numbers to be generated (one each for K1,, K2,, and Kai). There is no evidence at present, how- ever, to suggest that K3 is a random variable. Hence, K3 was considered to be deterministic in the present study. The determination of the probability distribution of the oxygen deficit at any time of travel, T, by Monte Carlo simulation involves the following steps: 1. Experimentally determine the mean and vari- ance of K1 and K2 and the correlation coeffi- cient between K1 and K2; or by experimentally determining the means, estimate the variance of K1 and K2 from equations 12 and 16, respec- tively, and the correlation coeflicient between K1 and K2 from equation 20. 2. Choose an integer n and determine AT from AT=T/'n. 3. Determine AK1 and AK2 from equations 31 and 38, respectively, as AK 1 = Vm AKfl/m where Bl=variance of K1; and ,82=variance of K2. 4. If the mean values of K1 and K2 are assumed to be time-independent, then a1=a2=0 and p1=p2 =0.5. For time-dependent mean values of K1 DEVELOPMENT OF THE PROBABILISTIC MODEL 11 and K2, determine a1 and 012 from mean (K1) =K1—‘7‘1/31 and mean (K2) =K2—a2,32 (K1 and K2 will be estimated as the values of K1 and K2 at T =0), and compute p1 and m from equa- tions 30 and 37 as p1=(1—011AK1/T)/2 p2=(1—a2AKz/T)/2. 5. Compute B from equation 40 with known values of C8, F(§) , DB, 0,, La, K3, and mean values of K1 and K2, and consider B as deterministic at any time of travel, T. If B is not deterministic, see below for the procedure to use. 6. Generate two sets of n uniformly distributed random numbers, R1]. and R2], between 0 and 1, and compute KleT and szAT as KleT=K1AT+AK1 R1j. 3 F l J I t Lu 0 I i g i i 7 i l 8 0 1 2 3 0 1 2 3 O 1 2 3 ‘3? “- I l l l l I Lu E 1.50— 7:4 Days T=5 Days _ 5 W Lu 0: 1.00" — 0.50~ m 0 i l ____________ l r I : O 1 2 3 0 1 2 3 OXYGEN DEFICIT, IN MILLIGRAMS PER LITRE FIGURE 6.—-—Distribution of the oxygen deficit as estimated from equation 41; Sacramento River data. ample (recall figs. 1, 2, and 3). This difference is apparently a result of differences in the values of the coefficients and parameters used in the computations. There is considerable scatter in the oxygen-deficit variances plotted in figure 7, and as a result, it is difficult to decide which of the two models best fits the data. A disadvantage of the Thayer and Krutch- kofl" (1966) model is that the parameter A must be determined from the measured variance at some time of travel. They used a time near zero; however, be- cause of the small variances and considerable scat- ter in the data at this time, it seems likely that A cannot be estimated very accurately. On the other hand, the stochastic model developed in the present study does not depend on a measurement of the deficit variance for predicting the deficit variance as a function of travel time. Thayer and Krutchkofi‘ (1966) concluded that the oxygen-deficit variance is a maximum at the travel time when the deficit is maximum. For the Sacra- mento River data (see fig. 5), the maximum deficit occurred between 1 and 2 days. On the other hand, figure 7 suggests that the deficit variance is a maxi- mum for a time of about 3 days. The stochastic mod- el predicted a similar behavior in that the variance increased rapidly for 3 days, then increased little be- tween 3 and 4 days (see table 3) . The occurrence of the maximum deficit variance at a time larger than the critical time of travel did not appear to cause To to shift downstream relative to the deterministic critical time of travel. However, the 10 and 20 per- centile limits of dissolved-oxygen concentration be- come rather flat between the second and third days. ESTIMATION OF THE ACCURACY OF THE MONTE CARLO PROCEDURE The accuracy obtained from the Monte Carlo sim- ulation procedure depends on the values chosen for m and n. Recall that m is the number of times the computations are repeated and that n, the number of steps in the random walk process, appeared in several equations where integrals were approxi- 18 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS 0-18 l T l l l 0.16 0.10 0.08 Thayer and Krutchkoff (1966) 0.06 VARIANCE, IN (MILLIGRAMS PER LITRE)2 0.04 e W . Observed . variance 0.02 — _ o l ‘ . l 0 1 2 3 4 5 6 TIME, IN DAYS FIGURE 7.—Observed and predicted variance of the oxygen- deficit distributions as a function of time; Sacramento River data. mated by finite sums. The accuracy obtained in these equations depends on the functional form of K1 and K2; if they are constants, then the accuracy will be 100 percent; if they are functions of time, then some error will result from the approximation of the inte- grals. Although the accuracy of the simulation pro- cedure increases as m and n are increased, the com- putation time and resultant computer costs also increase. The development of equations for estimat- ing the accuracy of the Monte Carlo method is de- scribed in the following paragraphs. The comparison of the central moments of K1T with the central moments of the limiting normal dis- tribution was used as the criterion for estimating the accuracy of the Monte Carlo procedure. Moments of KIT rather than K1 were compared because deter- mining the value of KIT as the result of a random walk process is physically more meaningful. The probability that the diffusing particle of a simple random walk will be at some location after n steps is distributed according to the binomial dis- tribution (Bailey, 1964). Hence, it follows analo- gously that the probability density function of KIT is n P[K1T= .21 KjAT— (72—20 An] =( g? )(qum-i .7 = (46) Previously it was established, in the limit as AT'>O, A_K1->0, that K1 was normally distributed with mean Kl—al/S’, and variance ,8, for constant K1. Analo- gously, it can be slgwn that KIT is normally dis- tributed with mean K,T — alfilT and variance ,8sz. Equation 46 gives exact values for the mean and variance of KIT even for very large values of AT because the mean and variance depend only on an and ,81 computed from equations 30 and 31. The third central moment of the limiting normal distribution is zero and therefore the error in the third central moment of KlT could not be computed as a percent- age difference. As an alternative, the difference be- tween the fourth central moments of KIT obtained from the limiting normal distribution, ”4(K1T, N), and from the probability density function defined by equation 46, [14(K1T, AT), was used as an indicator of the effect of the value of n on the accuracy of the Monte Carlo simulation procedure. The fourth central moment of the normal distribu- tion is ”4:304 hence u4(K1T, N) =3B§T4 Similarly from equation 46, M4(K1T, AT) = (2AK1)4IL;(.71‘) (47) where 9‘ has a binomial distribution with parameters n and 101. The fourth central moment of a binomial distribution is h=np1q1[1+3(n—2)201q1] hence ,u4(K1T, AT) = (2AK1)4[3 (”11141) 2 +np1q1-6np'jqi] (48) For small AT/ T, it follows from equation 30 and 31 that afBaAT 1.. p1q1= ”£11 exp (—affilAT/T). (49) DEVELOPMENT OF THE PROBABILISTIC MODEL 19 Substituting equation 49 into equation 48 and not- ing that n = T / AT we obtain “4(K1T, AT~ 3BfT4 exp (—2affilAT/T) +,8jT3AT [4 exp (—af,81AT/T) —6 exp (—2a:/31AT/T)]. (50) The error, E', in the fourth central moment of KIT is defined as E=M(K1T, AT) —,1,(K1T, N) ydKlT. N) ' Then, using equations 47 and 50, (51) E 1 5:5 2 4 _ 2 _ 3 3,8:T4{3’81T [exp ( 2a1/31AT/T 1] +,8§T AT [4 exp (—a;",31AT/T) —6 exp (—2af/31AT/T)]} Expanding the exponentials and taking the first two terms gives GaEfiIAT SaifllAT ___,_ _ 2 3 __ _______ T 231T AT (1 T ) E . 3,8:T4 For the special case when (11 = 0, the expression for E becomes exact, and E='§—T‘ Thus, if n= T/AT= 100, the error in the fourth central moment of KT is less than 1 percent. The same arguments are also valid for the fourth central moment of K2T. The effect of the choice of the integer m on the accuracy of the Monte Carlo procedure was esti- mated on the basis of the weak law of large numbers (Mood and Graybill, 1963) , or P[IZm—ul1—8 In this expression, Z, is the sample mean of a ran- dom sample of size m from a probability density with a mean of a and a variance of 0'2, m is any in- teger greater than (72/528, and s>0 and 0<8< 1. Chebyshev’s inequality states that P[g (Z) >s] — — £2 1 8 (53) for m> 02/528. Thus, the error in an answer is inversely propor- tional to the square root of the sample size m. For example, m has to be >160 in order that one is 90 percent certain that Z, is within 0/4 of ,1, that is [m>02/ (0.250) 2 (0.10) = 160] In the hypothetical example and the treatment of the Sacramento River data described in the previous section, 72: T/AT was taken as 100. Therefore, from equation 52 it follows that there is less than 1 per- cent error in the fourth central moments of KIT and KZT for an infinitely large sample size. The computa- tions were repeated 200 times, that is, m= 200. From equation 53, we are 92 percent certain that the esti- mated mean value of the oxygen deficit will be with- in 0/4 of the true mean where a' is the standard de- viation of the oxygen deficit. ESTIMATION OF THE VARIANCE OF THE OXYGEN-DEFICIT DISTRIBUTION The random walk model and Monte Carlo simula- tion procedure described in previous sections can be used to predict the variance of the dissolved-oxygen deficit at any time of travel for a stream for which Lo, K1, K2, the coefficients of variation of K1 and K2, and the correlation coefiicient between K1 and K2 are known. In practice, however, it may not be feasible to compute the oxygen-deficit variance for all possi- ble combinations of these variables because of the computer expenses involved in the simulation proc- ess. Hence, an approximate procedure for estimating the variance was developed. This development is described in the following paragraphs. The basic procedure consists of expanding in a Taylor series the random part of the fundamental equation of the stochastic model, equation 41, and determining the expected values of the first few terms of the series. From equation 41, it follows that 08— C—B=g (K11, K12, ..., K1", K21, K22, ---’K2n) (54) Expanding g in a Taylor series gives =g(I? ff)+ {I (K —K)fl g 1; 2 i=1 1,- 1 3K1,- K1i=K1 + 20 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS _ 82g —K K2 —K2 ___._..__ 1)( 1; )aKl aKziK1i= K2:= _ +... n + Z (1-{10J i=1 K(55) The last term on the right hand side of equation 55 has only a_ single summation because the terms E[(K1i_K1) (K1j_K1)], E[ (Kai—1T2) (Ky—ED ’E[K1i_1?1) (K2;— are zero for iyéj. The mean value of 9, that is E'[g], is obtained from equation 55 as 172)] K1 K12 E[g]~E g(K1, K2)+ Z n(__‘__l__._ a g _ $=1 2 3K1: K16=K1 i=1 2 aKZi2 K2i=K2+ i=1 _ ._ 629 (KIi—K1)(K2i_K2)mK1i=I:1 - (56) 2i=K2 Computing the terms in equation 56 with 9 given by equations 41 and 54, and taking the limit as n -> 00 gives E[g]~g(K, K2) +G1+G2+G3 (57) where _ _ _ K, 9(K1, K2) =f(§)e—K2T[—(e“T-1)] (58) G1=E£g£§l¢r 221' TLK :(1— eaT)+KTe“T] (59) G2=fle—Kzr[;(ear_1) _£1£] (60) 2 a2 a G3=T(K17 K2)\/Bl—£2Tf(€)e_ffl' _ K1 [2a, (61) a=K2—K,—K3 (62) and MK” K2) is the correlation coeflicient between K1 and K2. Note that E[(K1,—K1)2] =BlT/AT and ET (Kzi‘Kz) 2] =BzT/AT- The mean value of 03— C, that is E[C,—C], is E'[Cs-C]=B+Elgl (63) Then by definition, the variance of g, or equivalently the variance of C,- C, is var(g) =E[(g—E(g) )2]; hence, from equation 55, it follows that n n var~E Emu—Emir _+ i=1 K1,. K1.-=K1 i231 n — (Kg—[(2)2 -'—g-)2 _+2 Igu—IZ1 (BKZiK K2 =K2 i;1( ) (K2- K2) (#1) (——~) K =K1 (64) 515K 8K2,- K2: —K2 Computing the terms in equation 64 and taking the limit as who gives var(g) ~G4+G5+G6 (65) where 2 ZaT __ _ _ “—2“3 (e 1 )— 2K1K2 KjT (6111— eZaT) + eZaT] (66) a2 =BlT[f(€) Jae—232i + a3 Gs =52T[f(§) ]23_2K2T K2 2K: KZT l:___(ezaT_ 1)+__(1_ea'1‘)+___ a2 :l(67) 2 a ‘IKZ =27‘(K1, K2) (18132)1/2[f(€)]2T e‘_2K2T|:_ K2a3 KK — K3 K2; (eZaT— 1)+———2(euT.—1)+3(eza1'_ea,q-)_ a2 6311'] (68) and a is defined by equation 62. Equations 63 and 65 for estimating the mean and variance of the oxygen-deficit distribution are based on only the first and second order terms in the Taylor series expansion, hence the estimates of the vari- ance are less than the actual variance. No attempt was made in the present study to improve the ac- curacy of the estimated variance by considering higher order terms of the expansion; this is possible but tedious. Qualitative inspection of the next higher order terms in equation 65 shows that they are of the order and form E2 1 B:T2[f12 as which indicates that the accuracy obtained in the variance of the deficit estimated by equat_ion 65 de- creases as 31, ,82, and T increase and Kz—Kl—K3 decreases. e—ZKzT DEVELOPMENT OF THE PROBABILISTIC MODEL 21 For constant values of C, (K1), 0.,(Kz) , ’I'(K1, K2). and K3, equation 65 can be put into the form var(Ca—C)~[f(c)]2‘1’(1?n I?” T)- (69) To facilitate obtaining first estimates of the variance of the oxygen-deficit distribution, «Ir-values were com- puted from equations 66, 67, and 68 for K values ranging from 0 to 0.5 days—1 and IE values ranging from 0 to 2.5 days*1 for times of travel of 1, 2, and 3 days. In these computations, it was assumed that C,(K1) =0.35, 0,,(K2) =0.3, 7*(K1, K2) =0.5, argi K3=0. Values of «II are plotted as a function of K1 and if, in figures 8, 9, and 10 for travel times of 1, 2, and 3 days, respectively. The dashed line indicates the discontinuity at the point where K =I?2. The \Ir-values are underestimated if 0,,(K1) and (or) C,(K2) are less than the true values and con- versely. At small times of travel, 0,, (K1) is most im- portant in the determination of \P and at large times of travel, C,(K2) is most important. 0n the other hand, ‘1’ decreases at all times as r(K1, K2) increases. Computations were not carried out for travel times of 4 and 5 days because equation 65 becomes less re- liable at large T. The oxygen-deficit variances at T = 1, 2, and 3 days for cases 1 and 3 of the hypothetical example de- scribed previously were compared with the variances estimated from equation 65 and figures 8, 9, and 10. For case 1, equation 65 was used with G6=0, that is, r(K1, K2) =0; for case 3 the variance of the deficit was estimated from figures 8, 9, and 10 and the relation var (D) =[f(§)]2~1r=100~1r. The results presented in table 4 show that the esti- mated variances are comparable to those deter— mined from the stochastic model for the first 2 days. For T =3 days, however, the estimated variances are ./ ,/ / / 8° 6° °/s Q DEOXYGENATION COEFFICIENT, IN RECIPROCAL DAYS 0 0.5 1.0 1.5 2.0 2.5 REAERATION COEFFICIENT, IN RECIPROCAL DAYS FIGURE 8.—‘I'-para.meter as a function of the reaeration and deoxygenation coefficients; time of travel of 1 day. 22 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS 0.5 6; /T l i 0». 0.0 1x0 090 U) 0 O >- 0.4— N no < H o 3‘) O O 0 cs 0' 09° i 8 30 E 090 o g 03 Z 6 _ 002 ,_~ 0' E Q / o t 0907’ “‘ / 8 0.2 / 2 0X5 ' 9 / 0.0 p— ‘z‘ / “J / 4/ ‘00010 ‘5 / § 0 ' / 0.0005 ' o i l l l 0 0.5 1.0 1.5 2.0 2.5 REAERATION COEFFICIENT. IN RECIPROCAL DAYS FIGURE 9.;—‘1’-parameter as a function of the reaeration and deoxygenation coeflicients; time of travel of 2 days. TABI_4E 4.—Variances of the oxygen-deficit distributions deter- mined from the stochastic model and the Taylor series ap- proximation (eq 6‘5 ) Time, T Variance, in (mg/1)? (days) Stochastic model Equation 65 Case 1 1 ____________ 0.121 0.135 2 ____________ .315 .307 3 ____________ .528 .442 Case 3 1 0.101 0.105 2 .191 .186 3 .266 .230 smaller than the variances from the stochastic model. This is to be expected because the higher order terms of the Taylor series expansion were neglected in the development of equation 65, as discussed previously. Despite this restriction, however, figures 8, 9, and 10 can be used to obtain a first estimate of the variance of the deficit distribution for those situations in which TCéS days and when the deficit profile ob- tained by deterministic procedures is changing rapidly around T=Tc. EXTENSION OF THE MODEL In the previous sections, a stochastic model based on Monte Carlo simulation was developed for pre- dicting the variance of the distribution of the dis- solved-oxygen deficit at points downstream of a waste source for a stream system in which K1 and K2 are random variables. In addition, a simplified pro- cedure based on Taylor series expansion of the equa- tion for the stochastic model was developed for esti- mating the variance of the oxygen-deficit distribu- tion. In this section, the model was extended such that four parameters could be considered as random variables: the BOD of the waste at the upstream end EXTENSION OF THE MODEL 23 o 9020 o .0015 ‘4: $09010 DEOXYGENATION COEFFICIENT, IN RECIPROCAL DAYS 0.0005 / 0.0002 0 I | I I o 0.5 1.0 1.5 2.0 2.5 REAERATION COEFFICIENT, IN RECIPROCAL DAYS FIGURE 10.—‘I’-parameter as a function of the reaeration and deoxygenafion coefficients; time of travel of 3 days. of the reach (L0), the deoxygenation coefficient at the upstream end of the reach (K1) , the deoxygena— tion coefl‘icient along the reach (k), and the reaera- tion coefficient (K2). In addition, correlation was assumed between K2 and k and between K1 and L0. The fundamental equation of the stochastic model (eq 41) was used with three changes: B was as- sumed negligible, K3 was assumed zero, and the. variation of K1 along the reach was separated from the variation of the initial K1 values by the addition of the k parameter. These changes are for the pur- pose of simplification only; they do not in any way change the basic procedures used in the computa- tions. With these changes, equation 41 becomes K1+lc1 (K21_K1—k1) ”—1(K.+k.> (K.,+,)—(K2,>] (71) and for a stochastic critical time of travel, T08, ob— tained from the Taylor series expansion approxima- tion of equation 70. These critical times of travel in general Will not be the same; results presented pre- viously suggested that the critical time of travel for the stochastic model was larger than the determin- istic critical time of travel. The procedure used to estimate the stochastic critical time of travel will be described shortly. MONTE CARLO COMPUTATIONS The basic Monte Carlo simulation procedure was as described previously and as shown in figures 22, 23, and 24; however, because of the several changes in this section, the procedure will be outlined quali- tatively here. Input data consisted of the coefficients of variation of the reaeration coefficient (K2) , of the deoxygenation coefficient along the reach (k) , of the total deoxygenation coefl‘icient (K1+k), and of the BOD at the upstream end of the reach (L0) ; the cor- relation coefficients between K1 and L0 and between K2 and k; and the critical time of travel. Steps in the procedure were: 1. Selection of L0 from a normal distribution. 2. Selection of K1 from a normal distribution with the specified correlation coefficient between LJ and K1. 3. Selection of K1+k and K2 to be used in the ran— dom walk computations (eq 70). Computation of the oxygen deficit. Repetition of the procedure m times. Computation of the mean and the variance from the m values of the oxygen deficit. 9‘51"!" The Monte Carlo computations were completed for 25 sets of K1 and K2 values with the following conditions: C, (K1+k) =0.35 C, (k) = 0.19 0, (K2) = 0.30 0,, (L0) = 0.20 MK” L0) = —0.67 r(K2, k) =0.50 f. = 1.0 mg/l D, = 0 n = 100 m= 800 for most computations, 200 for some com- putations. These conditions were also used in all computations in this section with the Taylor series approximation of the stochastic model. ESTIMATION OF THE MEAN AND THE VARIANCE OF THE OXYGEN-DEFICIT DISTRIBUTION As in previous sections, the mean and the vari- ance of the distribution of the oxygen deficit may be estimated from a Taylor series approximation of the ' stochastic model. Basically it is assumed that D=h(K11, K12, ...K1n; K21, K22, K2"; L01, L02, "-Lon; k1, 152! “'k") - (72) Expanding in a Taylor series, taking expected values, and assuming correlation only between K1, and L0, and between K 21. and k,- gives Mean D=h(I?1, IE, :0, E) + var ‘L" 21 2 3L: + 33' 3.22.5: i ii" —2”,-=1 AT ale: 2 i=1 AT ax; _ n 1/2 _________.+______ +1 (K2, 16) (131/32) 1:; AT aKziak, 2 3K: 62h I 1 2 ____ 73 +r(K1,L0) (Blvar Lo) / aLoaK1 ( ) where var implies variance and ,8’1=,81 — Bf. Determining the derivatives from equation 70, evaluating at the mean values of the variables, and substituting into equation 72 gives where I-l—ofl . — — h: _ e—K1T_e—K2T K.2—K1L ] (75) EXTENSION OF THE MODEL 25 BZT L—(Je—EzT I: — (1_ e(K2—K1)T)+ K1T 2 (K2— 2K1)2 K2_K1 H1: en‘s—Elm] (76) _ M foe-ET Kn —_1—< KT ] H2——__2— (Tris—W“ ”‘D'Kz—Kl (77) H— K is (8’3 W2“. jfl[ Z 3—7‘( 2; ) 1 2 ~ 06 2(K2—K1)2 (e(K‘2—K1)T_1)+ (1—6‘22—E1)T):l (78) 2(1{2_-K1)2 Bil‘EOe—KgT 2K _______l:__2___2__._(e(K2—-K1)T_ 1)_ 2 (K2 K03 K1T2 e(I?s—K1)T+ _ (K2_K1) H5=’I‘(K1, Lo) (B;)1/2EOC1)(L0) e—K‘ng: K1T 2 1 The terms H1, H 2, and H 3 represent the effects on the deficit of the variation of the deoxygenation coeffi- cient along the reach, the variation of the reaeration coefficient, and the correlation between these coeffi- cients, respectively; allowing for slight changes in nomenclature and the assumption that K3= 0 in the present section, H1, H 2, and H 3 are equivalent to G1, G2, and G3 (eqs 59, 60, and 61), respectively. The terms H4 and H 5 represent the effect on the deficit of the variation of the initial value of the deoxygena- tion coefiicient and the correlation between the ini- tial deoxygenation coefficient and upstream BOD, respectively. Variations in the upstream BOD do not affect estimates of the mean deficit because aZh/aL:=O. When the variables K1,,Kgi, L0,, and k,- are assumed independent of time of travel, then H1, H2, H 3, H4, and H5 are zero and equation 73 reduces to 2111‘ HE (Kl—Km 6(K2—I—(1)T] (79) K2 (K2 _K1) 2 (e(ff2—I?1)T_1)_ _e(K2—-K1)T:|_ (80) Mean D = h (81) or the classical oxygen sag equation of Streeter and Phelps (1925) (see eq 4). By similar arguments, it may be shown that n Var D= ,8” i2 ah " —<——>2+ 1AT 57.31 i=1AT 5K2, n +2'r K2,k ” 1/2 __ ( )(8132) 2:21” 8K2, h h +VarLo +B;(— a TIL) +2r(K1, Lo)(B’ varLo )1/2(——— ah XX :1}: )- (82) Evaluating derivatives as before, it follows that VarD=H6+H7+H8+H9+H10+H11 (83) Where IT? = usz —2K2T —2 (2172—1701.. 31 °e 2(K —K )3 (e 1) __ E217 + 3K1£2 (e(172—I?1)T_62(I-fz—§1)T)+_—1__ (1(2"1z1)3 (1(2"I{1)2 eZ(I?2—I—{1)T:l (84) BTITZe 2371: K: — “ = — 2 __ _ 2(K2—K1)T_ 2 ° 2(K2—K1)3(e 1) 2?: ET ] +__=__=_ _ (RE—Kim —# 85 (K,—K,)3(1 e ”(m—Km ( ) _ _ KI? H =2 K2, k // 1/2L2T —2K2T _—_,..2.1_— — — KK eZ(K2—-K1)T_1 2 1 6(K_2—K1)T_ 1 ( I? ) (K —K,)3 ) 2 PT _=_.;__<62(E2—E1)T_e(f2—E2)T>_.—_-IT ”(z—1(1)3 1 (1(2“I{1)2 Mia—EDT] (86) K2 H =52 Co L 2 —21?2T 1_ 203—71” 9 0|: ( 0)] e [(K2_K1)2(e -26(K'2—Kx)1'+1):l (87) K2 W(ez(1—(z—E1)T_29(E2—E)T+ 1) 2— 1 = /_2 —21?T IBILOe 2 [ _ _—62(K2_K1,T_e(x2—K1)T)+2K2KT K2T2 (K —1{1)3 (K2— gum—Elm] (88) H11=2T(I?1, Lo) (18,)1/211201, (L0) 3— 2K:TTl:(K2 K2 _:___I§_{11)3 (e2(K2—K1) T_ 29(K2—K1)T+ 1)__ (K2 _K) 2 (62(22—1—{1} T— e(7{‘2—E1)T)] (89) The terms H6, H7, and H8 represent the effects on the variance of the distribution of the deficit of the 26 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS variation of the deoxygenation coefficient along the reach, the variation of the reaeration coefficient, and the correlation between these coefficients, respec- tively; again allowing for slight changes in nomen- clature and the assumption that K3=0 in the present section, H 6, H 7, and H g are equivalent to G4, G5, and G6 (eqs 66, 67, and 68), respectively. The terms H9, H10, and H11 represent the effects on the variance of the variation of the initial BOD, the variation of the initial value of the deoxygenation coefficient, and the correlation between these variables, respectively. ESTIMATION OF A STOCHASTIC CRITICAL TIME OF TRAVEL As mentioned previously, it would be very diffi- cult to determine explicitly the critical time of travel for the stochastic model; also the results of previous sections suggest that in general the stochastic criti- cal time is larger than the deterministic critical time. To get an approximate estimate of T08, equation 74 was used. Recall that this equation is the result of a Taylor series expansion of the equation for the stochastic model; because higher order terms of the expansion were neglected, the resultant To, will be approximate only. In theory, equation 74 can be differentiated with respect to T and the result set equal to zero and solved to determine Tc. When this is tried, however it is found that a complicated expression containing exponentials and quadratic terms must be solved by a trial-and-error iterative type of solution to deter- mine Tc. Instead of this procedure, equation 74 was programed for a desk-top computer so that the oxy- gen deficit was computed as a function of T. After each computation, the time was incremented by 0.01 day and the computation was repeated. Because of the print-out feature of the computer, it was possi- ble to follow the computations and to stop the pro- gram after the deficit passed through a maximum. In this manner, T63 could be estimated to the near- est 0.01 day for any combination of K1 and K2 values. PRESENTATION AND DISCUSSION OF RESULTS The results presented and discussed in this section are directed toward the practical determination of a first estimate of the variance of the distribution of the oxygen deficit at the critical time of travel. All graphs are for a critical time of travel (To, or TOD) and are for a mean BOD at the upstream end of the reach (ITO) of 1.0 mg/l. For other values of Le, the estimated mean deficit is a simple multiple of the value of Do (recall eqs 70 and 74) and the variance is a multiple of 1:: (recall eqs 70 and 83). Values assumed for the coefficients of variation and the cor- relation coefficients were presented previously. Also recall that the oxygen deficit at the upstream end of the reach was zero. CRITICAL TIME OF TRAVEL The critical time of travel for the deterministic model was computed from equation 71 with D0=0 for the same 25 sets of K1 and K2 values used in the Monte Carlo calculations, and the results are pre- sented in table 5. Figure 11 is a plot of the deter- ministic critical time of travel as a function of the deoxygenation and reaeration coefficients. The dashed line indicates the discontinuity at the point where K1=K2. The critical time of travel for the stochastic model was estimated from the Taylor series expression for the oxygen deficit by the procedure described previ- ously. The T08 values for the 25 sets of K1 and K2 values are presented in table 5 and figure 12 is a plot of the stochastic critical time of travel as a func- tion of the deoxygenation and reaeration coefficients. The critical times presented in table 5 show that the stochastic T. was greater than the deterministic To for all combinations of K1 and K2 considered, in agreement with previous results. The difference, however, was not constant but was largest for the large KZ/K1 ratios and in general decreased .as Kz/K1 decreased for most of the range of conditior)s considered. The percentage difference, defined as (Tcs—TCD) 100/ T0,), is presented in table 5 and plot- ted in figure 13 as a function of the ratio of the re- aeration and deoxygenation coefl‘icients. Where more TABLE 5.—Crttieal time of travel for the deterministic and stochastic models and percentage difl'erence To, in days Percent- K2 (.1 K‘ 1) 1%“)? T—_— a“? (days—1) ays— 2 1 e er- . i er- ministic Stochastic ence 0.60 0.025 24.0 5.53 8.21 +48.5 1.20 .050 24.0 2.76 4.10 48.6 1.80 .075 24.0 1.84 2.74 48.9 2.40 .100 24.0 1.38 2.05 48.6 .60 .045 13.3 4.67 5.90 26.3 1.20 .095 12.6 2.30 2.87 24.8 1.80 .140 12.9 1.54 1.93 25.3 2.40 .190 12.6 1.15 1.44 25.2 .60 .080 7.50 3.87 4.50 16.3 1.20 .160 7.50 1.94 2.25 16.0 1.80 .240 7.50 1.29 1.50 16.3 2.40 .320 7.50 .97 1.13 16.5 .40 .075 5.33 5.15 5.78 12.2 2.00 .375 5.33 1.03 116 12.6 .40 .100 4.00 4.62 5.07 9.7 2.05 .500 4.10 .91 1.00 9.9 50 .150 3.33 3.44 3.73 8.4 1 65 .500 3.30 1.04 1.13 8.7 1 00 .370 2.70 1.58 1.69 7.0 1 00 .445 2.25 1.46 1.55 6.2 .50 .259 1.93 2.73 2.89 5.9 80 .415 1 93 1.70 1.80 5.9 30 .375 800 2.98 3.10 4.0 20 325 615 3.88 4.05 4.4 10 250 400 6.10 6.41 5.1 EXTENSION OF THE MODEL 27 / / 0.2 — // §\\ \/ / .00 96° / / o‘ 8’00 / 6“ // “900:0 \ | | DEOXYGENATION COEFFICIENT, IN RECIPROCAL DAYS // ‘ / . / / 0.4 — / x / . / e. / z ’0 / ‘90 0.3 — / / .2 / ‘ 0 06 / /"f60 V. 6‘0 // l | O 0.5 1.0 H U! 2.0 2.5 REAERATION COEFFICIENT, IN RECIPROCAL DAYS FIGURE 11.——Critica1 time of travel (deterministic model) as a function of the deoxygenation and reaeration coefficients. than 1 percent difference exists for a ratio, the aver- age is plotted. The percentage difference decreases as Kz/K1 decreases for ratios larger than one, and con- versely, increases as K2/K1 decreases for ratios smaller than one. This behavior suggests that the percentage difference between the critical times of travel is a minimum when the reaeration and de- oxygenation coefficients are approximately equal. The K1 and K2 values used in the computations were chosen so as to cover the range of coefiicients ex- pected for natural streams and rivers; figure 13 can be used to estimate the percentage difference in the critical times for any ratio of coefficients for a spe- cific situation. Figures 11 and 12 are graphs for estimating the critical time of travel for specific values of K1 and K2; figure 11 is for T0,, and figure 12 is for T03. Com- parison of these graphs shows in a general manner that T08 is larger than T these graphs also show a CD, the large dependence of T, on one of the coefficients when the other coefficient is small. MEAN OXYGEN DEFICIT The oxygen deficit at the critical time of travel was computed from the deterministic model for TcD and from the stochastic model for both T0,, and To. and the results are presented in table 6. The oxygen deficit is referred to as the mean oxygen deficit be- cause the values reported for the stochastic model are the arithmetic means of the results of the repe- titions of the Monte Carlo procedure for each set of conditions. For the To, computations, the number of repetitions was 800; for the Tc!) computations, the number was either 200 or 800. Use of 200 repetitions for the 24.0 and 7.50 ratios probably explains why the variability of these means is larger than the variability of the means for these ratios when 800 repetitions were used. 28 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS 0.5 /I / / .0 4; | P w I .0 N 5.’ ‘50 5300 e DEOXYGENATION COEFFICIENT, IN RECIPROCAL DAYS / / / / 7 / . / (P | I \ 200 \ “we (“’0 \ ['50 3‘ 1‘90 / / / / / / / / / / / d‘ _ / €00 _ 0.1 — ‘00 ‘0 6‘ 0 / 0 I | I I 0 0.5 1.0 1.5 2.0 2.5 REAERATION COEFFICIENT, IN RECIPROCAL DAYS FIGURE 12,—Critical time of travel (stochastic model) as a function of the deoxygenation and reaeration coefficients. The oxygen deficits are plotted in figure 14 as a function of the ratio of the reaeration and deoxy- genation coefl‘icients. The results in figure 14 and table 6 show that the oxygen deficit for the deter- ministic model tends to be slightly larger than the deficit for the stochastic model for small ratios, but in general the differences among the oxygen deficits are negligible. Also, the relatively large difference in T08 and TOD for large ratios had relatively little effect on the computation of the deficit by the stochastic model. The oxygen deficit was also computed from the Taylor series approximation (recall eq 73) for T68 and the results are presented in table 7 together with stochastic model results for T08. The oxygen deficits are plotted in figure 15 as a function of the ratio of the reaeration and deoxygenation coefficients. The results show that the oxygen deficits computed for TABLE 6.—Me(m oxygen deficit for deterministic and stochastic models Mean oxygen deficit, in mg/l K2 K1 Ratio Deter- Sto- Sto- (days—1) (days-‘1) K2/K1 ministic chastic chastic Tap T0,, To, 0 60 0.025 24.0 0.0363 0.0343 0.0380 1 20 .050 24.0 .0363 .0362 .0380 1 80 .075 24.0 .0363 .0377 .0380 2 40 .100 24.0 .0363 .0388 .0380 60 .045 13.3 .0608 .0632 0625 1 20 .095 12.6 .0637 .0644 0647 1 80 .140 12.9 .0627 .0651 0628 2 40 .190 12.6 .0637 .0641 .0653 60 .080 7.50 .0978 .0962 .0970 1 20 .160 7.50 .0978 .0968 .0975 1 80 .240 7.50 .0978 0944 0996 2 40 .320 7.50 .0978 0898 100 40 .075 5.33 .127 122 127 2 00 375 5.33 .127 125 121 40 100 4.00 .158 153 152 2 05 500 4.10 .155 148 153 50 150 3.33 .179 162 172 1 65 500 3.30 .180 171 176 1 00 370 2.70 .206 195 .200 1 00 445 2.25 .232 219 .223 50 259 1.93 .256 225 .245 80 415 1.93 .256 ______ .242 30 375 .800 .410 382 .386 20 325 .615 .460 ______ .436 10 250 .400 .543 506 507 EXTENSION OF THE MODEL 29 50 40-— 30— 20— 8.0 —— DIFFERENCE IN CRITICAL TIMES, IN PERCENT 4.0 — 3.0 lllllll | IIIIIIII I 0.3 0.4 0.6 0.8 1.0 2.0 3.0 4.0 6.0 8.0 10 20 30 RATIO OF REAERATION AND DEOXYGENATION COEFFICIENTS FIGURE 13.—Percentage difference in deterministic and stochastic critical times of travel as a function of the ratio of the re- aeration and deoxygenation coefficients. TABLE 7.——Mean oxygen deficit for stochastic model and Taylor VARIANCE OF THE OXYGEN DEFICIT series approximation, with T98 _ _ . . Mean oxygen deficit, in . The varlance of the oxygen defiCIt at the critical K2 K1 Ratio ___flil___ time of travel was computed from the stochast1c (days-1) (days—1) K2 / K 1 Sto- Taylor . . . chastic series model and from the Taylor serles approx1mat10n of 0.60 0.025 24.0 0.0380 0.0387 1.20 .050 24.0 .0380 .0387 the model tor both TOD and T0. and the results are 1.80 .075 24.0 .0380 .0387 presented 1n table 8. The eifect of usmg 800 repetl— 2.40 .100 24.0 .0380 .0387 ~60 ~045 13-3 -0625 0628 TABLE 8.—VariarLce of the oxygen deficit from the stochastic 1-20 ~095 12-6 -0647 -0655 model and the Taylor series approximation 1.80 .140 12.9 .0628 .0646 2.40 .190 12.6 .0653 .0655 _ Variance, in (mg/1)2 '60 '080 7'50 '0970 '0983 (11152—1) (dig—1) gflfi Stochastic Taylor series 1.20 .160 7.50 .0975 .0983 T... T. To” 2. $.20 .24210 7.50 .0996 .0983 a l . 0 .3 0 . .10 .098 0.60 0.025 24.0 0.00024 0.00055 0.00020 0.00026 _40 .075 $33 129 .1263 1 2 .050 24.0 .00023 .00055 .00020 .00026 2-00 375 5.33 .121 .126 3.33 333 333 33333 33333 33333 33333 0-40 10° 4-00 152 155 333 33333 33333 33333 33333 2.05 .500 4.10 .153 .152 ' ' ' ' ‘ ' ‘ .8 .1 12. .00072 .00094 .00049 .00058 .50 .150 3.33 .172 .175 2.43 .133 12.3 .00097 .00105 .00051 .00059 1... 50° 3-30 .17. m 333 333 33333 33333 33333 3333 1'83 312 it? €33 332 3133 3333 3333 333333 333333 333333 133333 -50 259 1393 3245 .24. 33 3333 133333 333333 333333 333333 .80 .415 1.93 242 .246 ' ' ' ' ' ' ‘ - .40 .100 4.00 .00274 00265 .00205 00220 .30 .375 .800 .386 .389 2.05 .500 4.10 .00271 200273 .0 20 200214 2° 325 £15 .43. .43. .33 333 33333 33333 33333 '1” '25” '4‘” '5‘” '5” 3333 1313 3333 133333 133333 33333 13333: T98 from the stochastic model and the Taylor series :50 2259 1393 :00469 :00480 I00393 100411 . . . .80 .415 1.93 ______ .00462 .00393 .00410 appr0x1mat10n are essentlally the same over the en- .30 .375 .800 .00940 .00959 .00696 .00719 . . . .20 .325 .615 ______ .01135 .00799 .00828 tlre range of ratios CODSIdered. .10 .250 .400 .0147 .01322 .00992 .0104 30 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS OXYGEN DEFICIT. IN MILLIGRAMS PER LITRE 0-80 I IfI I I I I I I I I I . Deterministic model (TCD) “ A Stochastic model (TCD) T E] Stochastic model (Tcs) 0.60—- _ 0.40 ~ ' , ‘ \ 0.20m _ ‘0 I I I I I I I I I I I I I I I I I ‘ I I 0.3 0.4 0.6 0.8 1.0 2.0 3.0 4.0 6.0 8.0 10 20 30 40 RATIO OF REAERATION AND DEOXYGENATION COEFFICIENTS FIGURE ILL—Oxygen deficit as a function of the ratio of the reaeration and deoxygenation coefficients; deterministic and stochastic models. 03° IIIIII I I IIIIIII I ' Stochastic model(Tcs) _ E A Taylor series approximation (Tcs) .— 3 II: 0.60— — Lu 0. (n 2 < _ _ o: (D 3 :’ E 0.40— — E \ I: \ _ __ \ _ o \ II Lu 0 Z 0.20— — Lu (5 >. X o _ _ O l | l I l l I l l | l | | l I l l l 0.3 0.4 0.6 0.8 1.0 2.0 3.0 4.0 6.0 8.0 10 20 30 40 RATIO OF REAERATION AND DEOXYGENATION COEFFICIENTS FIGURE 15.——0xygen deficit as a function of the ratio of the reaeration and deoxygenation coefficients; stochastic model and Taylor series approximation. EXTENSION OF THE MODEL tions in the computations for the stochastic model with To, as compared with 200 repetitions for T60 is again noticeable, particularly for the 24.0 and 7.50 ratios. The variability of the variances within a group is much larger for the calculations with 200 repetitions. It is also interesting that the largest variances for the 7 .50 ‘ratio occurred for different combinations of K2 and K1. This probably occurs be- cause of the random nature of the simulation process. The variances computed from the stochastic model and the Taylor series approximation with the deter- ministic critical time of travel are plotted in figure 16 as a function of the ratio of the reaeration and 31 'deoxygenation coefficients. The average variance was plotted for those ratios where several variances were computed. The variance from the stochastic model is larger than the variance from the Taylor series approximation over the entire range of ratios considered. This difference undoubtedly occurred because higher order terms in the Taylor series ex- pansion were neglected. Figure 16 also shows the random nature of the simulation process in that the points for the stochastic model tend to scatter about the trend line whereas the points for the Taylor series approximation all tend to lie on the trend line. The variances computed with the stochastic criti- 0.0150 I I I I I I 0.0100 4 _ 0.0080 — \ _ _ \ ‘ 0.0060 H \ fl \ \ 0.0040 — 0.0030 — 0.0020 — 0.0010— 0.0008 — VARIANCE, IN (MILLIGRAMS PER LITRE)2 0-0006 _ . Stochastic model (TCD) _ — A Taonr series approximation (TCD) — 0.0004 — — 0.0003 _ — 0.0002 I I I I I I I I I I I I I I I I _ 0.3 0.4 0.6 0.8 1.0 2.0 3.0 4.0 6.0 8.0 10 20 RATIO OF REAERATION AND DEOXYGENATION COEFFICIENTS FIGURE 16.——Variance of the oxygen deficit as a function of the ratio of the reaeration and deoxygenation coefficients; stochastic model and Taylor series approximation with the deterministic critical time of travel. 32 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS cal time of travel are plotted in figure 17 as a func- tion of the ratio of the reaeration and deoxygena- tion coefi‘icients. The dependence of the variance on the ratio is essentially identical to the behavior for the deterministic critical time of travel shown in figure 16, with the exception that the curve for the Taylor series approximation falls away more rapid- ly than the stochastic model curve for large ratio values. The effect on the variance of the type of critical travel time used in the computations is shown in figures 18 and 19. Figure 18 shows the variation 0.0150 with the ratio of the reaeration and deoxygenation coefficients of the variance computed from the sto- chastic model; figure 19 shows the variation with the ratio of the coefficients of the variance computed from the Taylor series approximation. Figure 18 shows that the variances are essentially identical for the two critical times up to a ratio of about 4 at which point the variances computed for To” drop off more rapidly than the variances computed for Tea. Figure 19 shows essentially the same behavior for the Taylor series approximation except that there is less scatter of the points around the curves. Also the 0.0100 — 0.0080 * 0.0060 w 0.0040 * 0.0030 * 0.0020 ~ 0.0010— VARIANCE, IN (MILLIGRAMS PER LITRE)2 0.0008 W 0.0006 ~ i 0mm - $demmwdQQ 0.0004 i 0.0003 * 00002 I l l l i i l l A Taylor series approximation (Tcs) lllllill l 0.3 0.4 0.6 0.8 1.0 2.0 3.0 4.0 6.0 8.0 10 20 30 RATIO OF REAERATION AND DEOXYGENATION COEFFICIENTS FIGURE 17.—Variance of the oxygen deficit as a function of the ratio of the reaeration and deoxyg‘enation coefficients; stochastic model and Taylor series approximation with the stochastic critical time of travel. EXTENSION OF THE MODEL 33 0.0150 0.0100 — 0.0080 — 0.0060 v 0.0040 ~— 0.0030 — 0.0020 — 0.0010~ 0.0008 — VARIANCE, IN (MILLIGRAMS PER LITRE)2 0.0006 — q Stochastic model TCD 0.0004 — A res 0.0003 ~ Illlllll | 0.0002 1 I l l l l | | 0.3 0.4 0.6 0.8 1.0 2.0 3.0 4.0 6.0 8.0 10 20 30 RATIO OF REAERATION AND DEOXYGENATION COEFFICIENTS FIGURE 18.—Variance of the oxygen deficit as a function of the ratio of the reaeration and deoxygenation coeflicients; stochastic model with the deterministic and stochastic critical times of travel. T03 results are always larger than the Tab results and the difference increases as the ratio of the reaeration and deoxygenation coefficients increases, in agree- ment with the dependence of the difference of the critical times of travel on the ratio (recall fig. 13). Consideration of figures 13, 18, and 19 suggests that the variance for a specific set of conditions tends to increase as the estimate of the critical time of travel increases and the amormt of the increase increases as the ratio of the coefficients increases. This ob- servation is qualitatively in agreement with the re- sults of the analysis of the Sacramento River data Where it was concluded that the variance was a maxi— mum for some time larger than the critical time of travel. The Taylor series expression for the variance (re- call eq 83) may be used to determine the relative contributions of the various terms to the total vari- ance. The terms H6, H7, H8, H9, H10, and H11 were computed for each of the sets of reaeration and de- oxygenation coefficients (see table 5) with T08 and the other coefficients given previously. The results are presented in table 9. Because the results were the same for each ratio of K2 and K1, only one result for each ratio is given in table 9. Recall that H 6, H 7, and H 8 represent the effects on the variance of the varia- 34 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS ”150 lllllll | 0.0100 — 0.0080 — 0.0060 R 0.0040 — 0.0030 — 0.0020 s 0.0010 -- 0.0008 7 VARIANCE, lN (MILLIGRAMS PER LlTRE)2 0.0006 * 0.0004 — 0 TCD A Tc, 0.0003 * 0.0002 — 0.00015 1 | l l | | l I Taylor series approximation ill llll llllll'l l 0.3 0.4 0.6 0.8 1.0 2.0 3.0 4.0 6.0 8.0' 10 20 30 RATIO OF REAERATION AND DEOXYGENATION COEFFICIENTS FIGURE 19.—Variance of the oxygen deficit as a function of the ratio of the reaeration and deoxygenation coefficients; Taylor series approximation with the deterministic and stochastic critical times of travel. tion of the deoxygenation coefficient along the reach, the variation of the reaeration coefficient, and the correlation between these coeflicients, respectively. Also the terms H 9, H10, and H11 represent the effects on the variance of the variation of the initial BOD, the variation of the initial value of the deoxygena- tion coefficient, and the correlation between these coefficients, respectively. The H terms are plotted in figure 20 as a function of the ratio of the reaeration and deoxygenation co— efficients. The terms H g and H11 were negative; hence, the negative of these two were plotted in figure 20. The variation of the total variance, that is, the sum of H6, H7, H8, H9, H10, and H11, with the ratio of the deoxygenation and reaeration coefficients was presented previously in figure 19. Figure 20 and table 9 show that H 7 changes the largest amount over the range of ratios considered; that is, H 7 is largest for the small ratios and smallest for the large ratios. Recall that H 7 gives the effect on EXTENSION OF THE MODEL 35 TABLE 9.—Distribution of the variance among the terms making up the variance estimated by the Taylor series approxima- tion of the stochastic model Ratio H terms. in (milligrams per litre)2 Variance (K2/K1) Ha H1 H3 H9 H10 H11 (mg/1)” 24.0 0.0000744 0.0000493 —0.0000812 0.0000990 0.000280 —-0.000166 0.000256 12.9 .000212 .000153 —.000241 .000208 .000599 —.000352 .000579 7.50 .000464 .000376 —.000560 .000383 .00111 —.000650 .00112 5.33 .000718 .000642 —-.000910 .000544 .00157 —.000921 .00164 4.00 .000997 .000985 —.00133 .000710 .00203 —.00120 .00219 3.33 .00120 .00128 —.00166 .000827 .00236 —.00140 .00261 2.70 .00147 .00170 —.00211 .000971 .00275 —-.00163 .00315 2.25 .00171 .00215 —.0025'7 .00110 .00310 —.00184 .00365 1.93 .00191 .00260 —.00299 .00120 .00340 —.00202 .00410 .800 .00282 .00670 ——.00582 .00161 .00460 ——.00272 .00719 .615 .00288 .00844 —.00660 .00162 .00470 —.00275 .00829 .400 .00269 .0118 —.00753 .00148 .00454 —.00259 .0104 .___‘ 0.02000 7 7 7 7 7 7777 , 7 . 7 . 7 .77 . 0.400, the percentages are 25.9, 113.6, -72.5, 14.2, E H 43.7, and —24.9, respectively. Thus, the contribution 3 0.01000» 7 \\ ; of H 7 increases from 19.3 percent to 113.6 percent as T -H i / . 772J 000600 _ 8 \§ ‘ K2, It, decreases from 24.0 to 0.400. <5 ’ H w—o—o\ - Flgure 20 and table 9 show that H6, H9, H10, and ;N 0.004007 10 4 _ 7 _ . o a 1 H _ H11 all have max1mum contr1but10ns to the variance ESE 000200? 44:51 ' §‘ _ at a ratio of 0.615 for the specific values of ratios L77 "’ ' H W\\ considered. However, for ratios between 0.4 and g E 9 about 2, these four factors do not change appreci- < CL 0.00100 : : 7 E 7,, : : ably. Also, as the ratio decreases, the dependence of g E 000060 : : the total variance on the ratio decreases (see fig. 19). 7.71 5 0.00040 L 1 Recall that H .7, H 9, H10, and H 11 give the effect on the E 3 ’ variance of random variations in to, random varia- g E 0.00020 7 4 tions in L0, random variations in K1, and correlation <5 2 between L0 and k, respectively. Z - . . 2 000010; : Figure 20 shows that H 9 contributes the least to E 000006: 3 the total variance over most of the ratio range con- 2 0000047” 7 7 7 7 7 7 7 77 7 7 7 7 7 7 777 7 c sidered. On the other hand, H17, contributes the most 5 0.1 0.2 0.4 0.6 1.0 2.0 4.0 6.0 10 20 30 to the total variance for ratios larger than 1.0. Thus, *- RATIO OF REAERATION AND of the two factors describing the waste at the up- DEOXYGENATION COEFFICIENTS FIGURE 20.-—Distribution of the variance among the terms making up the variance as a function of the ratio of the reaeration and deoxygenation coefficients. the variance of random variations in the reaeration coefficient. Thus, in a deep, slow-flowing stream with a small reaeration coefficient, random variations in K2 make the largest contribution to the variance of the oxygen deficit. On the other hand, for a shallow, rapidly flowing stream with a large reaeration coefii- cient and the same deoxygenation coefficient, random variations in K2 make the smallest contribution to the variance of the oxygen deficit. This difference in contribution may also be seen by computing the per- centage of the total variance that each of the six terms contributes. For Kg/K1=24.0, the percentages are 29.1, 19.3, —31.7, 38.7, 109.4, and —64.8 for H6, H7, H8, H9, H10, and H11, respectively; for K2/K1= stream end of the reach, the variations in the deoxy- genation coefficient, K1, contribute the greater amount to the total variance. Figure 20 shows that the six terms are most near- ly equal in the middle part of the range of ratios considered, that is, between ratios of about 2 and 6. For larger and smaller ratios, the values of the dif- ferent terms diverge. However, for all ratios con- sidered, all the terms contribute significantly to the total variance with the minimum contribution being the 19.3 percent of H 7 for a ratio of 24.0. In considering the results presented in table 9 and figure 20, it should be remembered that these re- sults are specifically only for the values of the co- efficients of variation and the correlation coefficients used in the computations. For other values of these coefficients, the distribution of the total variance among the six terms would undoubtedly be different. 36 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS Further work is needed to determine the sensitivity of the equation to these coefficients. The variances computed from the stochastic model with Tc, were used to prepare a graph showing the variance as a function of the reaeration coefficient and the critical time of travel. The result is pre- sented in figure 21. Because the mean value of the BOD at the upstream end of the reach was assumed to be 1.0 mg/l, figure 21 is similar to the graphs pre- sented previously for the «Ir-parameter (recall figs. 8, 9, and 10). The one important difference is that figure 21 is specifically for the critical time of travel Whereas figures 8, 9, and 10 are for general times of 1, 2, and 3 days, respectively. Note the difference in the figures, however. The additional constraint of a critical time results in the deoxygenation coefficient varying linearly with the reaeration coefficient for a specific value of the variance; on the other hand, the dependence shown in figures 8, 9, and 10 in general is not linear but depends on the values of the vari- ables under consideration. EVALUATION OF THE MODEL The problem of predicting the response of the dis- solved-oxygen concentration of a stream to the addi- tion of biodegradable wastes has been of much inter- est ever since the pioneering work of Streeter and Phelps (1925). Much of this interest has been di- rected toward predicting the dissolved-oxygen con- centration or the oxygen deficit at downstream points as a function of the hydraulic properties of the stream and the deoxygenation coefficient and BOD of the waste. With the development of com- puters, however, the use of mathematical techniques that previously would not have been practical has be- 0.5 / l / / 2? 9, / § / ‘0 04— / ' / 2 / g / g / EOSF / a ' / p / / / '8' 02— / ‘z’ ' / g / é / / é / _ g o.1— / 0.0005/ / 00 0‘5 110 ll.5 2i0 2.5 REAERATION COEFFICIENT, IN RECIPROCAL DAYS FIGURE 21.—-Variance of the oxygen deficit as a function of the deoxygenation and reaeration coefficients; stochastic model with the stochastic critical time of travel. EVALUATION OF THE MODEL 37 come common. With these techniques, it has become possible to predict the variances of the oxygen deficit and the biochemical-oxygen-demand, in addition to the mean values, at downstream points. This ability adds a new dimension to water-quality standards in that it is possible to estimate the confidence limits for the mean oxygen deficit and hence the proba- bility of failing to meet the standard by any speci- fied amount. The stochastic model of this study uses Monte Carlo simulation in which a complex system with random components is operated by random numbers chosen so that they simulate the physical behavior of the components. The distribution of the dissolved- oxygen deficit and consequently the mean and vari- ance of the distribution are estimated by repeating the process a large number of times. The model has the capability of considering the reaeration and de— oxygenation coefficients and the BOD at the up- stream end of the reach as random variables; in ad- dition, correlation between pairs of these variables can be considered in the model. The model does not require actual measurements of the variance for pre- diction of other variances, as does the model of Thay- er and Krutchkoff (1966). However, considerable information is needed. For each coefficient consid- ered to be a random variable, the variance and the mean value, or equivalently, the coefficient of varia- tion, must be known. For each pair of variables for which correlation is assumed, the correlation coefl‘i- cient must be known. Information of this type for the most part is not available; it was necessary to assume values for the coefficients of variation of the reaeration and deoxy— genation coefficients and for the correlation coeffi- cient between these variables for the test of the model with the Sacramento River data. The values assumed, however, were derived from data on other rivers; the coefl‘icient of variation of the deoxygena- tion coefficient was determined from the Ohio River data and the coefficient of variation of the reaeration coefficient was estimated from data for Tennessee Valley streams. The value used for the correlation coefficient was arbitrarily assumed. Consideration of the physical processes involved in the reaeration and deoxygenation process suggests that a positive cor— relation would be expected but gives no indication of the magnitude of the coefficient. It is well known that the deoxygenation and reaeration coefficients deter- mined for one stream or one particular reach of a stream should be used for other streams or other reaches only with caution because of the sensitivity of these coefficients to changes in conditions. But note that in this instance it is the variability of the coefficients with respect to the mean values that is transferred between streams rather than the mean values. Thus, the assumption is that variations in the coefficients of variation for different streams will be considerably less than variations in the mean values. This assumption is supported by the data for the Tennessee Valley rivers presented in table 11. Note that the reaeration coefficients vary over about a 22-fold range, whereas the coeflicients of variation vary only over about a 6—fold range. The tests of the stochastic model with the hypo- thetical examples, the Sacramento River data, and the computations of the variances at the critical time of travel for the most part gave reasonable values for the variance of the oxygen deficit. This result suggests that the values assumed for the coefficients of variation and the correlation coefficients were reasonable. As discussed in the previous section, the stochastic model also has the capability of handling the division of random variations in the deoxygenation coeffi- cient between those of the coefficient at the upstream end of the reach and those of the coefficient along the reach. The variances of these two coefficients may be significantly different, as the analysis of the Ohio River data showed; furthermore this differ- ence may significantly affect the calculation of the variance of the deficit distribution. The stochastic model in general can consider any coefficient as a normally distributed random variable if the mean and variance are known; additionally, correlation between any two variables can be treated if the cor- relation coefficient is known. The problem is the de— termination of the coefficients of variation and the correlation coefficients for the variables of interest. Further work in this area is needed. Comprehensive studies of water quality usually require segmenting the stream when conditions change appreciably with distance downstream. In segmenting, the conditions at the downstream end of each segment serve as the input conditions to the next segment. The stochastic model can be used in this situation, however, a small modification is needed. Recall that in the development of the basic equation of the stochastic model, equation 41, it was assumed that the variance of the initial dissolved- oxygen deficit, C,—F(§), was small. With segment- ing, this assumption is probably valid only for the first segment because in reality it is the variance of this term, C,—F (g) , which the stochastic model esti- mates for the downstream end of each reach. Hence, the input deficit to the next reach (or segment) is a 38 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS random variable with some variance that cannot be neglected. This modification requires that C,—F(§) be removed from the B factor (recall equation 40) and added to the right hand side of equation 41 as n the term[C,—F(g)]exp (— Z KziAT). 7. = 1 Another approach that could be used in applying the stochastic model to a segmented stream is to use equations 42, 43, 44, and 45. These difference equa- tions segment the stream at AT intervals and there- fore could be used recurrently in the computation procedure, as described previously. Similarly, the Taylor series approximations of the stochastic mod- el could be applied readily to a segmented stream system. The basic equation of the stochastic model can be expanded in a Taylor series to give an approximate equation for the variance. This equation involves sums and differences of exponentials and can be solved without a computer; the Monte Carlo com- putations, on the other hand, can be done efficiently only on a computer. It is possible to prepare charts for estimating the variance (recall figs. 8, 9, 10, and 21) ; these, however, are for specific conditions. Thus, in the event that a preliminary estimate of the variance is desired for conditions not covered by the available charts, the approximate equation may be used. The approximate equation may also be used to indicate how the variance is divided among the various terms describing the total variance, as described previously. The stochastic model is a step toward the fulfill- ment of the need for a procedure for predicting the variances, in addition to the mean values, of the bio- chemical-oxygen-demand and the dissolved-oxygen deficit at points in a stream downstream from a waste source. Further work is needed, however, on both the experimental and theoretical aspects of the problem. Experimental aspects needing further study are the: 1. Coefficients of variation of the various coefficients and the dependence, if any, on the type of stream. 2. Correlation coefficients between the reaeration and deoxygenation coefficients and possible de- pendence on the time of travel. 3. Possible correlation between BOD and the oxy- gen deficit and between BOD and the deoxy- genation coefficient. 4. Possible variations in K3, the rate constant for removal of BOD by sedimentation to the stream bottom. An experimental problem fundamental to all aspects of dissolved-oxygen balance studies is the time inter- val required for the BOD determinations; for reaches where the input BOD varies with time, the standard procedure is not satisfactory. Work is needed in the area of correlating BOD With some quantity that can be measured rapidly, for example, total organic carbon. Some work of this type has been done, however, considerable additional effort is needed. Theoretical aspects needing further study are: 1. Possible analytical methods for determining the probability distribution of the oxygen deficit. 2. Methods for considering lateral variations in the rate coefl‘icients and the concentrations of BOD and dissolved oxygen. 3. Sensitivity analysis of the stochastic model to determine which of the coefficients of variation and correlation coefficients are most important in the determination of the variance. 4. Accelerated Monte Carlo techniques to reduce the computer time required in the determination of the oxygen-deficit distributions. SUMMARY A random walk model was developed for predict- ing the distribution of the biochemical-oxygen-de mand for points downstream from a waste source for a stream system in which the deoxygenation co- efficient is a normally distributed random variable. The model has the capability of considering both the mean and variance of the deoxygenation coefl‘icient as functions of the time of travel through the reach. A stochastic model using a Monte Carlo technique for simulating a random walk process was developed for estimating the distribution of the dissolved- oxygen deficit for points downstream from a waste source for a stream system in which both the deoxy- genation and reaeration coefficients are normally distributed random variables. The model has the capability of considering the mean and variance of the two coeflicients as functions of the time of travel through the reach. The model has the additional capabilities of considering the biochemical-oxygen- demand at the upstream end of the reach as a nor- mally distributed random variable and of dividing random variations of the deoxygenation coefficient into variations of the deoxygenation coefficient at the upstream end of the reach and of the deoxygena- tion coefficient along the reach. Equations for approximating the mean oxygen deficit and the variance of the oxygen deficit were SUMMARY 39 developed by expanding the basic equation of the stochastic model in a Taylor series. Principal conclusions for the range of conditions considered are: 1. Random variations in the type and concentration of wastes discharged into a stream and random variations in the hydraulic conditions within a stream provide ample basis for considering the deoxygenation and reaeration coefficients and the biochemical-oxygen-demand at the up- stream end of the reach as random variables. 2. The distribution function for the biochemical- oxygen-demand derived from the random walk model shows that the BOD is distributed ac- cording to the lognormal distribution. 3. The random walk model simulated by the Monte Carlo technique efficiently estimates the vari- ance of the dissolved-oxygen deficits for points in a stream downstream from a waste source. The error in the simulation process was found to be inversely proportional to the number of steps used in the random walk process and in- versely proportional to the square root of the number of times the process is repeated. 4. The predicted frequency distributions of the oxy- gen deficit became flatter and skewed to the right as time of travel increased. This type of skewness is favorable in the determination of probabilistic water-quality standards because the percentile limits of the oxygen deficit will be less sensitive to errors in the values esti- mated for the coefficients of variation of the deoxygenation and reaeration coefficients and the correlation coefficient between these two coefficients. 5. The critical time of travel estimated from the stochastic model was always larger than the critical time of travel computed from the de- terministic model; the difference decreased as the ratio of the reaeration and deoxygenation coefficients decreased. 6. The critical time of travel for both the stochastic and deterministic models was extremely sensi- tive to the reaeration coefficient when the de- oxygenation coefficient was small and extreme- ly sensitive to the deoxygenation coefficient when the reaeration coefficient was small, for the range of coefficients considered. 7. Differences among the mean oxygen deficits com- puted from the deterministic model, the sto- chastic model, and the Taylor series approxi- mation of the stochastic model with both the deterministic and stochastic critical time of travel were essentially negligible. 8. The variance of the oxygen deficit seems to be maximum for some time of travel larger than the critical time of travel. The Sacramento River data, although containing considerable scatter, tend to demonstrate this effect. 9. The variance estimated from the Taylor series approximation of the stochastic model was comparable to the variance obtained from the stochastic model for small times of travel; as the time of travel increased, the Taylor series approximation underestimated the variance because of the neglecting of higher order terms. 10. The variance at the critical time of travel esti- mated from the Taylor series approximation was less than the variance of the stochastic model over the entire range of deoxygenation and reaeration coefficients considered. The same behavior was found for both the deter- ministic and the stochastic critical times of travel. 11. The variance at the critical time of travel showed the greatest dependence on the ratio of the re- aeration and deoxygenation coefficients at large values of the ratio and the smallest dependence at small values of the ratio. On the other hand, the variance was largest for the small ratios and smallest for the large ratios. 12. The distribution of the variance among the six terms making up the Taylor series approxima- tion of the stochastic model shows that the term giving the effect of random variations in the reaeration coefficient varies the largest amount over the range of conditions consid- ered. Of the two factors describing the waste at the upstream end of the reach, the deoxy- genation coefficient contributes the greater amount to the variance. 13. The stochastic model developed in this study re- quires estimates of the coefficient of variation of the deoxygenation and reaeration coefficients and of the correlation coefficient between these variables for estimating the variance of the oxygen deficit as a function of the time of travel. In comparison, the model of Thayer and Krutchkofl" (1966) requires the measurement of the variance at some time of travel for pre- dicting the variance at other times of travel. 14. There is considerable need for a procedure for predicting the variances, in addition to the mean values, of the biochemical-oxygen-de- 40 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS mand and the dissolved-oxygen deficit at points in a stream downstream from a waste source. The stochastic model developed in this study is a step toward the fulfillment of this need. LITERATURE CITED American Public Health Association, 1969, Water quality standards of the United States, territories, and the Dis— trict of Columbia: Report of the Subcommittee on Water Quality Control. Bailey, N. T. J., 1964, The elements of stochastic processes: New York, John Wiley and Sons, Inc., 249 p. Batchelor, G. K., 1959, The theory of homogeneous turbulence: Cambridge, England, University Press, 197 p. Bennett, J. P., and Rathbun, R. E., 1972, Reaeration in open- channel flow: U.S. Geol. Survey Prof. Paper 737, 75 p. Churchill, M. A., Elmore, H. L., and Buckingham, R. A., 1962, The prediction of stream reaeration rates: Am. Soc. Civil Engineers Jour., v. 88 no. SA—4, p. 1—46. Dobbins, W. E., 1964, BOD and oxygen relationships in streams: Am. Soc. Civil Engineers Jour., v. 90, no. SA—3, p. 53—78. Esen, I. 1., 1971, Probabilistic analysis of dissolved oxygen: Colorado State University, Ph.D. Dissertation, 104 p. Feller, W., 1968, An introduction to probability theory and its applications: New York, John Wiley and Sons, Inc., v. 1, 3d ed., 509 p. Hammersley, J. M., and Handscomb, D. C., 1964, Monte Carlo methods: New York, John Wiley and Sons, Inc., 178 p. Kahn, H., 1957, Use of different Monte Carlo sampling tech- niques, in Symposium on Monte Carlo methods [Florida Univ., Gainsville, 1954], ed. H. A., Meyer: New York, John Wiley and Sons, Inc., p. 147—190. Kothandaraman, Veerasamy, 1968, Probabilistic analysis of wastewater treatment and disposal systems: Research rept. no. 14, University of Illinois, Water Resources Center, 158 p. 1970, Probabilistic variations in ultimate first stage BOD: Am. Soc. Civil Engineers Jour., v. 96, no. SA—l, p. 27—34. Li, Wen-Hsiung, 1962, Unsteady dissolved oxygen sag in a stream: Am. Soc. Civil Engineers Jour., v. 88, no. SA— 3, p. 75—85. Liebman, J. C., and Lynn, W. R., 1966, The optimum alloca- tion of stream dissolved oxygen: Water Resources Re- search, v. 2, no. 3, p. 581—591. Loucks, D. P., and Lynn, W. R., 1966, Probabilistic models for predicting stream quality: Water Resources Research, v. 2, no. 3, p. 593—605. Matalas, N. C., 1971, Introduction to random walk theory and its application to open channel flow, in Proc. First Internat. Symposium on Stochastic Hydraulics [Univ. of Pittsburgh, Pittsburgh, Pa., 1971], ed. Chao—Lin Chiu: p. 56—65. Mood, A. M., and Graybill, F. A., 1963, Introduction to the theory of statistics: New York, McGraw-Hill Book Co., 443 p. Moushegian, R. H., and Krutchkoff, R. G., 1969, Generalized initial conditions for the stochastic model for pollution and dissolved oxygen in streams: Water Resources Re- search Center Bull. 28, Virginia Poly. Inst. Reed, L. J., and Theriault, E. J ., 1931, Least squares treat- ment of the unimolecular expression Y=L (l—e'K‘), part II of The statistical treatment of reaction—velocity data: Jour. Phys. Chemistry, v. 35, pt. 2, p. 950—971. Streeter, H. W., and Phelps, E. B., 1925, A study of the pollution and natural purification of the Ohio River: Public Health Bull. 146, Washington, U.S. Public Health Service, 75 p. Thayer, R. P., and Krutchkoff, R. G., 1966, A stochastic model for pollution and dissolved oxygen in streams: Water Resources Research Center Bull. 3, Virginia Poly. Inst., 130 p. U.S. Department of Health, Education, and Welfare, Public Health Service, 1960, Ohio River—Cincinnati Pool: Part I, 1957, Survey, Robert A. Taft Sanitary Eng. Center, 65 p. Yotsukura, N., Fischer, H. B., and Sayre, W. W., 1970, Meas- urement of mixing characteristics of the Missouri River between Sioux City, Iowa, and Platbsmouth, Nebraska: U.S. Geol. Survey Water-Supply Paper 1899—G, 29 p. SUPPLEMENTAL DATA SUPPLEMENTAL DATA TABLE 10.——Blochemical—oxygtan-demand data for the Ohio River (from Kothandaraman, 1968) and results of the data analysis 43 TABLE 10.—Biochemical—oxygen-demand data for the Ohio River (from Kothandaraman, 1968) and results of the data analysis—Continued Observa- . T BOD K1 L. 1?. V” Ob‘ierva' T BOD K1 Lo 1?. V“ 13:? (days) (mg/1) (dawn (mg/1) (days—1) “(5:12) ‘11,? (days) (mg/1) (days-1) (mg/l) (days-1) ”‘59., 1 _-- 0.99 2.42 0.286 9.8 0.280 0.0013 19 ___ .98 1.76 .237 8.5 .218 .0003 1.86 4.19 .315 _-.. ___- _____ 1.85 2.84 .201 ..._ _ ___..- ..... 2.95 5.60 .266 ..-_. ___- _____ 2.79 3.96 .235 _..-— ———— ————— 4.85 7.02 .217 ___ ___- _____ 4.84 5.49 .200 ..-_ _...-- ..... 2 -_.. .85 2.01 .317 8.5 .284 .0009 20 _. .89 1.47 .242 7.6 .211 .0006 1.52 3.18 .298 ___ ---_ ..... 1.55 2.23 .201 ___ ___- ..... 2.60 4.39 .239 ___ ___- _____ 2.64 3.37 .219 _-- ..-_- _____ 4.51 6.06 .273 ___ ___- ..... 4.54 4.56 .174 ___ ___- ..... 3 ..-_. 1.02 2.49 .281 10.0 .240 .0011 21 _.. 1.02 1.32 .174 8.1 .164 .0002 2.21 4.21 .219 -.._ ___- _____ 2.21 2.33 .136 ___ ..-_- ..... 3.12 5.24 .215 _-.. ___- ..... 3.12 3.14 .166 ___... ---_ _____ 4.91 6.63 .193 -.._. ---_ ..... 4.91 4.30 .149 ___. ___- ..... 4 ___ .95 2.44 .298 9.9 .277 0009 22 __ .96 1.84 .217 9.8 .181 .0007 1.96 4.35 .293 ___. ___- ..... 1.96 3.01 .159 ___ ___- _____ 2.88 5.40 .228 _-- _...-- _____ 2.88 3.86 .145 _-- ___- ..... 4.83 7.12 .247 ___ ___- _____ 4.90 5.59 .170 ___ _....- ..... 5 _-.. .68 1.73 .286 9.8 .254 0013 23 __ .64 1.23 .224 9.2 .193 .0012 1.68 3.55 .256 -.._ ___- _____ 1.64 2.72 .207 ___ _.——— ————— 2.59 4.84 .254 ___ ___- ..... 2.55 3.45 .131 _-- ___- ..... 4.89 6.58 .188 ___ __ ..... 4.57 5.34 .197 ___. ---_ _____ 6 _-.. .96 2.35 .267 10.4 239 0007 24 -_ 1.01 1.26 .141 9.5 .127 .0004 1.88 3.91 .234 ___ ___- ..... 1.93 2.16 .126 _..- ___- ..... 2.86 5.18 .222 ___ ___- ..... 2.92 2.81 .094 ___ ___- ..... 4.81 6.80 .191 ___ ___- _____ 4.86 4.40 .140 ___ -.._- ..... 7 ___ 0.97 2.62 .317 9.9 .275 0013 25 __ 1.00 1.55 .204 8.4 .195 .0001 1.84 4.01 .244 ___ ___- _____ 1.88 2.60 .189 ___. ___..- ..... 2.93 5.54 .276 ___ ___- ..... 2.96 3.73 .201 ___. ___... ..... 4.85 7.05 .221 ___ ___- ..... 4.86 5.05 .175 ___ ___- ..... 8 ___ .86 1.49 .227 8.4 .216 0007 26 _.. .90 1.13 .127 10.5 .105 .0002 1.53 2.53 .243 ___ ___- ..... 1.56 1.71 .097 _-.. ___- ..... 2.59 3.52 .174 ___ _.-..- ..... 2.65 2.62 .100 ___ ___- ..... 4.52 5.18 .215 ___ ___- ..... 4.55 3.88 .092 ___. ___- ..... 9 ___ 1.02 2.25 .229 10.8 .201 0006 27 __ 1.02 1.22 .152 8.5 .135 .0002 2.21 4.03 .196 _..- ___- ..... 2.21 2.20 .121 ___ ___- ..... 3.12 5.02 .174 ___ ___- ..... 3.10 2.85 .122 _.-- _--- _____ 4.91 6.49 .164 ___ ___ _____ 4.91 4.13 .142 ___ _...-- ..... 10 ___ .96 2.53 .269 11.1 .218 0014 28 __ .96 1.34 .160 9.4 .142 .0003 1.96 3.96 .183 ___ ___- ..... 1.96 2.37 .137 ___ ___- _-..-- 2.88 5.03 .176 ___ ___- ..... 2.88 3.20 .137 ..-_ ---_ ..... 4.83 6.98 .199 ___ ___- ..... 4.90 4.44 .110 ..._- ---_ ..... 11 _-.. .67 1.48 .268 9.0 .233 0010 29 _.- .62 .82 .144 9.6 .120 .0002 1.67 3.08 .239 ___ ___- ..... 1.62 1.79 .117 ___- ___- ..... 2.58 4.09 .206 ___ ___- ..... 2.54 2.62 .122 ___ -.._- ..... 4.60 5.63 .186 ___. ___- _____ 4.56 3.92 .102 ___ ___- ..... 12 ___ 1.03 2.04 .259 8.7 .226 0011 30 .._ .96 1.11 .131 9.4 .122 .0001 1.95 3.30 .228 _-- ___- _____ 1.88 2.06 .132 ___ ___- ..... 2.94 4.13 .169 ___ ___- ..... 2.86 2.78 .105 ___ ___- _____ 4.89 5.62 .202 ___ ___- ..... 4.81 4.16 .120 ___ ___- ..... 13 ___ .96 2.29 .230 9.7 .241 0009 31 _.. .87 1.61 .316 6.7 .207 .0052 1.83 3.66 .235 ___ ___- _____ 1.87 2.41 .171 ___. ___- ..... 2.92 4.84 .199 ..-_ ____ _____ 2.92 3.07 .159 _-- ___- ..... 4.82 6.52 .223 ___ ___- _____ 4.75 3.84 .130 ___ ___- ..... 14 ___. .88 1.52 .230 8.3 .207 0005 32 __ .96 1.86 .241 9.0 .185 .0021 1.54 2.35 .198 ___ ___- _____ 1.56 2.62 .188 -.._ ..-_- ..... 2.62 3.35 .170 ___ ___- ..... 2.60 3.42 .129 _.-- ___- _____ 4.53 4.91 .198 ___ __ ..... 4.44 4.64 .134 ___. ___- ..... 15 ..-_ 1.02 1.79 .178 10.8 151 0005 33 __ 1.19 2.06 .155 12.2 .128 .0003 2.21 3.07 .129 ___ ___- _____ 2.54 3.48 .118 ___ ___- _____ 3.12 4.09 .156 ___ _....- _____ 3.23 4.17 .119 ..-_ ___- ..... 4.91 5.39 .120 ___ ___- _____ 5.10 5.62 .106 ___ ___- ..... 16 ___ .96 2.35 .292 9.6 .254 0018 34 -_ .83 1.59 .221 9.5 .154 .0022 1.96 4.01 .260 ___ ___- _____ 1.98 2.77 .140 ___ ---_ ..... 2.88 4.84 .175 -.._. ____ ..... 2.87 3.42 .114 ___ ..-_- ..... 4.83 6.64 .244 ___. ___- _____ 4.75 4.48 .102 ___ -.._- _____ 17 -.._. .66 1.23 .210 9.5 .186 0006 35 __ .46 .87 .190 10.4 .157 .0014 1.66 2.67 .191 ___- ___- _____ 1.60 2.74 .192 ___ ____. ..... 2.57 3.72 .183 ___ ___- _____ 2.50 3.50 .116 ___ ---_. _____ 4.59 5.15 .141 ___ ___- _____ 4.38 4.86 .117 ___ ____ _____ 18 ___ .98 1.73 .232 8 5 197 0008 36 _.. 1.13 1.82 .246 7.5 .196 .0018 1.90 2.81 .189 ___ ___- _____ 2.02 2.64 .175 ___ ___- ..... 2.89 3.64 .159 ___ ____ _____ 2.85 3.29 .173 ___ ___- _____ 4.83 4.99 .168 ___ ___- _____ 4.77 4.20 .129 44 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS TABPE 10.—Biochemical-oxygen—demand data for the Ohio TABLE 10.——Bzochemical-oxygen-demand data for the Ohio Rwer (from Kothandaraman, 1968) and results of the River (from Kathandaraman, 1.968) and results of the data analysw—Continued data analysis—Continued 03’9”” T BOD K1 L. ”If; V“ 0‘99”” T B D K L 1?. V” tit? (days) (mg/1) (days—1) (mg/1) Maya") (4532) 131%? (days) (mg/1) (“vi“) (mo/1) Mays") («12532) 37 ——- .87 1.61 .416 5-3 286 -0095 55 ___ .96 1.28 .201 7.3 .146 .0020 1.87 2.34 .220 ___ ___- _____ 1.56 1.60 .091 ___ __-_ _____ 2.92 2.79 .157 ___ __-_ _____ 2.60 2.16 .099 ___ ___- ..... 4.75 3.62 .219 ___ ___- _____ 4.44 3.46 .158 ___ ___- _____ 38 ___ .96 1.68 .264 7.5 .208 -0019 56 ___ 1.19 1.40 .130 9.8 .097 .0005 1.56 2.33 .197 ___ ___- _____ 2.54 2.22 .076 ___ ___- _____ 2.60 3.03 .140 ___ ___- _____ 3.23 2.66 .087 ___ ___- _____ 4.44 4.36 .192 __- ___- _____ 5.10 3.62 .077 ___ ___- _____ 39 ___ 1.19 1.70 .118 13.0 .090 .0005 57 ___ .83 1.28 .207 8.1 .165 .0013 2.54 2.82 .077 ___ ___- _____ 1.98 2.46 .165 _.._ ___- _____ 3.23 3.40 .085 ___ ___- _____ 2.87 3.21 .160 ___ __-_ _____ 5.10 4.34 .055 ___ ___- _____ 4.75 4.08 .104 -__ ___- _____ 40 ___ .83 1.44 .253 7.6 .174 .0037 58 ___ .54 .59 .199 5.8 .152 .0018 1.98 2.46 .157 ___ ____ _____ 1.69 1.58 .183 ___. ___- _____ 2.87 3.13 .157 ___ ___... _____ 2.58 1.90 .089 ___ ___- _____ 4.75 3.76 .081 ___ ___- _____ 3.42 2.36 .149 ..__ ___- _____ 41 ___ .46 .82 .221 8.5 .163 .0026 59 ___ 1.13 1.44 .164 8.5 .125 .0009 1.60 2.23 .178 ___ -__.. _____ 2.02 2.16 .121 ___ ___- _____ 2.50 3.02 .150 ___ ___- _____ 2.85 2.66 .099 ___ ___- _____ 4.38 4.07 .113 ___ ___- _____ 4.77 3.54 .085 __- ___- _____ 42 ___ 1.13 1.79 .196 9.0 .168 .0014 60 ___ .88 2.10 .119 21.1 .093 .0005 2.02 2.86 .180 ___ ___- _____ 1.88 3.42 .072 ___ ___- _____ 2.85 3.63 .161 _-- ___- _____ 2.75 4.72 .088 ___ ___- _____ 4.77 4.54 .097 _-_ ___- _____ 4.88 6.75 .062 ___ -___ _____ 43 ___ .87 1.57 .370 5.7 .274 .0044 61 ___ .90 4.09 .186 26.6 .122 .0001 1.87 2.34 .206 ___ ___- _____ 1.81 6.19 .108 ___ ___- _____ 2.92 3.10 .244 ___ ___- _____ 2.40 7.20 .086 ___ ___- _____ 4.75 3.93 .210 __- ___- _____ 4.52 10.43 .086 -__ ___- _____ 44 ___ .96 1.39 .173 9.1 .119 .0012 62 ___ 2.04 4.05 .101 21.8 .084 .0003 1.56 1.85 .103 ___ ___- _____ 3.23 5.43 .068 ___ ___- _____ 2.60 2.45 .083 ___ ___- _____ 5.00 6.95 .055 -__ ____ _____ 4.44 3.51 .094 ___ ___- _____ 6.96 9.07 .079 ___ ____ _____ 45 __- 1.19 1.74 .185 8.8 .154 .0007 63 ___ 1.04 2.60 .199 13.9 .191 .0030 2.54 2.98 .143 ___ ____ _____ 1.79 4.61 .261 ___ ___- _____ 3.23 3.55 .149 ___ ____ _____ 2.98 5.88 .124 ___ ____ _____ 5.10 4.52 .109 ___ ___- _____ 4.75 7.60 .136 ___ ___- _____ 46 ___ .82 1.34 .320 5.8 .260 .0166 64 ___ .73 2.97 .244 18.2 .173 .0034 1.98 2.48 .254 ___ ____ _____ 1.35 4.88 .216 ___ _-__ _____ 2.87 2.66 .063 ___ ___- _____ 2.54 6.72 .125 ___. ___- _____ 4.75 4.36 .415 ___ ___- _____ 4.31 8.71 .108 ___ ___- _____ 47 ___ .54 .57 .144 7.6 .119 .0005 65 __- 1.04 3.42 .137 25.7 .121 .0005 1.69 1.54 .129 ___ ___- _____ 2.23 6.32 .117 -__ ___- _____ 2.58 1.96 .081 ___ ___- _____ 3.17 7.74 .081 ___ ___- _____ 3.42 2.48 .115 ___ ___- _____ 4.98 11.57 .133 ___ ___- _____ 48 ___ 1.13 1.94 .226 8.6 .182 .0015 66 ___. .94 1.54 .052 32.4 .048 .0001 2.02 2.89 .173 ___ ____ _____ 1.88 2.60 .037 ___ ___- _____ 2.85 3.62 .165 ___ ____ _____ 2.81 3.57 .036 ___ ___- _____ 4.77 4.63 .118 ___ ____ _____ 4.62 6.50 .059 ___ ___- _____ 49 ___ .87 1.56 .315 6.5 .252 .0037 67 n. .88 1.50 .071 24.8 .051 .0002 1.87 2.68 .257 ___ ____ _____ 1.88 2.65 .051 ___ ___- _____ 2.92 3.50 .230 ___ ____ _____ 2.75 3.51 .046 ___ ___- _____ 4.75 4.20 .145 ___ ____ _____ 4.88 5.07 .036 ___ ___- _____ 50 ___ .96 1.37 .234 6.8 .165 .0022 68 ___ 2.04 6.03 .105 25.2 .116 .0004 1.56 1.74 .118 ___ ____ _____ 3.23 7.72 .078 ___ ___- _____ 2.60 2.34 .121 ___ ___- _____ 5.00 11.18 .125 ___ ___- _____ 4.44 3.36 .141 ___ ____ _____ 6.96 13.14 .077 ___ ___- _____ 51 ___. .82 1.28 .164 10.2 .117 .0009 69 ___ 1.01 2.30 .228 11.2 .204 .0006 1.98 2.40 .116 ___ ___- _____ 1.76 3.59 .209 ___ ___- _____ 2.87 3.00 .090 ___ ____ _____ 2.95 4.98 .169 ___ ___- _____ 4.75 4.11 .089 ___ ____ _____ 4.72 6.65 .177 ___ ___- _____ 52 ___ .54 .60 .185 6.3 .130 .0021 70 ___ .78 2.35 .180 17.9 .143 .0011 1.69 1.45 .140 ___ ___- _____ 1.35 3.67 .156 ___ ___- _____ 2.58 1.83 .092 ___ ___- _____ 2.54 5.82 .138 ___ ___- _____ 3.42 2.15 .088 ___ ____ _____ 4.31 7.61 .091 -__ ___- _____ 53 ___ 1.13 1.73 .147 11,3 .107 .0009 71 ___ 1.04 2.76 .094 29.7 .084 .0003 2.02 2.52 .097 ___ ___- _____ 2.23 5.22 .080 ___ ___- _____ 2.85 3.03 .072 ___ ___- _____ 3.17 6.42 .053 ___ ___- _____ 4.77 4.13 .106 ___ ___- _____ 4.98 10.23 .099 _-_ ___- _____ 54 ___ .87 1,58 .351 6,0 .313 .0336 72 __- .94 1.18 .051 25.4 .053 .0004 1.87 2.52 .239 ___ ___- _____ 1.88 2.26 .049 ___. ___- _____ 2.29 3.36 .658 ___. ___- _____ 2.81 3.22 .046 -__ _-__ _____ 4.75 4.32 .184 _-_ ____ _____ 4.62 5.44 .095 ___ SUPPLEMENTAL DATA 45 TABLE 10.—Biocherm'cal-oxygen-demand data for the Ohio TABLE 11.—-Meom, variance, and caefi‘ieient of variation of the River (from Kothmwlararnan, 1.968) and results of the reaeratlon coefilcient data of Churchill, Elmore, and Buck- data analysis—Continued 771ng (196.2) Obsgrva— T BOD K1 Lo f1 var . Num- i‘fi: (days) (mg/'1) (days'l) (mg/1) ( days—1) ( (1:53“) Eggs:- 1:? “310:1 ) FIE") C” ( K2) No. observs- ys ( day’s-3 ) 73 ___ .59 1.38 .184 13.4 .142 .0008 tions 1. 3 2.76 .130 ___ ___- _____ 21517 3.78 .108 ___ ____ ______ 1 ______ 19 2.920 2.741 0.56 4.28 5.94 .140 ___ ____ _____ 2 ______ 19 1.449 .027 .11 74 ___ .88 1.39 .086 19.0 .065 .0002 3 ------ 25 1.061 .152 .37 1.88 2.56 .069 ___ ___- _____ 4 ______ 29 .550 .051 .40 2.75 3.31 .054 ___ ____ _____ 5 ______ 29 .842 .139 .43 4.88 4.79 .047 ___ _-__ _____ 6 ______ 30 1.170 .092 .26 75 ___ 2.04 3.55 .110 17.7 .096 .0003 7 ------ 26 .315 .018 .43 3.23 5.03 .093 ___ ____ _____ 8 ______ 30 3.422 .414 .19 5.00 6.35 .062 ___ ____ _____ 9 ______ 16 2.819 .146 .14 6.96 8.19 .090 -__ ____ _____ 10 ...... 5 1.574 .023 .10 76 ___ .94 2.01 .233 10.2 .226 .0137 11 —————— 31 .505 .043 .41 1.69 3.75 .318 ___ ____ _____ 12 ______ 26 .420 .023 .36 3.88 5.32 .127 -__ ____ _____ 13 ...... 27 .300 .015 .41 4.65 6.74 .467 ___ ___- _____ 14 ______ 26 .660 .114 .51 77 ___ .73 1.63 .113 20.6 .082 .0003 15 —————— 18 .559 .017 .23 1.35 2.65 .089 ___ ____ _____ 16 ______ 7 .670 .085 .43 2.54 4.16 .074 ___ ____ _____ 17 ______ 20 1.309 .184 .33 4.31 5.92 .064 ___ ___- _____ 18 ______ 8 0.284 .006 .28 78 ___ .59 1.06 .089 20.7 .061 .0005 19 ------ 8 0.261 .016 .49 1.53 2.24 .066 ___ ___- _____ 20 ______ 8 1.896 .063 .13 2.47 2.70 .027 ___ ___- _____ 21 ______ 8 .870 .051 .26 4.28 4.38 .054 ___ ___- _____ 22 ______ 6 .934 .101 .34 79 ___ .94 2.39 .372 8.1 .314 .0029 23 —————— 7 .983 .135 .37 1.69 3.26 .220 ___ ____ _____ 24 ...... 7 1.006 .025 .16 2.88 4.66 .287 ___ ____ _____ 8 .557 .012 .20 4.65 6.06 .295 ___ ___- _____ 8 .903 .043 .23 80 ___ .73 1.20 .041 40.4 .036 .0000 16 5.858 3.289 .31 1.35 1.86 .027 ___ ___- _____ 16 1.812 .336 .32 2.54 3.29 .032 ___ ____ _____ 19 3.265 .285 .16 4.31 5.55 .035 ___ ___- _____ Average C.(K2)=0.307 81 ___ .94 1.34 .187 8.3 .178 .0005 1.69 2.30 .198 ___ ___- _____ 2.88 3.32 .157 ___ ___- _____ 4.65 4.42 .141 ___ ___- _____ 82 ___ .73 1.32 .170 11.3 .159 .0014 1.35 2.58 .218 ___ ___- _____ 2.54 3.98 .147 ___ ___- _____ 4.31 5.33 .115 ___ ___- _____ 83 ___ 1.04 2.96 .305 10.9 .295 .0045 2.23 5.20 .279 ___ ___- _____ 3.17 6.38 .247 ___ ___- _____ 4.98 8.80 .424 ___ ___- _____ 46 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS TABLE 12.—Correlation coeflicient between the biochemical-oxygen-demand and the dissolved-oxygen concentration (from Moushegian and Krutchkoff, 1.969) Correlation coefficient, 1(BOD, C) K2 ............ 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 K. T : 1 day 0.1 __________ .70 .59 .52 .49 .46 .45 .43 .43 .42 .42 .41 .41 .41 .2 __________ __ .69 .61 .56 .53 .51 .50 .48 .47 .46 .46 .45 .44 .3 __________ .80 .70 .63 .59 .56 .53 .51 .50 .48 .47 .46 .45 .44 .4 __________ .81 __ .64 .60 .56 .53 .51 .50 .48 .47 .46 .44 .43 .5 __________ .81 .71 .64 .59 .56 .53 .51 .49 .47 .46 .44 .43 .42 K1 T : 2 days 0.1 __________ .67 .56 .51 .48 .46 .45 .44 .43 .42 .41 .40 .39 .38 .2 __________ __ .60 .53 .50 .47 .44 .43 .41 .39 .38 .36 .35 .34 .3 __________ .70 .58 .52 .47 .44 .41 .39 .37 .36 .34 .32 .31 .30 .4 __________ .68 _- .49 .41 .41 .38 .36 .34 .32 .30 .29 .27 .26 .5 __________ .65 .53 .45 .41 .37 .34 .37 .30 .28 .27 .25 .24 .23 K1 T : 3 days 0.1 __________ .63 .53 .48 .45 .43 .41 .39 .37 .35 .34 .32 .31 .29 .2 __________ __ .52 .46 .41 .38 .36 .33 .31 .29 .28 .26 .25 .24 .3 __________ .60 .47 .41 .36 .33 .30 .28 .26 .24 .23 .22 .24 .20 .4 __________ .54 __ .35 .31 .27 .25 .23 .21 .20 .19 .17 .17 .16 .5 __________ .49 .36 .30 .23 .21 .19 .19 .17 .16 .15 .14 .13 .13 K1 T = 4 days 0.1 __________ .60 .50 .44 .41 .38 .35 .33 .31 .29 .27 .26 .25 .24 .2 __________ __ .44 .38 .34 .30 .27 .25 .23 .22 .21 .20 .19 .18 .3 __________ .49 .37 .31 .27 .24 .21 .19 .18 .17 .16 .15 .14 .13 .4 __________ .41 __ .24 .21 .18 .16 .15 .13 .12 .12 .11 .10 .10 .5 __________ .35 .24 .19 .16 .14 .12 .11 .10 .09 .09 .08 .08 .07 K. T : 5 days 0.1 __________ .56 .46 .40 .36 .32 .30 .27 .25 .24 .23 .21 .20 .20 .2 __________ __ .3'7 .31 .27 .24 .21 .19 .18 .17 .16 .15 .14 .14 .3 __________ .40 .29 .23 .19 .17 .15 .14 .12 .12 .11 .10 .10 .09 4 __________ 31 __ 17 14 12 10 09 09 08 07 07 07 06 I5 __________ I24 .16 :12 :10 :08 :07 :06 :06 :05 :05 I05 :04 :04 SUPPLEMENTAL DATA 47 TABLE 13.—Dissol'ved-oxygen concentration data for the Sacra— TABLE 13.—Dvissolved-owygen concentration data for the Sacra- mento River (from Thayer and Krutchkofi’, 1.966) mento River (from Thayer and Krutchkofl', 1966')—Continued Dissolved Dissolved oxygen oxygen concentra- concentra- River T tion Rivet T tion mile 1 (days) (mg/l) mile 1 (days) (mg/l) 46.3 0.00 8.3 28.4 1.33 7.4 8.6 7.5 8.9 7.6 45.1 .05 8.2 27.4 1.52 7.3 8.7 7.0 8.6 7.6 43.4 .14 8.0 26.8 1.63 7.1 8.6 7.3 8.3 6.8 8.4 7.4 42.1 .22 8.1 25.5 1.96 7.3 8.0 7.5 8.4 6.8 8.5 24.3 2.09 7.2 41.1 .29 8.2 7.5 8.0 6.8 8.2 7.7 39.8 .39 7.9 23.3 2.25 7.6 7.9 7.1 8.0 7.3 8.3 22.3 2.43 7.2 38.6 .47 7.7 7.3 8.0 7.3 37.2 .58 7.9 21.1 2.65 7.2 7.9 7.7 35.9 .68 7.8 8.0 34.4 .80 8.1 7.5 33.5 .87 8.0 20.1 2.83 7.2 7.9 7.7 32.5 .95 7.8 8.2 31.6 1.02 7.9 7.6 7.7 18.8 3.03 7.2 30 1 1 15 5'3 5'8 . . . .2 7.5 17-5 4.2 8.1 7.7 3.5 .7 15.1 5,6 8.1 8.4 8.8 1 River kilometre=lriver mile X 1.609. MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS Read I?” '11, ,81, K2! “2! ’82, T(K1’ K2)! L0! La: 009 Ca, 089 DB; K3, T, m9 7" AT =T/n AK1 =\/,31TAT AKz =VBZTAT l alAKl (1~ ) P1 l KlLa _ _ = —-————— -K T —(K +K )T -3 7' B (Ea—El—K3)(I?1+K3)[e 2 — e 1 3 ]+ (08 — Co)e 2 DB _ Ca KlLa _ + _ + _ _ - 1 _ ‘K2T K2 K2(K1 + K3) 8 ] Call random number generator for R1; 89 FIGURE 22.——Flow chart for the computer program for the stochastic model for estimating the variance of the oxygen deficit. SUPPLEMENTAL DATA $3 Call random number generator for R2‘ 49 V V KM AT = K AT — AK1 KM AT = [Z AT + AK1 K21 AT = E AT + AK2 K2,. AT = 11 AT — AK2 K21 AT = E AT — AK2 K21 AT = 11 AT + AKz _ A K11 + ”—1 K11: (K%+1—Ka)-(K2«;—K3) Kw i= 1 K21_I(11_'I{3 n Dk=B+LoeXp[— 2 KZiAT] i=1 (Kai—Ku—Ks) (Kzi+1—K1¢+1—Ks) ' i K1,, n K2,_K,_K,, AT + —__—_ - ex K2,.—K1.—Ka AT exp [Efl . , ) ] KVKWKS P[.2< , a > H (15 FIGURE 23.—Continuation of the flow chart for the computer program for the stochastic model for estimating the variance of the oxygen deficit. 50 MODEL FOR PREDICTING THE DISSOLVED-OXYGEN DEFICIT IN STREAMS 9 G m Dk MeanD = 2 Ic=1 m {r m (D 2 Var (D) = 2 ———k)—— — [mean D]‘-‘ Ic=1 m Order Dk such that Dk < DH, for all k Write mean I), var (D), and Dk for k= 1, m Stop FIGURE 24.-——Completion of the flow chart for the computer program for’ the stochastic model for estimating the variance of the oxygen deficit. Z DAY (P O o 0 31/4 Geologlc ConSIderauons for Redevelopment Planning of Managua, Nicaragua, Following the 1972 Earthquake GEOLOGICAL SURVEY-PROFESSIONAL PAPER 914 Prepared in cooperation with the Government of Nicaragua and the Agency for International Development, U.S. Department of State OCT 16 1975 j; / \\ § 6" 4 - 3,. '77” SCIENCE “Bl/ . ,3“ .13UMENTS DEPARTMEN3~,3 OCT 1 5 1975 , i ufilfiiyfllqfim’fifikfym ream-”>53. '5 > d 1975 Geologic Considerations for Redevelopment Planning of Managua, Nicaragua, Following the 1972 Earthquake By HENRY R. SCHMOLL, RICHARD D. KRUSHENSKY, and ERNEST DOBROVOLNY GEOLOGICAL SURVEY PROFESSIONAL PAPER 914 Prepared in cooperation with the Government of Nicaragua and the Agency for International Development, US Department of State The geologic framework of the area around Managua provides the basis for identifying potential geologic hazards that should be considered in redeveloping the city UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON’: 1975 UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress Cataloging in Publication Data Schmoll, Henry R. Geologic considerations for redevelopment planning of Managua, Nicaragua, following the 1972 earthquake. (Geological Survey Professional Paper 914) Includes bibliographical references. Supt. of Docs. No.2 1 19.16:914 l. Seismology—Nicaragua—Managua (Dept.) 2. Volcanism—Nicaragua—Managua (Dept.) 3. Sediments (Geology)—Nicaragua— Managua (Dept.) I. Krushensky, R. D., joint author. 11. Dobrovolny, Ernest, 1912- joint author. III. United States. Agency for International Development. IV. Title: Geologic considerations for redevelopment planning of Managua, Nicaragua . . . V. Series: United States Geological Survey Professional Paper 914. QE53S.2.NSS35 624'.151’o97285 75—619060 For sale by the Superintendent of Documents, US. Government Printing Office Washington, DC. 20402 Stock Number 024—001—02709—7 CONTENTS Page Page Abstract ........................................................................................... 1 Physical characteristics of geologic materials—Con. Introduct1on """"""""""""""""""""""""" 1 Alluvium .................................................................................... 1 1 Geologic selling~ ----- 3 5011 .................................................................... 12 Volcanism """""""""""""""""""""""""""""""""" 3 Summary ........................................................... 12 Centers 0‘ volcamc eruptions. """""""" 3 Faults and other linear geologic features.... 12 Masaya ................................................................................ 3 Linear features in the Managua area... 14 Apoyeque...: _ ' 5 Tiscapa set ..................................... 15 Asososca, Tiscapa, and the Nejapa-Ticomo collapse Las Nubes set ............ 17 pits and associated cones ................................................ 6 Nejapa—Tipitapa set.. 18 TalpelaS-Miraflores --------------------------- 7 Cofradia set ...... .................................. 19 Chronologic relationship 0f volcanic “1’05“? """"""""""""""" 7 Northwest-trendlng sets ....................................... 19 Historical and present volcanic act1v1ty 1n N1caragua """"""" 8 Major lineaments in Nicaragua ................................................ 20 Possible future volcanic activity ................................................ 8 Summary ________________ 21 Physical characteristics of geologic materials .................................. 9 Conclusions... 21 Loose tephra .................................................................... : ......... 9 Recommendations... 21 Partly indurated volcanic mudflow and ash-flow deposus. 10 References cited ................................................................................. 22 Hard volcanic rock .................................................................... ll ILLUSTRATIONS Page PLATE 1. Map of Nicaragua showing lineaments and other features seen on side-looking imagery .......................................................... In pocket FIGURE 1. Sketch map showing approximate extent of deposits from various volcanic centers ..................................................................... 2 2. Sketch map of southern Nicaragua, showing physiographic provinces and geologic units“ 4 3. Schematic and composite stratigraphic column illustrating the kinds and complexity of deposits 1n the Managua area 7 4. Same schematic and composite stratigraphic column as shown 1n figure 3, with the materials according to physical character- istics and, in part mode of deposition .................................................................................................................................... 9 5. Map of the Managua area showing faults, other linear geologic features, and damage zone .................. 13 6. Diagram showing approximate orientation of sets of linear geologic features 1n Nicaragua and surrounding areas .................. 14 7. Map of the Managua area showing zones dominated by sets of linear geologic features ............................................................... l5 8. Map of Managua and vicinity, showing location of faults and fractures that developed during the 1972 earthquake ................ 16 TABLES Page TABLE 1. Grain-size distribution and selected distribution parameters for three samples of partly indurated mudflow or ash-flow deposits. 10 2. Selected physical properties for two samples of partly indurated mudflow or ash-flow deposits .................................................. 10 III GEOLOGIC CONSIDERATIONS FOR REDEVELOPMENT PLANNING OF MANAGUA, NICARAGUA, FOLLOWING THE 1972 EARTHQUAKE By HENRY R. SCHMOLL, RICHARD D. KRUSHENSKY, and ERNEST DOBROVOLNY ABSTRACT A brief reconnaissance investigation of the geology in the vicinity of Managua, Nicaragua, was undertaken by the U.S. Geological Survey at the request of the Agency for International Development, U.S. Depart- ment of State. The objective was to determine whether any areas within about 15 km of the present city are better suited than others as sites for reconstruction, taking into consideration volcanism, geologic materials and structures, and seismicity. During the time that man has inhabited the area, several metres of materials from four volcanic centers have been deposited in the vicinity of Managua. These deposits include lava flows, mudflows, ash flows, and ash falls. Volcanic activity continues today; widespread and destructive deposition of volcanic materials similar to that of the past can occur at any time. Most of the materials underlying the Managua area range from loose or poorly consolidated pyroclastic deposits to partly indurated mudflow and ash-flow deposits; lava flows, alluvium, and soil are dominant in only small areas. The partly indurated deposits are widespread and have sufficient strength and stiffness to provide adequate foundations in most places; these characteristics are not likely to be altered by seismic shock. Linear geologic features, including faults, fractures, and lineaments, are present throughout the Managua area; they can be grouped, on the basis of orientation, into seven sets, most of which are directly related to the regional geologic structure. Faulting has occurred during the last 10,000 years along nearly all these trends; faulting and concomitant seismic activity, with varying frequency and magnitude along different trends, can be expected to continue. We conclude that no site within about 15 km of Managua is substantially better suited for the capital than the present site, because all such sites share to a significant degree the geologic hazards of the present city. Clearly, it would be desirable to locate a major city in a geologically more stable part of Nicaragua, and search for such a site with a long-term goal of at least partial relocation should be undertaken. For the immediate future it may be possible to cope successfully with the geologic hazards of the Managua area if appropriate reconstruction plans are developed and rigorously carried out. INTRODUCTION A major part of Managua, Nicaragua, was destroyed by a severe earthquake, associated aftershocks, and extensive fires early in the morning of December 23, 1972. Because of a history of seismic and volcanic activity in the region, the U.S. Agency for International Development (USAID) asked the U.S. Geological Survey to evaluate, within approximately 15 km of the city, geologic hazards that might be important in the reconstruction or relocation of the city. This report is the result of a 3-week reconnaissance field investigation in an area that extends roughly from Mateare on the west to Tipitapa on the east, and from Lake Managua south to the Cordillera del Pacifico and the Masaya caldera (fig. 1); this area will be referred to as the Managua area. Fieldwork involved (1) study and mapping of volcanic centers and their associated deposits, (2) study of the physical properties of these deposits, and (3) study and mapping of faults and other linear geologic features. A major part of the study was a review of all available geologic reports concerning the December 1972 earth- quake, most of which were only in manuscript form. Also used were an unpublished report and associated maps prepared by the Parsons Corp., Marshall and Stevens, Inc., and International Aero Service Corp. (1972) under contract to Catastro e Inventario de Recursos Naturales. The existing geologic literature was examined and is cited throughout the present report. Colored aerial photography of the Managua area at various scales was furnished by the National Aeronautics and Space Administration; black-and-white aerial pho— tography of the city, at scales of 127,000 and l:10,000, was furnished by the Inter-American Geodetic Survey (IAGS). Side-looking radar imagery, which was taken before the earthquake under contract to Catastro, was also used in the regional study. Nicaraguans who assisted the writers in the completion of this study are too numerous to list individually, and the names of many who assisted us in the field are unknown to us. But without their help a valid picture of events during and after the earthquake would not have been possible. Among those who contributed most significantly were Ing. Arturo Aburto Q., Servicio Geolégico Nacional de Nicaragua (SGNN), who gave his enthusiastic and unstinting assistance in the field; Ing. Juan Kuan S., Geologist, Catastro, who guided us on fieldtrips in the area; and Ing. Orlando Rodriguez M., Director, SGNN; Ing. Fernando Montiel, Director, Catastro, and Ing. Humberto Porta, Director, Instituto Geogrfifico Nacional de Nicaragua, who all gave full cooperation and invaluable logistic support. Mr. Leroy Anstead (IAGS) l 2 GEOLOGIC CONSIDERATIONS, MANAGUA, NICARAGUA, FOLLOWING THE 1972 EARTHQUAKE :550 000 mE 12 15 12 00’ 1320000 mN Lagu na (18 Acahualinca VD O‘Haiwujrm waist/y 860116, I 58 E 59 Puma Huete Punta Huete 0 Peninsula Lago de Managua Aeropuerto Las Mercedes Sabana OGra H e OCofradta Las Nubes {___ Masaya O caldera _ :»\\ Df/" 4 C \lC'EI‘RV '1184y4 / P l - 2 .3 [asaya \\\- 039‘, C o , ~ av \ 17 San Juana . /’ Laguna Apoyo \\ '7 de la Concepcion V “AP / Granrid‘y _ ‘ l 10 20 KILOMETRES I ‘ | I r . ,E' gsgg? Basaltic lava flows of 1670, and an earlier prehistoric Pumiceous unconsolidated ejecta, generally thicker than ’ lava flow, from Masaya caldera—youngest in age 1 metre, from Apoyeque ,1 4‘ f : Unconsolidated pyroclastic ejecta and local basaltic 32.11115: Unconsolidated pyroclastic ejecta from the Cerro Talpetas- ‘ 1 lava flows from Asososca, Motastepe, and Santa ' Miraflores center—oldest Anita AP Apoyo Volcano MV Masaya Volcano and associated craters AQ Apoyeque Volcano ND Nejapa depression AS Asososca Volcano NV Nindiri Volcano and associated craters; from west to east, CH Chiltepe Volcano San Pedro crater, N indiri crater, and Santiago crater CM Cerro Motastepe SA Santa Anita Volcano CT Cerro Talpetas TD Ticomo depression MF Miraflores TI Tiscapa Volcano FIGURE 1.—Sketch map showing approximate extent of deposits from General de Cartografi’a, Managua, Nicaragua l:250,000-sca1e topo- various volcanic centers. Unconsolidated pyroclastic ejecta, ash-flow graphic quadrangles: Managua, 1969, and Granada, Nicaragua; tuff, and mudflow deposits from the Masaya volcanic center, to— Costa Rica, 1968. Internal numbers around edge of map are the gether with minor lacustrine and alluvial deposits, extend over the 10,000-metre Universal Transverse Mercator grid, zone 16, Clarke entire map area; these deposits underlie and in part overlie deposits 1866 spheroid. represented by the map units shown in the explanation. Base from VOLCANISM 3 supplied us with aerial photographs, logistic support, and endless hospitality, and Ing. Glen Hodgson, Catastro, ably assisted us in the field for a few days. Particular thanks are extended to the Oceanic Exploration Co. of Denver, Colorado, which contributed copies of reports containing information not otherwise available. GEOLOGIC SETTING Nicaragua has been divided into four physiographic provinces that approximately coincide with major geologic units. These are the Atlantic Coastal Plain, the Interior Highlands, the Nicaragua Depression, and the Pacific Coastal Province (fig. 2). The Managua area, in central-western Nicaragua, lies at the southwestern edge of the Nicaragua Depression, which trends northwest and extends across Central America from the Caribbean Sea to the Pacific Ocean. A chain of active volcanoes, of which the northwestern part is designated the Cordillera Los Marrabios (Parsons Corp. and others, 1972, p. IV—15; McBirney and Williams, 1965, p. 29), lies within the, Nicaragua Depression near its southwestern margin. The volcanic chain is parallel to the trend of the depression and passes through the Managua area. The chain extends for over 300 km, from Cosegiiina Volcano in the northwest to Madera Volcano in Lake Nicaragua on the southeast (pl. 1); it may be considered part of a segmented volcanic arc that extends into Costa Rica. Of the 18 major volcanoes in the chain in Nicaragua, 11 have been active historically. Momotombo Volcano, about 40 km to the northwest, and Masaya-Nindiri Volcano (hereafter referred to as Masaya), 18 km southeast of Managua, are the active volcanoes nearest the city. Asososca and Tiscapa Volcanoes and a number of unnamed cones and collapse pits, all currently inactive, lie within the Managua area. (See fig. 1.) The Pacific Coastal Province lies southwest of the Nicaragua Depression. In the Managua area, this complex province includes the Cordillera del Pacifico, which lies about 10 km from the city (fig. 1). The Interior Highlands border the Nicaragua Depression on the northeast side, and lie about 40 km northeast of Managua. The Nicaragua Depression is underlain mainly by unconsolidated pyroclasts of various sizes (tephra) and by partly indurated deposits, such as ash-flow tuffs and mudflow deposits. Some of the material accumulated in streams and lakes. Hard volcanic rocks, such as lava flows, occur at the surface only in the immediate vicinity of the volcanoes. All the_volcanic materials are of Quaternary age—that is, younger than about 2 my (million years). Many of the surface materials are as young as several thousand years, and some are no more than a few hundred years old. The deposits of the Cordillera del Pacifico are similar in type and age to those in the Nicaragua Depression; hard volcanic rocks are generally absent from surface exposures. The Interior Highlands are composed largely of older volcanic rocks of various kinds and are Tertiary in age. The principal geologic structures of western Nicaragua are parallel to the trend of the Nicaragua Depression and the volcanic chain. These regionally important structures include faults, fractures, and undetermined linear features that may be faults or fractures, here called lineaments; they form both margins of the depression as well as the lineament—possibly a fault—along which most of the volcanoes are alined. A second major trend, known chiefly from radar imagery, strikes northeast. The principal lineament of this trend extends between the Pacific and Caribbean coasts and may mark a major fault zone. Other locally important structural features, including known faults, trend more nearly north. The 1972 earthquakes resulted from movement along northeast-trending structures; many of these structural trends appear to be of geologically recent development, and seismic activity along them may continue. VOLCANISM Volcanism has been the dominant geologic process in the Tertiary and Quaternary history of Nicaragua, and it continues unabated to the present. Emphasis in this report is placed on the volcanic geology of the Quaternary in the Managua area. The subject is treated under the following headings: (1) Centers of volcanic eruptions, (2) chronologic relationships of volcanic deposits, (3) histOrical and present volcanic activity, and (4) possible future volcanic activity. CENTERS OF VOLCANIC ERUPTIONS Material from four volcanic centers forms the bulk of the material visible on the surface in the Managua area. These centers, in order of the extent of their deposits, are (l) Masaya, (2) Apoyeque, (3) Asososca, Tiscapa, and the Nejapa-Ticomo collapse pits and associated cones; and (4) Talpetas and Miraflores. (See fig. 1.) MASAYA At present, the volcanic center at Masaya consists of a composite basaltic cone (Masaya Volcano on the east and Nindiri Volcano on the west). Interbedded basaltic lava and minor ash (pyroclastic material under 2.0 mm in di- ameter) largely fill the 11.6- by 6-km collapse depression, named the Masaya caldera by McBirney (1956, p. 83). The cone is capped by five vents, all of which have been active at some time since the early 16th century. Historic activity has been largely effusive, and all but one surficial lava flow—that of 1670—have been restricted to the caldera. 4 GEOLOGIC CONSIDERATIONS, MANAGUA, NICARAGUA, FOLLOWING THE 1972 EARTHQUAKE ‘ ,NICARAGUA 50 100 KILOMETRES EXPLANATION Qal ' I { Quaternary alluvium ”0v; Quaternary volcanic rocks, pyroclastic sediments, and minor alluvium Tertiary volcanic rocks—includes minor areas of Quater- nary volcanic rock Tertiary sedimentary rock Mesozoic and Paleozoic sedimentary, volcanic, and metamorphic rocks Contact between geologic units —— Approximate boundary of physiographic provinces :1: Principal Quaternary volcanoes ff) Area mentioned in text as site for further investigation for possible urban development —— D, Darl’o Plains; T, Teustepe—Monte Grande area Area of figures 1, 5, and 8 FIGURE 2.—Sketch map of southern Nicaragua, showing physiographic provinces and geologic units. Generalized from McBirney and Williams (1965), and selected additional data. VOLCANISM 5 Extra-caldera lava flows may be present in the subsurface, but are unknown. The flow of 1670 and an earlier pre- historic flow in about the same area moved to within 10 km of the present metropolitan area. Historic eruptions of gases and ash, although minor, have forced the intermit- tent abandonment of agricultural activities to the west of Masaya caldera. In contrast to deposits exposed at the surface, those of the precaldera volcano; here called proto-Masaya, comprise much of the substrate of the Managua area and the bulk of the Cordillera del Pacifico. The deposits can best be seen in the wall of the Masaya caldera, particularly the east wall near Laguna de Masaya and the southwest wall. The sequence in the eastern part of the caldera wall, from the lake level up, consists of (1) a slightly indurated massive tuff (consolidated ash), (2) a dense and massive basalt, and (3) a three-part sequence of bedded tephra. The term “tephra” is applied to all unconsolidated pyroclastic material ejected from a volcano and transported through the air and includes ash, Cinders, lapilli, bombs, and blocks, regardless of size and composition. Material exposed at the lake level consists of about 15 m of massive unsorted tuff composed of about 60 percent ash- size material (under 2.0 mm in diameter) and about 40 percent glassy, very finely vesicular basalt lapilli (2—64 mm in diameter), bombs (rounded fragments >64 mm in di- ameter), and blocks (angular fragments generally as large as bombs). All the larger fragments (0.5 to 40 cm) are dark gray, aphyric (without visible crystals), and angular. The lack of sorting and the massive character suggest torrential ash fall near a source vent. This tuff is overlain by about 30 m of dense dark-gray aphyric basaltic lava which is essentially like that of the larger pyroclastic fragments in the underlying tuff. Overlying the basalt is a sequence of about 30 m of unconsolidated lapilli- and ash-size tephra which can be divided into three units, each with a soil zone at the top. In each of the units, bedding is well developed and varies widely in thickness; each unit shows prominent crisscross bedding from bottom to top. Generally, near the base there are abundant dark-gray finely vesicular aphyric and angular bombs and blocks which occur in a matrix of poorly sorted ash and lapilli. Each unit becomes finer up- ward, ending in a very fine grained poorly consolidated pale-yellow-brown pisolitic tuff (containing concentri- cally layered, round accretionary pellets 2—10 mm in diam- eter). Upward graduation from coarse to fine tuff apparently indicates decreasing energy of eruption; the soil horizons, indicated chiefly by color changes at the tops of the pisolitic zones, suggest long periods of inactivity. The lowest soil zone lies about 7 m above the basalt unit and is about 1 m thick. The second tuff unit is 10-15 m thick and clearly was deposited on a deeply eroded surface; its soil zone is about 1.5 m thick. The third tuff unit, as much as 8 m thick, was also deposited on an eroded surface; its soil is about 60 cm thick and generally comprises the present surface. The upper two soil zones and intervening tephra can be traced by nearly continuous outcrops from Masaya Volcano into the Managua metro- politan area. The three soil zones can be traced westward into the San Juan de la Concepcién area (fig. 1) in the eastern part of the Cordillera del Pacifico. The bulk of the western wall of the Masaya caldera consists of medium to thick layers of well-sorted sharply angular, finely vesicular dark-gray glassy basalt cinders. This material can be traced directly into the Cordillera del Pacifico. Outcrops suggest that these mountains consist chiefly of tephra; they also contain irregularly distributed but generally poorly indurated blanketlike ash—flow tuff, mudflow, and minor alluvial deposits. This sequence of unconsolidated pyroclasts, ash-flow tuff, and alluvium was named Las Sierras Series by Zoppis Bracci and del Giudice (1958, p. 44). The name was later called the Las Sierras Formation (Karim and Chilingar, 1963, p. 104); present usage is Las Sierras Group (Parsons Corp. and others, 1972, p. IV—66). APOYEQUE Apoyeque Volcano lies approximately 10 km north- northwest of Managua in about the center of the Chiltepe Peninsula (fig. 1). Its low composite cone has a vertically walled summit caldera about 2.8 km in diameter. The youngest deposits on the rim of the caldera and the deposits blanketing the surrounding area consist of massive unsorted pumiceous tephra, typically pale gray, glassy, and highly vesicular. At the caldera rim the tephra commonly contains abundant clasts (fragments) ‘of tubular pumice as much as 25 cm across, and abundant clasts of dark-gray dense and apparently fresh basalt as much as 2 m across. The matrix consists chiefly of finer pumice clasts; it characteristically also contains common to abundant angular clasts of altered basalt that are limonite coated and veined. These angular clasts were apparently altered in the volcanic pile prior to the forma- tion of the caldera. By these characteristics the pumiceous tephra of Apoyeque can easily be distinguished from the pale-orange pumiceous tephra of Apoyo caldera (south- east of Masaya caldera); the Apoyo tephra does not con- tain basalt clasts. The Apoyeque pumiceous tephra crops out continuously westward to Mateare (fig. 1); it is as much as 4 m thick. Near Mateare it overlies a very well sorted thin-bedded medium-gray tephra that commonly shows graded bedding (beds that grade from coarser at the base to finer at the top). This graded material apparently was deposited in ancestral Lake Managua. West of Quinta Eva, and apparently west of the Mateare fault, the Apoyeque pumiceous tephra crops out near the top of the section and near the top of the ridge of the Cordillera del Pacifico. It can also be traced by continuous outcrops 6 GEOLOGIC CONSIDERATIONS, MANAGUA, NICARAGUA, FOLLOWING THE 1972 EARTHQUAKE southward to San Francisco, where it is 1.25 m thick. In the San Francisco area, soil at the present surface is developed to a depth of about 50 cm on the Apoyeque pumice. The same Apoyeque pumice in the Laguna de Acahualinca area is about 2 m thick. There, it overlies lacustrine tephra like that near Mateare and is overlain by tephra like that occurring in the Motastepe cone and the Asososca crater. Although the Apoyeque pumice is largely covered in the Nejapa-Ticomo area, it crops out again in the small cone at Santa Anita under about 20 m of slightly scoriaceous basalt and cinders. It was observed cropping out beneath the youngest pisolitic soil and tephra sequence of proto- Masaya as far south as the areas of El Hawi and Santa Ana, about 5.5 km south of Managua in the foothills of the Cordillera del Pacifico, and in “quarry hill,” southeast of Asososca Volcano. The easternmost outcrops of the pumice also lie beneath the youngest proto-Masaya deposits in the La Argentina area southeast of Managua. There the pumice is less than 1 m thick. ASOSOSCA, TISCAPA, AND THE NEJAPA-TICOMA COLLAPSE PITS AND ASSOCIATED CONES Asososca Volcano (fig. 1) lies on the western edge of Managua and appears to be closely related to a line of cones and volcanic collapse features in that area. Deposits of Asososca are largely covered by later deposits of proto- Masaya and by the ejecta of small nearby cones. The best outcrops lie north of Asososca in the unnamed ridges east of Bella Cruz and San Francisco. The lower part of the Asososca sequence consists of unconsolidated thinly bedded, generally well-sorted, well-rounded light-gray cinders with an abundant pale-gray matrix of volcani- clastic silt. Graded bedding is commonly present, and, with the characteristics noted above, suggests that this part of the sequence was deposited in a lake, probably ancestral Lake Managua. Interfingering with these lacustrine beds and higher in the section are medium- to light-gray poorly sorted, crudely bedded angular cinders—typical air-fall tephra. This material can be traced from the ridge east of Prinzapolka into the crater wall of Asososca. Pumiceous tephra of Apoyeque lies on the Asososca deposits at Xavier, on the north end of the ridge east of Prinzapolka. A small unnamed volcanic cone, now partly destroyed by quarrying operations, lies 550 m southeast of the Asososca crater. It consists, on the northwest and west sides, of a thick sequence of northwest-dipping well- bedded, well-sorted dark-gray and hematite-red cinders. The sequence is well exposed in roadcuts along Highway 2 and in quarries operated by the Ministry of Public Works. A l-m-thick bed of the Apoyeque pumiceous tephra occurs within a thick sequence of dark-colored cinders about 20 m below the top. As in the Talpetas area to the north and in the ridges east of Bella Cruz and Prinza- polka (fig. 1), a soil zone is developed on the cinders beneath the pumice. This soil in the cone, here called “quarry hill,” is about 3 m thick; because the soil zone at Talpetas is only 75 cm thick, and because conditions of soil formation for both appear to have been similar, a greater age for the “quarry hill” cone is suggested. The western surface of the “quarry hill” cone is covered by the younger pisolitic soil and tephra that extends from the rim of the Masaya caldera into the Managua metropolitan area. The bulk of the “quarry hill” cone consists of an olivine-pyroxene basalt or trachyandesite plug. In the quarries that border the hill on three sides, the rock is dense at depth and is vesicular within only 2—3 m of the generally flat top of the plug. A small cone composed of vesicular basalt, like that in the plug, caps the top of the western part of the plug. The bulk of the plug appears to have filled a volcanic vent about one-third the size of the present Asososca crater. Outcrops between the “quarry hill” cone and Highway 2 consist of interbedded dark-gray and hematite-red cinders and somewhat vesicular basalt lava flows like those in the plug. These dip southeast- ward, away from Asososca and into the “quarry hill.” They probably are part of the Asososca cone structure and suggest that the “quarry hill” and Asososca vents were active at the same time. Deposits of Tiscapa Volcano (fig. 1) are largely unknown outside the crater itself because the sur— rounding surface is covered by deposits of proto-Masaya and Apoyeque and because relief in streambanks cutting the cone is less than 2 m._Deposits in the crater walls, apparently from the Tiscapa vent, are predominantly poorly sorted thin- to thick-bedded angular cinders and ash of air-fall origin. One distinctive unit crops out at lake level in the northwestern half of the crater. It consists of a poorly consolidated massive and unsorted pale-gray tuff. The Nejapa-Ticomo pits and associated cones lie along a line about 600 m south-southwest of the Asososca crater; the line strikes north-northwest from the Ticomo depression through the Lake Nejapa depression and the small unnamed depression to the north (fig. 1). These depressions have been well described by McBirney (1955, p. 150—152) and McBirney and Williams (1965, p. 33-34) as collapse features that resulted from subterranean magma withdrawal. Examination of the surrounding generally flat area reveals no quantity of ejecta even approximately equivalent to the volume of the depressions. The Cerro Motastepe cinder cone, two smaller cones north of Colonia Molina, and a small composite cone at Santa Anita were the only eruptive centers late in the history of this area. Of these, the Motastepe vent had probably ceased eruption before collapse of the northernmost pit of the Nejapa depression because the cone surrounding that vent was partly destroyed by the collapse. Dark—gray cinders above the Apoyeque pumice at Xavier thicken southward and are not found north of Xavier, suggesting a source in the VOLCANISM 7 Asososca or Motastepe vents. The small composite cone at Santa Anita includes both dark-gray cinders of air-fall origin and slightly scoriaceous basalt flows above the Apoyeque pumice. TALPETAS-MIRAFLORES The Talpetas-Miraflores centers (fig. 1) consist of only two cinder cones. The trend, if such can be established by only two vents, is north-northeast; it appears distinctly offset from the Nejapa-Ticomo trend. The Talpetas cone is very small and consists of only dark-gray and hematite- red cinders with very few ash-size clasts. The cone is over- lain on the northwest by about 2 m of the Apoyeque pumice. At Miraflores, a low, apparently partly destroyed cone of dark-colored cinders is intruded by a light- pinkish-gray sparsely vesicular dacite(?) plug. The plug shows surface striations and deep channels or furrows typical of volcanic rock intruded in a nearly solid state. Cinders near the intrusive contact were fused and brecciated, apparently by the force and heat of the continuing intrusion. Evidence for the subsequent extrusion of this plug is lacking. Although no analyses are available, this intrusive may prove to be from the same magma source as the pumiceous tephra of Apoyeque 4 km to the northwest. CHRONOLOGIC RELATIONSHIP OF VOLCANIC DEPOSITS The relationship in time of the activity of volcanic centers and their deposits in the Managua area is indicated in the schematic and composite stratigraphic column (fig. 3). The thickness of units is largely unknown because stream gullies have barely incised most of the deposits. The duration of activity from any one center, with the possible exception of Masaya, and the age of the youngest deposits in years before the present are also unknown. Although carbonized wood is known (Juan Kuan S., oral commun., 1973) both above and below the Apoyeque pumiceous tephra, we found none during the course of our study. A total rock potassium-argon age (Parsons Corp. and others, 1972, table IV—lA) on the lowest basalt over— lying the rim of the Masaya caldera is given as 95,000+45,000 years B. P., but the date is imprecise because of the low total potassium content (2.501 percent). Activity from the Masaya center appears to have continued from the late Pliocene to the present, if basal deposits of the Las Sierras Group, composed of proto-Masaya tephra, are correctly correlated with the upper part of the El Salto Formation (Parsons Corp. and others, 1972, p. IV—66) of late Pliocene age. The age of the three pisolitic soil zones at the top of the proto-Masaya section is unknown. Because individual soil horizons were formed over wide areas at a particular time, both the overlying and underlying units can be stratigraphically related to the soil zones and to each other. Formation of a Krakatoan-type caldera like v Ill-I'll.- ”‘VVVVV" " _a -. --.---...—- C— °o°g= a- abs "0' .° —C x ”//r I 1° 6— ///,,I/ ’11” d_v\/Vvv1/ / , 1,1/’ 6 —-—IA, _ //////’/' Ill/n/ll/ __ 1H ,, WI 6 //,/’ I1‘,¢"ll// _ , ////// I; .o‘.°'0._ f ,, .,0‘,,, 9 z- . M “‘72; ——h I I O ' W. .7— [0 o‘o,° -°/ +n1., , 'IZOI/I/ —k ~O'G-D‘oioio-D‘K3 V V VV .0' .0. m’/ //%c' k_////////.-»'O'° o' 0. "“.°"°'a- '°'° EXPLANATION Soil zone Alluvium Pisolitic tuff Tephra Mudflow and ash-flow deposits v v Basalt flow Intrusive plug FIGURE 3,—Schematic and composite stratigraphic column illustrating the kinds and complexity of deposits in the Managua area; neither thickness of deposits nor extent of time is implied in this diagram. Deposits from various volcanic vents combined into one eruptive phase are separated by heavy lines in the diagram and are listed here in approximate chronologic order, youngest first; numbers in parentheses refer to representation in map units of figure 1, “0” indicating not shown in that figure and “5” indicating present every- where in that figure: a, postcaldera lava flow from Masaya and Nindira (1); b, alluvium along present-day streams, generally re- worked deposits from Masaya (0); 6, upper tephra with pisolitic tuff and soil zone, from proto-Masaya (5); d, basaltic lava flow from Santa Anita (2); e, tephra from Motastepe, Asososca, and proto- Masaya (2 and 5); f, pumiceous tephra from Apoyeque (3); g, tephra, mudflow, and ash-flow deposits from proto-Masaya (5); h, fossil footprints of Homo sapiens (0); i, tephra from Talpetas, Asososca, and “quarry hill” (2 and 4); j, Miraflores intrusive plug (4); k, tephra, ash-flow tuff, mudflow deposits, and basalt from proto—Masaya (5); also includes minor alluvial deposits. that of Apoyeque—that is, formation of a summit depression through the explosive eruption of voluminous ash and the consequent collapse of the upper part of the cone into the evacuated magma chamber within a period of a few days—suggests that the pumiceous tephra of Apoyeque was deposited over a short period of time throughout its area of deposition. The date, in years before the present, of the formation of the Apoyeque caldera is also unknown, but it was apparently within Holocene time (about the last 10,000 years) because the Apoyeque pumice overlies fossil footprints of Homo sapiens by 8 GEOLOGIC CONSIDERATIONS, MANAGUA, NICARAGUA, FOLLOWING THE 1972 EARTHQUAKE about 1.3 m. One occurrence of these footprints has been preserved by the Nicaraguan Government near Lake Acahualinca on the northwestern margin of Managua (Williams, 1952). Volcanic deposits that postdate the presence of man in what is now Nicaragua are at least 3.3 m thick near Laguna de Acahualinca and are marked at the top by the uppermost pisolitic soil that makes up the top of the proto-Masaya deposits. A rough estimate of the age of the youngest pumiceous tuff from Apoyeque caldera was made by Virginia Steen- McIntyre (U.S. Geological Survey, written commun., 1974), who used the degree of superhydration of the glass. In this method the amount of liquid in selected vesicles of glass shards from the tuff is estimated and compared with the amount in glass from dated tuffs of similar compositon. In general, the older the glass, the more water present in the vesicles (Steen—Mclntyre and others, 1973; Steen-Mclntyre, 1975). Hydration and superhydration of the youngest pumiceous tuff from Apoyeque suggest an age between 1,000 and 5,000 years, probably closer to 1,000 years. For a similar-appearing tuff from Santa Anita that shows a slightly different set of heavy minerals, the suggested age is between 5,000 and 15,000 years, probably closer to 5,000 years. HISTORICAL AND PRESENT VOLCANIC ACTIVITY IN NICARAGUA An excellent and detailed summary and bibliography of historical volcanic activity has been given by McBirney (1958, p. 109—134). This short summary is, in part, abstracted from his compilation. The volcanoes discussed below are shown on plate 1. Cosegiiina Volcano is the northernmost volcano in Nicaragua. According to word-of—mouth history of the family that had owned the surrounding area prior to 1830, the volcano showed no signs of activity preceding the Krakatoan-type eruptions and subsequent collapse of the summit in mid-1835. Activity since then apparently has been restricted to fumarolic emissions within the crater lake. San Cristobal and Casita were known to be active volcanoes in the early 16th century, but there are no con- firmed reports of activity other than emission of steam and other gases since that time. Telica has shown almost continuous fumarolic activity and occasional eruptions of ash since the early part of the 16th century. Santa Clara, an apparently parasitic cone on the southern flank of Telica, was also active in the early 16th century. Cerro Negro began erupting ash and cinders in April 1850 and has had at least 10 major eruptions since that time. The last one was in 1971, when spectacular explosive eruptions of ash blanketed Leon, 22 km to the west—south- west. Cerro Negro has been in a fumarolic stage since then, but between September 1972 and February 1973 (Alain Creusot-Eon, SGNN, oral commun., 1973) gas temperatures within the crater increased from 500° to 1200°F. (245° to 635°C). The complex of cones and vents that caps El Hoyo (Las Pilas) Volcano showed no eruptive activity prior to October 1952, at which time a north—trending fissure opened in the summit area and emitted minor quantities of lithic(?) ash. A second ash eruption occurred in 1954. Current activity is limited to strong emission of sulfurous gas from one restricted part of the fissure on the southern lip of the cone, near the pit that gave the volcano its name. Momotombo Volcano, on the northwest shore of Lake Managua, has been active at least since the time of European settlement in the 16th century. It has had about eight major periods of explosive eruption; that of 1905(?) also effused a major basaltic lava flow from the summit vent. Activity is now limited to fumarolic emissions from the summit crater. Masaya Volcano also has had a long history of activity. Lava eruptions from summit craters occurred in the 16th, 17th, and 18th centuries. The lava flow of 1670, or perhaps an earlier prehistoric flow, moved to within 10 km of the present metropolitan area. Chief activity for the last 300 years has been emission of gases. Lava lakes are apparently a recurring feature of the various vents (McBirney, 1956, p. 88—90). The last known lava lake, observed by one of us (RDK), filled the lower pit of the Santiago crater in early 1967. The lake surface has since solidified, and the column of magma has been withdrawn to a lower level. A circular collapse pit now occupies the center of the old lake, where strong fumarolic activity originates. Mombacho is known to have erupted in 1560, and it may have erupted as late as 1850. Concepcion Volcano on the island of Ometepe has produced numerous gas eruptions and one lava flow on the northeast flank of the cone within historic time; at present the volcano shows mild but nearly continuous fumarolic activity from the summit vent. POSSIBLE FUTURE VOLCANIC ACTIVITY None of the Quaternary volcanoes in the northwest- trending chain of volcanoes in western Nicaragua can be considered extinct, and all the major cities in the country must be considered potentially endangered by volcanic activity. Future explosive eruptions from the currently active Masaya Volcano will doubtless recur, and the cities of Masaya and Managua could be made uninhabitable by deposition of only a few metres of ash and cinders from this or other volcanoes. Similarly, renewed eruptions of molten lava from the summit vents or from the floor of the Masaya caldera could easily destroy structures over large areas and make parts of the Managua metropolitan area at least temporarily uninhabitable. Ash flows produced during nuée ardente eruptions—that is, explosive eruptions of predominantly fine gas-buoyed, very hot (as much as 800°C.) ash, commonly move at great speed. V PHYSICAL CHARACTERISTICS OF GEOLOGIC MATERIALS 9 Movement of such flows over the metropolitan areas of either Managua or Masaya would probably occur too rapidly to allow any significant evacuation of either city. Losses in life and property in such event would be very high. Similarly, mudflows, either in conjunction with explosive ash eruptions or following such eruptive periods, especially during times of hard rains, would probably move too rapidly to permit significant evacuation of the threatened areas. Recent examples of mudflow activity were seen in Costa Rica during and following the eruptions of Irazfi Volcano in 1963-65 (Krushensky, 1972). The city of Cartago was flooded by a number of mudflows, with a consequent great loss in property and life. During nuée ardente eruptions of his- torically quiescent Arenal Volcano in 1968, a large number of people were killed, and nearby structures and crops were burned (Melson and Saenz R., 1968). Asososca, Tiscapa, and the small vents associated with the Nejapa-Ticomo depressions lie within and close to Managua, and the possibility of renewed eruptions from them cannot be discounted. It should be emphasized that deposits from these vents and from Apoyeque Volcano postdate the presence of man in what is now the metro- politan area. Activity in these vents is therefore probably no older than 25,000 years and possibly is as young as a few thousand years. Eruptions that deposited the pumice of Apoyeque in the metropolitan area, and eruptions that led to the formation of the Apoyo caldera, 35 km southeast of Managua, were of the same type as the 1835 eruption of Cosegiiina Volcano, which “was perhaps the mostviolent eruption within historic times in the Americas” (Williams, 1952, p. 7—8). Ash blotted out the sun within an approximate radius of 150 km around the volcano, and the noise was heard in Jamaica and as far south as Bogota, Colombia (McBirney, 1958, p. 110). Such eruptions do take place in apparently extinct volcanoes and could occur in‘or near the Managua area again. PHYSICAL CHARACTERISTICS OF GEOLOGIC MATERIALS The geologic materials in Managua and the surrounding region, previously discussed in terms of their origin as deposits from various volcanic centers, may also be grouped according to their principal physical characteristics. These characteristics determine, among other things, how the deposits perform as foundation materials and how they respond to seismic shock. Five categories of materials may be differentiated; they are dis- cussed in the following sequence: (1) Loose tephra, (2) partly indurated volcanic mudflow and ash-flow deposits, (3) hard volcanic rock, (4) alluvium, and (5) soil. The first three of these materials are, by far, the most prevalent; alluvium is limited in occurrence, and soil is generally thin. The relationships of these five categories of deposits are illustrated in figure 4. In this figure the deposits shown in figure 3 are regrouped according to physical characteristics and, in part, mode of deposition. v’o‘o » v - .:.3.:.2.2.:.3.:.2Wo2o:« .¢ EXPLANATION % Residual soil and some windblown material — Generally fine-grained soft material Alluvial deposits along present-day streams — Fine and coarse-grained material that is generally loose .. 0 ° Loose tephra of various kinds 'l/A Partly indurated volcanic mudflow and ash-flow depos- its equivalent to soft rock — Known locally as “piedra de contera” (quarry rock) Hard volcanic rock FIGURE 4,—Same schematic and composite stratigraphic column as shown in figure 3, with the materials here shown by pattern ac- cording to physical characteristics and, in part, mode of deposition. LOOSE TEPHRA Tephra deposits are present in most of the exposures that we visited. Most of these materials were deposited directly from the air, after having been emitted from volcanic vents. These deposits generally range in thick- ness from 2 to 4 m, but closer to the vents from which they erupted, the thicknesses increase to a few tens of metres The deposits range widely in grain size and in sorting (that is, degree of sameness of grain size; well-sorted deposits have grains of nearly the same size, whereas poorly sorted deposits exhibit a wide variety of grain sizes in a single deposit). Some relatively well sorted beds consist of ash that is dominantly silt size, and others are lapilli in which particles range in size from a few to a few tens of milli- metres, somewhat like fine gravel. The more poorly sorted materials may contain scattered lapilli in a matrix of ash. Varying degrees of cohesiveness are characteristic of these deposits, but they are dominantly fairly loose. Some of the well-sorted deposits are very loose, having practically no cohesion between particles. Other well- sorted deposits stand in vertical faces, perhaps because of greater packing and interlocking of the rough surfaces of ‘ 10 GEOLOGIC CONSIDERATIONS, MANAGUA, NICARAGUA, FOLLOWING THE 1972 EARTHQUAKE the grains, although they probably have little chemical cementation. The poorly sorted deposits also generally have some degree of cohesiveness, perhaps caused in part by the wide range of grain sizes, fine grains filling in voids between the coarser grains. While there is a range of degree of cohesion in the deposits considered here, as a group they are considerably less cohesive than the somewhat indurated mudflow and ash-flow deposits discussed below; they are probably somewhat more cohesive than the better rounded, better sorted alluvial deposits of comparable grain size. The bearing capacity of these materials is likewise inter— mediate between that of alluvium of comparable grain size and that of the more indurated deposits; test data on the bearing capacity of these materials are not now available to us. PARTLY INDURATED VOLCANIC MUDFLOW AND ASH-FLOW DEPOSITS Generally underlying the relatively noncohesive tephra deposits are the more indurated volcanic deposits that are known locally by the term “piedra de cantera” (quarry rock). They represent principally mudflow and nonwelded ash-flow tuff deposits. They are similar in some ways to the more poorly sorted materials described under “Loose Tephra,” in that they also range widely in grain size. In places they contain rounded pebbles and cer- tainly represent mudflows. Elsewhere, angular volcanic fragments are dominant, and coarse fragments of any kind are not common; these deposits may represent either mud- flows or ash flows. Table 1 shows size distribution of three samples of this material. A plot of mean grain size (Mde) and sorting (a d>) parameters shows that these materials have characteristics similar to materials that probably were deposited by flushing of an ash cloud by rain, according to Walker (1971, p. 701). Sample C contains accretionary lapilli, which also tend to support this mode of origin. TABLE 1.—-Grain-size distribution and selected distribution parameters for three samples of partly indurated mudflow or ash—flow deposits [Determined by P. 5. Powers, US. Geological Survey, Denver, Colo.] Grain-size distribution Distribution (percent) parameters2 Sample‘ Clay Silt Sand Gravel Mean grain$ 10g" (181‘); A VO'Cano 1 d 11 u llll I Jiloa //‘Q, [ago 6 . anag a \ I M. 4,9 YR 4 I . ll? \\ I ' 43‘ , v, Rt? 6° a 31* 311.321 { / Laguna TISCAPA 5;. \O \n $63119 500 de Acahualinca FAULT a O \\ \ \\ Bella CruzOX 4' kgfl \ ~ g. xx \ . 4 an errfido I x \ . 3% \ \ ”1&1”qu 1)} £4 3/ ; a x \,\( \\ \\ \\ y “will... ’ J . ’3' fi/\ Xx ”4 A? \\ \ Mia/fie A???“ 1172/, / X} Mira/defi/I'Q‘sczpa/ K - / ( \\/ .'//// // ’/ 6‘ \\ 9‘ ' flit/m)?” olilna.'/////// / Lagtafz‘a de ’9 / ,” Q J WWII/51L; Arge [final/HI / I zsma ‘7 / \\ /// H7 {all asay a, 6‘ (2 fieldera/ £|\\ [97a M saya k vuzzsaze‘evi'm #4 1.2341494? 12 00 ’ / ~ 3' J - ' , 0“ ,/ :-. an uan o LAN .... L3 2" ”"0 W I// ...;:/;1‘P‘5 gfiafi, xx xde la Concepcion / f“ P 46:.— x "u x . i 0 1O 20 KILOMETRES IIII1 l | EXPLANATION FAULTS — Dashed where inferred or concealed + From previous mapping — Chiefly by Kuang and Williams (1971), Ferrey (1971), Ferrey and Williams (1971), and Sultan (1931). U, upthrown side; D, downthrown side —-0-\—o— Nejapa fault zone — Described but not mapped by McBirney and Williams (1965) Faults mapped following 1972 earthquake — In and near Managua generalized from figure 8; elsewhere, from field mapping by us OTHER FRACTURES AND LINEAMENTS -'"'-"‘“°" Fractures — From mapping by Kuang and Williams (1971), Ferrey (1971), and Ferrey and Williams (1971) Lineaments — Mapped by us from mosaics of side-looking radar imagery at scales of 1:100,000 and 121,000,000 (pl. 1) ————— Punta Huete lineament (pl. 1) — Mapped by us from 1:1,000,000-scale radar imagery. Extends for hundreds of kilometres but is not identifiable at every point along the line shown here . , .. DAMAGE ZONE OF 1968 EARTHQUAKE —— Described but not mapped by R. D. Brown, Jr. (written commun., 1968) FIGURE 5.—Map of the Managua area showing faults, other linear geologic features, and damage zone, compiled from various sources. 14 GEOLOGIC CONSIDERATIONS, MANAGUA, NICARAGUA, FOLLOWING THE 1972 EARTHQUAKE NORTE NORTH Tupitapa Iineament Neiapa—Tipitapa ESTE OESTE WEST EAST 3 2 1 1 2 3 FIGURE 6.—Approximate orientation of sets of linear geologic features in Nicaragua and surrounding areas. (1) Faults, fractures, and other lineaments in the Managua area, named sets as shown and described in figures 5 and 8. (2) Lineaments seen on 121,000,000-scale countrywide mosaic of side-looking radar imagery lettered sets (A—F) as shown in figure 7. (3) Major tectonic features shown on tectonic map of North America (King, 1969). very brief field reconnaissance that extended beyond the Managua area. Faults, fractures, and other lineaments appear in great abundance throughout most of the Managua area. They are illustrated in figure 5, where they have been dis- tinguished according to type of feature and source of information. The linear features are not randomly dis— tributed in orientation but may be grouped into seven sets, each with a characteristic orientation distributed over a range of about 10°. This grouping of linear features into sets is illustrated in figure 6; for convenience of reference, each set has been identified informally by a local name. To compare the features in the Managua area with those throughout Nicaragua, we examined the l:l,000,000-scale countrywide mosaic of side-looking radar imagery. The pattern of lineaments found is shown on plate 1. As in the Managua area, the Nicaragua lineaments may be grouped according to orientation. These groups are likewise shown in figure 6; there are six sets, referred to by letter. Numerous circular to elliptical features are also shown; some of these are the modern volcanoes, others may represent older volcanic centers, and the remainder are probably of other origins. We have not attempted further identification of these features. Plate 1 is not intended to be a thorough analysis of the imagery but merely indicates the kinds of patterns present. Before the data shown on plate 1 are used in the resolution of specific problems, they should be checked thoroughly on the ground and related to geologic maps to establish their geologic significance. Figure 6 also shows the orientation of major structural elements of the Earth’s crust in the region of Central America surrounding Nicaragua, taken largely from the tectonic map of North America (King, 1969). These features are indicative of the pattern within which con- tinuing major geologic activity in the region occurs, and local geologic structures that can be related to them may be interpreted in terms of the regional geologic activity. We will discuss first the linear features in the Managua area shown in figure 5, and then some of the major linea- ments shown on plate 1, as well as the relationship between the local, national, and regional features illustrated in figure 6. LINEAR FEATURES IN THE MANAGUA AREA The linear features of each set in the Managua area are dominant in one or more parts of the area and generally not present in the other parts. Figure 7 is a map showing the distribution of zones in which each set is dominant. There is some overlap of the sets, and in places some of them seem to grade into others, but in general each set dominates distinctly separate areas. This relationship makes it possible to apply the characteristics of each set to particular parts of the Managua area. The discussion that follows will consider each set-in terms of its charac- FAULTS AND OTHER LINEAR GEOLOGIC FEATURES 15 ssjws' 86"}00' as oooom 15 :5 l5 16 j 6 E , 8 9 0 {J 1 s Punta Huete // Punta Huete Peninsula W/ 0 $7) 1;- ,, , ...... Chiltepe . Volcano l , Lago de Managua l 4’4) If a 819 a, “a 9a g < a. < ': E r.- E at < < < 9'3 n. G ‘0'” < m U 2 0 <2 3 l l— 3 4 2 w 0 j j 12,_ .L 00 1320000 MN 0 10 FIGURE 7.—Map of the Managua area showing zones of these zones are based on the linear features shown in figure 5. teristics, its relationship with the comparable country- wide set and with regional features, and its significance in terms of local seismicity and possible effect on man’s development of the area. TISCAPA SET After the 1972 earthquake several workers independently identified lines of ground breakage extending across and south of the city of Managua. These lines constitute a set of fractures which, together with related features discussed below, we here term the “Tiscapa set,” after the Tiscapa fault mapped by Kuang and Williams (1971). In figure 8 we have compiled the results of four studies, by Plafker and Brown (in Brown and others, 1973), by Rodriguez and Martinez in an un- dominated by sets of linear and lineaments with the characteristic orientation shown here and in figure published report by the Servicio Geolégico Nacional de ORIENTATION OF SETS OF LINEAR GEOLOGIC FEATURES 1 2O KILOMETRES | geologic features. Each set consists of faults, fracfures, 6. The establishment of these sets and the construction Nicaragua (1973), by Juan Kuan S., Catastro (written commun., 1973), and by Federico Fiedler, Oficina Nacional de Urbanismo (written commun., 1973; the last two references are manuscript maps. Although there was some disagreement among workers concerning precise locations of postearthquake features shown on the various maps, our compilation shows that there is, in fact, general agreement on location of these features. Areas of disagree- ment are restricted chiefly to places where the evidence is inconclusive. It is also likely that while each of the workers saw most of the same evidence, there was some evidence seen by only some of them. At least nine zones of fracturing have been identified; on four of these, displacement of the ground has been deter- mined, making them faults (Brown and others, 1973). Minor displacement may have occurred along some of the I6 GEOLOGIC CONSIDERATIONS, MANAGUA, NICARAGUA, FOLLOWING THE 1972 EARTHQUAKE 86715' S I 1 I I 12° 4345 75 10’ Laguna de ;' A cah ualinca O ,' .' Lago de Managua , Laguna de Tiscapa 5801 | 1 1 2 KI LOMETR ES ‘ _ ' |__I_I__I__I_L_.__—l ”\ l I I I | I I I I I Internal numbers around the edge of the map are the 1000—metre Universal Transverse Mercator grid, zone 16, Clarke 1866 spheroid EXPLANATION FAULTS AND FRACTURES -—- From maps by George Plafker and R. D. Brown, IL, US. Geological Survey --—- From maps by Ing. Orlando Rodriguez M. and Ing. Maximiliano A. Martinez H., Servicio Geolégico Nacional de Nicaragua -—---— From maps by Ing. Juan Kuan S., Catastro From maps by Ing. Federico Fiedler, Oficina Nacional de Urbanismo AREA WHERE TWO OR MORE FAULTS AND FRACTURES ARE LOCATED WITHIN ABOUT 100 METRES OR LESS OF EACH OTHER --------- BOUNDARY OF BUILT-UP AREA - As shown on 1:50,000 scale Managua quadrangle, second edition, 1971, Instituto Geografico Nacional, from which the base for this was prepared FIGURE 8.—Map of Managua and vicinity, showing location of faults and fractures that developed during the 1972 earthquake. others, or, in this earthquake, they may have served only as zones of fracturing rather than of faulting. Displacement along one of the zones is known to have occurred during the 1931 earthquake (Sultan, 1931); we consider it probable that all nine zones are true fault zones along which displacement has occurred at some time and along which it may occur in the future. These features generally trend about N. 30° to 40° E., are spaced from slightly less than one-half to somewhat more than 1 km apart, and have been mapped on the surface for lengths of about 4—6 km. Geophysical evidence indicates that at least one zone extends beneath Lago de Managua for another 6 km (Brown and others, 1973); it is likely that all the faults and fractures actually extend for distances greater than those FAULTS AND OTHER LINEAR GEOLOGIC FEATURES 17 mapped. Horizontal displacement of 2—38 cm and very much smaller vertical displacement have been reported by Brown, Ward, and Plafker (1973). We will not discuss these features further here, because they have been mapped and (or) described fully in the reports cited. Beyond the limits of figure 8, we found lineaments in contiguous areas to the east and south that trend approximately parallel to the fractures and faults shown in figure 8. These lineaments, together with the features in the city and vicinity, define the principal zone of the Tiscapa set. The lineaments are also similar in spacing and length to the faults within the city. We found no evidence of faulting along these lineaments, but there are few places where such evidence could be observed. At one place we observed small ground cracks similar to those found in the city; at a nearby exposure there was slightly disturbed bedding that might suggest the presence of a fault along which there could be minor horizontal dis- placement but no vertical displacement. This relative difference between amount of horizontal and vertical dis- placements is the same as that observed in the city and vicinity. The kind of evidence used in the city to establish the presence of the fault and fracture lines is quite ephemeral, and it is likely that evidence of fracturing and minor faulting that occurred during older earthquakes would be difficult to find; the lineaments that we did find could be all that remains to be seen on the surface of the ground. Figure 7 shows other places in which lineaments parallel to those of the Tiscapa set are found. One zone is on line with the trend of the linear features in the city but is southwest of the crest of the Cordillera del Pacifico. Only a few lineaments were found here, but the position of this zone suggests that it is a continuation of the principal Tiscapa zone, interrupted by zones of linear sets trending northwest, parallel to the Cordillera del Pacifico. If this relationship is valid, it would seem that the Tiscapa set is older than the three northwest-trending sets; thus, activity in the city of Managua may represent renewed activity along an old established pattern of fractures. Two small zones near the Masaya caldera that show lineaments parallel to the Tiscapa set are mapped as part of that set. The relationship here is less clear, but if the Tiscapa set is part of an old major system of faulting, these zones could also be remnants of it. The structural homogeneity of the principal zone occupied by the Tiscapa set of linear features, both within the city and in contiguous areas, suggests that seismic activity similar to that of the 1972 earthquake within the city area has a good likelihood of occurring anywhere within this zone. One of the 1972 fracture zones coincided with the fault that was active during the 1931 earthquake. Ground breakage was not observed following the smaller 1968 earthquake, but the area of damage, as described by R. D. Brown, Jr., (written commun., 1968), occurs along a trend that coincides with lineaments of the Tiscapa set. (See fig. 5.) We thus conclude that any place within the entire principal zone of the Tiscapa set has about the same chance of being subjected in the future to the type of seismic activity that occurred during the 1972 earthquake. On the basis of the historical record of seismic activity, this zone probably will continue to have a higher frequency of occurrence of earthquakes than any other part of the Managua area. LAS NUBES SET The Las Nubes fault (fig. 5), shown on the 1150,000-scale geologic map of the Managua quadrangle (Kuang and Williams, 1971), is the principal structural element of what is here termed the “Las Nubes set.” There are three zones in which the Las Nubes set is dominant: (l) in the same general area as the Las Nubes fault, (2) in a nearby zone to the northeast, and (3) in another zone southwest of Mateare. The geologic map of the San Rafael del Sur quadrangle (which we have seen only in a reduced version) shows a series of closely spaced fractures in the vicinity of Casa Colorada; these fractures are also roughly parallel to the other elements of the Las Nubes set. All of these fea- tures trend about N. 60° to 70° E. We have relatively little information pertaining to the significance of this set. The Las Nubes fault has not been observed by us on the ground. It occupies a topographic setting similar to, and cuts the same rocks as, the Mateare fault (discussed below). By analogy with that fault, the Las Nubes fault may also be considered active, but we have no record of historic activity on this structure. Southeast 0f the area that we examined, for example, east of Casa Colorada, there is some indication that the lineaments of the Las Nubes set and those of the Tiscapa set may each change orientation somewhat and merge to become parallel to set B (pl. 1). Thus, the Las Nubes set may essentially be a variant of the dominant set B and its local equivalent, the Tiscapa set. The Las N ubes set differs somewhat in orientation from a minor set of lineaments shown on plate 1 as set A, which trends about N. 55° to 60° E. Set A occurs northeast of Lago de Nicaragua and near the Honduras border; its relation- ship to the Las N ubes set is not clear. A major segment of the Cayman Trough in the Caribbean Sea north of Nicaragua trends in the same general direction; the Cayman Trough probably marks the boundary of the Caribbean plate. The relationship of this major feature to the minor lineaments discussed here is also obscure. We can prognosticate seismic activity with much less certainty in the zones dominated by the Las Nubes set than in several of the other zones. However, future activity along the Las N ubes fault probably can be expected, and it is possible that similar activity to that found along the l8 GEOLOGIC CONSIDERATIONS, MANAGUA, NICARAGUA, FOLLOWING THE 1972 EARTHQUAKE lineaments of the Tiscapa set may also occur along those of the Las Nubes set. Thus, the Las Nubes zones are by no means free of the threat of future seismic activity. NEJAPA-TIPITAPA SET Faults, fractures, and lineaments that trend about due north to N. 10° E. occur in two large zones and in a few smaller ones within the Managua area. West of the city there is a belt of volcanic features that occurs along such an alinement, including the Ticomo depression, Nejapa pit, Asososca, and Jiloa. McBirney (1955, p. 148) and McBirney and Williams (1965, p. 43) described and named, but did not map, the Nejapa fault zone as extending through these features (fig. 5). They cited evidence for right-lateral dis— placement along this fault zone in the segment south of Laguna de Jiloa; they also speculated that the offset in the line of major active volcanoes, mentioned previously, may also be due to such right-lateral displacement. We did not find evidence for horizomal displacement along the Nejapa fault zone, but do not rule out this possibility. A very cursory look at the outcrops on the north side of Laguna de Asososca suggests the possibility of vertical dis- placement, the western block up relative to the eastern; further work at that locality should confirm or deny this possibility. We did find several lineaments that parallel the presumed trace of this fault zone. East of the city there are mapped faults (Ferrey, 1971) south of Tipitapa that trend just east of north. Linea- ments also occur along this trend and extend along the east shore of Lago de Managua. These faults lie along scarps in which rocks similar to those that underlie the area to the west are exposed at the tops of the scarps. The evidence for vertical displacement on these faults, with the eastern block up relative to the western, appears convincing, although we did not observe actual displacement of an individual identifiable rock unit. The displacement is probably on the order of 10-15 m. West of these faults, southeast of the Aeropuerto Las Mercedes, is another fault which has similar but smaller displacement. Near Sabana Grande a prominent fault cuts the surface of the ground as a straight scarp with relief of about 2 m; there, the western block has moved up, and so the area between these two faults is a small graben which appears to be partly filled with alluvium. Lineaments permit extending this zone south to the Masaya caldera, but evidence for faulting is lacking south of about Vera Cruz. Smaller zones also containing linear features that trend nearly due north occur in the central part of the Managua area. One, south of Laguna de Tiscapa, is dominated by a mapped fault (Kuang and Williams, 1971) that exhibits primarily vertical displacement, the western block up relative to the eastern. Two other zones, south of Esquipulas and along the Carretera Interamericana, consist chiefly of lineaments. These two zones border the Las Nubes zone on the east and west sides, respectively, suggesting that they might be zones of rotational faulting, although this relationship has not been established. McBirney and Williams (1965, p. 43) indicated that the north-trending faults are among the youngest structural features in Central America. Southwest of Managua this seems to be borne out by the pattern of linear features illus- trated in figure 5, where the faults and fractures of the Nejapa-Tipitapa set seem to truncate those of the Tiscapa set. Near Sabana Grande and Tipitapa, faults of the Nejapa-Tipitapa set cut the youngest rocks in the area, and the scarps by which they have been recognized are straight and relatively uneroded; this is particularly true of the scarp «near Sabana Grande, which appears as if it could have formed during historic time. This evidence notwith- standing, the fact remains that the most recent fault activity, that of 1972, has been along the northeast- trending faults of the Tiscapa set. The Nejapa-Tipitapa linear features exactly parallel a set of lineaments seen on the countrywide radar imagery identified in figure 6 and on plate 1 as set D. This is not an extensively developed set; it consists chiefly of the prominent lineament that marks the straight eastern edge of Lago de Managua. McBirney and Williams (1965, p. 43) pointed out other north-south features associated with active volcanoes in El Salvador and Guatemala as well as Nicaragua; the prominent fissure which opened on the side of E1 Hoyo (Las Pilas) Volcano (pl.1) during 1952 is an example. There is little direct evidence on which to base predic- tions for future fault activity within the area of the Nejapa- Tipitapa set. However, the recency of the faulting suggests that movement is likely to continue into the future. The magnitude of the displacements suggests that earth- quakes of larger magnitude, if lesser in frequency of occurrence, than those in the zone of the Tiscapa set may be expected. There have been reports that, following the earthquake of 1844, the Rio Tipitapa was blocked by faulting and no longer was able to serve as an outlet for Lago de Managua. This effect is not mentioned by Squier (1852, p. 85, 114) and is doubted by most later workers, but it is given credence by Froebel (1859, p. 62—63). The dry channel of the Rio Tipitapa between Lago de Managua and Bahos Termales (at the northeast corner of Tipitapa), as well as the fault scarps and associated hot springs in the area, suggests that blockage indeed took place; if blockage of the outlet did not happen as a single event during the 1844 earthquake it probably did take place as the result of a series of similar events over a long period of time. We must conclude that the Nejapa-Tipitapa zones may well be the sites of continued faulting at any time in the future. The further development of volcanic collapse pits such as the Nejapa and Ticomo depressions is also a distinct possibility. FAULTS AND OTHER LINEAR GEOLOGIC FEATURES 19 corRADfA SET The principal zone of the Cofradia set extends from Tipitapa through the village of Cofradfa to the Masaya caldera (fig. 7). It is dominated by faults that are a continuation of faults of the Nejapa-Tipitapa set that extend south from the town of Tipitapa; the Cofradia faults, however, trend about N. 10° to 20° E. They have characteristics similar to those of the Tipitapa faults in that they occur along fresh scarps that are as much as 15 m high and appear to exhibit vertical displacement in which the eastern block has moved up relative to the western block. The faults occur en echelon, each continuing for 2—4 km before dying out and being replaced by another fault several hundred metres distant. Another zone extending southwest from near Esquipulas is marked by a fault along which the western block has moved up; this relationship to the faults at Cofradia further expands the graben formed by the faults near Sabana Grande. Both of these zones also contain lineaments parallel to the mapped faults. An entirely separate zone of small lineaments that trend parallel to the Cofradia features occurs southwest of Apoyeque. The relationship of this zone to the others is obscure, and the small number of lineaments is rather inconclusive. The Cofradfa linear features are subparallel to a set of lineaments on the countrywide radar imagery identified as set C (fig. 6; pl. 1), which trends about N. 20° E. These lineaments are more extensively developed than the D set and traverse much the same part of the country as the prominent B set, becoming dominant in some places. The Cofradfa set appears to be the local representation of the extensive C set. Prognostication for the two eastern zones of the Cofradia set is similar to that for the Nejapa-Tipitapa zones—continued faulting can be expected, and earth- quakes of larger magnitude than those in Managua are possible. Little can be said with certainty about the western zone. NORTHWEST-TRENDING SETS These three sets, identified on figures 6 and 7 as the Mateare, Las Uvas, and Ticomo sets, will be considered together, because they seem to make up a single series of closely related zones differentiated chiefly by small differences in orientation of the features. These features are dominated by the Mateare fault zone (fig. 5), mapped by Kuang and Williams ( 1971). Although shown on their map as a single fault, there may in fact be a zone of faults, some of which appear on figure 5 as lineaments and some as the mapped fault. The fault zone extends along the abrupt northeast-facing escarpment of the Cordillera del Pacifico and apparently marks the southwestern boundary of the Nicaragua Depression in this area. This is the single largest fault zone in the area. Displacement appears to have been largely vertical and probably is on the order of 900 m. This figure represents the approximate difference in elevation between the crest of the mountains and Lago de Managua. We have not identified the same rock units at the crest of the range and near the lake and, thus, cannot give an exact figure; however, if the same rock unit that occurs at the crest of the range also occurs at depth near the lake, the displacement exceeds 900 m. Because the rocks that have been displaced are probably tens of thousands rather than hundreds of thoUsands of years old, the dis- placement must have occurred in a relatively short time and at a fairly rapid rate. The Mateare fault zone, thus, must be regarded as active, and activity will very possibly occur in the future, although no rate can be given with any certainty. What is here termed the “Mateare set” of faults and lineaments dominates the northwestern end of the series of zones and trends about N. 20° to 30° W. The set is made up of several closely spaced lineaments, most of which are likely to be individual faults that make up the fault zone. Near the crossing of the Carretera a Le6n (fig. 7), the dominant trend changes to about N. 45° to 50° W.; these lineaments are here termed the “Las Uvas set,” from a locality along the crest of the Cordillera del Pacifico. The lineaments here are more widely spaced; they continue intermittently along this trend to the Masaya caldera, as indicated by isolated small zones in figure 7. It is by no means certain that the Mateare fault zone extends this far southeast; these lineaments may merely mark the con- tinuation of the trend. The change in trend that marks the break between the Mateare and Las Uvas sets occurs at about where the Punta Huete lineament intersects the Mateare fault zone. (See fig. 5.) Nearby there is a small zone of lineaments that trends about N. 10° to 15° W., referred to as the Ticomo set because it extends into the Ticomo depression. The significance of this small set is not clear, but it does Occur where the N ejapa-Tipitapa set intersects the Mateare fault zone and may be a type of resultant set influenced by forces responsible for both the Mateare and Nejapa sets. The Las Uvas set is almost exactly parallel to set F, a per- vasive set of lineaments that occurs throughout Nicaragua, but particularly in and adjacent to the Nicaragua Depression (pl. 1). These features are also nearly parallel to axes of folding in the Rivas Formation and related rocks along the Pacific coast, and to the seg- ment of the Middle America Trench that lies off the Pacific coast of Nicaragua. The Las Uvas set is thus part of the major regional structure. The Mateare set closely parallels a set of lineaments that is prominent locally in northern Nicaragua, set E; this set is not well developed in the rest of Nicaragua, however, and the Mateare set is probably a local variant of the major set F. There is no direct evidence of current activity along the 20 CEOLOGIC CONSIDERATIONS, MANAGUA, NICARAGUA, FOLLOWING THE 1972 EARTHQUAKE Mateare fault zone; however, the youth of the faulting and the probable magnitude of the displacement suggest that faulting is likely to continue and that earthquakes of large magnitude may occur, possibly at widely spaced intervals of time. MAJOR LINEAMENTS IN NICARAGUA Some of the lineaments shown on plate 1 appear more prominent than others and extend for considerable dis- tances across Nicaragua. Three such major lineaments intersect in a crudely triangular area in which Managua and vicinity is located; these lineaments will be discussed briefly below. One of these major lineaments is the 10cus of most of the active volcanoes of Nicaragua (pl. 1); it trends northwest- ward and is part of set F; it will be called the Marrabios lineament, after the Cordillera Los Marrabios. The Marrabios lineament parallels the Nicaragua Depression (fig. 2), which traverses Nicaragua from northwest to southeast. Active faulting has not been identified along this lineament, but because so many active volcanoes occur along this line it must represent a major zone of weakness in the Earth’s crust. The nearby and subparallel Mateare fault, however, appears to have been active. A second major lineament trends northeastward across all of Nicaragua; in the Managua area it extends along the southeast sides of the Chiltepe and Punta Huete Peninsulas (fig. 5) and is here termed the Punta Huete lineament. It is part of set B, although farther northeast it deviates somewhat from this general trend, both to the north and to the east. A number of lines of evidence suggest that the Punta Huete lineament may be a major transcurrent fault zone that separates two segments of the Caribbean plate. It is clear that the Punta Huete lineament intersects and off- sets the line of volcanoes that defines the Marrabios linea- ment by about 10 km in a right—lateral sense (pl. 1). It also appears to offset lineaments of the Mateare set, but in the opposite, left—lateral sense. As previously noted, offsets along faults active in the 1931 and 1972 earthquakes were also left lateral. Oil seeps have been reported (Parsons Corp. and others, 1972, p. IV-197; J. C. Zahn, Oceanic Exploration Co., written commun., 1973) along the trend of the Punta Huete lineament both onshore and offshore in the Pacific Ocean, suggesting active faulting there. The Punta Huete lineament has not been studied elsewhere along its length, and additional geologic work will be necessary to verify its tectonic significance. Although a recent reversal of movement of the two plate segments southeast and northwest of the Punta Huete lineament might be assumed, such an assumption is un- necessary to explain the observed relationship. Rather, a decrease in the velocity of the southeastern plate segment relative to the northwestern plate segment would have produced the left—lateral offsets of 1931 and 1972 (Decker, 1973; Krushensky and others, 1974). This probably recent decrease in velocity, together with the greater degree of dif- ferentiation of the volcanic rocks southeastward along the volcanic arc, suggests a recent and probably progressive northward stalling of subduction along the Middle America Trench (Malfait and Dinkelman, ' 1972; Krushensky and others, 1974) prior to its complete stagna— tion, as is evident in the Panama Basin (Van Andel and others, 1971, p. 1506). The Punta Huete lineament is approximately parallel to the trend of an apparent offset between the southern Nicaragua part of the volcanic arc and the Guanacaste part in adjacent Costa Rica (King, 1969; a major lineament in this position also appears on pl. 1 at about lat 12° N., long 85° W.). In central Costa Rica the end of the volcanic arc abuts against another northeast—striking tectonic trend. Both of these trends may also mark major transcurrent fault zones. A third major lineament in the Managua area is here termed the Tipitapa lineament. It is part of set D, trending nearly north-south. The Tipitapa lineament does not have as great an extent as the two other major lineaments discussed here, and it may be subsidiary to them, but it is clearly marked by active faulting in the Managua area; faults of the north-south Nejapa-Tipitapa set and the sub- parallel Cofradia set make up part of this lineament. It has been suggested that the southern shore of Lago de Managua might also mark the site of a fault (map in Viv6, 1973, p. 6). We have found no evidence for a fault there and think it more likely that the south shore of the lake, which in detail is not a very straight east-west line, is the product of differential erosion, where linear features of the north- east—trending Tiscapa set intersect the lake. Our analyses of linear features both in the area surrounding Managua and in Nicaragua as a whole show very few features alined in this east-west direction; one such feature is shown in figure 5, and a very small number of them appear on plate 1. However, Rinker (1972) shows a series of linear features in an approximate east-west direction derived from an analysis of ERTS multispectral imagery. This direction is almost exactly parallel to the scanning direction of the radar imagery we used, and it is possible that there is a set of lineaments trending in this direction that is masked on our imagery. At the local, national, and regional levels shown in figure 6, all the sets and major linear features occur within a range of about 120°, extending between about N. 60° W. and N. 60° E. Although there is not an exact coincidence between the features at the three levels and no statistical comparisons were made, we believe that the coincidence of features is sufficiently strong to conclude that the linear features in the Managua area reflect part of the tectonic structure of the Central America region and are not fea- RECOMMENDATIONS 21 tures of purely local genesis dependent solely on local vol— canoes or volcanic activity. The intersection of three major lineaments near Managua further enhances the concept that the Managua area faults and other linear features are regionally controlled, and suggests that the Managua area may be located at a critical focus of major structural ele- ments, a focus that may even be unique “within the country. The Managua area is thus a likely site for an un- usual concentration of seismic, tectonic, and volcanic activity that is likely to continue unabated in the fore- seeable geologic future. SUMMARY j The Managua area has been divided into a number of zones, each dominated by a set of linear geologic features (including faults, fractures, and lineaments) that trend in one direction. The geologic structure in each of these zones has different characteristics, but each of the zones shares a common likelihood of having some degree of continued faulting and accompanying seismic activity. The Tiscapa zone seems most likely to be the site of future activity with the highest frequency of occurrence, but the other zones are also likely to be the sites of future activity, perhaps of greater magnitude but lower frequency of occurrence. We thus cannot recommend, from the standpoint of future faulting and seismic activity, that any of these zones represents a significant improvement over the Managua city area as a site for future development. It would be desirable to monitor, on a continuing basis, the seismic activity (including microseisms) in each of these zones, and particularly along the known faults associated with them. Such monitoring might aid in deter— mining whether there is any indication that some of these zones are more active than others, whether these zones of linear features have any use as zones of predictably different levels of seismic activity, and, thus, whether some of them may prove to be more favorable than others as areas for development than is now apparent from the analysis we have made. CONCLUSIONS Managua and the surrounding area are subject to an ever-present threat of the effects of volcanic and seismic activity. The geologic materials that underlie the area, however, generally do not pose major problems to development. Renewed volcanism beyond the present levels of fumarolic emission should be expected, but kinds of activity and time of events cannot now be forecast. Deposition of ash and Cinders is the most probable activity but one that is not likely to cause high loss of life; how- ever, property damage may be very widespread. Eruptions of molten lava flows may destroy property in parts of the area. Swift-moving volcanic mudflows or hot-ash nuée ardente flows are likely to occur, less frequently but are possible, and when they do occur are likely to be destruc- tive of both life and property. Any of these volcanic activities can render parts of the area uninhabitable, at least temporarily. The geologic materials in the area generally provide high bearing capacity for foundations of buildings. In their response to seismic shock they are likely to behave more as solid rock than as loose alluvium and are thus relatively unproblematical in this regard. The deposits are cut by many faults and related geo- logic structures throughout the area. These features testify to a long history of seismicity even before records were kept by man. Recorded earthquakes demonstrate that seismic events are frequent, and there is no evidence to suggest that they will diminish. Some parts of the area exhibit more clear-cut evidence for continued seismic activity at the level that has prevailed in the city of Managua than do other parts. Movement is likely to continue throughout the area, both along existing faults and along new faults. In places within the Managua area, earthquakes of larger magnitude than those in the city may possibly occur, perhaps more widely spaced in time. Parts of the Managua area may be subject to different kinds of volcanic activity. There are variations in the distribution and thickness of geologic materials; and expectations of seismic activity also vary from place to place. However, we cannot determine that any major parts of the area are significantly superior to others as sites for urban development. The entire area is probably among the poorer sites in Nicaragua for the location of a major c1ty. RECOMMENDATIONS Two alternative choices must be considered by re- sponsible public officials: (1) To relocate Managua in another part of Nicaragua, or (2) to rebuild the city partly on its present site and generally within the Managua area as defined in this report. Because the present site of Managua is in a geologically hazardous environment, one solution that frequently has been suggested is to relocate the city elsewhere in Nicaragua. Selection of a site elsewhere implies that the new location would be free of overriding geologic con- straints and have other attributes of a positive nature, such as adequate water supply, acceSsibility, usable terrain, and the like, that are essential to the development of a large city. Comments are made about two areas that have been suggested as possible locations. These localities were studied briefly and in reconnaissance fashion only. The first one is on Highway 7 between Teustepe and Monte Grande, and the second is the Dario Plains (fig. 2). The Teustepe—Monte Grande site appears to offer sufficient space for a large city. The valley is broad and 22 GEOLOGIC CONSIDERATIONS, MANAGUA, NICARAGUA, FOLLOWING THE 1972 EARTHQUAKE relatively flat, and partly filled with alluvium. It is apparently not cut by major active faults, and active vol— canic centers are some distance away. The surrounding bedrock hills are low, steplike, and suitable for develop- ment; they would provide good foundations for struc- tures, and would tend to diminish the damaging effects of earthquakes more than alluvium would. The site would need to be studied thoroughly to insure that no unrecognized geologic hazards, such as active faults, are present. According to Juan Kuan S. (oral commun., 1973), available water for domestic and industrial uses is in short supply, but hydrologic data were not available to us. Addi- tional consideration of the location is pointless if water is not readily and economically available. The second site, referred to as the Dario Plains (Parsons Corp. and others, 1972), was the only area being investigated by Kuan under the auspices of Catastro (Juan Kuan S., oral commun., 1973) as a possible alternate relocation site for Managua. The general area has many desirable attributes, but it also has unsatisfactory ones. In the vicinity are a major river, a hydroelectric power dam currently under construction, and connecting highways. The southern border of the plains is defined by a Holocene fault zone, and the eastern margin is defined by a north- northeast-trending zone of faulting (Parsons Corp. and others, 1972, p. IV—23). Much of the area is underlain by flat-lying lake beds, and the water table is relatively shallow. The lake deposits consist of many beds of very plastic and sticky, silty clays that contain a high per- centage of expansive montmorillonite (Parsons Corp. and others, 1971, p. II—685—699). These clays pose severe siting problems for most structures. Obviously, more detailed investigations should be made of these two areas to supply the data required in deciding their suitability for relocation of Managua. Similarly, other sites that might be suggested for relocation will require in-depth studies before they can be considered seriously. While authorization of our study in Nicaragua clearly did not include investigation of potential alternate relocation sites, we would recommend that criteria be established for such an inquiry. The criteria should include all aspects involved in the siting of a large modern city—not just the geologic framework. Because relocation of Managua involves greater financial resources than are likely to be available in the near future, because no such move, even if financially feasible, should be undertaken without full assurance that the new site has characteristics of such superiority over the old as to be commensurate with the cost of the move, and because no such suitable site is clearly identified at present or is likely to be identified without expensive and time- consuming investigation, it is likely that reconstruction will indeed take place within the area of the present city and its environs. The following recommendations apply to reconstruc- tion in the Managua area and also to development in Nicaragua as a whole: 1. Construction over known fault zones should be avoided. 2. Fault zones should be reserved for low-density land use, such as recreational open space. 3. Where service facilities must cross known faults, or buildings must be located over them, the design should provide accommodation for anticipated horizontal and vertical movement, as well as for shock. 4. Seismic activity of the fault zones, as delineated in fig- ure 7, should be systematically monitored on a continuing basis. 5. In conjunction with seismic monitoring of fault zones shown in figure 7, detailed mapping should be continued, using drilling and geophysical methods to more clearly define the location and existence of known faults, and to determine the existence of other faults not now known. 6. Seismic activity of distant as well as nearby volcanoes should be instrumentally recorded systematically on a continuing basis. Recording of temperature variations and other observations that give evidence of magma movement should also be made, so that a body of data may be accumulated which might serve as a basis for predicting physical changes in the volcanoes that could lead to eruptions. 7. A nationwide network of seismic stations should be installed at selected locations. It should be pointed out that in such an area even the most careful geologic appraisal and consideration will be ineffective without adequate attention to decentralization of public service facilities, formulation of appropriate building codes, and careful surveillance of construc- tion practices. REFERENCES CITED Brown, R. D., Jr., Ward, P. L., and Plafker, George, 1973, Geologic and seismologic aspects of the Managua, Nicaragua, earthquakes of December 23, 1972: U.S. Geol. Survey Prof. Paper 838, 34 p. Decker, R. W., 1973, Inconsistent faultSP: Geol. Soc. America Abs. with Programs, v. 5, no, 7, p. 595. Ferrey 0., C. J., 1971, Mapa geolégico de Las Mercedes, Nicaragua: Catastro e Inventario de Recursos Naturales, sheet 2952 II, scale 1:50,000. Ferrey 0., C. J., and Williams, R. L., 1971, Mapa geolégico de Chiltepe, Nicaragua: Catastro e Inventario de Recursos Naturales, sheet 2952 IV, scale 1250,000. Froebel, Julius, 1859, Seven years’ travel in Central America, northern Mexico, and the Far West of the United States: London, Richard Bentley, 587 p. Inman, D. I_.., 1952, Measures for describing the size distribution of sediments: Jour. Sed. Petrology, v. 22, no. 3, p. 125—145. Karim, M. F., and Chilingar, G. V., 1963, Exploracion geoqufmica para petroleo en la costa del Pacifico de Nicaragua: Nicaragua Servicio Geol. Nac. 801. 7, p. 97—136. REFERENCES CITED 23 King, P. B., compiler, 1969, Tectonic map of North America: U.S. Geol. Survey, 2 sheets, scale l:5,000,000. Krushensky, R. D., 1972, Geology of the Istarfi quadrangle, Costa Rica: U.S. Geol. Survey Bull. 1358, 46 p [1973]. Krushensky, R. D., Schmoll, H. R., and Dobrovolny, Ernest, 1974, The 1972 Managua earthquake—Geologic constraints on planning for the reconstruction or relocation of Managua, Nicaragua: Caribbean Geol. Conf., 7th, Pointe a Pitre, Guadeloupe, Trans. (in press). Kuang 5., Juan, and Williams, R. L., 1971, Mapa geolégico de Mana- gua, Nicaragua: Catastro e Inventario de Recursos Naturales, sheet 2952 III, scale 1:50,000. Malfait, B. T., and Dinkelman, M. G., 1972, Circum-Caribbean tectonic and igneous activity and the evolution of the Caribbean plate: Geol. Soc. America Bull., v. 83, no. 2, p. 251—271. McBirney, A. R., 1955, The origin of the Nejapa pits near Managua, Nicaragua: Bull. Volcanol., ser. 2, v. 17, p. 145—154. 1956, The Nicaraguan volcano Masaya and its caldera: Am. Geophys. Union Trans, v. 37, no. 1, p. 83—96. 1958, Active volcanoes of Nicaragua and Costa Rica, in Central America, pt. 6 of Catalogue of the active volcanoes of the world including solfatara fields: Naples, Italy, Internat. Volcanol. Assoc, p. 107—146. McBirney, A. R., and Williams, Howel, 1965, Volcanic history of Nicaragua: California Univ. Pubs. Geol. Sci., v. 55, 73 p. Melson, W. G., and Saenz R., Rodrigo, 1968, The 1968 eruption of Volcan Arenal, Costa Rica—Preliminary summary of field and laboratory studies: Cambridge, Mass., Smithsonian Inst. Center for Short-lived Phenomena (interim rept.), 35 p. Parsons Corp., Marshall and Stevens, Inc., and International Aero Service Corp., 1971, Genesis and classification of soils, pt. 3 of Soil Survey of the Pacific region of Nicaragua, v. 2 of Final tech- nical report: Nicaragua, Tax Improvement and Nat. Resources Inventory Proj. 1972, The geology of western Nicaragua, v. 4 of Final technical report: Nicaragua, Tax Improvement and Nat. Resources In- ventory Proj., 221 p. Rinker, J. N., 1972, Image showing fracture pattern analysis, in NASA earth resources technology satellite (ERTS—l) image northwest of Managua, Nicaragua—Image and fracture patterns, 1972: US. Geol. Survey, 1 sheet. Servicio Geolégico Nacional de Nicaragua, 1973, Estudio preliminar del terremoto del 23 de Diciembre de 1972: Servicio Geolégico Nac. Nicaragua, referencia—P. 2—1973, [prelim. rept.] 6 p. Squier, E. G., 1852, Nicaragua; its people, scenery, monuments, and the proposed interoceanic canal, v. 2: New York, D. Appleton, 452 p. Steen-McIntyre, V. C., 1975, Hydration and superhydration of tephra glass—a potential tool for estimating age of Holocene and Pleis- tocene ash beds: Royal Soc. New Zealand Bull. (in press). Steen-McIntyre, V. C., Fryxell, Roald, and Malde, H. E., 1973, Unex- pectedly old age of deposits at Hueyatlaco archaeological site, Valsequillo, Mexico, implied by new stratigraphic and petro- graphic findings: Geol. Soc. America Abs. with Programs, v. 5, no. 7, p. 820—821. Sultan, D. I., 1931, The Managua earthquake: Military Engineer, v. 23 no. 130, p. 354—361. Van Andel, T. H., Heath, G. R., Malfait, B. T., Heinrichs, D. F., and Ewing, J. 1., 1971, Tectonics of the Panama basin, eastern equatorial Pacific: Geol. Soc. America Bull., v. 82, no. 6, p. 1489— 1508. Vivo E., J. A., 1978, E1 terremoto del 23 de Diciembre de 1972 y la ciudad de Managua: Panamericano Geografia e Historia Inst. Aéreo Bol. 128, 10 p. Walker, G. P. L., 1971, Grain-size characteristics of pyroclastic de- posits: Jour. Geology, v. 79, no. 6, p. 696—714. Williams, Howel, 1952, Geologic observations on the ancient human footprints near Managua, Nicaragua: Carnegie Inst. Washington Pub. 596, Contr. Am. Anthropology and History, no. 52, p. 1—31. Zoppis Bracci, Luigi, and Giudice, Daniele del, 1958, Geologia de la costa del Pacifico de Nicaragua: Nicaragua Servicio Geol. Nac. Bol. 2, p. 19—68. fiUS. GOVERNMENT PRINTING OFFICE: 1975—677-340/7 15° 14° UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY 87° 86° PREPARED IN COOPERATION WITH THE GOVERNMENT OF NICARAGUA AND THE AGENCY OF INTERNATIONAL DEVELOPMENT US. DEPARTMENT OF STATE 85° PROFESSIONAL PAPER 914 PLATE 1 83° EXPLANATION w Lineaments classified according to sets designated by letters 7 Each has the approximate orientation shown. (See also fig. 6.) Selected major lineaments indicated by heavy line; those discussed in text labeled by name - --------~ Other lineaments (":1 Circular to elliptical features >I< Principal Quaternary volcanoes — Modified from McBirney and Williams (1965) ._. O o m 10. Apoyeque . s 'ina 2. San C istob I ll. Jiloa 3. Casit 12. A sssssss 4. Telica l3. Tiscapa 5. Santa Clara l4. Masaya g 6. Cerro Negro 15. Apoyo 7. E1 Hoyo (Las Pilas) l6. Mombacho 8. Momotombo I 7. Concepcion 9. Momotombito l8. Madera 1973) ___ — Edge of mosaic — It overlaps national boundary by about 5—10 km and includes adjacent parts of Honduras and Costa Rica I 87° Base from radar imagery furnished by Instituto Geografico Nacional, Ministerio de Obras Publicas, Gobierno de Nicaragua Coordinates modified to approximately correct position :5 Oil seep (J. C Zahn, Oceanic Exploration Co., written commun., ' ' ‘_ TAPA TI/PI j/\/\, 86° HONDURAS I 1 I ,1 , 1 I SCALE APPROXIMATE I 85° COSTA RICA 100 KILOMETRES | MAP OF NICARAGUA, SHOWING LINEAMENTS AND OTHER FEATURES SEEN ON 121,000,000—SCALE UNCONTROLLED MOSAIC OF SIDE-LOOKING RADAR IMAGERY, WITH SELECTED DATA FROM OTHER SOURCES _ 15° 12° 11° 83° fiUS. GOVERNMENT PRINTING OFFICE: l975-—677-340/7 9576 a (a W? v V, q [5" > ? £3“!er at S Plutonic Rocks of the Santa Rita Mountains, Southeast of Tucson, ' Arizona GEOLOGICAL SURVEY PROFESSjONAL PAPER 915 gNTOF, i Y6 I 6/ Plutonic Rocks of the Santa Rita Mountains, Southeast of Tucson, Arizona By HARALD DREWES GEOLOGICAL SURVEY PROFESSIONAL PAPER 915 A petrographic description, augmented by many modal and chemical analyses and radiometric age determinations, of a series of platonic rocks and related h ypabyssal intrusive rocks, and remarks on their relation to some mineral deposits UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON 11976 UNITED STATES DEPARTMENT OF THE INTERIOR THOMAS S. KLEPPE, Secretary GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress Cataloging in Publication Data Drewes, Harold, 1927— Plutonic rocks of the Santa Rita Mountains, southeast of Tucson, Arizona. (Geological Survey Professional Paper 915) Bibliography: p. Includes index. Supt. of Docs. no.: 119.16:915 1. Rocks, Igneous. 2. Petrology—Arizona—Santa Rita Mountains. 3. Intrusions (Geology)—Arizona—Santa Rita Mountains. 1. Title. II. Series: United States Geological Survey Professional Paper 915. QE461.D65 552’.3 75—619358 For sale by the Superintendent of Documents, US. Government Printing Office Washington, DC. 20402 Stock Number 024-001—02865—4 CONTENTS Page Page Metric—English equivalents .................................. V Cretaceous rocks—Continued Abstract ................................................. 1 Josephine Canyon Diorite—Continued Introduction .............................................. l Modal and chemical summary ....................... 37 Objectives ............................................ 2 Age and correlation ................................ 39 Acknowledgments ..................................... 3 Madera Canyon Granodiorite --------------------------- 40 Geologic setting ....................................... 3 Petrography -------------------------------------- 41 Precambrian rocks ........................................ 6 Gr anodiorite ------------------------------- 41 Final Schist .......................................... 6 Porphyritic granodiorite ------------------------ 42 Granite gneiss ........................................ 7 Melanocratic granodiorite ---------------------- 42 Continental Granodiorite ............................... 9 Modal and chemical summary ----------------------- 43 Petrography ______________________________________ 9 Age and correlation ................................ 44 Granodiorite _________________________________ 9 Elephant Head Quartz Monzonite ....................... 47 Aplitic rocks ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, ll Petrography ...................................... 48 Lamprophyre ..... , ............................ l3 Quantrell stock, coarse—grained quartz monzonite . . 49 Modal and chemical summary ....................... 13 Quantrell stock, fine-grained quartz monzonite - - . . 49 Age and correlation ................................ 15 Yoas stock, coarse-grained quartz monzonite ------ 50 Triassic and Jurassic rocks .................................. l7 Modal and Chemical summary ----------------------- 51 Piper Gulch Monzonite ................................ 17 Age and correlation -------------------------------- 51 Quartz diorite ........................................ 23 Tertiary rocks ............................................ 54 Squaw Gulch Granite .................................. 24 Rocks of the Gringo Gulch pluton and other rocks .......... 55 Petrography ...................................... 25 Granitoid rocks of the Helvetia stock ..................... 57 Granite and quartz monzonite ................... 26 Petrography ...................................... 58 Aplite and lamprophyre ........................ 27 Modal and chemical summary ....................... 60 Modal and chemical summary ....................... 28 Age and correlation ................................ 61 Age and correlation ................................ 28 Quartz latite porphyry of the Greaterville plugs ............ 62 Cretaceous rocks ........................................... 29 Petrography ...................................... 65 Granitoid rocks of the Corona stock ...................... 29 Modal and chemical summary ....................... 66 Josephine Canyon Diorite .............................. 33 Age and correlation ................................ 67 Petrography ...................................... 34 Granodiorite of the San Cayetano stock ................... 67 Dioritic rocks ................................. 34 References cited ........................................... 71 Fine—grained quartz monzonite .................. 36 Index .................................................... 73 ILLUSTRATIONS Page FIGURE 1. Index map showing location of Sahuarita and Mount Wrightson quadrangles and the Santa Rita Mountains ..................... 2 2. Index map of the Santa Rita Mountains ............................................................................... 4 3. Map showing distribution of Precambrian rocks and specimen collection sites ............................................... 8 4. Photograph of specimen 17, Continental Granodiorite .................................................................. 10 5. Photomicrograph of specimen 1‘9, slightly metamorphosed Continental Granodiorite ........................................ 10 6. Photomicrograph of specimen 12, Continental Granodiorite ............................................................. 12 7. Photomicrograph of specimen 17, Continental Granodiorite ............................................................. 12 8. Isometric diagram of a modal tetrahedron ............................................................................ l3 9. Modified triangular diagram showing the modal quartz, potassium feldspar, plagioclase, and femic minerals of Continental Granodiorite ................................................................................................. 14 10. Histograms showing average chemical composition of each group of plutonic rocks .......................................... 18 11. Map showing distribution of Triassic rocks and specimen collection sites ................................................... 20 12. Photograph of specimen 37, Piper Gulch Monzonite .................................................................... 21 13. Photomicrograph of specimen 37, Piper Gulch Monzonite ............................................................... 21 III IV CONTENTS Page FIGURE ‘14. Modified triangular diagram showing the modal quartz, potassium feldspar, plagioclase, and femic minerals of Piper Gulch Monzonite and related rocks .................................................................................... 22 15. Map showing distribution of Squaw Gulch Granite and sample collection sites .............................................. 25 16. Photograph of specimen 45, Squaw Gulch Granite ...................................................................... 26 17. Photomicrographs of specimen 45, Squaw Gulch Granite ................................................................ 27 18. Photomicrograph of Specimen 48, Squaw Gulch Granite ............................................................... 28 19. Modified triangular diagram showing modal quartz, potassium feldspar, plagioclase, and femic minerals of Squaw Gulch Granite . . 30 20. Map showing distribution of granitoid rocks of the Corona stock and specimen collection sites ................................ 32 21. Photomicrograph of sepcimen 64, quartz monzonite of the Corona stock .................................................. 32 22. Modified triangular diagram showing modal quartz, potassium feldspar, plagioclase, and femic minerals of granitoid rocks of the Corona stock .................................................................................................. 33 23. Map showing distribution of Josephine Canyon Diorite and specimen collection sites ........................... , ............. 35 24. Photograph of specimen 80, Josephine Canyon Diorite .................................................................. ’36 25. Photomicrographs of specimen 81, Josephine Canyon Diorite ............................................................ 37 26. Modified triangular diagram showing modal quartz, potassium feldspar, plagioclase, and femic minerals of Josephine Canyon Diorite ...................................................................................................... 38 27. Map showing distribution of Madera Canyon Granodiorite and specimen collection sites ..................................... 41 28. Photograph of specimen 113, nonporphyritic type of Madera Canyon Granodiorite .......................................... 42 29. Photomicrographs of specimen 113, nonporphyritic type of Madera Canyon Granodiorite .................................... 43 30. Photograph of Specimen 127 of the melanocratic type of Madera Canyon Granodiorite ................................... 44 31. Modified triangular diagram showing modal quartz, potassium feldspar, plagioclase, and femic minerals of Madera Canyon Granodiorite ................................................................................................. 45 32. Map showing distribution of Elephant Head Quartz Monzonite and specimen collection sites .................................. 47 33. Photograph of Elephant Head. . . .3 ................................................................................... 49 34. Photograph of specimen 134, Elephant Head Quartz Monzonite .......................................................... 50 35. Photomicrograph of specimen 132, coarse-grained Elephant Head Quartz Monzonite ...................................... 50 36. Modified triangular diagram showing modal quartz, potassium feldspar, plagioclase, and femic minerals of Elephant Head Quartz Monzonite ........................................................................... _ ........................ 52 37. Modified triangular diagram comparing the range of distribution of the modes of the Josephine Canyon Diorite, Madera Canyon Granodiorite, and Elephant Head Quartz Monzonite ............................................................... 53 38. Map showing distribution of rocks of the Gringo Gulch pluron and other rocks and specimen collection sites ..................... 56 39. Modified triangular diagram showing modal quartz, potassium feldspar, plagioclase, and femic minerals of rocks of the Gringo Gulch pluton and of the San Cayetano stock ....................................................................... 53 40. Map showing distribution of granitoid rocks of the Helvetia stocks and porphyry of the Greaterville plugs and specimen collection 60 sites ......................................................................................................... 41. Photograph of specimen 174, quartz monzonite of the Helvetia stocks ................................................... 61 42. Photomicrograph of specimen 174, quartz monzonite of the Helvetia stocks ................................................ 61 43. Modified triangular diagram showing modal quartz, potassium feldspar, plagioclase, and femic minerals of the granitoid rocks of the Helvetia stocks ............................................................................................ 63 44. Photomicrograph of specimen 185, quartz latite porphyry of the Greaterville plugs .......................................... 66 45. Modified triangular diagram showing modal quartz, potassium feldspar, plagioclase, and femic minerals of quartz latite porphyry of the Greaterville plugs ........................................................................................ 68 TABLES Page TABLE 1. Stratigraphic summary of rocks of the Santa Rita Mountains ........................................................... 5 2. Modes of Continental Granodiorite .................................................................................. l4 3. Chemical and spectrographic analyses and CIPW norms of Continental Granodiorite ........................................ 16 4. Summary of radiometric age determinations ........................................................................... l7 5. Modes of Piper Gulch Monzonite .................................................................................... 22 6. Chemical and spectrographic analyses and CIPW norms of Piper Gulch Monzonite .......................................... 23 7. Modes of Squaw Gulch Granite ..................................................................................... 30 8. Chemical and spectrographic analyses and CIPW norms of Squaw Gulch Granite ........................................... 31 9. Modes of granitoid rocks of the Corona stock ......................................................................... 33 10. Chemical and spectrographic analyses and CIPW norms of granitoid rocks of the Corona stock ............................... 33 11. Modes of Josephine Canyon Diorite .................................................................................. 38 12. Chemical and spectrographic analyses and CIPW norms of Josephine Canyon Diorite ....................................... 40 13. Modes of Madera Canyon Granodiorite .............................................................................. 44 14. Chemical and spectrographic analyses and CIPW norms of Madera Canyon Granodiorite .................................... 46 15. Modes of Elephant Head Quartz Monzonite ........................................................................... 52 CONTENTS V Page TABLE 16. Chemical and spectrographic analyses and CIPW norms of Elephant Head Quartz Monzonite ................................. 54 17. Modes of the Gringo Gulch pluton ................................................................................... 59 18. Chemical and spectrographic analyses and CIPW norms of rocks of the Gringo Gulch pluton ................................. 59 19. Modes of granitoid rocks of the Helvetia stocks ........................................................................ 62 20. Chemical and spectrographic analyses and CIPW norms of granitoid rocks of the Helvetia stocks ........................... 64 21. Modes of quartz latite porphyry of the Greaterville plugs ................................................................ 68 22. Chemical and spectrographic analyses and CIPW norms of quartz latite porphyry of the Greaterville plugs ...................... 69 23. Modes of granodiorite of the San Cayetano stock ...................................................................... 70 24. Chemical and spectrographic analyses and CIPW norms of granodiorite of the San Cayetano stock ............................ 70 Metric unit English equivalent Metric unit English equivalent Length Specific combinations—Continued millimetre (mm) : 0.03937 inch (in) litre per second (l/s) : .0353 cubic foot per second metre (m) = 3. 8 feet (ft) cubic metre per second kilometre (km) = mile (ml) per square kilometre [(m3/s)/km9] : 91.47 cubic feet per second per Area square mile [(fta/s)/ini3] metre per day (m/d) 3.28 feet per day (hydraulic square metre (m2) : 10.76 square feet (ft?) conductivity) (ft/d) square kilometre (km?) : .386 lquare mile (mifi) metre per kilometre _ hectare (ha) 2 2.47 acres (111/ m =: 0.28 feet per mile (ft/mi) kilometll'le) per hour 9113 f (1 (ft/ > m/ = . oot per secon 8 Volume metre per secdond (fin/S) : 3.28 feet per second . 3 _ 3 me re square per ay 33:; (clentimetre (cm ) ; 69:3? 333:3 1135398011 ) (mz/d) : 10.764 feet squared per day (ftg/d) cubic metre (m3) = 35.31 cubic feet (rte) (transmiss‘my’ cubic metre = .00081 acre-foot (acre-ft) cubicametre per second cubic hectometre (hm3) =810.7 acre—feet (m /s) = 22-826 million gallons per day litre : 2.113 pints (pt) (Mga /d) litre = 1.06 quarts (qt) cubic metre per minute litre : .26 gallon (gal) (ma/min) =264.2 gallons per minute (gal/min) cubic metre = .00026 million gallons (Mgal or litre per second (1/8) = 15-85 gallons per minute 10° gal) litre per second per _ cubic metre : 6.290 barrels (bbl) (1 bb1=42 gal) mEtre [(I/s)/m] 4.83 gallqnslr/Jerinil/nflge per foot ga m 11 Weight kilometre per hour (t m/h) d ( / ) = 2:37 mile per 11(1):" (mi/h) gram (g) = 0.035 ounce, avoirdupois (oz avdp) me re per secon m s = ' m es per our gram = .0022 pound, avoirdupois (lb avdp) gram a" eubic 3 _ 2 43 d 3 tonne (t) = 1.1 tons, short (2.000111) grgfunpgesa‘fmg/cm) ‘ 6 ' 1’0““ 8 per ““0 f°°t (lb/ft) t = .98 t , l 2,24 lb onne on 0’13 ( 0 ) centimetre (g/cmz) : 2.048 pounds per square foot (lb/ft“) . . . gram per square Specrfic combinations centimetre .0142 pound per square inch (lb/in”) kilogram per square centimetre (kg/cm!) : 0.96 atmosphere (atm) Temperature kilogram per square 0 _ centimetre = .98 bar (0.9869 atm) degree 09181115 ( C) = 1.8 degrees Fahrenheit (°F) cubic metre per second degrees Celsius (m3/s) = 35.3 cubic feet per second (fta/s) (temperature) :[(1.8>< °C) +32] degrees Fahrenheit PLUTONIC ROCKS OF THE SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA By HARALD DREWES ABSTRACT Plutonic masses and some related hypabyssal bodies were intruded into the rocks of the Santa Rita Mountains during Precambrian, Triassic, Jurassic, Cretaceous (2 episodes), Paleocene (3 episodes), and Oligocene times. The plutonic rocks range in composition from diorite to granite; most of them are granodiorite or quartz monzonite. The intrusive masses range in size from a small batholith to plugs and related dikes; most of them underlie areas of l to 10 square miles. Some of the intrusive masses are locally associated with contact metamorphic mineral deposits, and plugs of quartz latite porphyry of late Paleocene age are closely associated with hydrothermal deposits of base and noble metals in the Helvetia and Greaterville mining districts. Petrographic descriptions of the plutonic rocks are augmented by nearly 200 modal analyses, more than 50 chemical and spectrographic analyses, and more than 25 radiometric age determinations. The oldest of the plutonic rocks, the Continental Granodiorite, of Precambrian Y age, is a coarse-grained, very coarsely porphyritic, partly» metamorphosed and much-altered granodiorite and quartz monzonite that forms a mass of unknown size. Its host rocks are intensely metamorphosed, but they are not mineralized near the intrusive mass. During Triassic time, the Piper Gulch Monzonite was intruded, probably along a major northwest-trending fault zone. It consists mainly of very coarse grained, dark—gray, anorthositic-looking rock in large inclusions, septa, and wallrock to younger intrusive masses. Its host rocks are metamorphosed and locally mineralized; the mineralization in the host is probably related to nearby younger intrusive masses. Squaw Gulch Granite forms a partly concealed batholith or very large stock of Jurassic age. The rock is coarse grained and pinkish gray, and it ranges in composition from granite to quartz monzonite. At least some of its host rocks are contact metamorphosed, and those which are Paleozoic limestone, near the Glove mine, contain contact metamorphic ore deposits, chiefly of base metals and silver. Magmatic activity was particularly widespread during the Laramide Orogeny extending through much of Late Cretaceous time and through Paleocene time. Quartz monzonite and granodiorite of the Corona stock were intruded into the rocks of the northern end of the Santa Rita Mountains about 73 to 74 my (million years) ago. Three composite stocks were emplaced in the central and southern part of the mountains about 67 to 69 my ago. Of these three, the Josephine Canyon Diorite ranges widely in composition from diorite, syenodiorite, and quartz diorite to granodiorite and quartz monzonite. It is generally a fine- to medium-grained light- to dark-gray rock, and locally it may be associated with base-metal mineral deposits. The Madera Canyon Granodiorite is a coarse—grained light- to medium-gray porphyritic and locally metamorphosed rock not obviously associated with mineral deposits. The Elephant Head Quartz Monzonite is also coarse grained, is pinkish gray, and is only locally associated with a little mineralization. Scattered small stocks and plugs, such as the Gringo Gulch pluton, were intruded into rocks of the southern part of the Santa Rita Mountains, at least in part during the early Paleocene. They include some hypabyssal rocks of uncertain composition, age, and geologic association, some hornblende dacite porphyry plugs, and a small micro- granodiorite plug. During the late Paleocene, one group of small barren stocks and a group of ore—associated plugs were intruded into complexly faulted rocks of the Helvetia and Greaterville mining districts of the northern part of the Santa Rita Mountains. The stocks of Helvetia are mainly of granodiorite and quartz monzonite; they are elliptical in plan, their position and shape is only slightly controlled by faults, and the host rocks are widely metamorphosed. The plugs of Greaterville are of quartz latite porphyry, are commonly irregular in plan, and are strongly controlled by faults. Hydrothermal fluids associated with these plugs deposited base and noble metals in favorable host rocks along faults near the plugs. A small stock of granodiorite, of late C?) Oligocene age, was intruded into the rocks‘ of the San Cayetano Mountains. This stock is believed to occupy part of a magma chamber associated with an extensive volcanic pile in the nearby Grosvenor Hills area. INTRODUCTION Plutonic rocks and related hypabyssal intrusive rocks underlie much of the Santa Rita Mountains and adjacent mountains of southeastern Arizona. Significant intrusive events are recorded during the Precambrian, Triassic, Jurassic, Late Cretaceous, Paleocene, and Oligocene. The plutonic rocks intrude sequences of Precambrian metamorphic rocks, Paleozoic sedimentary rocks, and several sequences of Mesozoic and Cenozoic sedimentary and volcanic rocks. As a result of this extensive geologic record, the Santa Rita Mountains arethe best referencearea from which to develop the regional geologic hisrory. The plutonic rocks deserve special attention not only because they are closely related to the depositional and tectonic development of the area, but also because they provide most of the radiometrically datable rocks, and because some of them seem to be genetically related to mineral deposits. The Santa Rita Mountains are the first range southeast of Tucson (fig. 1). They extend more than 25 miles southward from Pantano Wash, along which the main railroad and highway east of Tucson lie, to Sonoita Creek, about 12 miles from the Mexican border. The mountains commonly reach heights of 6,000 to 7,000 feet and the highest, Mount Wrightson, is 9,453 feet high. The broad 1 2 valley east of the mountains lies at an elevation of about 4,500 feet, whereas the valley west of the mountains is only 3,000 feet high. The mountains are high enough to receive an amount of precipitation adequate to support an extensive, though largely scrubby, forest, which contrasts strikingly with grasslands in the high valley to the east and with the thorn brush and cactus vegetation typical of the Sonoran Desert plant assemblage in the low valley to the west. OBJECTIVES A geologic investigation of the Santa Rita Mountains was carried out by the U.S. Geological Survey during 1962— 68 to help determine the geologic history of a part 111°OO’ PLUTONIC ROCKS. SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA of southeastern Arizona having the potential of being one of the major sources of copper in the country. Several mountain ranges were studied during part of the 1950’s and 1960’s by eight geologists, J. R. Cooper, S. C. Creasey, T. L. Finnel, P. T. Hayes, E. R. Landis, R. B. Raup, F. S. Simons, and the author. The results of the investigation in the Santa Rita Mountains are presented mainly on two geologic maps (Drewes, 1971b, c) and in a report on the structural geology of the area (Drewes, 1972b). Details on field and laboratory observations and analyses that support the main interpretations and the results of subordinate studies on particular topical problems are presented in separate reports, referred to 110°00’ TyCSON MOUNTAINS L/g; TUCSON \ ' v6 i. "' RINCON § 10 VNEY QUAD /‘3000 R» m a 32°, 34 oo' TWIN BUT ES 'SAi-tugtarfg. EMF! 0 N M4000 UAD $61 ‘ " QUAD 3 4° Twin 4“,” 4% Buttes 33 Q&b° \ 0 ““0 m \e s i a {0/ _ A3»! \ 00 : l x: K 1‘ DRA\GOON QUAD \ _/ s-fi E RAGOQ DRAGOON MOUNTAINS QOOOCERRO COLOR D Kin/ITS Uz'co= i O I_MA co $3? *4 ~-"MOU NTAINS o (g HUACHUCA MUS‘FAN az 0“ D 9f“ ~ Babacomari ’ -... n. [-3 o * 7 t}; Q» 0:: Goo ’Ir Hua F35: Sierra ”00' 2“ mom" [a '0 Vim . . . , ’ n1 _ ‘Wmmso ., . Q 31° 0 o ‘ 0.1.3.13 ' GI 30' R b ‘9 LI Yo Y? 3 D \goégig \ AR] . \ “VOA/019:4--- goggles: : _ $OQL59U . IiNYSIDE 2!.9 1o 20 MILES | 4 °T° I 10 I 20 KILOMETRES CONTOUR INTERVAL 1000 FEET DATUM IS MEAN SEA LEVEL FIGURE 1.—Location of the Sahuarita and Mount Wrightson quadrangles and the Santa Rita Mountains, southeastern Arizona. Major copper mines: 1, Mission-Pima; 2, Twin Buttes; 3, Esperanza; 4, Sierrita. Published geologic maps: Twin Buttes quadrangle (Cooper, 1974), Empire Mountains quadrangle (Finnell, 1973), Benson quadrangle (Creasey, 1967), Happy Valley quadrangle (Drewes, 1974), Nogales and Lochiel quadrangles (Simons, 1974), Sunnyside quadran- gle and parts of Hereford and Fort Huachuca quadrangles (Hayes and Raup, 1968), Bisbee quadrangle (Hayes and Landis, 1964), and Dragoon quadrangle (Cooper and Silver, 1964). INTRODUCTION 3 herein where pertinent. The objective of the present report is to support and augment the study of the structural history of the Santa Rita Mountains by presenting abundant data on the plutonic and related rocks. The petrographic studies of the plutonic rocks of the Santa Rita Mountains include both field and laboratory observations. Almost 200 modal analyses were made by petrographic microscope. More than 50 chemical analyses of the major constituents of rocks and spectrographic analyses of the trace constituents of the rocks are presented, and 27 radiogenic age determinations are included. This considerable effort in describing and defining the plutonic rocks of the area was made for two reasons. First, the evidence was clear early during the field investigations that the area contained plutonic rocks of many ages, among them some new additions to the then-available geologic knowledge of the region. Second, some plutonic rocks were more closely related to mineralization than others, which suggested that distinguishing between the plutonic rocks was desirable for purposes of mineral exploration. Furthermore, the conditions of emplacement of the plutons and their subsequent alteration are also of potential economic interest. For the purpose of this report, the area referred to as the Santa Rita Mountains coincides with the Mount Wrightson and Sahuarita 15-minute quadrangles (fig. 2). A small part of the northern end of the mountains is therefore omitted; it lies in the Empire Mountains quadrangle mapped by T. L. Finnell (1971). At the southwest flank of the Santa Rita mountains the outlying hills named the Grosvenor Hills and the San Cayetano Mountains are included in this study. Included in this study of plutonic rocks are some granitoid rocks believed to be intruded near the surface and some fine-grained hypabyssal intrusive rocks believed to be genetically related to the plutonic rocks. Precambrian regionally metamorphosed rocks are also mentioned herein simply for convenience and in order to complete the general description of the rocks of the Santa Rita Mountains. Hereafter in this report general references to plutonic rocks of the area are used in the sense of including these few hypabyssal granitoid rocks and aphanitic rocks and the Precambrian metamorphic rocks, unless specifically stated otherwise. ACKNOWLEDGMENTS The geologic investigation of the Santa Rita Mountains was carried out approximately simultaneously with the other studies of the larger program of the US Geological Survey, mentioned in the preceding section. As a result, the investigation benefited through field consultations and from frequent discussions with my colleagues during the preparation of all the reports. I am particularly grateful for the benefits derived from the colleagues who have worked in the adjoining areas (fig. 1)—J. C. Cooper for the Sierrita Mountains, T. L. Finnell for the Empire Mountains, R. B. Raup for the Canelo Hills, and F. S. Simons for the Patagonia Mountains. Undoubtedly some geologic interpretations were developed jointly with them, yet the following presentation of these ideas and their validity in the context of the geology of the Santa Rita Mountains remain my own responsibility. The success of the field study was facilitated by many people. Invaluable assistance in mapping and sampling was provided by G. C. Cone, Bruce Hansen, C. W. Norton, F. W. Plut, J. R. Riele, R. A. Rohrbacker, F. Sutheimer, and W. M. Swartz. Discussions in the field benefited from the experience of R. E. Wallace, M. D. Crittenden, Jr., J. H. Courtright, and many others. The courtesies of George and Sis Bradt, Roy Green, Dewey Kieth, and George Yakobian, all residents near the Santa Rita Mountains, and those of Professors John Anthony, Evans Mayo, and P. E. Damon of the University of Arizona are appreciated. Invaluable support was obtained from many laboratories. One radiometric age determination was obtained from P. E. Damon before its publication, and the others were provided by R. F. Marvin, T. W. Stern, Z. E. Peterman, S. C. Creasey, and their many colleagues. Chemists and spectrographers who have contributed are acknowledged in the tables of analytical data. G. C. Cone also assisted in many phases of the preparatory work on samples. Previous geologic studies of the Santa Rita Mountains are few, and most of them consider very local areas or topical problems. The study by Schrader (1915) represents the pioneer effort at geologic mapping in the area and the influence of his work is still noticeable on the latest edition of the geologic map of Arizona (Wilson and others, 1969). Several Master of Science theses on file at the University of Arizona were also consulted. The present geologic study is thus most accurately viewed as a second-generation effort, even though in some respects this is an overstatement. GEOLOGIC SETTING The geologic setting of the plutonic rocks of the Santa Rita Mountains has been extensively described in earlier reports (Drewes, 1972a, b), but for convenience of the reader, the geologic history is briefly summarized here. The names, stratigraphic positions, and ages of the plutonic rocks and the other rocks of the area are shown in table 1. 1 The oldest known rocks of the area are regionally metamorphosed rocks Conventionally assigned to the Pinal Schist. In most nearby areas the formation is a muscovite- chlorite or muscovite-biotite schist, but locally this formation consists largely of biotite gneiss and granite gneiss and of only a small amount of schist, which are described in the following section of this report. These rocks were deformed, metamorphosed, and intruded, PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA 111° 00 4 T0 TUCSON 110°45’ R TO TUCSON 32° 00' T0 NOGALES \ 45' \ 311.1111.” 1 QUADRANG LE EMPIRE MOUNTAINS l OUADRANGLE Ripe, . -—\~ T0 BENSON 1y Sahuarita 0 .——_/“~ Santa Cruz Gre-n / Val‘Iey I 3|an q. A 31°30’ ELGIN OUADRANGLE 1'0 MILES \I 5 10 KILOMETRES CONTOUR INTERVAL 1,000 FEET DATUM IS MEAN SEA LEVEL 0 | | 0 FIGURE 2. -—The Santa Rita Mountains, showing areas covered by geologic maps in this report: A, figure 3; B, fig ure 20; C, figure 40; D, figure 27; E, figure 32; F, figure 11; G, figure 23; H, figure 15; I, figure 38. INTRODUCTION 5 TABLE l.—Stratigmphic summary of rocks of the Santa Rita Mountains, indicating the rocks described in this report Age Rocks described in this report Other rocks (mainly plutonic rocks) Quaternary and Tertiary ........................ Tertiary: Miocene and Pliocene .................. Oligocene ..................... Granodiorite of San Cayetano Mountains. Paleocene to Oligocene ......................... ............... Gravel. ............... Nogales Formation. l Grosvenor Hills Volcanics, lacco- liths, and dike swarms.’ ............... Rhyolite, andesite, and quartz veins. l Paleocene ..................... Aphanitic rocks, Greaterville in- trusives; granitic rocks, Hel- vetia stocks; rocks of Gringo Gulch pluton. Paleocene(?) ................................................. Gringo Gulch Volcanics, volcanics of Red Mountain.I Tertiary or Cretaceous ......... Quartz latite porphyry intrusives. Late Cretaceous ................ Elephant Head Quartz Monzonite. Do ....................... Madera Canyon Granodiorite. Do ....................... Josephine Canyon Diorite. Do ....................... Granitoid rocks of Corona stock. Salero Formation.l Do ..................................................... Fort Crittenden Formation.‘ Early Cretaceous .............................. ............... Bisbee Group. Do ..................................................... Bathtub Formation. | Do ..................................................... Temporal FormationJ Jurassic ....................... Squaw Gulch Granite. Triassic ....................... Piper Gulch Monzonite- Do .......................... - ........................... Gardner Canyon Formation. 1 Do ..................................................... Mount Wrightson Formation.‘ Permian .................................................... Rainvalley Formation. Do ..................................................... Concha Limestone. Do ..................................................... Scherrer Formation. Do ..................................................... Epitaph Dolomite. Do ..................................................... Colina Limestone. Permian and Pennsylvanian ..................... Pennsylvanian ................................ Mississippian ................................. Late Devonian ................................ Late and Middle Cambrian ..................... ............... Earp Formation. ............... Horquilla Limestone. ............... Escabrosa Limestone. ............... Martin Formation. ............... Abrigo Formation. Middle Cambrian ............................................ Bolsa Quartzite. Precambrian Y ................. Continental Granodiorite. Precambrian X ................ Pinal Schist and granite geneiss. 1 Formation that contains volcanic rocks. either during one geologic event or separately in the given sequence, by a large pluton of Continental Granodiorite (Drewes, 1968). The rocks were uplifted and deeply eroded before the deposition of a sequence of Paleozoic rocks; indeed, considering evidence from the Rincon Mountains (fig. 1) 10 miles north of the Santa Rita Mountains, this uplift predates the deposition of the Apache Group of Precambrian Y age. The Paleozoic rocks of the Santa Rita Mountains are believed to have lain unconformably on the Precambrian crystalline rocks, although they are now invariably faulted upon them. The Paleozoic rocks form a concordant marine sequence that begins with the Bolsa Quartzite of Middle Cambrian age, contains several disconformities, and ends with the Rainvalley Formation, of late Early Permian or slightly younger age. Nearby areas have no Permian rocks younger than the Rainvalley Formation. This marine sequence was deposited in a shallow continental sea and its internal disconformities mark a succession of epeirogenic fluctuations of an otherwise tectonically stable region. Moderate tectonic activity marked the interval between the deposition of Permian rocks and the onset of the Laramide orogeny in mid-Cretaceous time. Toward the end of the Permian the region was uplifted and the sea retreated; in Early Triassic time, the area now occupied by the Santa Rita Mountains was faulted, an uplifted block was deeply eroded, and a thick pile of volcanic and sedimentary rocks of the Mount Wrightson and Gardner 6 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA Canyon Formations were deposited in a fault-block basin (Drewes, 1971a). Near the end of Triassic time a stock of the Piper Gulch Monzonite was intruded, perhaps along an old fault zone, the Santa Rita fault scar (Drewes, 1972b). During the Jurassic Period a large stock or small batholith of granite intruded the southern part of the area. Faulting and erosion followed, and during Early Cretaceous time the Temporal Formation, Bathtub Formation and Bisbee Group were deposited in a continental environment. Both the local relief and tectonic activity decreased during this interval and, before its close, a shallow sea briefly encroached upon the area for the last time. The initial impulse of the Laramide orogeny occurred during mid-Cretaceous time, for the Upper Cretaceous Fort Crittenden Formation was deposited unconformably upon the Lower Cretaceous Bisbee. The deposition of coarse clastic rocks and resumption of volcanism mark the upper part of the Fort Crittenden and widespread volcanism is recorded in the succeeding Salero Formation. During this time, the Corona stock was emplaced in the northern part of what is now the Santa Rita Mountains. Major northeastward-directed thrust faulting marked the tectonic climax of the Piman phase of the Laramide orogeny during the time of deposition of the Salero Formation (probably Campanian time). The Piman phase culminated a few million years later with a series of multiple intrusions of rocks known as the Josephine Canyon Diorite, Madera Canyon Granodiorite, and Elephant Head Quartz Monzonite. During early Paleocene time, about 63 to 58 my (million years) ago, the area was tectonically fairly quiet. The Josephine Canyon stock, a large epizonal body, was partly exposed by erosion and the volcanic and sedimentary rocks of the Gringo Gulch Volcanics and the volcanic rocks of Red Mountain were deposited in the southern part of the area. The Gringo Gulch pluton and other small bodies were probably emplaced during that time. In late Paleocene time, tectonic and volcanic activity was renewed in the northern part of the area to record the Helvetian phase of the Laramide orogeny (Drewes, 1972b). Minor northwestward-directed thrust faulting was penecontemporaneous with the emplacement of a group of small stocks named the Helvetia stocks. Quartz latite porphyry intrusives that are locally referred to as “ore porphyry” because of their association with mineralization (Drewes, 1970) were intruded at the close of the Helvetian phase of the Laramide orogeny. During the time between the Paleocene and Oligocene the local geologic record is fragmentary. Mineralized quartz veins were emplaced in the southern part of the area and some rhyolite volcanic rocks and andesitic intrusive rocks were emplaced in the northern part. The area was strongly eroded and possibly block faulted to form the ancestral Santa Rita Mountains. During late Oligocene time magmatic activity was renewed in the southern part of the area. Rhyolite and rhyodacite of the Grosvenor Hills Volcanics were extruded as flows and tuffs. Vitric laccoliths and dikes were intruded into these tuffs (Drewes, 1972a), and at a considerably greater depth the feeder dikes of the volcanic rocks and hypabyssal intrusive rocks are believed to have emanated from small stocks such as that of the granodiorite of the San Cayetano Mountains, the youngest pluton of the area. The area was block faulted several times during the late Tertiary and Quaternary, and gravel of several formations filled the basins next to the raised and gently southeastward tilted Santa Rita Mountains. The latest magmatic activity is recorded by basalt flows, of Miocene or Pliocene age that are intercalated in the oldest unit of gravel along the Santa Cruz River a few miles south and west of the Mount Wrightson quadrangle. In the above review of the geologic setting of the plutonic rocks, as well as in the summary in table 1, the association of plutonic events with volcanic and tectonic events seems clear. Likewise, some of the magmatic events are shown to be related, at least in time if not also in origin, to mineralization. PRECAMBRIAN ROCKS Rocks of Precambrian age compose an older group of regionally metamorphosed granite, gneiss, and schist and a younger granodiorite pluton. As mentioned before, the metamorphic rocks are described with the plutonic rocks because they have not been previously described in the topical papers on the Mesozoic or Cenozoic rocks of the area. The metamorphic rocks include rocks assigned to the Pinal Schist and to an associated granite gneiss, and they are intruded by the Precambrian Continental Granodio- rite. PINAL SCHIST The oldest rocks of the Santa Rita Mountains are foliated rocks that sufficiently resemble the Pinal Schist of some nearby areas to be assigned to that formation. The Pinal Schist, as it has been used in southeastern Arizona in recent decades, includes a wide variety of metasediment- tary, metavolcanic, and other metaigneous rock types, probably of significantly different ages from place to place. Cooper and Silver (1964, p. 23, 24) described in detail a variety of these rocks that are extensively exposed 30 miles east-northeast of the north end of the Santa Rita Mountains. Somewhat more gneissic varieties are being studied in the Rincon Mountains (Drewes, 1974). In the Santa Rita Mountains, Pinal Schist appears in a few small widely scattered areas, shown in detail on the geologic map of the Sahuarita quadrangle (Drewes, 1971b) and shown in a more general way, as rocks older than the Continental Granodiorite, in figure 3 of this report. Biotite gneiss and schist are the most common lithology, and PRECAMBRIAN ROCKS 7 chlorite schist and phyllite appear in a few localities. The Final Schist in the large outcrop areas, lying athwart the crest of the mountains west of Greaterville, is biotite gneiss and schist that is intruded to the north and east by the Continental Granodiorite and is faulted against younger rocks to the southwest. The intrusive contact is broad and gradational, and inclusions of Final Schist occur in the granodiorite as far north as Box Canyon (fig. 3). The rocks of the main outcrop areas consist of alternating layers, typically a half an inch to many inches thick, of biotite gneiss or biotite schist and of feldspathic rock. Less commonly the rocks also include hornblende- biotite gneiss and some pegmatite bodies and lamprophyre dikes that may be related to the Continental Granodiorite. Weathered outcrops of biotite gneiss are dark grayish brown and have the sheen of a micaceous surface. The relatively unweathered outcrops, such as those along the bottom of Sawmill Canyon, are medium dark gray and the gnessic layers are more conspicuous. In places these layers are strongly deformed on a small scale, in the manner of ptygmatic folds. Elsewhere the layers are cut by many small faults and intrusives that are internal features of the formation which may be related to the emplacement of the nearby Continental Granodiorite. Locally the attitude of the foliation appears to be erratic, but over the outcrop area foliation strikes northwesterly and dips steeply southwest. Under the microscope, the grains of a typical dark or micaceous layer are seen to be about 0.5 mm long and those of the adjacent light-colored felsic layers 0.5 to 1.5 mm. Alined grains of biotite control the strong schistosity of the rock; grain alinement is fairly subtle in the felsic layers. Dark layers have an estimated mode of: quartz, 5; andesine plagioclase, 65; biotite, 30; ilmenitic magnetite, 3; and traces each of sphene, apatite, zircon, and allanite(?). Light layers have an estimated mode of: quartz, 50; andesine, 40; biotite, 5; and traces each of magnetite, sphene, and apatite. Alteration minerals are scarce and they consist of a little chlorite and epidote. The anorthite content of plagioclase ranges from 38 to 42 percent. Biotite is pleochroic in pale yellow brown to moderate brown. Chlorite schist and phyllite are mapped (Drewes, 1971b) as small slivers along a northwest-trending tear fault about half a mile east of Helvetia. The rocks crop out on dark-gray low knolls or ledges in many places surrounded by slopes of light-brownish-gray Continental Granodio- rite. The schist consists of microscopic layers of alined mica and of quartz and feldspar with a cataclastic texture. Biotite, plagioclase, and quartz each make up about 30 percent of the rock, sericite and chlorite each almost 5 percent of the rock, and ilmenitic magnetite, zircon, and apatite trace amounts. The chlorite- and biotite-rich layers have a felty texture, indicative of recrystallization after shearing. The quartz forms mosaic-textured or interlocking aggregates of crystals, also indicative of postcataclasis recrystallization. The origin and metamorphic development of the biotite gneiss and schist and of the chlorite schist and phyllite are uncertain. The biotite gneiss may have been a sedimentary rock, whose alternating beds of siltstone and arkosic sandstone were transformed by dynamothermal (load?) metamorphism into a rock in the biotite-chlorite subfacies of the greenschist facies (Turner and Verhoogen, 1951, p. 446). Many of the aplitic sheets intruded subparallel to the foliation could be orogenic features, perhaps related to metamorphism itself. These events may have occurred during the Mazatzal Revolution (Wilson, 1939, p. 1161). The chlorite schist and phyllite were tectonically deformed by movement along the tear faults in which the slivers lie. These faults were active during the Helvetian phase of the Laramide orogeny (Drewes, 1972b). The posttectonic recrystallization was probably concomitant with the emplacement of the youngest quartz latite porphyry plugs toward the end of the Paleocene. Alternatively, the recrystallization may be still younger and related to the mid-Tertiary thermal event recorded mainly in the Rincon Mountains. GRANITE GNEISS Granite gneiss crops out over an area of about 3 square miles near Cottonwood Canyon on the west flank of the southern part of the Santa Rita Mountains; it is shown as schist and granite gneiss in figure 3 and is shown in greater detail on the geologic map of the Mount Wrightson quadrangle (Drewes, 1971c). The gneiss is intruded by granitoid rocks that resemble the Continental Granodiorite or a strongly metamorphosed equivalent of that granodiorite. The foliation of the gneiss and the geologic relation with the granodiorite suggest that the gneiss is probably of the same general age as the Final Schist. Rocks of Paleozoic or younger age overlie, or are faulted against, the granite gneiss. The granite gneiss underlies low hills, which are uniformly colored a pale brownish gray, that contrast subtly with the more varicolored hills of volcanic rocks to the east and west. A northwest-trending grain, reflecting the strike of the foliation, helps to distinguish this terrain from the nearby hills of Jurassic granite. Although generally striking northwestward, the foliation of the gneiss is arcuate, concave to the northeast, and dips moderately to the southwest. In detail, the lithology of the granite gneiss varies as much as does that of the Final Schist. In the northeastern part of the outcrop area the rock is a faintly gneissic quartz monzonite. The central part of the area contains the typical granite gneiss, and the southwestern part contains a biotite-granite gneiss. Shearing and chemical alteration of .the rocks increases toward the north and northeast, where faults bound the gneiss. PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA 110°45' EXPLANATION Younger rocks Igneous and sedimentary rocks of Phanerozoic age Continental Granodiorite and related rocks of Precambrian Y age Ye, granodiorite, quartz monzonite, and aplite Yee, Continental Granodiorite, con- cealed beneath pediment gravel Older rocks Schist and granite gneiss of Precambrian age Contact Dotted where concealed Dotted where concealed . 9 01 1 *1 Moded Chemically Radiometrically analyzed dated Specimen collection sites and numbers k Maum‘ H Wkiahmn 110°55’ l 2 3 4 SMILES | l l | | l l 1 2 3 4 SKILOMETRES OVAO FIGURE 3.—Distribution of Precambrian rocks and specimen collection sites. PRECAMBRIAN ROCKS 9 The granite gneiss has a hypidiomorphic-granular to tabular microscopic texture that is modified by an alinement of mineral grains and by a slight zoning of light- and dark-colored grains. In this section, some specimens show a seriate texture. Mineral grains are commonly 0.2—3.0 mm long, and in some specimens the grain size distribution is bimodal with the smaller grains included in larger poikilitic grains. Essential minerals of the gneiss typically are quartz, microcline, plagioclase, and biotite, but some rocks have only one of the feldspars. The microcline contains very fine perthitic intergrowths and indistinct grid twinning that appears to be partly obliterated. The plagioclase commonly has a calcic oligoclase composition, but in some rocks it has an albite composition, presumable as a result of albitization. Biotite forms discrete subhedral plates that are pleochroic in yellowish brown to moderate brown and that are partly chloritized. In addition, the rock contains 0.5 to 3 percent each of ilmenitic magnetite and apatite and trace amounts of sphene and zircon. This entire mineral assemblage is slightly to moderately altered to epidote, chlorite, sericite, kaolinite, and leucoxene. The granite gneiss presumably was formed by the metamorphism of granitoid rock. This metamorphism probably is part of the regional metamorphism affecting the Final Schist, for it is associated with a well-developed foliation and shows no increase in intensity near the contact with younger plutonic rocks. This rock, then, may well be the oldest plutonic rock of the Santa Rita Mountains. CONTINENTAL GRAN ODIORITE The Continental Granodiorite is a coarse-grained and coarsely porphyritic rock that forms part of a single large stock (or batholith?) in the northern part of the Santa Rita Mountains; it also forms some small intrusive bodies on the southwest flank of the mountains (fig. 3). The formation was defined and was briefly described by Drewes (1968); further details on its distribution are shown on the geologic maps of the area (Drewes, 1971b and 1971c). The granodiorite is generally massive, but locally it is faintly foliated. Dark minerals are fairly abundant in this rock, and they typically form aggregates that form a meshwork around the light-colored minerals. The Continental Granodiorite intrudes the Final Schist, which forms roof pendants or remnants of wallrock, as well as some scattered inclusions. Small bodies of granodiorite also intrude the granite gneiss of the southwestern flank of the mountains. Bolsa Quartzite, of Cambrian age, is the oldest rock overlying the granodiorite, but, inasmuch as these rocks seem to be separated everywhere by a fault, the geologic relations have been variously interpreted as a depositional contact and as an intrusive one. The fault separating the granodiorite from the Bolsa is believed to be a regional thrust fault of the Piman phase of the Laramide orogeny (Drewes, 1972b), which followed a major unconformity. The complexities of this contact led to conflicting interpretations on the age of the granodiorite, as reviewed in the section on the age of that rock. PETROGRAPHY The Continental Granodiorite is made up mainly of granodiorite, but it includes small aplitic dikes and sills and small lamprophyre dikes. These small intrusive bodies presumably are genetically related to the granodiorite pluton because they are virtually coextensive with it and are absent in the younger rocks around it. The included dikes and sills are too small to be shown in figure 3 but the distribution of some of the larger aplitic masses are shown on the geologic maps of the northern part of the mountains (Drewes, 1971b; 1972b, pl. 5). GRANODIORITE The dominant rock type of the Continental Granodiorite is mostly biotite granodiorite, but in places it grades into biotite quartz monzonite, and it includes some biotite-hornblende granodiorite. The rock of the main outcrop area in the northern part of the mountains (fig. 3) is almost free of hornblende and grades, seemingly at random, from the biotite granodiorite to the biotite quartz monzonite. The rock of the small outcrop areas, to the southwest, contains several percent of hornblende. Slopes underlain by the granodiorite are somber colored and mostly a light brownish gray. Outcrops are small and irregular; even in the deepest canyons, such as Box Canyon, they rarely form extensive cliffs because the rock is abundantly fractured and disaggregates readily. As a result of these internal weaknesses, the local relief on slopes underlain by granodiorite is generally more subdued than that on nearby slopes underlain by Paleozoic rocks or by some of the younger plutonic rocks. In areas of low relief the granodiorite is covered thickly by grus and in all areas the freshest exposures are in the narrowest and deepest gullies or canyons. The one exception to this weak weathering habit occurs along the crest of the mountains southeast of Helvetia, where the granodiorite is very massive and less friable than elsewhere, and so it forms large rounded knobs and bosses which dominate the landscape. Unweathered specimens of granodiorite (fig. 4) are dark gray, commonly with a slight greenish tinge in rocks having perhaps as much as 20 percent dark minerals, but with a pinkish cast in rocks having only about 10 percent dark minerals. Biotite and chlorite, most abundant of the dark minerals, form clusters between or around the quartz and feldspars. Phenocrysts, which in part are actually porphyroblasts, are of potassium feldspar and are as much 10 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA as 4 cm long. In most rocks they make up about 5 percent of the volume, but they are scarce or absent in some of the small intrusive masses on the southwest flank of the mountains. Compared to the phenocrysts in the younger plutonic rocks, they are mostly smaller and less abundant. The large phenocrysts of the Continental Granodiorite are found in the grus mantle on weathered terrane, and they persist as clasts in some Mesozoic conglomerates that overlie the granodiorite. The granodiorite locally contains small widely scattered xenoliths that are mostly of uncertain origin. A few of them shown in the areas of Box Canyon and Enzenberg Canyon (Drewes, 1971c; and about 1 mile south of collection site 3, fig. 3, this report) are from the Final Schist. Elsewhere the granodiorite contains smaller xenoliths that resemble the granite gneiss unit. Still others may be remnants of a hornblende gneiss or an amphibolite not known at the surface. Zones of sheared rock and veinlets of aplitic material and of quartz are also common features of the granodiorite. Some of the sheared zones are obviously related to mapped faults, but most of them are small and so widely scattered and randomly oriented as to indicate little more than that the rocks apparently have had a long and involved geologic history. The aplitic veinlets may be genetically related to the larger mapped bodies of aplite, which are described in the following section of this report. In thin section, the granodiorite is seen to have a hypidiomorphic-granular texture in which the groundmass FIGURE 4.—Specimen 17. Continental Granodiorite, showing porphyritic texture and clustered dark minerals. grains are 3 to 7 mm in diameter. A cataclastic texture is superposed on the hypidiomorhpic-granular texture of those specimens from the eastern bodies of hornblende- biotite granodiorite, which occur near the fault that bounds the Precambrian rocks. The samples from in, and north of, the Box Canyon area have a mosaic to sutured or seriate texture superposed on the granular one. Traces of microgranophyric texture appear in some specimens and the rims of many large crystals are poikilitic. Poikilitic grains and a bimodal distribution of groundmass grain size are particularly common in the southern part of the outcrop area of hornblende-biotite granodiorite, which appears to be metamorphosed by the nearby pluton of Squaw Gulch Granite. Modes of 24 samples of Continental Granodiorite, grouped to show the mineral composition of the unmetamorphosed and slightly metamorphosed rocks of the main granodiorite stock separately from the other rocks, are listed in table 2. The collection sites of the FIGURE diorite showing indistinct grid twinning in microcline. Crystals: micro- cline (M), microcline or orthoclase (mo), plagioclase (P), quartz (Q), biotite ( B); ilmenitic magnetite (‘im), and sphene (S). Crossed nicols; X 20. 5.—Specimen l9. Slightly metamorphosed Continental Grano- PRECAMBRIAN ROCKS l 1 samples are shown in figure 3; and additional data in field notes and maps, filed in the U.S. Geological Survey Records Center in Denver, are keyed to this report through the field numbers. Standard techniques were used in moding the plutonic rocks of the area; many thin sections were preferred to a few rock slabs. In such coarsely porphyritic rocks as the Continental Granodiorite modal data involving potassium feldspar are most reliably obtained by using the mean of many counts, and even then the relatively large value of the standard deviation suggests that the data are insufficient. Point counting, using an 0.5- mm grid, was done on thin sections that were stained with sodium cobaltinitrate, as needed. Separate counts of 1,000 points each were made of the top and bottom halves of thin sections in an attempt to check the modal uniformity over a thin section area of 7 to 8 square centimetres. The relative abundance of quartz between the halves of a moded thin section, reported as the quartz index, gives an approximate measure of the mineralogic uniformity of a rock at this scale. Thus, a quartz index of 1 implies perfect mixing, and the smaller the value of the index, the poorer the mixing. To some extent, of course, variations in grain size between rock types interfere with this attempt, the finer grained rocks giving the visual impression of being better mixed. However, the mean quartz index of rock types having similar grain size varies sufficiently to suggest that the degree of mixing is an intrinsic feature of the rock. The main mineral constituents of the granodiorite are quartz, two feldspars, biotite, and, in some specimens, hornblende. Among the light-colored grains, quartz is anhedral and bimodally sized in those rocks which have the mosaic or seriate texture. Its undulatory extinction is commonly large, but that of quartz of the strongly metamorphosed granodiorite of the southwest flank of the mountains is faint or absent. Plagioclase forms subhedral tabular crystals. The anorthite content of plagioclase from the biotite granodiorite is 0 to 30 percent; in about half the specimens crystal cores are of oligoclase and the rims albite, whereas in the other specimens only albite is present, suggesting that the granodiorite is incompletely albitized. The anorthite content of plagioclase from the hornblende-biotite granodiorite is 35 to 50 percent, in the range of andesine. The plagioclase of most specimens is moderately to intensely altered to sericite and clay minerals, which are stained by some iron oxide; the specimens from the strongly metamorphosed hornblende- biotite granodiorite are relatively unaltered. The potassium feldspar of most specimens forms anhedral subequant grains. In the unmetamorphosed rocks the potassium feldspar is microcline; in the other rocks it is either orthoclase or microcline or possibly both. Typically the microcline grid twinning is indistinct or blurred and discontinuous through a crystal (fig. 5) of the slightly metamorphosed rock. Most orthoclase and microcline is perthitic, with lace and patch perthite types and coarse and fine perthite sizes equally common (fig. 6). The albite phase of the perthitic crystals are moded separately; in this report such albite is considered to be part of the plagioclase component, but its abundance is tabulated independently to permit an alternate grouping. Albite in perthitic grains makes up only 2 to 3 percent of most specimens but ranges from 0.8 to 8.4 percent. Kaolinite alteration of potassium feldspar is ubiquitous and in many specimens is intense. Dark minerals are more abundant in the Continental Granodiorite than in most other plutonic rocks of the area; the color index of the rock ranges from 10 to 20 percent. Biotite, or its chlorite alteration mineral, is the most abundant dark mineral. It forms clusters of relatively large subequant grains in unmetamorphosed rocks (fig. 6) or felty aggregates of relatively small tabular grains in metamorphosed rocks (fig. 7). Biotite is pleochroic in yellowish brown to moderate brown. The alteration of biotite to chlorite is a common feature; relatively unaltered biotite is most common in the metamorphosed rocks, and the more intense the metamorphism, the more abundant the biotite seems to be, no doubt as recrystallized or secondary micas. Hornblende is pleochroic in pale yellowish green to pale olive green to pale bluish green. Some hornblende grains are poikilitic, containing quartz and biotite. Magnetite is invariably ilmenitic and commonly is clustered with biotite and sphene. The abundance of zircon is generally greater in this granodiorite than in other plutonic rocks of the area, and the zircon of specimen 19 (fig. 5) is probably hyacinth zircon. Alteration minerals are chiefly sericite, clay minerals, and chlorite; epidote, uralite, calcite, iron oxide, leucoxene, and penninite are also present in smaller amounts and in various combinations. APLITIC ROCKS Aplite, alaskite, and fine-grained leucocratic quartz monzonite bodies in the Continental Granodiorite are hereinafter collectively referred to as aplitic rocks of the granodiorite. The aplitic rocks form many intrusive masses too small to be shown either in figure 3 of this report or on the geologic maps, but the larger masses are shown on the map of the Sahuarita quadrangle (Drewes, 1971b). The quartz veinlets that were mentioned in the preceding section of this report as being scattered throughout the granodiorite may be related to the aplitic rocks but are unmapped because of their small size. The intrusive masses of aplitic rocks commonly are lenticular or tabular, a few feet to a» few tens of feet thick, and hundreds of feet long. Some of the masses are pluglike(?) and are irregular in plan; others resemble closely spaced dikelets. Many of the tabular intrusive 12 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA FIGURE 6.—Specimen 12. Continental Granodiorite showing a cluster of large chloritized biotite crystals and part of a perthitic microcline phen— ocryst that has a eoarsedextured lace to patch perthite. Crystals: microcline (M) and altered albite (ab) of the perthitic phenocryst, quartz (Q), chloritized biotite (cb), ilmenitic magnetite (im), calcite (ca) derived from alteration of biotite, apatite (ap) leucoxene (L) derived from sphene, and possible zircon (Z). Crossed nicols; X 20. masses are clustered in belts of subparallel-striking sheets in which gently inclined masses are as abundant as steeply inclined ones. The contact of most of the aplitic rocks with the granodiorite is sharp, but in a few places, such as the hills a mile southwest of the cemetery below Helvetia and around the large aplitic mass 1 to 1.5 miles south of Box Canyon (Drewes, 1971b), the contact is apparently gradational. The aplitic rocks weather yellowish gray rather than the more somber brownish gray of the granodiorite, and outcrops of aplitic rocks are more resistant to weathering and hence are more abundant and somewhat more rugged than those of the granodiorite. Outcrops of aplitic rocks commonly form narrows and falls where they cross the bottoms of canyons. Small angular blocks of aplitic rocks are strewn around the outcrops, along with a grus that does not contain large chips of feldspar phenocrysts. FIGURE 7.—Specimen 17. Continental Granodiorite showing part of a phenocryst or porphyroblast and small felty-textured recrystallized biotite. Crystals: orthoclase or microcline (om), quartz (Q), plagio- clase (P), biotite (B), and altered albite (a b) in fine lacy perthitic intergrowths, ilmenitic magnetite (im), sphene (S), apatite (a p), and zircon (Z). Crossed nicols; X 25. Mineral grains of the aplitic rocks are mostly 1 to 2 mm in diameter and are arranged in an idiomorphic-granular texture. Most of the aplitic masses have a felty or granular nonporphyritic texture; some have a few phenocrysts, which are generally less than 7 mm long. The assemblage and abundance of minerals in the aplitic rocks differ slightly from those of its granodiorite host. Modal analyses of two rocks, specimens 23 and 24, are given in table 2. Estimated modes of three additional rocks, not presented in the table, show some variations from the tabulated modes; the mica of one rock is biotite, that of another is muscovite, and the third contains both micas. Likewise, apatite and zircon are in only some of the rocks and not consistently together. And finally, the potassium feldspar is orthoclase in some rocks and microcline in others. Throughout the aplitic rocks, though, the plagioclase is either albite or oligoclase. The albite in perthitic grains is finely lacy and rarely exceeds a few PRECAMBRIAN ROCKS 13 percent by volume. Biotite occurs either as widely scattered chloritized flakes or as unaltered inclusions in quartz. Felty aggregates of small biotite crystals, typical of recrystallized rocks, are found in the aplitic rocks 1 to 2 miles southwest of Helvetia. Sericite and clay minerals are abundant alteration products in the aplitic rocks. LAMPROPHYRE Small dikes of dark-greenish-gray rock are sparsely scattered throughout the Continental Granodiorite. Some dikes occupy dilation fractures as much as 3 feet wide and are a fine-grained dioritic rock. Other dikes form short tabular bodies as much as 10 feet wide and are a finely porphyritic dacite or andesite. The outcrops of both rock types are inconspicuous and generally lie in small sags and low areas. A few specimens of lamprophyre have an ophitic to subophitic texture, with crystals 0.1 to 0.3 mm long. A few acicular crystals and rare phenocrysts are as much as 4 mm long. Plagioclase, potassium feldspar, and an amphibole are the most abundant minerals; quartz and pyroxene(?) are present in small amounts, and magnetite, apatite, and zircon occur in trace amounts. The plagioclase has a composition of albite, probably as a result of diagenetic(?) albitization. Alteration minerals, such as sericite, clay minerals, chlorite, and epidote, are abundant. In the Santa Rita Mountains, many of the plutons and their host rocks contain lamprophyre dikes. Although slightly variable in habit, texture, and phenocryst mineralogy from dike to dike, the lamprophyre dikes of some plutons are uniform. In the absence of a considerably more detailed study of these dikes, they are considered simply as lamprophyres from late-phase magmas of the plutons with which they are associatied, rather than as the result of distinctly later and genetically unassociated magmatic events. MODAL AND CHEMICAL SUMMARY For ease of comparison of the abundant analytical data on the plutonic rocks, the tables of analyses are augmented by a series of modified triangular diagrams summarizing the modes and by a series of histograms summarizing the chemical analyses. The triangular diagram summarizing the modes consists of two parts (fig. 9): the left half is part of a standard quartz—potassium feldspar—plagioclase diagram, following the general method of Johannsen (1939, p. 152) and as used by Bateman (1961, p. 1524) and by Ross (1969, p. 7). The quartz-rich corner is eliminated, as it is unoccupied by analyses of the plutonic rocks of the Santa Rita Mountains suite; and the right half is a similar but smaller segment of a quartz-plagioclase-femic mineral triangular diagram. Together, these two parts of the diagram represent the front and right side of a tetrahedron (fig. 8) showing the Quartz Femic minerals Potassium feldspar P Plagioclase FIGURE 8.—Isometric diagram of a modal tetrahedron whose apices QK P F represent, respectively, the quartz, potassium feldspar, plag- ioclase, and femic minerals. The solid A B C D K P is a hemitetrahe- dron. The solid EG H J K P is the body whose front face, E K PJ, and right side, J P H, are projected into the plane of the modified tri- angular diagrams, such as that of figure 9. The true mode of hypotheti- cal specimen, 8, (having 16 percent quartz, 16 percent orthoclase, 48 percent plagioclase, and 20 percent femic minerals) is first plotted as 8’ on the leucocratic face, QK P, of the tetrahedron in the usual manner, the femic component having been subtracted and the others recalculated to 100 percent. The projection of the true mode on the right face, QPF, of the tetrahedron is then plotted as 8 along the edge, LF, of plane LFM, on which 8, s', and s”all lie. three major leucocratic components of granitoid rocks plus the combined femic mineral component. The method of plotting the femic component is illustrated in figure 8; the resulting diagram slightly exceeds in size a hemitetrahedron in the example shown in figure 9 but equals a hemitetrahedron in all following examples. In figure 9 the right side of the oversized hemitetrahedron is rotated into the plane of its front face. Each mode is represented by two points on the diagram, one showing the relative abundance of the leucocratic components and the other the actual abundance of the femic component. The true composition, in terms of these four components, must be visualized as a projection within the tetrahedron controlled by the two points shown. The method of projection shown in figure 8 gives both the recomputed data shown on the conventional triangular diagram plus actual data on the femic component, which the recomputed data on a triangular diagram showing quartz, plagioclase, and femic end members would not. A composite of the modal analyses of the Continental Granodiorite (fig. 9) shows the unmetamorphosed rocks to be scattered in a fairly compact subspherical body centered 14 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA TABLE 2.—Mode (in percent) [Field numbers are abbreviated, symbols showing year of collection and collector's initial omitted. Full field number of specimen 1 thus . . . . Slightly (regionally?) Rock type ............... Granodlorite (Includes quartz monzomte) metamorphosed granodiorite and quartz monzonite Specimen No ............. 1 2 3 4 5 6 7 8 9 10 11 12 113 1—13 14 15 16 Field No. ,,,,,,,,,,,,,,, 914 1034 1014 997 1022 913 994 899b 898 100 897 104 627 Mean 5 1220 1191 1142; Quartz .......... ,. 15.8 28.1 19.9 30.5 21.5 24.4 24.5 22.8 29.0 32.2 25.1 16.4 20 5 23.9 5.1 28.2 32.8 46.3 Plagiclase, total .......... 32.8 35.1 52.6 43.4 46.9 41.0 45.8 42.2 39.6 39.4 42.8 26.7 56.5 41.9 7.9 35.4 32.9 27.6 (plagioclase in perthite) (8.4) (1.0) (0.2) 0 (1.2) (0.8) (0.8) (3.5) (2.6) 0 (1.5) 0 0 (1.5) (2.3) 0 (5.1) (2.0) Microcline .............. 36.0 19.8 12.4 8.1 19.1 25.0 12.7 21.8 15.3 13.7 11.1 38.6 9 0 18.7 9.7 0 17.5 17.4 Orthoclase .............. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ... 15.2 0 0 Biolilc .................. 12.4 12.8 13.2 13.0 9.1 7 8 13.2 9.8 12.1 10.7 16.2 11.7 10.1 11.7 2.2 15.4 14.0 6.6 Hornhlende . .. , 0 0 0 0 0 0 0 0 0 0 0 0 .2 Tr. - - 0 0 0 Magneliic ..... 1.5 2.0 1.7 3.9 1.4 .4 1.5 1.7 .9 0 2.4 2.9 2.2 2.0 .9 4.4 1.9 1.6 Apatitc ..... ... 6 4 Tr. 6 .4 4 .5 .4 .6 8 1.0 1.0 .4 .5 .3 1.0 .9 .15 Sphenc ............ .7 1.8 0 Tr 1.6 .5 1.8 1.2 2.4 0 1.4 0 9 .9 8 .05 Tr. -4 Zircon ............. 0 .05 .2 .05 .05 Tr. Tr. .05 .05 .1 .05 .2 2 .1 1 .4 Tr. Tr. Allanile. .. ....... 0 0 0 0 0 .5 0 0 .05 0 0 0 0 Tr. 0 0 0 Rutile .... ....... .05 0 0 0 0 0 0 0 0 0 0 0 0 Tr. O 0 0 Muscovilc ............... 0 0 0 .5 0 0 0 0 0 0 0 0 0 Tr. 0 0 0 0 Pyrite .................. .1 0 0 O 0 0 0 0 0 0 0 0 0 Tr. 0 0 0 Total ............. 99.95 100.05 100.0 100.05 100.05 100.0 100.0 99.95 100.0 99.9 100.05 97.5 100.0 99.7 ... 100.05 100.0 100.05 licmic ............ 15.4 17.0 15.1 18.1 12.5 9.6 17.0 13.2 16.1 14.6 21.0 16.8 14.0 15.2 2.9 21.2 16.8 8 7 Pcrccm amhorilc in plagioclasc .......... 5—10 0—10 5—10 5—10 0 0 0—5 0—5 0 0—5 5—10 3 27—30 0—10 ... 0 0—5 Quart/ mixing index (see text) .......... .88 .79 .92 .86 .77 .89 .91 .93 .91 .82 .95 ... .92 .88 .06 .89 .86 .77 lMode 01‘ specimen 13 has affinity to mctagranodioritc. FEMIC MINERALS QUARTZ QUARTZ {6 of ,‘E\ Granite [luartz monzonite ‘Grapodiorite (quartz latlte) (rhyodKite) A. .2 24 El 12 r// 20 Monzunite Syenodiorite Diorite (dacite) V V POTASSIUM 50 PLAGIDCLASE FELDSPAR EXPLANATION /.\ E1 Granodiorite and quartz Fine~grained quartz monzonite monzonite, unmetamorphosed and aplite and weakly metamorphosed ._.._ _ __. Estimated shift in composition resulting from adjustment of phenocryst abundance / G7 Granodiorite and quartz monzonite, strongly metamorphosed FIGURE 9.—Modified triangular diagram showing modal quartz, potassium feldspar, plagioclase, and femic minerals of Continental Granodiorite. Abundance of quartz in specimen 16 may be anomalously large. near a point about 15 percent of the distance from the of the metamorphosed rocks is larger and less regular. In granodiorite field toward the femic point of the subsequent sections of this report the skewness of the tetrahedron. The body circumscribing the modal analyses modal bodies thus represented will vary considerably. . PRECAMBRIAN ROCKS 15 of Continental Granodiorite is 65D9l4. Symbols 5, standard deviation; Tr., trace; .... not determined] Slightly (regionally?) metamorphosed Strongly contact Fine-grained granodiome and quartz metamorphosed quartz monzonite monzonite—Continued metagranodiorite and aplite phase 17 18 19 14—19 M 1—19 20 21 22 23 24 5 1152 1043 l046 Mean 5 can 620 6ll.686 614 1106 1104 31.0 27.7 27.8 32.3 6.3 26.8 6.8 10.2 4.3 15.8 21.8 36.0 42.8 41.5 31.0 35.2 7.8 39.3 7.8 36.3 55.4 42.4 46.4 19.8 (1.2) (TL) (4.9) (2.2) (2-31 (1.7) (2.3) (2-4) 0 (2.5) (5.4) (9.0) 9.4 0 7.6 8.7 10.1 15.5 10.1 42.0 0 0 27.5 42.7 0 18.0 18.0 8.5 6-4 2.7 6.4 0 .5 29.4 0 0 13.0 9.4 11.6 11.7 2.5 11.7 2.5 9.3 5.6 7.5 3.5 .7 0 0 0 0 Tr, .9 29.2 .1 0 0 1.9 1.8 1.9 2.3 1.0 2.1 1.0 1.0 2.5 3.4 .4 .5 .9 .6 .9 .7 .3 .6 .3 Tr. .7 .6 .2 Tr. 1.0 .9 1.1 .6 .7 .8 .7 .1 1.8 .3 .2 .1 .05 .1 .1 .1 .2 .1 .2 .1 0 Tr. .05 TI'. 0 0 Tr. Tl’. Tr. Tr. ... 0 0 0 0 .15 0 0 0 0 . . 0 . . . 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 O . . . 0 0 0 0 0 100.05 100.0 100 0 100.1 100.1 ... 99.9 100.0 100.0 100.05 99.95 16.8 12.8 15 6 15.4 4 2 15.3 3.2 11.4 39.8 12.4 4.3 1.5 5—15 0—5 5 10 0—10 . .. 0—10 ... 37—40 28—31 5—35 0—5 0 .93 .92 .98 .89 .06 .88 .05 .67 .58 .98 .83 .84 The chemical and spectrographic analyses and calculated CIPW norms are listed in table 3 and the chemical analyses of the Continental Granodiorite and of the other plutonic rocks are summarized in figure 10. Inasmuch as this study was part of a general geological investigation, rapid rock analytic techniques and semiquantitative spectrographic methods were used. Chemically, the Continental Granodiorite closely resembles the average granodiorites reported by Johannsen (1932, p. 344) and by Nockolds (1954, p. 1014). Most noteworthy of the spectrographic results is the decrease in the abundance of copper and lead in the metamorphosed granodiorite near Helvetia compared to the content in unmetamorphosed granodiorite near Greaterville. This change reflects the variations in the copper content of mica, described by Lovering and others, (1970), and studied in greater detail in some rocks of the nearby Sierrita Mountains by Banks (1974). AGE AND CORRELATION The Continental Granodiorite has not been precisely dated by radiometric means. Lead-alpha dates indicate an age of at least 1,450 my (Drewes, 1968, p. C5); rubidium- strontium and potassium-argon dates are discordant with the lead-alpha date, and they suggest that later thermal events have modified at least many of the available radiometric ages. The granodiorite is thus considered to be at least 1,450 my old and it may be 1,600 to 1,700 my. old. Geologic means do not permit the dating of the granodiorite with complete confidence either, but the field relations also suggest a Precambrian age. The fault separating the Cambrian Bolsa Quartzite from the granodiorite has been described in the introductory section on the Continental Granodiorite as following an unconformity. In addition, cobbles of granodiorite resembling the Continental occur in conglomerate lenses intercalated in the Triassic Mount Wrightson Formation (Drewes, 19713, p. 12); if the cobbles are derived from the Continental, the pluton must be at least as old as Triassic. Furthermore, the granodiorite has a distinctly older appearance through its alteration, shear zones, and local areas of faint foliation, than the plutonic rocks of Triassic and younger ages in the Santa Rita Mountains. Radiometric ages of 11 plutonic rocks of the Santa Rita Mountains are listed in table 4. The general locations from which the dated samples were collected are shown in figure 3; and detailed locations of two sample sites are shown on the geologic map of the Sahuarita quadrangle (Drewes, 1971b). The detailed collection site of the third sample is not shown on the geologic map of the Mount Wrightson quadrangle (Drewes, 1971c); it lies in the Mavis Wash area, just east of the middle one of three small areas mapped as Bolsa (?) Quartzite, in the larger outcrop area of contact- metamorphosed Continental (?) Granodiorite. Three samples of granodiorite are radiometrically dated, two of them by two methods each. Sample 21 is probably thoroughly recrystallized and sample 19 is slightly recrystallized, features which were not initially recognized in the field. The lead-alpha ages of zircons of specimens 1 and 19, 1,360 my. and 1,450 my, are probably nearest to the true age of the rock, for the zircon would be among the minerals most refractory to change by thermal metamorphism. The rubidium-strontium whole-rock age, 800 m.y., only corroborates a Precambrian age of specimen 1; perhaps a very mild metamorphism, petrographically not noticeable and so not considered in tables 2—4, has resulted in loss of radiogenic strontium from this rock. The potassium-argon age of 55 my of specimen l9 coincides closely with the ages of the Helvetia stocks, the nearest of which lies 2% miles to the north, and the biotite of the rock of the intervening area is subtly recrystallized. The other potassium-argon age, 159 my. (sample 21), dates the age of crystallization of an anomalously fresh-looking granodiorite variety as Jurassic, in an area of rock otherwise resembling the Continental Granodiorite. The granodiorite lies within a few hundred feet of a concealed contact of Squaw Gulch Granite, of Jurassic age, and apparently was contact metamorphosed by that granite. The laboratory measurements and constants used in calculating the ages shown in table 4 are listed by Marvin and others (1973). Plutonic rocks resembling the Continental Granodiorite occur in the Rincon and Sierrita Mountains and possibly 16 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA TABLE 3.—Chemical and spectrographic analyses and CIPW norms of Continental Granodiorite [Chemical analyses by rapid rock method (Shapiro and Brannock, 1962), with analyses of specimens 1, 11, 14, and 19 supplemented by atomic absorption method and others supplemented by X-ray fluorescence method. Chemical analyses by Lowell Artis, S. D. Botts, G. W. Chloe, P. L. D. Elmore, John Glenn, J. Kelsey, H. Smith, and Dwight Taylor. Spectrographic analyses (semiquamitative method) by W. E. Crandell, .1. L. Finley, J. C. Hamilton, and A. L. Sutton. Elements looked for but not found: As, Au, B, 131, Cd, Eu, Ge, Hf, Hg, In, Li, Pd, Pr, Pt, Re, Sb, Sm, Sn, Ta, Te, Th, T1, U, W, and Zn. Symbols: 5, standard deviation:<, less than; .. . ., not determined; Tr., trace; N, not detected] . ‘ . 5118le Granodiorite Strongly contact Rock type .................. Granod1or1te and quartz monzonite metamorphosed and metamorphosed granodiorite quartz monzonite metagranodiorite Specimen No ................. 1 11 12 13 14 1— 14 19 I— I9 20 21 Field No ............... 914 897 104 627 1220 Mean 3 1046 Mean 5 620 611 ‘ Chemical analyses (weight percent) SiO2 ---------------------- 59.0 67.4 69.4 64.6 67.8 65.6 4.1 67.4 65.9 3.7 64.3 59.7 A1203 .................... 17.4 14.4 12.4 17.6 14.0 15.2 2.3 14.3 15.0 2.1 17.4 17.4 F6203 ------------------- 3.4 2.2 3.6 1.0 1.8 2.4 1.1 3.0 2.5 1.0 2.1 3.3 3.0 2.1 1.3 1.2 3.0 2.1 .9 2.0 2.1 .8 2.3 3.3 1.2 12 61 12 1.3 1.1 .3 .91 1.1 .3 .90 2.0 2.9 2.7 1.7 5 2 2.0 2.9 1.4 3.3 3.0 1.2 4.5 4.9 3.7 2.8 3.8 4.2 2.3 3.4 .8 3.1 3.3 .7 4.6 3.8 6.9 4.1 4.5 2.2 5.3 4.6 1.7 2.8 4.3 1.7 2.2 2.9 .08 .11 .13 .33 .08 .15 .11 .14 .15 .09 .08 .16 1‘0 11 1.2 1-0 .42 1.0 .32 .82 .95 .29 .70 .76 .46 .18 .92 .74 .83 .63 .30 1.0 .69 .31 .52 1.0 .42 .29 .34 .28 .27 .32 .06 .43 .34 .07 .22 .47 .12 .12 .12 .08 .13 .11 .02 .08 .11 .02 .12 .13 (.05 .36 .34 .11 .05 .16 .18 .05 .14 .17 .05 .08 100 99 100 100 99 100 99 100 100 100 Speetrogmphic analyses (weight percent) 0 0 0.0005 0 0 Tr. ... 0 Tr. ... 0 0 .2 .15 .2 .2 .1 .17 .04 .07 .15 .06 .2 .15 .0001 .0002 0 .0002 .0003 .0002 .0001 0 .0001 .0007 .0002 0 .02 .02 .015 0 .05 .02 .02 N N N 0 .02 .002 .001 .001 .0007 .001 .001 .0004 .007 .002 .002 .001 .0015 .0015 .002 .001 .0005 .01 .003 .004 .002 .003 .004 .0003 .001 .003 .02 .1 .00015 .005 .03 .04 .007 .02 .04 .0005 .002 .0015 .0015 .003 .002 .0015 .002 .001 N N N .002 .001 .02 .01 .005 0 .01 .009 .007 .007 .009 .007 .01 .005 0 0 0 0 .0003 Tr. N 0 Tr N 0 .0003 .0005 .0007 .0015 0 .002 .0009 .0008 O .0008 .0008 0 0 .02 .015 .007 0 0 .008 .009 N N N .005 0 < .003 < .003 .0007 .0015 .002 .002 .003 .003 .002 .001 .0005 < .003 .007 .007 .07 .0015 .002 .02 .03 .0015 .01 .03 .0015 .001 .002 .001 .001 .0015 .002 .0015 .0005 .007 .002 .002 .002 .002 .0005 .0007 0 0 0 .0001 .015 0 .0002 .0003 0 0 .03 .02 .03 .1 .02 .04 .03 .003 .03 .03 .01 .07 .007 .007 .007 .01 .007 .008 .001 .02 .01 .005 .007 .015 .005 .005 .005 .002 .01 .005 .003 .0015 .005 .003 .003 .003 .0005 .0005 N .0003 .001 .0005 .0003 N N N .0007 .0003 .03 .05 .03 .015 .03 .03 .01 .007 .03 .01 .015 .03 CIPW norms 3.6 29.0 27.6 19.9 27.7 18.1 13.7 0 2.0 0 0 1.6 0 ,4 40.9 24.4 26.6 13.0 31.5 13.0 17.1 31.4 23.9 32.1 35.5 19.6 38.9 32.1 10.5 9.3 3.5 22.7 7.9 20.3 20.7 .35 0 .23 .24 0 .23 0 .l9 0 .20 .19 0 .12 0 .11 0 O .02 0 .10 0 2.8 3.0 1.3 2.8 3.3 2.1 5.0 2.0 2.0 0 .28 2.9 1.8 1.9 4.9 3.2 1.9 1.5 2.6 3.0 4.8 0 0 2.3 0 0 0 0 .87 .34 1.7 1.4 1.6 1.0 1.9 1.0 .69 .80 .66 .64 .52 1.1 .11 .82 .77 .25 .11 0 .18 98.7 98.7 99.0 98.4 99.5 99.2 98.9 12.3 10.1 9.2 7.3 11.2 8.9 14.9 also in the Empire and Patagonia Mountains (fig. 1). In the 1970) as granite and diorite of Precambrian age, and Rincon Mountains (Drewes, 1974) a gneissic granodiorite schist and gneiss of undetermined age resemble, porphyry that intrudes Pinal Schist and is intruded by respectively, the unfoliated and the foliated varieties of the other less foliated rock of probable Precambrian age is Continental Granodiorite, with which I have correlated believed to be a strongly metamorphosed equivalent of the them (Cooper, 1974). On the northern flank of the Empire Continental Granodiorite. In the Sierrita Mountains, rocks Mountains T. L. Finnell (oral commun., 1972;) has mapped originally mapped by J. R. Cooper (written commun., Precambrian plutonic rocks, some of which resemble a TRlASSIC AND JURASSIC ROCKS 17 TABLE 4.—Summary of radiometric age determinations of plutom'c rocks of the Santa Rita Mountains Specimen Field Formation Geologic Dating Radiometric Comment Collection No. No. (pluton) age method 388 (m-YJ site shownl 197 V 65D687 Granodiorite ofthe San (late?) Oligocene . . . . K—Ar 27.6 i 0.8 Genetically related to Map {—614. Cayetano Mountains. Grosvenor Hills Vol» canics (Drewes, 1972). 189 66D1185 55.8 i 21 Map 1—613. 192 67D1245 Aphanitic rocks (late?) Paleocene . . .. K-Ar 56.3 i 2 1 “Ore porphyry” quartz Map 1-613, 2 (Greaterville plugs). latite porphyry. ( ) 68D1472 55 71' l 9 Fig. 40, this report. 163 66D1051 53.5 1' 1.3 Map 1—613. 163 66D1051 Granitoid rocks (late?) Paleocene . . . . K-Ar 53.5 i 2.0 Barren stocks, geologi- Map 1—613. (Helvetia stockS). cally older than “ore 174 70D1612 53.9: 2.0 porphyry." Fig. 40, this report. 160 63D281 60.3 i 6.0 Microgranodiorite ..... Map [_514. (Gringo Gulch pluton) . .Paleocene .......... K-Ar 158 364D660 60.4 1 6.0 Hornblende dacite por- phyry ............... Map 1—614. 132 65D754 K-Ar 68.21" 3.0 Pb-alpha 170 i 20 Map [—614. Elephant Head Quartz Late Cretaceous. . . . Rock without recrystalli- 139 650876 Monzonite (Quantrell K-Ar 69.0 i 2.9 zation texture. Pb. Map 1—614. stock). Pb-alpha 188 i 40 alpha ages may reflect a Jurassic history, new 2 nearly obliterated. ( ) RM—6—63 Madera Canyon Late Cretaceous ..... K-Ar 67.9 1 2.1 Collected and dated by Map 1—614. Granodiorite. P. E. Damon (written 80 63 3 4 commun., 1964). D 16 K-Ar 67.1: 7.0} - - [—614 Pb»alpha 62 i 10 Quartz dlortte phase . . . . Map . 83 63R292 Josephine Canyon Late Cretaceous. . . Pb~alpha 63 i 10 Map 1—614. Diorite . 1201 563D507 Pb-alpha 6] 1' 20 Quartz monzonite phase ..... Map 1—614. ( ) F—33—68 Granitic rock (Corona Late Cretaceous ..... KAAr 73.8i 2.6 Quartz diorite collected Open-file map. stock). and correlated by T- L F' . - 54 640605 K-Ar 145 1 4 "me" Squaw Gulch Granite _ . Jurassic ........... Pb-alpha 160 i- 20 """""""""""""""""" Map 1‘51“ 45 63D282 Pb-alpha 161 1'20 Map 1—614. 34 63 D280 Piper Gulch Monzonite Triassic ............ Pb-alpha 184 t 20 Map 1—614. 1 65D914 Pb-alpha 14501 160 Map 1—613. Continental Grano- Precambrian ...... Rb-ST 300$ 30 Biotite has recrystalliza- 19 65D1046 diorim K»Ar 55 — 1.7 “on tenure; nearby Map 1—613. Pb—alpha 1360 1' 270 stocks 53—55 my. old. 21 665D686 Continental Grano- Precambrian (7) ..... K-Ar 159 i: 5 Nearby stock 145—161 Map 1—614. diorite (7). my. 1 Map [—614 (Drewes, 1971c); map'1—613(Drewes, 1971b). Not given a specimen number; no modal or chemical analyses available. Specimens 64D660 and 64D661 were collected from nearly the same outcrops; the dated rock has coarser phenocrysts than the moded one. nonporphyritic and finer grained variety of Continental Granodiorite and others of which resemble different plutonic rocks like ones found in the Rincon and Whetstone Mountains. Precambrian plutonic rocks of the Patagonia Mountains (Simons, 1974) also include coarsely porphyritic granodiorite with a general resemblance to the Continental. Although rocks similar to the Continental Granodiorite seem. to be fairly widespread, there is no evidence to suggest that all known occurrences are from the same pluton, or indeed are of identical age. TRIASSIC AND JURASSIC ROCKS Plutonic masses were intruded into the area of the Santa Rita Mountains during the Triassic and Jurassic Periods, approximately coinciding with times of abundant volcanic activity. This magmatic activity followed a long quiescent interval that began after the emplacement of the Continental Granodiorite in the Precambrian. The magmatic activity of the early Mesozoic began with the 4 Reported as 67:3 by Marvin and others (1973, table 1, no. 3). Specimen 63DSO7 and 630276, referred to in other tables, from same outcrop. Specimens 65D686 and 64D611, referred to in other tables, from same outcrop. locally prolific volcanism that formed the Mount Wrightson pile (Drewes, 1971a) and followed near the end of the Triassic with the intrusion of a stock of the Piper Gulch Monzonite and smaller masses of quartz diorite. About this time, or slightly later, volcanism is again recorded, in the Upper Triassic and Lower Jurassic Canelo Hills Volcanics. Near the middle of Jurassic time a batholith of Squaw Gulch Granite was emplaced in the southern part of the area. Finally, between the Middle Jurassic and the onset of the Laramide orogenic events in the mid-Cretaceous, magmatic activity dwindled; of three formations, only the oldest two contain volcanic rocks. PIPER GULCH MONZONITE The Piper Gulch Monzonite, defined by Drewes (1968), is a very coarse grained dark-gray rock almost unique in its Triassic age and monzonite to syenodiorite composition to this part of Arizona. The monzonite underlies many small areas and a larger one, together not exceeding 4 square 18 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA A B Piper Gulch Monzonite (3) Continental Granodiorite, exclusive of strongly metamorphosed grano- diorite (6) WEIGHT PERCENT 70 70 H Quartz latite porphyry of Greaterville plugs (6) G Granitoid stocks of Helvetia stocks, exclusive of Sycamore stock (9); 15 broken line, quartz diorite of Sycamore stock (2) WEIGHT PERCENT FIGURE lO.—-Histograms showing average chemical composition of each group of plutonic rocks. miles. These outcrop areas lie mainly in a belt that extends from the middle reaches of Temporal Gulch northwestward to the west wall of Madera Canyon, shown in general fashion in figure 11 and shown in greater detail on the geologic map of the Mount Wrightson quadrangle (Drewes, 1971c). Monzonite underlies part of the east wall of Madera Canyon, and very small masses of monzonite lie well southwest of the main belt of outcrops. The monzonite intrudes the Mount Wrightson Formation and is extensively intruded by the Squaw Gulch Granite and C Josephine Canyon Diorite, exclusive of fine-grained quartz 55 monzonite (6); broken line, fine-grained quartz monzonite (3) D Madera Canyon Granodiorite l5) 70 70 J Quartz monzonite of Corona stock (1) / Squaw Gulch Granite (4i C90m O '+~mo 9:;9%2§Q9Q9Qe6 (n an ..... 7.2 7.2 6.5 4.2 6.8 6.1 1.5 1.6 .25 "Its ...... 2.1 2.9 2.3 1.4 .17 3.4 0 0 0 m1 ........ 4.1 5.8 4.8 4.6 6.2 3.5 3_7 2.9 .98 hm 0 0 0 0 0 0 .55 1.1 1 6 il ...... 1.6 .19 1.7 1.7 1.8 1.7 .86 1.2 ]_1 ap ......... .66 .64 .69 .64 .62 .59 .35 _3] 1.] CC . .13 0 0 0 0 1.3 o o 11 Total 99.1 99.0 98.8 98.2 99.0 98.5 98.2 98.1 99.3 Femic 15.8 19.0 18.7 17.3 18.8 17.6 7,0 7.1 5.1 * T 11‘ Pairs of replicate analyses. both composition and age have not been reported in mountain ranges immediately adjacent to the Santa Rita Mountains. Somewhat similar rocks, which have been little MADERA CANYON GRANODIORITE The Madera Canyon Granodiorite is a coat se-grained, studied and are poorly dated, occur in the Ruby area, 17 miles southwest of the small stock of Josephine Canyon Diorite, and other such rocks occur in Sonora near Nogales about 15 miles south of that stock (fig. 1). light-gray to medium-gray speckled rock in which dark minerals are moderately to very abundant. The granodiorite is in a stock in the Madera Canyon area (fig. 27) that trends northwest and is wedge shaped, tapering to CRETACEOUS ROCKS 41 110°55' Wrightson X 0 1 2 3 MILES O 1 2 3 KILOMETRES EXPLANATION Younger rocks Igneous and sedimentary rocks Cretaceous Madera Canyon Granodiorite K m , granodiorite K m p , porphyritic granodiorite K mm , melanocratic granodiorite Older rocks Igneous and sedimentary rocks Contact Dotted where concealed Fault Dotted where concealed . 109 O l 10 *? Moded Chemically Radiometrically analyzed dated Specimen collection sites and numbers FIGURE 27.—Distribution of Madera Canyon Granodiorite and specimen collection sites. the southeast. The stock underlies an area of about 10 square miles; about 6 square miles of this area is directly underlain by granodiorite; the remainder is projected beneath a thin gravel cover on the pediment off the mouth of Madera Canyon. The granodiorite intrudes mainly the Mount Wrightsou Formation and the Piper Gulch Monzonite, both of Triassic age, and the southeastern end of the stock intrudes the Josephine Canyon Diorite. Sedimentary rocks intercalated in the Mount Wrightson Formation are strongly contact metamorphosed along the eastern margin of the stock, as shown on the geologic map (Drewes, 19710), but the volcanic rocks of the formation are no more altered near the stock than they are away from it. The contact zone, as observed at collection site 103 (fig. 27), is narrow and chilled. Inclusions of epidotized and chloritized quartz diorite, probably derived from a nearby Triassic intrusive mass, form inclusions in the granodiorite near the eastern margin of the stock, midway between collection sites 126 and 127. The rocks on both sides of the contact in this area are also slightly pyritized. In general the contact is sharply defined, steeply inclined, and regular in trend; but near the southeastern end of the stock it is steeply to gently inclined and, consequently, irregular in trend. Apparently the southeastern outcrops lie near the top of that part of the stock, whereas the northwestern outcrops lie a considerable distance beneath the original top of the stock. PETROGRAPHY The Madera Canyon Granodiorite consists of three rock types each distinctive in texture or in composition. A nonporphyritic granodiorite is the most widespread type; it crops out from the narrow end of the stock almost to the mouth of Madera Canyon, where it appears to grade into a porphyritic granodiorite. The porphyritic granodiorite underlies the northwestern part of the stock, largely in the pediment area. Melanocratic granodiorite underlies an elongate area along the northeastern side of the stock, where it is separated from the other two types of granodiorite by younger intrusive rocks of the Elephant Head Quartz Monzonite and by a prong of the host rocks. The term “melanocratic” will be used hereafter for a rock with an unspecified but relatively large amount of dark minerals. GRAN ODIORITE The nonporphyritic granodiorite, to be referred to hereafter simply as “granodiorite,” is much less resistant to weathering than the adjacent rocks, and so it underlies low areas, such as the bottom of Madera Canyon. This weathering characteristic is largely the result of the friability of the rock, which disaggregates readily to form extensive slopes of grus and a few small outcrops. Outcrops of granodiorite are most accessible in the 42 PLUTONIC ROCKS, SANTA RITA MLUNTAINS, SOUTHEAST OF TUCSON, ARIZONA roadcuts a few hundred feet south of the Santa Rita Lodge (area of specimen 113, fig. 27), as well as along trail cuts and in the bottoms of the narrowest canyons. Rocks from these relatively unweathered outcrops are illustrated by figure 28. The granodiorite typically has a hypidiomor- phic-granular texture and a 2- to S-mm grain size (fig. 29). Near the southeast end of the stock the grain size is slightly smaller and some specimens have a bimodal size distribution of grains, with the smaller ones only 0.1 to 0.5 mm long. Specimens from the northwest end of the granodiorite area, on the other hand, contain some crystals 6 to 7 mm long that resemble underdeveloped phenocrysts. Some granophyric texture is commonly evident in all specimens. Common minerals of the granodiorite are quartz, orthoclase, plagioclase, biotite, and hornblende (fig. 26). The quartz is anhedral and has little or no undulatory extinction; in rocks whose grains have a bimodal distribution, quartz is commonly among the smaller group of crystals. The orthoclase is anhedral or subhedral and contains a fine lacy perthitic intergrowth of albite. One specimen taken about 2 cm from the contact of the granodiorite with the Josephine Canyon Diorite contains sanidine instead of orthoclase. Plagioclase is subhedral; it generally has a composition of calcic oligoclase to sodic andesine, but in a few specimens it is albitized. The biotite is subhedral and is pleochroic in pale yellowish brown to moderate brown. Hornblende forms euhedral to subhedral crystals pleochroic in pale yellowish green to pale grayish green. In the northern part of the granodiorite body some of the biotite and hornblende are clustered, as they typically are in the nearby melanocratic granodiorite. Accessory minerals invariably include ilmenitic magnetite, apatite, and zircon, and some specimens also contain a trace of sphene, rutile, and beryl(?). Some magnetite grains are surrounded by sphene. The zircon is moderately abundant and fairly coarse for zircon, reaching a grain size of about 0.4 mm. Secondary minerals are moderately scarce and consist of kaolinite, sericite, chlorite, leucoxene, and epidote. PORPHYRITIC GRANODIORITE Aside from its phenocrysts, the porphyritic granodiorite closely resembles the nonporphyritic type. Indeed, the two units may be gradational, with the porphyritic rock characteristic of the core of the stock and the nonporphyritic rock present along an unrecognized narrow border zone and more extensively present in the narrow southeastern shoulder of the stock. Unweathered, nonfriable outcrops of the porphyritic granodiorite are scarce; some of the freshest material is available in a rock dump along the edge of Madera Canyon nearest to the collection site of specimen 120 (fig. 27). The texture of the rock is noteworthy for its scattered phenocrysts which are 3 to 5 cm long. Despite the FIGURE 28.—Specimen 113. Nonporphyritic type of Madera Canyon Granodiorite. occurrence of phenocrysts and the coarse-grained groundmass, the rock differs from the Continental Granodiorite by having fewer phenocrysts and ferromagnesian minerals and being much less altered. Very locally, 1/2 to 1 mile northwest of collection site 120, the porphyritic granodiorite contains some schlieren of more leucocratic quartz monzonite and some of more porphyritic rocks. In thin section, the porphyritic granodiorite is seen to differ from the nonporphyritic type in that the potassium feldspar commonly has indistinct or blurred grid twins typical of microcline. One specimen also contains accessory allanite. MELANOCRATIC GRANODIORITE The melanocratic granodiorite is a nonporphyritic dark-gray rock that contains abundant hornblende and biotite and a gray plagioclase (fig. 30). Clustered ferromagnesian minerals give the rock a faintly mottled CRETACEOUS ROCKS 43 FIGURE 29.——Specimen 113. Nonporphyritic type of Madera Canyon Granodiorite. Crystals: plagioclase ( P ), quartz ( Q), orthoclase ( O ), biotite ( B ), hornblende( H ), ilmenitic magnetite( im), and sphene( S ). x 20. A, Plain light; B, crossed nicols. appearance. It is most easily accessible near collection site 127 (fig. 27) in the northern part of the outcrop area. The clustering of small crystals of biotite and the bimodal distribution of grain size in many specimens of melanocratic granodiorite suggest mild recrystallization. Strongly sutured grain boundaries and a peculiar wormy-looking granophyric intergrowth also suggest recrystallization. The minerals in the melanocratic granodiorite are mostly the same as those in the other types of granodiorite. Most of the plagioclase has the composition of andesine and that of two specimens is calcic andesine, but some of the plagioclase is albitized. The biotite of one specimen is pleochroic in pale olive green to medium olive green, rather than in brown. Schorlite tourmaline is a common accessory mineral, and allanite is a rare one. MODAL AND CHEMICAL SUMMARY Modal analyses of 25 specimens of Madera Canyon Granodiorite are shown in table 13. The abundance of femic minerals in the nonporphyritic and in the porphyritic granodiorite is about 5 percent less than that in the melanocratic granodiorite. Two of the hybridized specimens, 116 and 122, that are excluded from the modal means in table 13 have probably been altered by the nearby intrusive body of Elephant Head Quartz Monzonite. The other hybridized rock, specimen 103, is from a contact chill zone along which there may have been some assimilation. The modes are plotted in a modified triangular diagram, figure 31. Hornblende and biotite together make up about 10 percent of the rock, with substantial amounts of both minerals present in most specimens. However, specimen 104 contains biotite and no hornblende, and specimen 105 contains hornblende and a very minor amount of biotite, suggesting that these minerals may proxy for each other. An enclosing shell drawn around the plotted modes in figure 31 is subspherical and is centered along the boundary between the granodiorite and quartz monzonite fields and about 15 percent of the way from the leucocratic face of the composition tetrahedron toward the femic apex. The modes for each of the three types of granodiorite 44 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA TABLE 13.—Modes of [Field numbers are abbreviated; year of collection and collector’s initial omitted. Full field number of specimen Rock type ....... Granodiorite Specimen No -------- 1103 104 105 106 107 108 109 110 111 112 113 114 115 104— 115 Field No. .......... 2403 363 765 1581 833 836 1123 769 837 843 826 310 347 Mean 5 Quartz .......... 24.8 24.1 19.1 23.1 21.4 23.7 17.5 21.9 16.0 18.3 22.5 17.9 24.2 20.8 2.9 Plagioclase, total . 52.7 47.9 40.3 45.2 36.2 39.0 44.6 41.8 42.8 47.5 47.1 48.7 43.4 43.7 3 9 (plagioclase in perthite) ..... 0 (1.7) (1.6) (2.0) (1.3) (2.0) (0.2) 0 (1.6) (2.9) (1.2) 0 (1.9) (1.4) (0.9) Orthoclase ....... 0 16.1 28.0 19.3 31.8 27.0 25.5 26.2 23.3 21.5 17 8 12.9 21.6 22.6 (5.5) Sanidine ......... 10.6 0 0 0 0 0 0 0 0 0 0 0 0 0 ...... Hornblende ...... 0 0 9.3 3.5 1.8 2.3 6.3 4.9 9.7 3.9 5.7 10.9 2.4 5.1 3.4 Biotite .......... 10.7 11.4 .5 5.4 6.4 5.7 3.7 3.7 5.6 6.5 4 3 6.2 6.4 5.5 2.5 Pyroxene ........ 0 0 0 1.3 0 0 0 0 0 0 0 0 0 .1 .4 Magnetite ....... .7 .5 1.9 1.5 1.6 1.5 1.2 1.1 1.4 1.2 1.7 2.2 1.4 1.4 .4 Apatite .......... .3 Tr. .3 .4 .2 .2 .5 1 .3 .4 .2 .1 .1 .2 .2 Sphene .......... 0 0 .6 .3 .5 .6 .6 .3 .9 .7 .7 1.1 .2 .5 .3 Zircon .......... .2 Tr. Tr. .05 .1 Tr. .1 Tr. r. .05 .05 0 Tr. .03 .04 Allanite ......... 0 0 0 0 Tr. 0 0 0 0 0 0 0 0 Tr. ...... Rutile ........... 0 Tr. 0 0 0 0 0 0 0 0 0 0 0 Tr. ...... Tourmaline ...... 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ..... Total ..... 100.0 100.0 100.0 100.05 100.0 100.0 100.0 100.0 100.0 100.05 100.05 100.0 99.7 99.93 ...... Femic ..... 11.9 11.9 12.6 12.4 10.6 10.3 12.4 10.1 17.9 12.7 12.6 20.5 10.5 12.8 3.2 Percent anorthite in plagioclase. .. 28 32—42 5—37 35—42 133—37 28—37 32—34 29-35 33—40 33 32-37 32—37 28—31 28—37 ...... (plagioclase rims) ..................................................................... (23) (20) ....... (20) (21) ,,,,,, Quartz mixing index (see text) . ....... .89 .88 .85 .85 .70 .96 .90 .97 .93 .97 .91 .94 .90 .07 lSpecimen is hybridized. ZAClual sample collection site is at dump '/a mile southwest of source site shown on map, figure 27. 3 Modal orthoclase includes a microcline (2’). 40mits hybridized specimens 103, 116, and 122. fall within smaller subspherical circumscribed figures. The modal ranges of the nonporphyritic and the porphyritic granodiorite overlap, bearing out the field evidence that these rocks are gradational. No such overlap occurs for the melanocratic figure and those of the granodiorite and the porphyritic granodiorite figures. The low value of the quartz mixing index of the melanocratic granodiorite relative to the other types may be the result of minor redistribution of quartz; this is compatible with the other evidence suggesting mild thermal metamorphism of the rock. Chemical and spectrographic analyses and CIPW norms of specimens of Madera Canyon Granodiorite are shown in table 14. The chemical data are summarized in the histogram of figure 10D, which shows the rock to be unique among the several granodiorites of the Santa Rita Mountains in having nearly equal amounts of CaO, NagO, and K20. The mean of the chemical analyses is fairly close to that of the average of hornblende-biotite granodiorite tabulated by Nockolds (1954, p. 1014). AGE AND CORRELATION The Madera Canyon Granodiorite is, on geologic evidence, COHSidered to be younger than the Josephine FIGURE 30.—Specimen 127. The melanocratic type of Madera Canyon Canyon Diorite and older than the Elephant Head Quartz Granodiorite. Madera Canyon Granodiorite 103 thus is 63D5103. Symbols: 5, standard deviation; Tr., trace; CRETACEOUS ROCKS not determined] 45 Porphyritic granodiorite Melanocralic granodiorite 1116 117 118 119 120 121 117—121 1122 123 124 125 126 127 l 122—127 4104 _77127 873 874 872 865 377 889 Me“ 3 825 908 324 867 869 902 Mean 5 Mean 5 37.5 26.0 21.8 28.5 27.3 20.7 24.9 3.4 16.3 12.2 12.1 8.2 11.0 9.3 10.6 1.8 19.4 5.8 28.9 37.8 47.6 46.0 42.0 43.2 43.3 3.8 34.1 37.3 50.1 47.6 45.5 49.0 45.9 5.1 43.8 5.3 0 (1.9) (1.4) (0.2) (1.1) (0.2) (1.0) (0.8) (3.0) (5.6) (4.8) (2.4) (3.3) (3.0) (3.8) (1.3) (1.7) (1.5) 29.9 21.3 18.2 15.3 223.7 14.8 18.7 3.8 23.7 34.7 16.9 22.1 23.6 28.6 25.2 6.8 21.7 7.3 0 O 0 0 0 0 0 ...... 0 0 0 0 0 0 0 ................. .7 4.0 5.5 .05 1.3 8.0 3.8 3.2 11 0 6.5 10.5 11.2 9.2 5.4 8.6 2.5 5.1 3.7 2.5 9.2 4.7 9.2 4.6 9.9 7.5 2.6 11.2 4.4 7.3 6.1 7.0 5.4 6.0 1.2 6.1 2.6 0 O 0 0 0 0 0 ...... 0 0 0 0 0 0 0 ................. .3 .9 9 .8 7 1.7 1.0 .4 1.6 2.3 1.2 2.9 1.7 1.6 1.9 .7 1.4 .6 .1 .4 .2 .1 3 3 .3 .1 .8 .6 .6 .5 .6 .3 .5 .1 .3 .2 .1 .3 1.1 .05 .1 1 3 .6 .6 1.1 1.9 1.0 1.4 1.3 .4 1.2 .6 .6 .5 Tr. .1 05 .05 Tr. .1 .06 .04 .15 .1 .15 Tr. .05 .05 .09 .05 .05 .05 0 0 0 0 Tr. 0 Tr ...... Tr. 0 .1 Tr. Tr. 0 Tr ...... Tr ...... 0 0 0 0 0 0 0 ...... 0 0 O 0 0 0 0 ...... Tr ...... 0 0 O 0 0 0 0 ...... 0 0 0 0 0 Tr. Tr. ...... Tr. ------ 100.0 100.0 100.05 100.05 100.0 100.0 100.16 ...... 99.95 00.0 99.95 100.0 99.95 100.05 99.99 ...... 98.45 ...... 3.7 14.9 12.4 10.2 7.0 21.3 13.1 5.4 25.9 15.8 20.9 22.1 19.9 13.1 18.3 3.8 13.6 4.4 0? 30—35 30 27—30 35 28—33 28—35 ...... 0? 37 38—49 36 40 27 28—49 ..... 28—49 ..... ..................... (20) (21) .......(20) .....(20) (20) .....(20—23)..... .97 .86 1.0 .63 .78 .92 .84 1 .60 .80 .81 .82 .80 .68 .78 .06 .86 .1 FEMIC MINERALS QUARTZ QUARTZ ‘éL Granite Quartz monzonite Granodiorite (quartz latite) (rhyodacite) A116 Monznnite Syenodiarite Diurite (dame) V V 50 V V 221322?" EXPLANATION PLAGIOCLASE ® «a Granodiorite; includes some quartz monzonite © Granodiorite; includes some quartz monzonite; hybridized ’AT‘ Porphyritic granodiorite A Porphyritic granodiorite; hybridized Melanocratic granodiorite; includes some quartz monzonite E1 Melanocratic granodiorite; includes some quartz monzonite; hybridized A Total Madera Canyon Granodiorite FIGURE 31.-—Modified triangular diagram showing modal quartz, potassium feldspar, plagioclase, and femic minerals ’of Madera Canyon Granodiorite. 46 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA TABLE 14.—Chemical and spectrograph ic analyses and CIPW norms of Madera Canyon Granodiorite [Chemical analyses by rapid rock method (Shapiro and Brannock, 1962), with specimen 113 supplemented by X-ray fluorescence method and specimens 118, 124, and 127 supplemented by atomic absorption method. Chemical analysts: Lowell Artis, S. D. Bolts, G. W. Chloe, P. L. D. Elmore, John Glenn. H. Smith, and Dwight Taylor. Spectrographic analyses by semiquantitative method. Spectrographic analysts: W. B. Crandell, J. L. Finley, and J. L. Harris. Elements looked for but not found: Ag, As, Au, Bi, Cd, Eu, Ge, Hf, Hg, ln, Li, Nd, Pd, Pr, Pt, Sb, Sm, Sn, Ta, Te, Th, Tl, U, W, and Zn. Symbols: 5, standard deviations;< , less than; . . ., not determined; N, not detected] Rock type .......... Granodiorite Porphyritic granodiorite Melanocratic granodiorite Specimen No ................. 110 113 110,113 117 118 117,118 124 125 127 124—127 110— 127 Field No. .................. 769 826 M53" 874 372 Me“ 824 867 802 Me“ Me“ 3 Chemical analyses (weight percent) Si02 ................ 67.9 65.7 66.8 68.6 67.5 68.1 60.5 ......... 60.4 60.5 65.1 3.7 A1203 ............... 15.6 16.3 16.0 15.2 15.1 15.2 16.5 ......... 16.1 16.3 15.8 .6 Fe203 ............... 1.7 2.1 1.9 2.1 2.2 2.2 2.5 ......... 2.8 2.7 2.2 .4 FeO ................. 1.4 1.7 1.6 1.2 1.4 1.3 3.3 ......... 3.2 3.3 2.0 1.0 MgO ................ 1 4 1.7 1.6 1.0 1.2 1.1 2.3 ......... 2.6 2.5 1.7 .6 C30 ................. 3 1 3.8 3.5 3.5 3.6 3.6 4.5 ......... 4.0 4.3 3.8 .5 NaZO ............... 3 6 3.9 3.8 3.6 4.0 3.8 3.7 ......... 3.8 3.8 3.8 .2 K20 ................. 3 8 3.0 3.4 3.2 3.3 3.3 4.1 ......... 4 6 4.4 3.6 .6 H20— .............. ll 12 .12 12 .14 13 09 ......... 07 .08 .11 .02 H20+ .............. 81 62 .72 51 .52 52 78 ......... 65 72 65 13 TiO 2 ................ 42 55 49 58 .66 62 98 ......... 99 99 72 24 P205 ................ 13 23 18 17 .19 18 44 ......... 49 47 28 15 MnO ................ 06 05 06 05 .07 .06 .17 ......... 19 18 10 06 C02 ................. < 05 11 06 < 05 <.05 < .05 <.05 ......... < 05 < 05 06 05 Total .......... 100 100 100 100 100 101 100 ......... 100 100 100 ......... Spectrographic analyses (weight percent) B ................... 0 0 0 O 0 0 0 0.007 0 0.002 0.001 0.003 Ba .................. .07 .1 .09 .1 .1 .1 .015 .015 .015 .015 .06 .04 Be .................. .0001 0 .00005 0 .00015 .0001 .0002 .0003 .0003 .0003 .00015 .0001 Ce .................. .01 0 .005 .015 .01 .01 N N .05 .05 .01 .02 Co .................. .0007 .001 .0009 .0005 .0007 .0006 .007 .005 .0015 .005 .002 .003 Cr .................. .001 .001 .001 .0007 .001 .0009 .0015 .003 .0015 .002 .001 .0007 Cu .................. .002 .02 .01 .0005 .007 .004 .015 .03 .005 .002 .01 .01 Ga .................. .001 .001 .001 .0015 .0015 .0015 N N .0015 .0015 .0009 .0007 La .................. .007 .005 .006 .01 .007 .009 .015 .015 .01 .015 .01 .004 Mo .................. 0 .0003 00015 0 0 0 0 0 0 0 .0004 .0001 Nb .................. 0 0 0 .0003 0 .00015 .003 .007 .0015 .004 .002 .003 N1! .................. 0 < .003 <.003 0 0 0 V .015 .003 O .006 .003 .006 Pb .................. .0007 .0005 .0006 .03 .0007 .015 .002 .001 .002 .002 .005 .01 Sc ................... .0007 .0007 .0007 .0007 .0007 .0007 .007 .015 .0015 .008 .004 .005 Sn .................. 0 0 0 .003 0 .0015 0 .0015 0 .0005 .0006 .001 Sr ................... .05 .07 .06 .07 .05 .06 .015 .01 .05 .03 .05 .02 V ................... .007 .007 .007 .005 .005 .005 .015 .03 .007 .02 .01 .009 Y ................... .002 .0015 .002 .0015 .001 .001 .015 .003 .005 .008 .004 .005 Yb .................. .0002 .0015 .0002 .00015 .0001 .0001 N N .0003 .0003 .0003 .0005 Zr .................. .01 .01 .01 .015 .02 .02 .007 .01 .05 .02 .02 .01 CIPW norms A ................... 23.9 21.6 ......... 27.1 23.4 ......... 11.0 ......... 9.4 ......................... C ................... .36 .54 ......... 0 O ......... 0 ......... 0 ......................... or ................... 22.4 17.7 ......... 18.9 19.5 ......... 24.2 ......... 27.2 ......................... ab .................. 30 4 33.0 ......... 30.5 33.9 ......... 31.3 ......... 32 2 ......................... an .................. l4 2 16.7 ......... 15 9 13.5 ......... 16.3 ......... 133 ......................... wo .............. 0 0 ......... 03 1.2 ......... 1 2 ......... 1.3 ......................... d1 {en ............... 0 0 ......... 03 1.0 ......... 75 ......... .87 ......................... fs ............... 0 0 ......... 0 0 ......... .35 ......... 37 ......................... h), en ............... 3.5 4.2 ......... 2 5 2.0 ......... 5.0 ......... 5.6 ......................... fs ............... .58 57 ......... 0 0 ......... 2.3 ......... 2.0 ......................... mt .................. 2.5 3 0 ......... 2.6 2.8 ......... 3.6 ......... 4 1 ......................... hm .................. 0 0 ......... . .48 25 ......... 0 ......... 0 ......................... il ................... .80 1.0 ......... 1.1 13 ......... 1.‘) ......... 19 ......................... ap .................. 31 .55 ......... 40 .45 ......... 1 0 ......... 1 2 ......................... cc ................... 11 .25 ......... 11 .11 ......... 11 ......... 11 ......................... Total .......... 99 1 99 1 ......... 99.6 99 4 ......... 99 0 ......... 99 6 ......................... Fem1c ......... 7 8 9 6 ......... 7.2 91 ......... 16 2 ......... 17 5 ......................... CRETACEOUS ROCKS 110°55’ x Mount Hopkins 0 1 2 3 4 4 KILOMETRES 47 EXPLANATION Younger rocks Igneous and sedimentary rocks Cretaceous Elephant Head Quartz Monzonite Key, quartz monzonite of the Yoas stock Keqc, quartz monzonite of the Quantrell stock, coarse-grained phase Keqf , quartz monzonite of the Quantrell stock, fine-grained phase Older rocks Igneous and sedimentary rocks Contact Dotted where concealed Dotted where concealed .128 Moded *132 Radiometrically dated Specimen collection sites and numbers 0130 Chemically analyzed MILES FIGURE 32.—Distribution of Elephant Head Quartz Monzonite and specimen collection sites. Monzonite, which are radiometrically dated as 67 to 69 m.y. old (table 4). In addition, the nonporphyritic type of Madera Canyon Granodiorite was dated by the potassium-argon method by P. E. Damon (written commun., 1964) as 67.9 $2.1 m.y. Because the precise site at which his specimen was collected is uncertain, the rock was not resampled for further dating. Granodiorite similar in both age and composition to the Madera Canyon Granodiorite has not been reported in ranges adjacent to the Santa Rita Mountains. Other granodiorite stocks of similar appearance occur in both the Sierrita Mountains (Cooper, 1974) and the Patagonia Mountains (Simons, 1973), but they are about 10 m.y. younger. ELEPHANT HEAD QUARTZ MONZONITE The Elephant Head Quartz Monzonite is a composite of the Quantrell and Yoas stocks and consists of coarse-grained pinkish-gray leucocratic rock that underlies much of the rugged terrain around Elephant Head and the mouth of Madera Canyon (fig. 32). It underlies several areas that total about 18 square miles, and it may also underlie a considerable area beneath the piedmont gravel northwest of Elephant Head. The main mass of quartz monzonite lies west of the upper reaches of Madera Canyon; it is slightly elongate to the northwest, a trend followed by most of the intrusive tongues and inclusions of host rocks in the quartz monzonite. Another mass of quartz monzonite lies northeast of the upper reaches of Madera Canyon and trends north, and small masses crop out between the larger ones. The attitude of the sides of the present plutonic masses suggests that in most places the original tops of the masses were a considerable distance above the present level of exposure. The trace of the eastern contact of the mass northeast of Madera Canyon is straight and the contact dips steeply outward; the quartz monzonite contact at the southern end of the mass probably also dips steeply outward. The other contacts are concealed by gravel. The 48 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, eastern contact of the larger mass of quartz monzonite is irregular but generally is steep. To the north the mass is faulted, and to the northwest it intrudes quartz diorite along a very irregular but mostly steep contact. However, to the south the contact dips gently outward and in places trends irregularly, and several large flat-bottomed roof pendants cap the highest hills. These features suggest that the upper contact of the stock plunges gently southward; to the south the top is exposed near the present surface, and to the north it has been removed by erosion. Small outcrops of quartz monzonite between the two larger masses of identical rock indicates that the septum of granodiorite along Madera Canyon probably does not extend far beneath the surface. Very likely then, the two masses are cupolas of a single stock. Generally the contacts of the quartz monzonite are fairly sharp and without noticeable chill zones or contact- metamorphosed aureoles, but the exceptions at several places are instructive. At one locality the contact of the smaller cupola, northeast of specimen collection site 141 (fig. 32), forms a gradational zone at least 400 feet wide. The rock toward the quartz monzonite is fine grained and light pinkish gray, and it contains only a few small clots of dark minerals. Toward the granodioritic host the rock gradually is grayer and the clots are progressively larger and more abundant, until they resemble the clusters of felty-textured biotite aggregates of the granodiorite, except that the hybrid rock has a sprinkling of small white feldspar crystals more typical of the quartz monzonite. The biotite in the nearby host rock is recrystallized, probably as a result of contact metamorphism produced by the emplacement of the small cupola of quartz monzonite. At a second locality, between the collection site of specimens 147 and 148 along Agua Caliente Canyon (fig. 32), the quartz monzonite intrudes a quartz diorite that is assigned to the Josephine Canyon Diorite. Not only is the contact very irregular, but also the quartz monzonite contains many small inclusions, and some larger ones of the quartz diorite. Some of these inclusions appear to be incompletely digested granodioritelike rocks. In both of these places, some of the host rocks seem to be assimilated, and in the Agua Caliente Canyon area some stoping may also have occurred along the sides of the stock. At a third locality, near the southern tip of the large cupola along Montosa Canyon, the wallrocks and roof pendants are increasingly intensely altered toward the quartz monzonite. These host rocks are arkose, dacitic volcanic tuff, and breccia of the Salero Formation, which are so widely argillitized and locally are so strongly pyritized that the primary minerals of the rock are almost entirely replaced. This alteration appears to be a hydrothermal rather than a thermal contact effect, but even so the alteration probably is genetically related to the Stock, which is the youngest intrusive rock of significant size in the area. Perhaps the alteration was particularly strong in this area of the fairly ARIZONA flat shoulder on the southern tip of the stock because upward-streaming fluids were more abundant than outward—moving fluids. Finally, a body of fine-grained Elephant Head Quartz Monzonite, erroneously shown as Josephine Canyon Diorite on the geologic map and section A—A’ of the Mount Wrightson quadrangle (Drewes, 1971c), intrudes a small unmapped inclusion of tuff breccia north of Montosa Canyon (specimen collection site 144, fig. 32). PETROGRAPHY The large cupola is a composite mass of two stocks; the larger and older Quantrell stock underlies a gentler terrain than the smaller and younger Yoas stock. The Quantrell stock includes the rocks of the small cupola, and it contains some fine-grained quartz monzonite,,as well as the common coarse-grained kind. Along the east side of the Yoas stock, which is named for Yoas Mountain near the center of the stock (fig. 32), the contrast in weathering characteristics is very striking, as shown in figure 33. Rocks of the Yoas stock underlie some of the most rugged country in the area, including the 25,000-foot-high rock dome of Elephant Head, for which the formation is named, but the eastern part of the stock near the edge of the pediment underlies more gentle terrain. Rocks of part of the Quantrell stock adjacent to the Yoas stock (see geologic map, Drewes, 1971c) also underlie gentle terrain, but some of the Quantrell stock in the high ridges east of the Quantrell mine underlies bold knobs. In most places the contact between the stocks is concealed, but east of Elephant Head it is exposed and the quartz monzonite of the Yoas stock forms a chill zone 5 to 10 feet wide against the Quantrell stock. The trend of this contact is straight and its dip is moderately steep to the east. Biotite in the rocks of the Quantrell stock is recrystallized or contact border of the Yoas stock is recrystallized or contact metamorphosed to felty aggregates of small crystals. The contact southwest of Yoas Mountain is irregular and in places is concealed by small intrusive bodies of lamprophyre and aplite, mostly too small to show on the scale of the geologic map of the Mount Wrightson quadrangle. At this locality the geologic map of the quadrangle erroneously shows two lamporphyre dikes as Josephine Canyon Diorite. The Quantrell stock is made up mostly of a coarse-grained quartz monzonite, but it also includes masses of fine-grained quartz monzonite or aplite, most of which are intrusive into the coarse-grained rock and are small and tabular. Larger masses of fine-grained rock lie along the edge of the stock on the west side of Madera Canyon and southwest of Yoas Mountain, where a few small bodies of aplite associated with the Yoas stock may have been mapped with the slightly older and petrographically identical aplite of the Quantrell stock. CRETACEOUS ROCKS 49 FIGURE 33.—View, toward northeast, of Elephant Head and the east-dipping contact between the Yoas stock, underlying the rugged terrain, and the Quantrell stock, underlying the gentle terrain. QUANTRELL STOCK, COARSE-GRAINED QUARTZ MONZONITE The coarse-grained quartz monzonite of the Quantrell stock underlies a varied terrain of gentle slopes partly veneered by grus and interspersed blocky debris, and of fairly rugged ridges and some chasmlike canyons. The rock is massive and contains at least three prominent sets of widely spaced joints. It is a light brownish gray to light pinkish gray and weathers to brownish gray. Specimens typically are coarsely granular and contain only a sparse scattering of ferromagnesian minerals (fig. 34). The quartz monzonite has a hypidiomorphic-granular texture and a grain size commonly 4 to 8 mm (fig. 35). Some specimens have poikilitic grains and fine-grained myrmekitic, granophyric, and perthitic textures, but the granophyric intergrowths of one specimen 'are coarse. Specimen 135 has a weak metamorphic texture superimposed upon the hypidiomorphic-granular one: small crystals of quartz are scattered between the more abundant coarse crystals, crystal boundaries are strongly sutured, and biotite occurs in clusters of small felty crystals. The strong undulatory extinction of quartz crystals in rocks of the Quantrell stock collected near the contact with the Yoas stock may reflect minor intrusive pressure during emplacement of the Yoas stock. The rock is made up mainly of quartz, plagioclase, orthoclase or microcline and contains small amounts of biotite, magnetite, apatite, sphene, and zircon. Some specimens also have a trace of amphibole, allanite, monazite, pyrite, and tourmaline. The quartz is anhedral and generally has a slight to moderate undulatory extinction. The plagioclase is subhedral, is strongly altered to clay minerals and sericite, and has a composition of albite. About 20 percent of the moded plagioclase occurs as perthitic intergrowths. Potassium feldspar is anhedral to subhedral and is moderately kaolinitized. Patch perthite is more common than lace perthite and is very coarse in some specimens. Biotite is pleochroic in pale yellow brown to moderate brown and is partly altered to penninitic chlorite, sericite, leucoxene, and iron oxide. Magnetite is probably ilmenitic, and zircon is euhedral, fairly large, and abundant when compared to that in most other plutonic rocks of the area. Amphibole is pleochroic in ‘bluish green. Alteration minerals include a trace of epidote, in addition to those already mentioned. QUANTRELL STOCK, FINE-GRAINED QUARTZ MONZONITE The fine-grained quartz monzonite of the Quantrell stock is generally a little more resistant to weathering than the adjacent coarse-grained rock. The rock is closely 50 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA FIGURE 34.—Specimen 134. Elephant Head Quartz Monzonite, a coarse- grained rock of the Quantrell stock. fractured, and the large mass on the west flank of Madera Canyon also has a weakly developed, steeply eastward dipping flow foliation and is pervasively silicified and pyritized. (See geologic map, Drewes, 1971c.) Specimens of fine-grained quartz monzonite are very pale grayish orange pink to light pinkish gray. Grain size is mostly 0.2 to 2 mm throughout the suite of specimens, although the range is much less in individual specimens. An idiomorphic- to hypidiomorphic-granular texture is dominant and parts of many specimens have a fine-grained granophyric texture. ’ The fine-grained rock type contains the same principal minerals as the coarse, but somewhat fewer kinds of accessory minerals are present. The characteristics of the minerals are substantially like those in the coarse-grained quartz monzonite. The alteration minerals are also the same but are more abundant in the fine-grained rock than in the coarse-grained. FIGURE 35—Specimen 132. Coarse-grained Elephant Head Quartz Mon- zonite from the Quantrell stock. Crystals: orthoelase (0), plagioclase (P ), quartz (Q), chloritized biotite (c b), magnetite (mt), apatite (a p), and allanite ( al), Plain light; X 20. YOAS STOCK, COARSE—GRAINED QUARTZ MONZONITE The quartz monzonite of the Yoas stock is almost entirely coarse grained, and it underlies some extremely rugged terrain. The rock is massive and is cut by several sets of joints, which are strikingly prominent on the many bold outcrops between Elephant Head and the southern end of the stock (fig. 33). Away from the sharp contact on the east side of the stock, the joints are fairly widely spaced, but in a zone 400 to 700 feet wide along this contact the joints are commonly l to 3 feet apart, with additional microfractures spaced at intervals of less than an inch. Viewed from a distance, the rock looks slightly pinker than the adjacent quartz monzonite of the Quantrell stock, perhaps because bold outcrops of the Yoas stock are less weathered than the gentler terrain of the Quantrell stock. The quartz monzonite of the Yoas stock is generally free of inclusions, but that in the canyon south of Elephant CRETACEOUS ROCKS 51 Head contains partly assimilated gray andesitic to dioritic inclusions mostly 2 to 12 inches across and rarely many feet across. Quartz veinlets, lamprophyre dikes, and a few aplitic bodies cut the quartz monzonite, apparently at random orientation. Grain size in quartz monzonite of the Yoas stock is typically 4 to 7 mm and in some specimens is as much as 10 mm. Hypidiographic-granular texture is ubiquitous and a few specimens also have some intergranular texture and some poorly developed granophyric texture. Lace type and patch type of perthitic intergrowths are equally common, and they vary widely in size. Mineralogically the quartz monzonite of the Yoas stock is indistinguishable from that of the Quantrell stock. MODAL AND CHEMICAL SUMMARY Modal analyses of 29 specimens of Elephant Head Quartz Monzonite are listed in table 15. The mean of these modes is much like that of the Squaw Gulch Granite (table 7), except that it has 36 percent (instead of 25 percent) plagioclase, slightly more biotite, and a greater variety of accessory minerals. The modified triangular diagram of figure 36 graphically summarizes the modal data. A figure drawn around the plotted modes has an hemielliptical shape, flattened against the leucocratic face of the composition tetrahedron and centered in-the quartz monzonite field. The ranges of the modes of the coarse-grained rocks of the two stocks are coextensive, and the modes of the fine-grained rock are centered slightly closer to the quartz-plagioclase edge of the composition tetrahedron than those of the coarse-grained rocks. A comparison of the modal ranges for the three large plutons of late Late Cretaceous age shows systematic changes. Only the figures circumscribed around the three groups of modes—for the Josephine Canyon Diorite, Madera Canyon Granodiorite, and Elephant Head Quartz Monzonite—are shown together in figure 37, along with the estimated positions of the centers of the figures. In the order from the oldest to the youngest pluton, as determined from field relations, the composition of the magma shifted from a femic-plagioclase-rich point to an alkali-quartz-rich point. In direction and amount the shift is similar to that which occurred between the phases of the Josephine Canyon Diorite. The circumscribed figures have also changed from a tilted discoidal body with a near-vertical short axis, to a su'bspherical body, to a hemielliptical body with a near-horizontal short axis. The trend of compositional change of the rocks is compatible with the typical model of fractionating magma in which the more “basic” fractions are emplaced before the more “acidic” fractions. However, some of the data on the age of these plutons, discussed in the next section, indicate that such a model may at least be oversimplified, if it is at all tenable. The change in the configuration of the figures drawn around the plotted modes is intriguing, but no speculation on its possible significance is offered here. Chemical and spectrographic analyses and CIPW norms of four coarse-grained specimens of Elephant Head Quartz Monzonite are shown in table 16 and are summarized in. the histogram of figures 10K— 10L. Chemically this rock is very similar to the Squaw Gulch Granite and the quartz monzonite of the Corona stock. The Elephant Head Quartz Monzonite has slightly more K20 and less CaO and Na‘20 than the granite, as reflected by the difference in the abundances of orthoclase or microcline and of plagioclase of rocks (tables 7, 15). Chemically, the rock resembles Nockolds’ (1954, p. 1012 — 1016) biotite adamellite and possibly some of the granite types. AGE AND CORRELATION The Elephant Head Quartz Monzonite is given a late Late Cretaceous age on the basis of interpretation of evidence from field relations and radiometric dating, some of which is conflicting. The field relations consistently show that the quartz monzonite is (1) younger than the Salero Formation, which is altered along the contact of the Quantrell stock, (2) younger than the inclusions of quartz diorite, which are correlated with the Josephine Canyon Diorite, and (3) younger than the Madera Canyon Granodiorite, which is metamorphosed near the contact with the small cupola. Because a low grade of metamorphism is a widespread feature in the Santa Rita Mountains, and because contact metamorphic zones are generally indistinct or erratic around stocks of all ages, determining age relations from this evidence requires caution. The evidence of the relationship to the quartz diorite inclusions is weak because it is based on a second interpretation of the identity of the rock. The combined field evidence is thus seen as consistent but not strong. Radiometric dates of two specimens of Elephant Head Quartz Monzonite, obtained from the coarse-grained rock of the Quantrell stock, are listed in table 4. Separate dates by the potassium-argon method on biotite and by the lead-alpha method on zircon are given for each specimen. The two ages obtained by each method are nearly identical, but the potassium-argon ages are about 68 my. and the lead-alpha ages about 180 my The close agreement of the dual results by each method is probably not fortuitous, although further dating information is certainly desirable, particularly in view of the uncertainties besetting the lead-alpha method. Because evidence of recrystallization in the quartz monzonite is lacking, except for the immediate vicinity of the Yoas stock, an explanation involving a late Late Cretaceous thermal 52 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA TABLE 15.—Modes of Elephant [Field numbers are abbreviated; year of collection and collector's initial omitted. Full field number lntrustve mass .................... Quantrell stock ROCK type ------------------- Coarse-grained quartz monzonite Specimen No ...................... 128 1129 130 131 132 133 134 135 136 137 138 139 140 '141 142 128— 142 Field No. . ,7 ........... 7. ........... 7 513 669 701 734 754, m 846 15,3 ”381 7777880 879 876 344 866 901 Mm“ 5 Quartz ........................... 32.1 22.9 23.7 27.2 27.4 32.2 33.3 26.4 24.0 29.4 27.0 18.7 34.8 29.5 27.7 27.8 4.3 Plagioclase, total ............ . 24.9 37.0 33.4 29.0 40.4 37.8 36.1 32.5 43.3 45.8 20.5 37.7 31.7 35.3 38.1 34.9 6.6 (in perthite) .......... . (14.7) (6.1) (7.5) (3.0) (3.9) (9.2) (15.1) (2.3) (13.1) (6.9) (2.2) (5.0) (7.6) (10.3) (8.2) (7.7) (4.2) K—feldspar, total .......... . 40.7 38.1 39.0 41.9 29.5 26.6 29.7 37.3 26.8 22.5 49.0 41.8 30.9 32.1 30.3 34.4 7.3 (orthoclase) .......... . (40.7) 0 (39.0) (41.9) (9.9) (26.6) (29.7) (26.1) (26.8) (22.5) (49.0) 0 0 0 0 (20.8) (17.8) (microcline?) ....... . 0 (38.1) 0 0 (19.7) 0 0 (11.2) 0 0 0 (41.8) (30.9) (32.1) (30.3) (13.6) (16.6) Biotite ................. 8 1.1 3.3 1.1 1.4 2.2 .8 2.6 4.6 1.3 2.4 1.0 1.4 2.1 2.7 1.9 1.1 Amphibole ............. 0 0 0 0 0 .3 0 0 0 0 .6 0 0 0 0 .06 .2 Magnetite ................ . 1 4 .8 .3 .5 .8 .6 .1 1.1 1 0 .6 .3 .8 .9 .6 .7 .7 .1 Apatite .......................... Tr. .1 .2 .05 .1 1 Tr. Tr. .2 .05 .05 Tr. .1 Tr. Tr. .07 .07 Sphene ................. Tr. Tr. Tr. 2 .3 1 0 .1 .1 .4 Tr. O .2 .2 .4 .l .1 Zircon .................. .1 .05 .05 1 .1 .1 Tr Tr. .05 Tr. .1 Tr. Tr. 0 .05 .05 .04 Allanite ................. 0 0 0 O Tr. O 0 0 0 0 0 Tr. 0 .2 .05 Tr. ...... Monazite(?) ............... Tr. 0 0 0 O .05 0 0 0 0 0 0 0 0 0 Tr. ....... Pyrite .................. 0 Tr. 0 0 0 0 0 0 0 0 0 0 0 0 0 Tr. ...... Tourmaline ...................... 0 0 0 0 Tr. 0 0 0 0 0 0 0 0 0 0 Tr. ...... Total ......................... 100.0 100.05 99.95 100.05 100.0 100.05 100.0 100.0 100.05 100.05 99.95 100.0 100.0 100.0 100.0 100.0 ...... Femic ........................ 2.3 2.0 3.9 1 9 2.7 3.4 .9 3 8 5 9 2.3 3.4 1.8 2.6 3.1 3.9 2.9 1.4 Percent anorthite in plagioclase ..................... 0 0—5 0—5 0 0—5 0—5 5—10 0—10 5—10 5 5—10 5—10 10 5—10 5—10 51 ...... (plagioclase rims) ............................................................................................................. (6) ............................. Quartz mixing index(see text) .......................... .93 .61 .74 .88 .73 .79 .57 .78 .67 .90 .65 .93 .90 .94 .79 .13 l Specimen 129 obtained from mine dump near shaft that bottoms in quartz monzonite. FEMIC MINERALS QUARTZ QUARTZ e/ 3e Quartz munzonite (quartz latite) Granodinrite (rhyodacite) Granite @143 140. 134 .128 1500 ' Q 145 144 152 ' 131, El jg 142-1921;? 13 147 130. 136' .129 a 15315, ‘5 Monzonite Syenodiorite Diorite (dacitel V V V POTASSIUM 50 PLAGIOCLASE FELDSPAR EXPLANATION ‘ (9 El Coarse~ Fine- Coarse-grained quartz grained grained monzonite of the Yoas stock Quartz monzonite of A the Quantrell StOCk Total Elephant Head Quartz Monzonite FIGURE 36.—Modified triangular diagram showing modal quartz, potassium feldspar, plagioclase, and femic minerals of the Elephant Head Quartz Monzonite. alteration of a Jurassic stock is as unsatisfying as an complex history of emplacement and subsequent explanation based on a loss of lead from the zircon. development of the stock than seems apparent from field Nevertheless, these conflictlng ages may indicate a more evidence alone. This matter deserves further study. CRETACEOUS ROCKS 53 Head Quartz Monzom‘te of specimen 128 thus is 64D579. Symbols: 5, standard deviation; TL, trace; .. .. not determined] Quantrell stock—Continued Yoas stock Fine-grained Quartz monzonite Coarse-grained quartz monzonite 143 144 145 146 147 2148 149 150 143— 150 151 152 153 154 155 156 151— 156 128—156 342 670 698 703 742 740 736 7_57 Mean 5 747 7a: 783 735 187 885 Mm 5 Mean 5 35.4 29.3 31.6 25.2 26.1 26.9 25.3 32.7 29.1 3.8 21.6 28.0 20.6 35.5 19.4 27.0 25.4 6.1 27.6 4.6 25.2 41.1 39.6 46.7 39.3 33.7 43.5 33.8 37.9 6.8 43.5 30.4 35.0 21.6 39.2 44.6 35.8 8.7 35.9 7.0 (2.5) (1.4) (2.1) 0 (3.6) (5.0) (0.4) (3.9) (2.4) (1.7) (18.5) (5.7) (5.4) (9.0) (11.4) (8.6) (9.8) (4.9) (6.6) (4.7) 38.5 25.2 26.0 20.3 31.4 35.8 24.1 32.3 29.2 6.3 27.7 39.8 41.9 42.2 38.8 26.7 36.2 7.1 33.3 7.3 (38.5) (21.4) (26.0) (20.1) (31.4) (35.8) (24.9) (32.3) (28.8) (6.7) (6.4) (37.7) (41.9) (42.2) (38.8) (26.7) (32.3) (13.9) (25.4) (15.1) 0 (3.8) 0 (0.2) 0 0 (0.1) 0 (.5) (1.3) (21.3) (2.1) 0 0 0 0 (3.9) (8.6) (8.0) (13.7) Tr. 2.5 1.2 5.5 1.7 1.9 5.2 .9 2.4 2.0 2.4 .3 1.2 .6 1.1 1.0 1.1 .7 1.9 1 4 0 0 Tr. 0 0 .2 0 0 .03 .1 2.9 0 0 0 Tr. 0 .5 1.2 .1 5 .8 1.8 .9 1.3 .9 .6 1.3 .1 1.0 .5 1.1 .6 1.1 Tr. .8 .7 .7 .4 .8 .4 .1 .1 .1 2 .2 .3 .2 .1 .2 .1 .2 .05 .1 Tr. Tr. .05 .07 .07 .09 .08 0 Tr. .4 .8 .3 .5 .3 .05 .3 .3 6 .3 Tr. .1 .6 Tr .3 .3 .2 3 Tr. Tr. Tr. Tr. .1 .1 .05 .05 .04 .04 Tr .05 .05 Tr. ,] Tr .04 04 .05 04 0 0 .2 0 Tr. 0 .05 0 Tr. ........ .05 Tr. 0 0 Tr‘ 0 Tr, ....... Tr. ........ 0 0 0 0 0 0 0 0 0 ........ 0 0 0 0 0 0 0 ....... Tr. ........ 0 0 0 0 0 0 0 0 0 ........ 0 0 0 0 0 0 0 ,,,,,,, Tr. ........ 0 0 0 0 0 0 0 0 0 ........ 0 0 0 0 0 0 0 ....... Tr. ........ 100.0 100.0 100.0 100.0 100.0 1WD 100.0 100.0 100.2 ........ 100.05 99 5 99.95 I00.0 100.0 100.05 100.1 ........ 99.9 ........ .9 4.4 2.8 7.8 3.2 3.6 7.1 1.2 4.0 2.5 7.2 1 3 2.5 .7 2.6 1.7 2.7 2.3 3.2 1.9 2 Specimen I48 obtained from a coarse-grained inclusion in a body of quartz diorite that intrudes a large mass of fine grained quartz monzonite. FEMIC MINERALS QUARTZ a/ QUARTZ Quartz munzonite (quartz latitE) Granite Granodiorite (rhvodacite) Monzonite Synodiorite Diorite (dacite V V POTASSIUM 50 FE LDSPA R EXPLANATION PLAGIOCLASE Elephant Head Quara Monzonite Madera Canyon Granodiorite _—\ Josephine Canyon Diorite, fine- grained quartz monzonite A <—A*A Estimated centers of circumscribed figures to modes. Arrow shows shift from oldest, A , to youngest,A Josephine Canyon Diorite, dioritic rocks FIGURE 37.—Modified triangular diagram comparing the range of distribution of the modes of the Josephine Canyon Diorite, Madera Canyon Granodiorite, and Elephant Head Quartz Monzonite. Quartz monzonite stocks around 68 my. old are not present in the Sierrita, Patagonia, Empire, and Rincon Mountains, or in the Canelo Hills, adjacent to the Santa Rita Mountains. Most of the plutonic rocks in these ranges have a granodiorite composition and are either older or younger by 10 my 54 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA TABLE 16.—Chemical and spectrographie analyses and CIPW norms of Elephant Head Quartz Monzonite, coarse—grained rocks [Chemical analyses by rapid rock method (Shapiro and Brannock, 1962), with analyses of specimens 139 and [55 supplemented by atomic absorption. Chemical analysts: Lowell Artis, S. D. Botts, G. W. Chloe. P. L. D. Elmore, John Glenn, H. Smith and Dwight Taylor. Spectrographic analyses by semiquantitative method. Spectrographic analysts: W. B. Crandcll and J. L. Harris. Elements looked for but not found: Ag, As, Bu, B, Bi, Cd, Eu, Ge, Hf, Hg, in, Li, Mo, Ni, Pd, Pr, Pt. Re, Sb. Sm. Sn, Ta. Te, Th, Tl, U, W, and Zn. Symbols: .5, standard deviation; < , less than, and .... not determined] Intrusive mass ........................... Quantreil stock Yeas stock Specimen No ............................. 130 139 130,139 151 155 151,155 130— 155 Field No. ............................... 701 876 Ma" 747 787 “‘3“ Mm ‘ Chemical analyses (weight percent) 5102 ........................... 72.2 73.7 73.0 67.2 74.7 71.0 71.6 3.3 13.7 14.1 15.5 13.6 14.6 14.3 9 1.5 1.3 1.9 .90 1.4 1.4 4 .56 .65 1.2 .36 .78 .72 36 .13 .27 1.4 .10 .75 .51 .61 .70 .73 2.9 .63 1.8 1.2 1.1 2.9 3.7 4.4 4.0 4.2 3.9 .7 5.7 5.2 3.8 4.7 4.3 4.7 .8 .09 .09 .22 .10 .16 .13 .06 .34 .58 .52 .47 .50 .54 20 .43 .34 .49 .38 .44 .39 11 .06 .03 .16 .02 .09 .06 07 .05 .09 .10 .05 .08 .08 .04 < .05 < .05 < .05 < .05 <.05 < .05 ...... Total ............................. 100 100 100 100 100 100 100 Spectrognphic analyses (weight percent) Ba ..................................... 0.05 0.07 0.06 0.07 0.05 0.06 0.06 0.01 Be ..................................... .0005 .0015 .0001 .0003 .0003 .0003 .0003 .0001 Ce ..................................... .01 .02 .015 .01 o 005 .01 01 0 0 .0007 0 .0004 .0002 .0004 0 0 .001 0 .0005 .0003 .0005 .00007 .0005 .01 .00003 .005 .003 .005 .0015 .001 .00I5 .0015 .0015 .002 .0003 .015 -001 4007 .003 .005 .008 .005 40003 .0009 .001 .001 .001 .001 .0005 .01 .005 0 o o .003 ,005 .003 .002 .001 .001 .001 .001 .001 0 -0002 .0007 0 .0004 .0003 .0003 ~01 .01 -05 .005 .03 .02 .02 0 0004 .005 0 .003 001 .002 -002 .003 -002 .002 .002 .002 .0005 .0002 .0003 11002 .0002 .0002 .0002 .0005 .003 ~03 -03 .01 .02 .03 .01 CIPW norm. 33.7 .......... 19.7 1.7 .......... o 33.7 22.5 24.6 37.3 2.8 11.3 o .71 0 .62 0 .002 .32 2.9 .32 .01 0 ............................ .72 2 8 .22 ............................ 1.0 0 .75 ............................ .82 .93 .72 ............................ .14 .38 .05 ............................ .11 .11 .11 ............................ 99 9 .......... 99.3 99 5 ............................ 3 4 .......... 8.5 2 1 ............................ TERTIARY ROCKS and so may well be the tops of larger stocks. Others are probably genetically related to contemporaneous volcanic The youngest plutonic rocks and related intrusive rocks rocks of nearby areas. in the Santa Rita Mountains are mainly 0f Tertiary age. These youngest plutonic rocks consist of four groups; They form small intrusive masses, SUCh as small stocks, the three oldest were emplaced late during the Laramide plugs, and dikes, and some of the rocks have an aphanitic orogeny, and the youngest is postorogenic. Rocks of the and POYPhYFltiC texture, rather than a granitoid one. All Gringo Gulch pluton and of some other unrelated masses these intrusive bodies were probably emplaced fairly close of quartz latite porphyry, latest Cretaceous or early to the surface. Some of them appear to spread downward Paleocene in age, are the oldest group. Granodiorite of the TERTIARY ROCKS 55 Helvetia stocks was emplaced next, followed shortly thereafter by quartz latite porphyry of the Greaterville plugs. Granodiorite of the San Cayetano stock, which was emplaced during the late Oliocene, is the youngest rock. ROCKS OF THE GRINGO GULCH PLUTON AND OTHER ROCKS Small stocks, plugs, and dikes of microgranodiorite, hornblende dacite porphyry, and quartz latite porphyry crop out at several places in the southern quarter of the Santa Rita Mountains (fig. 38). Many of these rocks are strongly altered; as a result, their original composition is uncertain and their radiometric age is unobtainable. Discussing these rocks together is done for convenience and is not meant to imply a demonstrable genetic association. Indeed, some of the rocks may even be of latest Cretaceous age, as shown on the geologic map of the Mount Wrightson quadrangle (Drewes, 1971c), and other rocks may be as much as 10 my younger. The least altered and, hence, the most thoroughly studied of these intrusive masses is the Gringo Gulch pluton. In plan it is an elliptical body covering almost half a square mile in the low hills 1 mile north of the junction of Gringo and Temporal Gulches. A smaller plug, cropping out near the Patagonia village garbage dump 2 miles south of that junction, is related to the pluton. The pluton is a composite body; two small areas in the core of the pluton contain light-brownish-gray microgranodiorite, and light-gray to medium-gray hornblende dacite porphyry forms most of the pluton, as well as all of the related plug. The large size, unaltered condition, and abundance of hornblende phenocrysts are the most striking features of this rock, and their uniqueness in the area suggest the correlation of the plug with the rock of the outer part of the pluton. The rocks of the Gringo Gulch pluton underlie a small basin and form only a few small outcrops as compared to the surrounding host rocks, which are mainly pyroclastic rocks of the Gringo Gulch Volcanics of probable Paleocene age. The plug, however, underlies a small knoll and forms outcrops slightly more resistant to weathering than its host rocks, pyroclastic and epiclastic rocks also of the Gringo Gulch Volcanics. In most places the host rocks next to the pluton and plug are not particularly strongly altered, but those west of the pluton are silicified and intensely argillitized, perhaps owing to the action of fluids from the pluton. The contact of the northern, topographically higher part of the pluton dips gently outward, but that along the southern, lower end of the pluton dips steeply. The southern part of the pluton contains inclusions of altered volcanic rocks. No chilled contact zones have been found along the borders of the hornblende dacite porphyry against the host rocks or on either side of the contact between the porphyry and the microgranodiorite. Several large intrusive masses that underlie areas of 1/4 to 1 square mile and many smaller masses consist of quartz latite porphyry or latite porphyry. A swarm of short dikes and pipes is associated with one large mass near Mansfield Canyon, 2 miles northwest of the Gringo Gulch pluton (fig. 38). Two other large masses of quartz latite porphyry lie between the Salero mine and Josephine Canyon, and a sill-like mass occurs in the northern end of the San Cayetano Mountains. In general, the rocks of these intrusive masses are about as resistant to weathering as the adjacent rocks. They form some small outcrops and a few cliffs, and they disintegrate to a blocky detritus. Both fresh and weathered rocks are mostly pale red to grayish orange pink. They are all finely porphyritic, and phenocrysts are mostly of a pink altered plagioclase and altered mica and, in comparison to other rocks in the area, relatively free of quartz. The edges of these intrusive bodies lack chilled margins, and the host rocks are not contact metamorphosed. In certain details, however, these intrusive masses differ from each other. The quartz latite porphyry, which intrudes the Mount Wrightson Formation near Mansfield Canyon, is ameboid shaped, is surrounded by the swarm of small dikes, and contains some bodies of brecciated latite porphyry. This intrusive mass may be a large breccia pipe formed from a highly fluid magma, and it could be of Mesozoic age. The intrusive masses between Salero mine and Josephine Canyon have a subelliptical shape, and they intrude rocks at least as young as the Salero Formation and possibly as young as the Josephine Canyon Diorite. These rocks are flow laminated and in many places the dips of the foliation are steep and the strike seems to be irregular. The contact along the west side of the northernmost of these masses dips moderately eastward and that on the east side is about vertical, so apparently the size of that mass diminishes downward. These intrusive masses were probably formed from a fairly viscous magma, and they may be plugs or laccoliths. The intrusive mass in the northern part of the San Cayetano Mountains forms a thick southward-dipping sheet of gray latite porphyry that intrudes the Salero Formation. Its geologic relations with an adjacent stock of Josephine Canyon Diorite are concealed. Under the microscope, the microgranodiorite of the Gringo Gulch pluton is seen to be made up of about equally abundant crystals of a larger size (1 — 4 mm) and a smaller size (0.03—0.1 mm), arranged in a hypidiomor- phic-granular texture. The rock contains mostly plagioclase, quartz, orthoclase, biotite, and hornblende, and trace amounts of ilmenitic magnetite, apatite, zircon, and allanite(?), as shown by the modes in table 17. The larger crystals of plagioclase have the composition of calcic 56 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA 110°55’ 110° 50’ " l as ./ | '75 fie/(17:53?» 4o C 2‘ o} 3 31° 35, A SALEHO MINE X ”"22..- . .. “fin ". , .n u u, u. n n r , a, .1' 0‘- «v, ‘ film ’w u.. , ’lvvv. 4' 1 ’2 3 4 MILES J 1 2 3 4 KILOMETRES EXPLANATION Contact Dotted where concealed Younger rocks Surficial deposits Fault Q V E Dotted where concealed \\\ < . . . . — .159 0196 *160 Grosvenor H1115 Volcamcs Granodionte If; M (1 Ch _ 11 Rad'ometricall Mainly rhyodacite and rhyolire Stock ofthe San Lu 0 ed em1ca y 1 y ruff, agglomerate, and flows Cayetano Mountains l— analyzed dated to E Specimen collection sites and numbers D o 3 L” l- lntrusive rocks 2 5 Includes hornblende dacite of the Gringo Gulch pluton E l— and quartz latite porphyry ofother plutons J 9: cc 0 O m, microgranodiorite of the Gringo Gulch pluton Older rocks Igneous and sedimentary rocks FIGURE 38.—Distribution of rocks of Gringo Gulch pluton and other rocks and specimen collection sites. andesine and the smaller ones are calcic oligoclase. Biotite actinolite and to a rim of iron oxide. Quartz and orthoclase is pleochroic in pale yellow brown to moderate yellow are typically smaller grains. . brown and is slightly altered to penninite. Hornblende Hornblende dacite porphyry of the Gringo Gulch pluton forms fairly long subhedral crystals largely altered to and of the associated plug has a felty and granular TERTIARY ROCKS 57 groundmass and phenocrysts as much as, 1 cm long. Phenocrysts make up 15 to 40 percent of the rock and are chiefly of hornblende and plagioclase, but some are augite and magnetite. The plagioclase phenocrysts have an andesine or labradorite composition, and in some specimens they are partly albitized. The hornblende phenocrysts are mostly euhedral, and in one specimen they form a sheathe around a xenocryst of quartz. They are pleochroic in light olive brown to dark olive brown or olive green and are slightly altered to chlorite, epidote, and iron oxide. The groundmass biotite forms small crystals pleochroic in yellowish brown that are largely chloritized, and the pyroxene is probably augite. The quartz latite porphyry of the other intrusive masses is a strongly altered rock that has an idiomorphic—granular groundmass. Phenocrysts are 4 to 7 mm long and make up 15 to 25 percent of the rock. They are much-albitized plagioclase and some chloritized and sericitized biotite; a few specimens contain a perthitic potassium feldspar. Trace amounts of magnetite, apatite, quartz, sphene, zircon, amphibole(?), and pyroxene(?) are present in various combinations. Alteration minerals are very abundant and include clay minerals, chlorite, sericite, leucoxene, epidote, actinolite(?), calcite, chalcedony, and iron oxide. The few available modes are plotted on the modified triangular diagram of figure 39. The hornblende dacite porphyry is a modal syenodiorite that is similar to the rocks of the Josephine Canyon Diorite. The chemical and spectrographic analyses and CIPW norms of two specimens of the Gringo Gulch pluton, which are shown in table 18 and are plotted in figure 10E, suggest the microgranodiorite to be compositionally similar to the granodiorite of the San Cayetano stock (table 23). The chemical analyses show that hornblende dacite porphyry is not simply a fine-grained variety of the microgranodiorite, as in a chilled-border phase. For example, the microgranodiorite contains at least 10 percent more Si02 than does the porphyry. Even assuming that this difference is due to a secondary enrichment, and so removing the excess and recalculating the analytical results to 100 percent, the microgranodiorite has much more K210 and less total iron, MgO, and CaO than the dacite porphyry. The rocks must have been crystallized from separate, but perhaps genetically related, magmas, even thOugh they were contemporaneously emplaced in the same pluton, as is shown by their radiometric ages. The age of the rocks of the Gringo Gulch pluton is more accurately determined by radiometric methods than by geologic relations. The pluton is known to antedate the Gringo Gulch Volcanics, which are younger than the Josephine Canyon Diorite. Most of the other rocks of the group are found to antedate the Salero Formation, and one of the intrusive masses may antedate the Josephine Canyon Diorite. An upper age limit cannot be determined from the geologic relations. Nearly identical radiometric potassium-argon ages of 60 my (Marvin and others, 1973) are obtained from biotite of the microgranodiorite and from hornblende of the hornblende dacite porphyry. No ages were determined on the other rocks of the group. The Gringo Gulch pluton and the related plug, and possibly also most of the other rocks, are considered to be emplaced penecontemporaneously with the deposition of the Gringo Gulch Volcanics, during the tectonically quiescent interval between the Piman and Helvetian phases of the Laramide orogenv. GRANITOID ROCKS OF THE HELVETIA STOCKS During the late Paleocene seven small elliptical stocks, mainly of granodiorite and quartz monzonite, intruded the complexly deformed rocks of the northern part of the Santa Rita Mountains (fig. 40). They underlie areas of about 1/4 to 2 square miles, and most of them are elongate in a northwest direction. Collectively I refer to the stocks as the Helvetia stocks; individually they include the Helvetia, Shamrod, South Johnson Ranch, and Johnson Ranch stocks of quartz monzonite and granodiorite, the Southeast stock entirely of quartz monzonite, the Huerfano stock wholly of granodiorite, and the Sycamore stock of quartz diorite. Compositionally, most of the rocks are nearly the same despite the different rock names that are applicable, but the quartz diorite is sufficiently different that the Sycamore stock may be a genetically distinct intrusive mass. The stocks were emplaced, approximately contemporaneously, during faulting of the Helvetian phase of the Laramide orogeny (Drewes, 1972b) and before the mineralization of the Helvetia mining district (Drewes, 1973). The stocks intrude a wide assortment of rocks of Precambrian and Paleozoic ages, and they are covered by much younger gravel deposits. They are also cut by faults, which are intruded by quartz latite porphyry of the Greaterville plugs. The host rocks are metamorphosed, and the zone of metamorphism is wider to the south than to the east, where unaltered Mesozoic rocks have been faulted against the older rocks. Although the intensity of metamorphism does not seem to increase immediately adjacent to the stocks, it does gradually decrease south- ward about 3 miles away from the stocks. This thermal event may have begun as early as the time of intrusion of the stocks of Helvetia but it probably extended through the time of injection of the plugs of Greaterville as well, and so it is not, strictly speaking, a contact metamorphic effect of one stock. The intrusive contacts of the Helvetia stocks are mostly straight and are vertical or dip steeply outward. In a few places apophyses of the stocks extend a short distance into 58 PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA QUARTZ EXPLANATION FEMIC 9. MIN E RALS Hornblende dacrte porphyry of the Gringo Gulch plugs “3/ he Granite Quartz monzonite Granodiorite Microgranodiorite of the (quartz Iatite) (rhvodacite) Gringo Gulch plugs A Granodiorite of the San Cayetano stock ‘o ,3; V .v is .2: 156 197 A A Estimated Syenodiorite position. , (dacite) ‘159 , . Monzonne O _r_ Dlonte 158 ( \ \\, 157 v v v v v v v 9 POTASSIUM 50 PLAGIOCLASE FELDSPAR FIGURE 39.—Modified triangular diagram showing modal quartz, potassium feldspar, plagioclase, and femic minerals of rocks of the Gringo Gulch pluton and of the San Cayetano stock. the host rocks, and, along the east side of the Shamrod stock, a large block of Cambrian Bolsa Quartzite is engulfed in the stock. Chill zones were not found along the margins of the stocks. PETROGRAPHY The granitoid rocks of the Helvetia stocks are generally light gray to yellowish gray or light brownish gray, coarse grained, massive, and slightly altered. The quartz monzonite and aplitic rocks of the Southeast stock (fig. 40) are slightly more orange or pink than are the rocks of the other stocks, and the rocks of the Sycamore stock are medium gray to medium greenish gray. Several of the stocks underlie, in part, the pediment at the foot of the mountains where they are deeply weathered. They also underlie some gentle slopes and low hills beneath the more rugged outcrops of Paleozoic rocks. Only the rocks of the Southeast stock and aplitic masses in the Huerfano stock form bold outcrops. The granitoid rocks are cut by widely spaced joints, and, locally, they weather to low rounded bosses or extensive fairly flat surfaces that are separated by abundant grus and overlain by scattered rounded residual blocks. The rocks of most of these stocks are nearly free of inclusions; exceptions are the Shamrod stock, which contains a large engulfed block on the east side, and the Sycamore stock, which contains large masses of Paleozoic rocks and granodiorite. The granitoid rocks are cut by a few widely scattered and mostly small dikes of lamprophyre and aplite (Drewes, 1972b, pl. 5). Huerfano Butte is underlain by the largest of the aplite masses and the nearby granodiorite is silicified. Other smaller aplite and lamprophyre dikes extend across the contacts of stocks or are entirely within the host rocks near the stocks. Most of the rocks have a hypidiomorphic-granular texture, a 3 to 7 mm grain size, and slight or no alteration (figs. 41, 42). A trace of microgranophyric texture is present in most specimens, and some of the finer grained rocks are idiomorphic granular or have a mosaic texture. A bimodal distribution of grain size occurs in the specimens of the Southeast stock, in the nearby end of the Helvetia stock, and in the southeast end of the South Johnson Ranch stock. Large irregular intergrowths of quartz in feldspar also are common in specimens from the east end of the Southeast stock. The texture of the Sycamore stock is subophitic and the grain size of the rock is only 0.5 to 3 mm. In a few specimens from the other stocks crystals are as much as 9 mm long, and one specimen has phenocrysts as much as 18 mm long. The dominant minerals in the granitoid rocks are quartz, plagioclase, potassium feldspar, biotite, and, in the Sycamore stock, hornblende. The quartz is generally anhedral and rarely subhedral, and that of most stocks has little or no undulatory extinction. However, quartz of the South Johnson Ranch stock has moderate to strong undulatory extinction, possibly suggesting that stresses along the faults where the highly elongated stock was emplaced had not been fully relieved at the time of emplacement. The Sycamore stock contains relatively minor amounts of quartz that has a moderate undulatory extinction. TERTIARY ROCKS 59 TABLE l7.—Modes of Gringo Gulch pluton and related plug [Field numbers are abbreviated; year of collection and collector’s initial omitted. Full field number of specimen 157 thus is 63Dl99. Symbols: Tr.. trace; . . ., not determined] R kt H bl d d ' h Mic” oc ype ............... orn en e ac1te porp yry grano- diorite Intrusive mass ........... Pluton Plug Pluton Specimen No ............. 157 158 159 160 Field No. ............... 199 661 257 281 Quartz .................. 1.3 1o 4.2 10.8 1o 19.8 Plagioclase .............. 73.5 5.0 58.6 21.8 10.8 52.1 Orthoclase .............. 9.9 0 18.9 0 0 14.9 Pyroxene(augite7) ....... 5.0 .5 .7 .3 .5 0 Hornblende ............. 3.3 3.3 9.0 5.3 5.8 5.4 Biotite. .. ...... .9 0 2.4 1.0 0 4.7 Magnetite ...... 5.5 1.2 5.9 2.6 1.4 2.5 Apatite .. .6 0 .3 0 0 .4 Sphene . . Tr. 0 0 0 0 0 Ziron. . .. 0 0 Tr. 0 0 .2 Allanite ....... 0 0 0 0 0 Tr. Total ............. 100.0 10.0 100.0 31.8 18 5 100.0 Femic ............ 15.3 5.0 18.3 9.2 7.7 13.2 Percent anorthite in plagioclase .......... 60—68 0 — 37 38 — 55 35 —47 (plagioclase rims) ..... (37) . . . . . . (24) I Phenocrysts. The plagioclase varies slightly in habit and composition from stock to stock. It typically forms subhedral grains. Plagioclase of the southern four stocks is slightly and finely perthitic, but that of the other stocks is rarely perthitic. The composition of the plagioclase of the Sycamore stock is andesine, except for the outer zones of some crystals, which are oligoclase. The composition of the plagioclase in the other stocks is oligoclase or albite. Several stocks have only albite or only oligoclase except for the rims of some crystals, but the Helvetia stock has oligoclase in some specimens and albite in others, perhaps suggesting an incompleted diagenetic alteration. Sericite and clay-minerals alteration products are generally sparse. The potassium feldspar also has variable features in the stocks of Helvetia. It generally forms anhedral to subhedral crystals. The Southeast stock has only orthoclase, the Huerfano and Helvetia stocks have both orthoclase and microcline(?), and the Shamrod, South Johnson Ranch, and Johnson Ranch stocks have only microcline(?), The identity of the microcline is in doubt because the grid twinning is very blurred, perhaps because it is partly obliterated. Both kinds of potassium feldspar occur in accessory amounts in the Sycamore stock. Lace and patch types of perthite are about equally abundant, and kaolinite alteration is typically slight. The biotite generally occurs in subhedral and partly chloritized crystals that are pleochroic in yellow brown to dark brown. Several specimens from the South Johnson Ranch stock that have a bimodal grain size contain small biotite crystals in aggregates, suggesting that they were probably recrystallized. The biotite of one specimen of the Sycamore stock is pleochroic in yellow brown to dark olive TABLE 18.—Chemical and spectrographic analyses and CIPW norms of the Gringo Gulch pluton [Chemical analyses by rapid rock method (Shapiro and Brannock. 1962), with specimen 157 sup- plemented by X-ray method. Chemical analysts: Lowell Artis, S. D. Botts. G. W. Chloe. P. L. D. Elmore. and H. Smith. Spectrographic analyses by semiquantitative method. Spectro- graphic analyst. J. C. Hamilton. Elements looked for but not found: Ag, As, Au. B, Bi. Cd, Ce, Eu, Ge, Hf, Hg, ln, Li. Mo, Nd, Pd. Pr. Pt. Re, Sb, Sm, Sn, Ta, Te, Th. Tl, U, W, and Zn] Rock type ................... Hornblende dacite Microgranodiorite porphyry Specimen No ................. 157 160 Field N0. ................... 199 281 Chemical analyses (weight percent) $102 ................... 59~0 66-9 A12 03 ................. 16'3 15'9 Fe 203 .................. 3.4 2.7 FeO ................... 2-8 1.4 MgO .................. 2-6 1-4 CaO ................... 4.7 2-8 NaZO ................. 4-4 4-0 K 20 ................... 2-4 3-2 H20- .................. L1 -36 H 20 + ................ l 4 -90 TiOz .................. ~98 -56 P 205 .................. 21 -18 MnO .................. 14 06 co 2 ................... 12 10 Total ............ 100 100 Spectrognphlc analyses (welglit percent) Ba ..................... 0.05 0.15 Be ..................... .0002 0 Co ..................... 0 .0007 Cr ..................... 0 .001 Cu ..................... .001 .01 Ga ..................... .002 .003 La ..................... 0 .005 Nb ..................... .003 0 Ni ..................... 0 .0007 Pb ..................... .002 .007 Sc ...................... 0 .0007 Sr ...................... .003 .07 V ...................... 0 .01 Y ...................... .0015 .002 Yb ..................... .0003 .0002 Zr ..................... .007 .007 CIPW norms Q ...................... l 1.5 24.1 C ...................... 0 1.0 or ...................... 14.2 18.9 ab ..................... 37.2 33.8 an ..................... 17.6 12.1 wo ................. 1.5 0 d1 {en .................. 1.1 0 fs .................. .17 0 en .................. 5.3 3.5 by {f5 .................. .80 0 mt ..................... 4.9 3.1 hm ..................... 0 57 11 ...................... 1.9 1.1 ap ..................... .50 .43 cc ...................... .27 .23 Total ............. 96.9 98.8 Femic ............ 16.4 8.9 7—7 60 _ PLUTONIC ROCKS, SANTA RITA MOUNTAINS, SOUTHEAST OF TUCSON, ARIZONA 110°50' 110°45’ EXPLANATION Younger rocks Igneous rocks and surficial deposits SOUTH JOHNSON ‘ ,, RANCH STOCK\ O H a: V-l C) CALIFORNIAN 30° 28°15’ o o m E SURIAN 24°45' -—-- —————— o p Sa n Laza ro 22°45’ 3.00 MILES l—T_'I_I—‘l—l—| 0 300 KILOMETRES 25° PANAMANIAN 20° coasts of the United States and Baja California (after Manar, 1953, and Lynn, 1967). FIGURE 5.—Ostracode fauna] provinces and the distribution of areas of intense upwelling (stippled pattern) along the Pacific 14 ZOOGEOGRAPHY 0F HOLOCEINE OSTRACODA OFF WESTERN NORTH AMERICA 1 30° 120° 115° 1 10° Cape Blanco PACIFIC \ : 40° \10 \ \ \ ~ ‘~—-- \ \s OCEAN \\ ‘I 35° \‘\ \1 1 30° 0 300 MILES O 300 KILOMETRES ' 25° azaro Cabo San Lucas: FIGURE 6.—Sea-temperature map (in degrees Celsius) for the shelf off the Pacific coasts of the United States and Baja Cali- fornia. Yearly minimum temperatures at —10 m from San Francisco south are after Lynn, 1967, and yearly minimum surface temperatures from San Francisco north are after Robinson, 1957. DISTRIBUTION PATTERNS 130° OIJIDVJ 300 MILES 300 KILOMETRES l FIGURE 7.—Sea-temperature map (in degrees Celsius) for the shelf off the Pacific coasts of the United States and Baja Cali- fornia. Yearly maximum temperatures at ——10 m from San Francisco, south are after Lynn, surface temperatures firom San Francisco north are after Robinson, 1957. 1967, and yearly maximum 16 ZOOGEOGRAPHY OF HOLOCENE OSTRACODA OFF WESTERN NORTH AMERICA Ostracoda, Foraminifera, and intertidal flora and fauna) are fairly coincident. (See references in J. W. Valentine, 1966; in Hazel, 1970; and in Lankford and Phleger, 1973.) This result lends weight to the hypothesis (J. W. Valentine, 1968, 1973) that provinces are composed of communities, the aspects (structure, composition, function) of which are characteristic of the provinces in which they occur. As expected, the ostracode provinces in the area un- der investigation mimic those based on the Mollusca. Subprovinc-es exist, however, which are new or which deviate from those of the Mollusca. In order to delineate the provinces, a cluster analy- sis of similarity coefficients was made. I consider this method to be the most expeditious and efficient means of comparing the faunal compositions of shelf areas. This tool has been utilized in previous studies (J. W. Valentine, 1966; J. W. Valentine and Peddi- cord, 1967; P. C. Valentine, 1971). The samples for this investigation were assembled from various sources and are not evenly distributed over the shelf area. The study, however, is based on the latitudinal ranges of ostracode species, and these ranges are adequately delineated by the samples analyzed. It is considered that samples composed of assemblages from consecutive latitudinal segments of shelf along the coast form a better basis for comparison than samples distributed nonuniformly over the shelf. If it is reasonable to conclude that species occurring at points along the shelf also occur on the shelf between those points, then all species occurring in or ranging into or through each 15-minute (latitude) shelf seg- ment (sample) can be considered to be a member of that sample (figs. 1—4; see table 2 for latitudinal ranges of 15-minute samples). Choice of 15—minute samples was made to achieve a balance between sam- ple size and sampling detail; the sample matrix was limited to manageable proportions, and samples re- mained small enough so that sample groups that might segregate over short geographic distances on the shelf would be detected. A Q-mode cluster analysis (defining groups of samples based on the similarity of their species com- position) was therefore based on a comparison of the faunal content of 110 15-minute samples. Degree of similarity between samples was calculated using the Otsuka similarity coefficient (C/\/N1N2) and presence—absence data, where C=number of species in common between two samples containing N1 and N 2 species. The Otsuka coefficient was chosen because it gives results which neither emphasize difference (as does the J accard coeflicient) nor similarity (as TABLE 2.—Latitudi71al ranges of 15-minute samples along the the west coast of the United States and Baja California, Mexico Sample Latitndinal range (north) Panamanian faunal province 001 21°30'—21°45’ 002 45’—22°00’ 003 22°00’— 15’ 004 15’— 30’ 005 30’— 45’ southern Surian faunal province 006 22°45’—23°00’ 007 23°00’— 15’ 008, 109 15’— 30' 009, 110 30'- 45’ 010 45’—24°00’ 011 24°00’— 15’ 012 15’— 30' 013 30'— 45’ northern Surian faunal province 014 24°45’—25°00’ 015 25°00’— 15’ 016 15’— 30’ 017 30'— 45’ 018 45’——26°00’ 019 26°00'— 15' 020 15’— 30’ 021 30’-—- 45’ 022 45’—27°00’ 023 27 °OO’— 15’ 024 15’— 30' 025 30’— 45’ 026 45'—28°00’ 027 28°00’— 15’ southern Californian faunal province 028 28°15’—28°30’ 029 30'— 45’ 030 45’—29°00’ 031 29°00’— 15’ 032 15’— 30' 033 30’—- 45’ 034 45’—30°00’ 035 30°00’— 15’ 036 15‘— 30’ 037 30’— 45’ 038 45’—31°00’ 039 31°00’~ 15’ 040 15’— 30’ 041 30’— 45' 042 45’—32°00’ 043 32°00’— 15' 044 15’— 30’ 045 30’— 45' 046 45'—33°00' 047 33°00’— 15’ 048 15’— 30' 049 30’— 45’ 050 45’——34°00’ northern Californian faunal province 051 34°00’—34°15’ 052 15’— 30’ 053 30’— 45’ 054 45’-—35°00’ 055 35°00’— 15’ 056 15’— 30' southern Oregonian faunal province 057 35°30’—35°45' 058 45’—36°00’ DISTRIBUTION PATTERNS 17 TABLE 2.—Lat1'tnd1'nal ranges of 15-m1'nute samples. along the west coast of the United States and Bow Cal1forn1a, Mexico—Continued Sample Latitudinal range (north) southern Oregonian faunal province 059 36°00’— 15 060 15’— 30’ 061 30’— 45’ 062 45’-—-37°00’ 063 37°00’— 15’ 064 15’— 30’ 065 30’— 45’ 066 45’—38°00’ 067 38°00’— 15’ 068 15’— 30’ northern Oregonian fauna] province 069 38°30’—38°45' 070 45’—39°00’ 071 39°00’— 15’ 072 15’— 30' 073 30’— 45’ 074 45’—40°00’ 075 40°00’— 15’ 076 15’— 30’ 077 30'— 45' 078 45’—41°00’ 079 41°00’— 15' 080 15’— 30’ 081 30’— 45’ 082 45'-—42°00’ 083 42°00’— 15' 084 15’— 30' 085 30’— 45’ 086 45’—43°00’ 087 43°00’— 15’ 088 15’— 30’ 089 30’— 45' 090 45’-44°00’ 091 44°00’—- 15’ 092 15’— 30' 093 30’— 45’ 094 45’—45°00’ 095 45°00'— 15’ 096 15'— 30’ 097 30’— 45’ 098 45’——46°00 099 46°00’— 15’ 100 15’— 30’ 101 30’— 45’ 102 45’—47°00’ 103 47°00’— 15’ 104 15’— 30’ 105 30'— 45’ 106 45’—48°00’ 107 48°00’— 15’ 108 15’— 30’ does the Simpson coefl‘icient) between the samples being compared. (See Cheethvam and Hazel, 1969, for a résumé of binary similarity coefficients.) The sam- ples were associated and ordered in a dendrogram using the unweighted pair-group method of cluster- ing (Sokal and Michener, 1958). This procedure al- lows all samples to have equal weight when being compared to another sample or samples (in an al- ready formed cluster). Since all samples in this study represent an equal geographic range (except where the coastline deviates from its approximate north-south trend), each sample should be consid- ered to be of equal importance in determining sample clusters (ultimately faun-al provinces). The matrix used was 192 species by 110 15-minute samples. (See table 3 for alphabetical list of ostracode species included in cluster analyses.) TABLE 3.—Alphabetical l1'st of Holocene and fossil ostracode species included in cluster analyses [Fzspecies occurs only in fossil deposits] Occur- Cluster rence m. Holo- No. Species cene and H010- Fos- Holo- fos- cene sil cene sil 150 Ambostmcon californicum (Hazel, 19 62) —————————————————————————————— X X X X 151 A. costatum Hazel, 1962 ______________ X X X X 094 A. diegoensis (LeRoy, 1943a) _________ X X X X 093 A. ylancum (Skogsberg, 1928) ___- X X X X 126 A. microreticulatum (LeRoy, 19433) X X X X 001 A. sp. A X —- X X 091 A. sp. X X X X 015F A. 51). —- X -— X 152 A. 31). X X X X 101 A. sp. X X X X 120 A. sp. X X X X 103 A. sp. X X X X 118 A. sp. X ._ X X 095 A. 51). X X X X 097 A. sp. X __ X X 099 A. sp. X X X X 154 A. 51). X X X X 098 A. sp. X X X X 116 A. sp. X X X X 114 A. sp. X __ X X 053 A. sp. Q X -- X X 002 Aurila lincolnensis (LeRoy, 1943a) ___ X X X X 007 A. montereyensis (Skogsberg, 1928) X X X X 004 A. sp. A X X X X 006 A. sp. B X X X X 009 A. sp. C X X X X 010 A. sp. X X X X 056 “Aurila” californica Benson and Kaesler, 1963 ________________________ X __ X X 113 “A.” drineri (LeRoy, 1943a) __________ X X X X 108 “A.” schumannensis (LeRoy, 1943a) ___ X X X X 008 “A.” sp. C - X X X X 110 “A.” sp. D _ X X X X 112 “A.” sp. E _ X __ X X 016F “A.” Sp. G _ -_ X _. X 024F “A ” 51). H __ X __ X 104 Basslerites delreyenais LeRoy, 1943a -._ X X X X 155 B. thlipsurmdea Swain, 1967 __________ X __ X __ 102 X —— X X 174 B. C X -_ X _- 171 “Bradleyaf’ pennata. (LeRoy, 1943a) X X X X 078 “B.” simiensia (LeRoy 1943a, b) X X X X 100F Buntonia sp. __ X -_ X 096 B. sp. B ______ X X X X 092E B. sp. C ......... _- X -- X 178 Bythoceratma sp ___ ___ X __ __ 04oF Bythocythe’re sp. A ___________________ -_ X _.. X 069 Catwella sem1translucens (Crouch, 1949) —————————————————————————————— X X X X 100 C. unitaria Swain, 1967 ___ X __ X __ 017 C. sp _______________ X __ X X 015 C. ................... X -_ X __ 020 Candites Bfragihs LeRoy, 1943B. _____ X X X X 023 C. purzi (McKenzie and Swam, 1967) X __ X X 121 C. sp. A X X X X 022 C. Sp. X, X X X 156 0. 31). X __ X X 025 C. 51). X __ X X 027 C. sp. X .- X X 181 0. sp. X —— X X 190 0. sp. X -- X X 177 C. sp. X .. X X 191 C. sp. X __ X X 021 C. 51). X __ X X 157 0. sp. X —_ X X 041 C. sp. X __ X X 075 C. X __ X X 087 Coqmmba pichelmguenais (Swain, —————————————————————————————— X _— X X 158 C schencln (LeRoy, 1943a) - __ X X X X 029 C. sp. A _____________________ _ X X X X 042F C. B ______________________ _ __ X __ X 092 Costa? sanfelipensis Swain, 1967 _ X __ X __ 035 Cythere maia (Benson, 1959) nu _ X X X X 037 C. sp. A _____________________________ X __ X X 039 ______________________________ X X X 138 Cytherella banda Benson, 1959 ........ -- X X 18 TABLE 3.—Alphabetical list of Holocene and fossil ostracode ZOOGEOGRAPHY OF HOLOCENE OSTRACODA OFF WESTERN NORTH AMERICA species included in cluster analyses—Continued TABLE 3.—Alphabetical list of Holocene and fossil ostracode species included in cluster analyses—Continued Occur— Cluster rence M Holo- No. Species cene and H010- Fos- Holo- fos- cene sil cene sil 140 C. 5.1) A ............................. X -_ X -_ 136 Cytherelloidea californica LeRoy, 19438- X X X 043 X X X 105 X .. X _. 044 X —— X X 187 X X X X 046 X —— X X 133 X X X X 129 C X X X X 048 Cytheromorpha sp. A X X X X 127 C. sp. B _____________________________ X X X X 106 Cytheropteron dobladoensis Swain, 1967- X -_ X __ 115 C. johnsonoides (Swain, 1967) _______ X __ X X 161 C. newportense Crouch, 1949 __________ X __ X X 050 sp. ..................... X __ X X 105F C. sp. B ............ __ X -_ X 164 Eucytherura sp. A ._ X _. X __ 011 Hemicythere sp. A ___ X __ X X 013 H. 51) _________ X __ X X 012 H.‘! sp. C ___________________________ X __ X X 122 “Hemicythere” culiforniensis LeRoy, 1943a ____________________________ x x x X 162 “H.” hispida. LeRoy, 19433 X X X X 123 “H.” sp. A X X X X 186 “H.” sp. X X X X 131 “H." sp. C ___- X __ X X 163 “H.” sp. X X X X 125 “H." sp. X __ X X 124 “H.” sp. F ___- X __ X X 165 “H.” :1). G __-- X X X X 106F “H. ” H __________ __ X __ X 054 Hemicsythe'rum sp. A X __ X X 055 H. sp. B ____________ X __ X X 057 H. sp. 0 __ X X X X 059 H. sp. D X __ X X 065 H. sp. E ______ X __ X X 060 H. sp. F X __ X X 063 H. sp. G X __ X X 062 H. sp. H X X X X 064 H. sp. I X X X X 066 H. sp. J X X X X 067 H. sp. K X X X X 068 H. SD. L ____________________________ x X X X 070 Hermamtes Icewi (LeRoy, 1943a) __ X X X X 132 H. sp. A ____________________________ X __ X X 071 H. sp. B _ X X X X 073 II. 513. C - X X X X 074 H. sp. D --_ X X X X 166 H. sp. E --_ X X X X 072 .H. sp. F ___- X __ X X 005 H. 5D. G ___ X X X X 14oF H. H __ X -- X 045 Kangarma aff. 003 K § __ § § 036 K. X X X X 038 K. X X X X 184 K . X X X X 185 K. . X _‘ X X 188 K. . X __ X __ 155}? K. . .- >< __ X 149 K. p. X __ X X 077 “Kangwrma X X X X 164F Krithe sp __ X __ X 081 Loxoconcha helenae Crouch 1949 _ X X X X 082 L. X X X X 079 L. X X X X 083 L. X X X X 085 L. X X X X 084 L. X __ X X 167 L. X __ X X 169 L. sp. F X __ X X 170 “Loxoconcha” emuciata Swain, 1967 ___ X __ X X 168 Loxocorniculum sculptoides, Swain, 1967 _______________________________ X __ X X 086 L. sp. A __ X X X X 080 L. sp. B X __ X X 058 Muneeyella pedroenais Triebel, 1957 -_-_ X X X X 174F M. similia? Triebel, 1957 ______________ __ X __ X X —— X X X X X X X —— X X - p —— X —- X 128 Neocaudites. " hen'ryhow ei (McKenzie and Swain, 1967) ___________________ X __ X X 172 New genus A sp. A .. X X X X 188F New genus B sp. A _- __ X __ X 193F New genus C so. A ___ ._ X __ X 160 New genus D sp. A X X X X 1941‘ New genus E sp. A __ X __ X 130 New genus F sp. A ___- X X X X 134 Orionina pseudovaughni S X -_ X X 195F Paijenborchella sp. A __________________ __ X -_ X Occur- Cluster rence analyses Holo- No. Species cene and H010- Fos- Holo- fos- cene sil cene sil 117 X X X X 119 X X X X 135 X X X X 137 X X X X 052 X __ X X 139 5.1) D X X X X 141 Palmenella caliform'ca Triebel, 1957 ___ X X X X 146 Paracytheridea granti LeRoy, 1943a _- X X X X 143 P. sp. A X X X X 142 P. sp. B X X X X 144 P. sp. C X —— X X 145 P. sp. D X -.. X X 173 P. sp. E X X X X 175 P. sp. F X -- X X 176 P. G X X X X 061 Peczocythere clavata (Triebel, 1957) ___ X X X X 148 P. tomalcnsis Watling, 1970) _________ X _- X X 147 P. 51) ___________________ X X X X 189 P. sp B X __ X X 196F P. sp. D - ._ X __ X 049 Pellucistomu and Swain 1967 ___ X X X X 179 P. scrippsi Benson, 19 X -_ X X 051 P. sp. A ____________ X __ X X 047 P. sp. B ________________________ X __ X X 180 Perissocytheridea pedroen is (LeRoy, 1943a) _____________________________ X X X X 159 Pontocythere sp. A __ X X X X 076 P. sp. B ____________ X X X X 033 P. sp. C ___ X __ X X 192 P. sp. D _________________ X __ X X 016 Pterygocythereis? sp. A ______________ X __ X __ 018 Pulmilocytheridea pseudoguardensis McKenzie and Swain, 1967 X X X X 019 Purirma pacified Benson, 1959 X __ X X 024 P. sp. A ___________________ X __ X __ 153 P. 5.12 B _____________________________ X __ X X 183 Radimella convergens (Swain, 1967) _- X __ X X 026 R. palosensis (LeRoy, 1943a) X X X X 030 X —— X X 931 p. -— >< -- X X 028 ‘ Radimellu" aurita (Skogsbe1g, 1928) __ X X X X 107 “R.” jollaensis (LeRoy, 1943a) ________ X X X X 109 “R.” pacifica (Skogsberg. 1928) X X X X 111 “R." sp. A ______________________ __ X X X X 182 “R." sp. B __-- X X X X 032 Suhnia sp. A _________ _ X X X X 040 “Trachyleberis” sp. A -_ X __ X -- 042 “T" sp. ____________________________ X ..- X -- 014 Triebelinu reticulopuncfata Benson, 1959 _______________________________ X X X X 034 X X X X New information has made it necessary to change the names of several of the species treated in this study. The new species designations are indicated below, and the reader may make these changes in tables 3 and 4 and in the plate explanations. 037 Cythere sp. A=C. alveolivalva Smith, 1952 046 “Cytheretto.” sp. A=“C.” minutimmctata Swain and Gilby, 1974 105 Cytherelloided sp. B=C. paratewwii Swain and Gilby, 1974 062 H emicytherura sp. H=H. santosensis Swain and Gilby, 1974 090 Mimseyella sp. B=M. parkeme (Swain and Gil- by, 1974) 105 Cythei’elloided sp. B=C. paratewarii Swain and Gilby, 1974 117 “Paijenborchella” sp. A=Acuminocythere cres- centensis Swain and Gilby, 1974 CLUSTER ANALYSIS 19 119 “Paijenborchella” sp. B=Acumz'nocythere sp. B 132 Hermam’tes sp. A=“Lucasocythere” sanmarti- nensis Swain and Gilby, 1974 133 “Cytheretta” sp. B=“C.” rothwelli Swain and Gilby, 1974 167 Loxoconcha sp. E=Palmoconcha laevimargina— ta Swain and Gilby, 197 4. SHELF ASSEMBLAGES DEFINED BY CLUSTER ANALYSIS The cluster analysis of Holocene samples (fig. 8, 9) indicates that the samples segregate into four major clusters on the basis of species content. These major sample clusters (actually provincial assem- blages) define ostracode faunal provinces. In gen— eral, they are areally equivalent to molluscan provinces (J. W. Valentine, 1966, figs. 1, 4), and rather than pro-pose new names for the ostracode provinces, molluscan provincial terminology will be retained. Latitudinal discrepancies between ostra- code and molluscan (sub) provincial boundaries are summarized below. Ostra- Molluscan Latitude code provinces North provinces Cluster 49 ______ 48°24’ (limit of 48 study area) 47°30' ___________ * northern * Mendocinan 39 ...... 38°30’ OREGON- 38 OREGONIAN A 37°30’ ___________ IAN 37 southern Montereyan 36 ——35°30' 35 34°30' northern 34 ______ 33 CALIFORNIAN 32 CALIFORNIAN B southern 31 30 29 28°15’ 28 27 26 northern SURIAN SURIAN C 25 ______ 24°45' 24 southern 23 22°45’ PANAMANIAN 22 PANAMANIAN D Cluster A (Oregonian ostracode province) con- tains samples occurring from the northern limit of the study area at Cap-e Flattery, Wash. (48°24’ N.) , south to latitude 35°30’ N. near Point Piedras Blan- cas, Calif. (35040’ N.). Further, this cluster con— tains two subprovinces, a southern and northern, which segregate at latitude 38°30’ N., south of Salt Point, Calif. (38°34’ N.). Cluster A corresponds to the Oregonian molluscan faunal province, although the subprovincial boundary within the ostracode province as well as the Californian—Oregonian ostra- code provincial boundary are shifted to the north 1° of latitude. Cluster B (Californian ostracode province) in- cludes samples extending from Point Piedras Blan- cas south to the Bahia Sebastian Vizcaino area (28° 15’ N.). This province, the approximate equivalent of theCalifornian molluscan province, also contains two subprovinces which segregate south of Point Conception in the region of the northern Channel Islands. The ostracode faunal nature of the Cali- fornian-Oregonian provincial boundary is discussed in the following section. Cluster C, the Surian ostracode province, is equiv- alent to the molluscan province of the same name which extends from Punta Eugenia south to latitude 22945’ N. south of Cabo San Lucas (22°52’ N.) ; two subprovinces segregate at latitude 24°45’ N., south of Cabo San Lazaro (24°48’ N.). The two subprov- inces are not discernible in molluscan studies. Cluster D is a small but distinctive group of sam— ples occurring at the southern end of the study area off the west coast of Mexico. They, together with samples 4 and 5 (which cluster at a low level with samples of the Surian province), demonstrate a faunal discontinuity at latitude 22°45’ N., south of Cabo San Lucas (22°52’ N.). Cluster D lies in the northernmost part of the Panamanian molluscan faunal province. The problem of deciding which clusters represent groupings of faunal-province magnitude is difficult and must be resolved subjectively. Delineation of known major provinces by cluster analysis provides a somewhat quantitative basis for the comparison of sample clusters and the detection of new provinces and subprovinces. Major provincial boundaries de- lineated by cluster analysis coincide in most cases with those previously determined by other means. Discussion of faunal provinces in the following sections refers to those based on ostracode distribu- tions (table 5 and figs. 5, 9, and 13) unless otherwise indicated. 20 15-MINUTE SAMPLE NUMBERS ZOOGEOGRAPHY OF HOLO‘CENE 0ST‘RACODA OFF WESTERN NORTH AMERICA OTSUKA COEFFICIENT OF SIMILARiTY 100 90 80 70 Cape Flattery —) 103 . . ‘ 48°24’ Salt Point —> 38°34’ \ Pt. Piedras Blancas —> 35°40’ Pt. Conception 34°30’ N. Channel Is. “> 34°00’ -—) Bahia S. Vizcaino 28°15’ Punta Eugenia ~—> 27°52’ Cabo San Lazaro —> 24°48’ Cabo San Lucas a 22°52’ 107 106 105 104 103 102 101 100 60 50 30 40 Northern subprovince Oregonian province A 20 n I I Southern subprovince Northern subprovince Californian province 8 Southern subprovince Northern subprovince Surian C province Southern subprovince Panamanian province D 60 50 40 30 l l l 20 10 0 FIGURE 8.——Dendrogram of Holocene sample clusters. Samples are compared on the basis of their ostracode species composi- tion using the Otsuka similarity coefficient (C/ VN1N2) X 100; clustering was by the unweighted pair-group method. FAUNAL CHARACTER OF PROVINCES 21 c3 (7 [€00 / ‘ I Cape Flatter OHIO Val '11, CALIFORNIAN Warm temperate Nvaoo SURIAN Subtropical PANAMANIAN Tropical O 300 MILES O 300 KILOMETRES FIGURE 9.—The geographic extent of faunal provinces and climatic zones of the eastern Pacific shelf based on distributions of ostracode species. FAUNAL CHARACTER OF PROVINCES AND PROVINCIAL BOUNDARIES As noted, 192 diagnostic species out of a total of 341 Holocene species have been used in this zoogeo- graphic study. The geographic ranges of Holocene species and the occurrence of species in Pliocene and Pleistocene units are shown in table 4, and some of the important members of the fauna are illustrated in the plates. Of the 56 genera treated in the total fauna (Holocene and fossil), 50 occur in the Holo- cene, and 9 are particularly diverse. Others, though less diverse, are important members of the fauna. The generic ranges which will now be discussed are based on the results of this study and, for occur- rences in the Panamanian province, the literature was a determining factor. The Gulf of California is considered to be part of the Panamanian province. The genus Ambostracon contains 21 species, 20 of which occur in the Holocene. Ambostracon is found in all provinces, but reaches its highest diversity in the Californian province. Loxoconcha, comprising 8 Holocene species, ranges throughout all four prov- inces. Another diverse genus, Caudites, is repre- sented by 15 species in the Holocene, all but one of which occur in the Surian province. Eleven species of Caudites are endemic to the Surian, and the genus is restricted to the south of Point Piedras Blancas. Caudz’tes is also recorded in the Panamanian province. Kangarina contains 9 species (8 occur in the Holo- cene) and is mainly an element of the Californian and Surian provinces; it is also known to occur in the Panamanian. Likewise, Paracythem’dea (8 H010- cene species) is found only in the Panamanian, Suri- an, and Californian provinces. H emicytherma is represented by 12 species which are rather evenly distributed throughout the study area, although it is not yet known from the Panamanian province. “Hemicythere” is an undescribed genus based on LeRoy’s (1934a) Hemicythere? califomiensis (two forms recognized by LeRoy) and H .? hism’da forms. This taxon includes 10 species, 9 of which occur in the Holocene, and it is found chiefly in the Californi- an province and the northern Surian subprovince, although it is reported fro-m the Panamanian prov- ince (Gulf of California; Swain, 1967). H ermam’tes includes 9 species, 8 occurring in the Holocene; the genus occurs in the Californian and Surian prov- inces, and it too has been reported from the Gulf of California. “Amila” is an undeScribed genus (or genera) and comprises 8 species, 6 of which occur in the Holo- cene. This group has representatives throughout the four provinces. Radimella (4 species occurring in the Holocene) occurs chiefly in the Californian and more southern provinces, but 1 species ranges into the southern Oregonian subprovince. “Radimella,” ZOOGEOGRAPHY 0F HOLO‘CENE OSTRACODA OFF WESTERN NORTH AMERICA TABLE 4.-—Holoeene geographic ranges and fossil occurrences of ostracode species in the study area [FISpecies occurs only in fossil deposits] Holocene Fossil occurrence geographic range gun Pedro San ormation Santa ‘ - _ ,.__ alos Foxen Caren a No. Species 15-minute Mafxmlum D1980 't Timms Xirdes Barbara Mud- Sandg latitudinal Forma- Lomi a P . t Forma- sample . arl om Sand . stone stone range range tion flt hon (north) Member Member 001 Ambostracon 51). A __________________ 053—108 34°30’—48°30’ __ __ _.- .__ .._ __ _- 002 Aurila lincolrbemis _ 028~108 28°15’—48°30' X X —— ~— X X __ 003 Kangarina sp. A __ _ 034—044 29°45’—32°30’ .__ __ __ __ __ __ __ 004 Aurila sp. A -__- _ 028-108 28°15’—48°30’ X X __ __ X X X 005 Hermam‘tea sp. G _ - 044 32°15’—32°30’ X X H -_ _.. X X 006 Aurila sp. B ...... - 026—057 27°45'—35°45’ X X __ __ __ X _.. 007 Aurila montereyensi -- 053—107 34°30’—48°15’ __ __ __ __ X _.- .__ 008 “Aurila” sp. C _______ -__ 0287063 28°15’—37°15' X X -_ X __ __ _.. 009 Aurila sp. C -__ _ 036—085 30°15’—42°45' __ X __ X X __ __ 010 A. sp. D ____________ _ 014-057 24°45’—35°45’ m. X __ X __ __ __ 011 Hemicythere 51). A __ _- 064—108 37015243030' __ __ __ __ __ .__ .__ 012 IL? sp. C ____________________________ 109, 110 22°45'—25°00' __ __ _.. __ _, .__ __ 006—014 013 H. Sp. B _____________________________ 081—108 41°30’-48°30' __ __ _.. __ __ __ __ 014 Triebelimt reticulopunctata —_ 021—051 26°30’—34°15’ X __ __ -_ __ _.. __ 015 Cati'uella 51). B ............ — 001—003 21“30’—22°15’ __ __ __ __ __ __ __ 015F Ambostracon 51). C ...... — ___________________ __ __ __ __ __ X X 016 Pterygacythereis? 51). A _ —— 001—003 21°30’—22°15’ __ __ __ .__ __ __ __ 016F “Aurila” sp. G __ _ _ ___________________ X __ __ __ __ __ __ 017 Cati’vella SD. A ______________ — 023—056 27°00’—35°30' __ __ __ __ __ __ .__ 018 Pulmilocytheridea pseudoguard m3 -__ 014—036 24°45’—30°30’ X __ __ __ __ .__ _.. 019 Puriana pacified _____________________ 109, 22°00’—35°30’ __ __ _.. _.. __ __ __ 003—056 020 Caudites fragilis 023—057 27°00'—35"45’ X X _.- X .__ __ __ 021 Caudit'x 51). J _ 109 23°15’—23°30’ __ __ __ __ __ _.- __ 022 Caudites sp. B __ 014—052 24°45’—34°30’ __ s_ __ X __ __ __ 023 Caudites purii __ 021—028 26°30’—28°30’ _._ __ __ __ __ __ __ 024 Puriana SD. A __ 001-003 21°30'—22°15' _.. _- __ _.. .__ __ __ 024F ”Aurila” sp. H __ ___________________ X _.. __ __ __ __ __ 025 Caudites sp. D ...... 109, 22°45’—28°00' __ __ __ _- __ __ __ 006-026 026 Radimella palosensis 014—068 24°45’—38°30’ __ X __ X __ __ __ 027 Caudites sp. E ________________________ 109, 22°45’—28“00’ __ __ __ _.. __ .__ .__ 006—026 028 “Radimellu” aurita ____________________ 014—062 24°45'——37°00' __ X __ X _... __ __ 029 Coquimba sp. A 022—051 26°45'734°15' X X X X X .__ __ 030 Radimella sp. A __________________ 109, 110 22°45'—32°45' __ __ __ __ __ .__ .__ 006—045 031 R. sp. B ____________________________ 109,110, 22°45'—27°15' __ __ __ _... __ __ __ 006—023 032 Sahnia SD. A ________________________ 023—107 27°00’-—48°15' __ __ __ X __ _.. __ 0‘33 Pontovythe’re sp. C - 053—108 34°30'—48°30' __ .__ _.. __ __ _.. __ 034 Triebelina 51). A _ 025—049 27°30’—33°45’ X __ __ __ __ __ _.. 035 Cythere maid. -__- 022—057 26°45’—35°45’ X X __ __ __ X __ 036 Kangarina 51). B - 034—045 29°45’—32°45' __ _.. _.. __ X __ __ 03'] Cythere sp. A _-. 069—108 38°30'—48°30’ __ __ __ __ _._ __ __ 038 Kangarina sp. C _ 021—067 26°30’—38°15’ __ X __ __ __ __ X 039 Cythere 51). B _________ 067—108 38°00'—48°30' __ __ __ __ X X _... 040 “Trachyleberis” sp. A . 001—003 21°30’—-22°15' __ _.. __ .__ __ __ __ 04oF Bythocythere sp. A -__- ................... __ __ __ __ X _.. __ 041 Caudites sp. L ________ _ 110 23°30'—23°45’ _.. __ __ __ __ .__ _.- 042 “Trachyleberis” sp. B _ _ 001—003 21°30'—22°15' __ __ __ __ _.- __ __ 042F Coquimba. sp. B _________ _ ___________________ X X __ __ a- __ .__ 043 Cytherelloidea 51). A __________________ 109, 22"45’-34°00’ __ __ _.. __ X _... .__ 006—050 044 C. 51). C ____________________________ 109,110 23°15'—23°45' .__. _.. __ _.- __ __ __ 045 Kangarina aft. K. quellita ____________ 109, 110, 22°45’—28°00’ __ __ __ __ __ __ .__ 006—02 046 “Cytheretta” 51). A ___________________ 052—108 34°15’—48"30' _.- __ __. __ .__ __ _.. 047 Pellueistama 51). B _________ 014—027 24°45'—28°15’ __ __ s- __ __ __ __ 048 Cytheromorpha sp. A -__ 056—108 35°15’748°30’ —_ __ __ _- __ X __ 049 Pellucistoma berLsorLi 023—051 27°00’—34°15’ __ X __ __ __ __ _.. 050 Cytheropteron sp. A -__ 067—108 38°00’—48°30’ _.. __ __ __ __ __ __ 051 Pellucistoma sp. A _ 025—049 27°30’—33“45’ __ __ __ __ __ __ __ 052 Palaciasa sp. C _____ 049—059 33°30’—36°15’ s- .__ __ w __ _.. __ 053 Ambostracon sp. Q -__- _ 021—025 26°30’—27°45’ —— _— —— __ __ __ _.. 054 Hemieytherura sp. A _ - 053—108 34°30’—48“30’ __ .__ __ __ __ __ _.. 055 H. sp. B _____________ 053—108 34°30’—48°30' —— —— —_ —— —— —— _— 056 “ urila” californica __________________ 109,110, 22°00’—32°45’ _._ __ __ _.- __ .__. _.. 003—045 057 Hemicytherura sp. C _________________ 021—052 26"30’—34°30' X X __ _-. X X X 058 Munseyella pedraensis 027—053 28°00’—34°45’ __ X X _.e X .__ __ 059 Hemicytherura sp. D __ _ 056—107 35°15’—48°15’ _.. __ _~ __ __ __ .__ 060 . 51). F _____________ _ 006—041 22°45’-31’45' __ _ __ __ _.. __ .__ 061 Pectocythere clmmta __ _ 049—052 33°30’—34°30' __ __ X __ X __ __ 062 Hemicytherura sp. H _- _ 034—059 29“45’—36°15’ X __ __ X _.. __ 063 H. Sp. G ___________ 006—014 22°45'—25°00’ __ _.- _.. __ __ .__ __ 064 H. sp. I __ 036—107 30°15’—48°15’ X __ .__ X _._ __ 065 H. 51). E _ 082—108 41°45'—48°30’ _.- __ __ __ __ __ __ 066 H. sp. J _ 034—107 29°45’—48°15' X __ __ _.. X __ __ 067 H. sp. K . 027—049 28°00’—33°45’ __ X —_ —— —— _.. _— 068 H. Sp. _______________ 034—081 29°45'—41°45' __ X __ —— X X —— 069 Cativellu semitranslucens 028—052 28°15’—-34°30’ __ X __ —— X —_ _— 070 Hermanites kewi _________ 046—057 32°45’—35°45' X X X X X X X 071 H. sp. B ________ 021—028 26°30'—28°30' X X __ __ __ __ __ 072 H. sp F ___ 025—034 27°30'—30°00’ __ __ __ __ __ __ .__ 073 H. sp. 0 --- 028—050 28°15’—34°00’ X x __ fl __ __ _— 074 H. SD. D ...... 026—056 270451455030, X X _1 X —_ —— _— 075 Caudites 31). M _____ 014—026 24°45’—28°00’ _.s __ __ __ __ -_ __ 076 Pontocythere 51). B - 028—068 28°15’—38°30’ __ __ X X —— —— —— 077 “Kungarina” sp. A -__- 036—108 30°15’—48°30‘ >4 __ __ ~s X __ -— 078 “Bradleya” simiensis _________________ 047—050 33°00’—34°00’ __ X X X X —— —— FAUNAL CHARACTER OF PROVINCES TABLE 4,—Holocene geographic ranges and fossil occurrences of ostracode species in the study area—Continued 23 Holocene Fossil occurrence geographic range Igun Pedro San ormation S n . , ' - —— . Palos Foxen Careaga N0. SDecxes 15-m1nute ggffirgiffi FDlego_ Lomim Tunms Verdes garb“? Mud- Sand- sample orma r1 Pomt Sand orma stone stone range range tlon ‘1 Silt “on (north) Member Member 079 Loxaconcha sp. A --__ -- _____ 056—108 35°15’—48°30’ __ __ __ __ __ X -— 080 Lozocorniculum 51). B ________________ 109, 110 22°45’—23°45’ __ __ __ __ __ __ __ 081 Loxovoncha hele’nae __________________ 051—085 34°00’—42°45’ __ __ X __ X __ __ 082 L. lenticulata ______ __ 041—052 31°30'—34°30' X X X X X __ __ 083 L sp. _____ _ 025—063 27°30’—37°15’ __ X __ X __ __ __ 084 L Sp. D __ _ 036—048 30°15’—33°30’ .. __ __ __ __ .. __ 085 L 51). C _____________ _ 034—052 29°45’-34°30’ X __ __ __ X X .. 086 Loxocorniculum sp. A _____ _ 023—049 27°00’—33°45’ X __ __ X .._ __ __ 087 Coquimba pichelinauemis ____________ 109,;10 22°45’—28°00’ __ __ __ __ __ __ __ 00 —026 088 Munseyellu sp. A ____________________ 036—045 30°15’—32°45’ __ __ __ __ __ __ __ 089 . sp. __________ _ 062—108 36°45'-48°30' __ __ __ __ __ __ __ 090 M. 513. B __________ - 045—095 32°30’—45°15’ __ X __ __ X X __ 091 Ambostmctm sp. B __ _ 045—108 32°30’—48°30' X __ __ _- __ X X 092 Coata? sanfelipensis __ _ 001-003 21°30'-22°15' __ __ __ .._ __ __ -— 092F Buntonia sp. C ______ - ................... _- __ X __ _.. __ __ 093 Ambostmcon glaucum _ 028—057 28°15’—35°45’ X __ __ X __ X __ 094 A. dieyoensia ________ _. 028—050 28°15'—34°00’ X X __ __ __ __ __ 095 A. 51). J ______ _ 014—057 24°45’—35°45’ X X __ X X __ __ 096 Buntonia Sp. B ______ _ 048-051 33°15'—34°15' X X —— _“ X —— ~— 097 Ambostmcon sp. K __________________ 306 014 22°45'—25°00' —— —— —— —— __ —— —— 098 A. sp, N ____________________________ 021—050 26°30’—34°00’ __ X X __ X __ __ 099 A. sp. L _______ _ 028—051 28°15’—34°15’ X X x __ >< __ X 100 Cuti’vella, unitaria _ 003 22°00’—22°15’ __ __ -_ __ __ __ -— 100F Buntoniu 51). A _____ — ------------------- —— —— —— —— —— —— X 101 Ambostmcon 5p, E _ _ 053—108 34°30'—48°30’ X _- __ __ __ __ __ 102 Bassle'rites 5p, B ____ _ 028—041 28°15’—31“45’ __ __ .. __ -_ __ __ 103 Ambostracon 31). G __ _ 021-108 26°30'—48°30' X X _— X —— X -— 104 Basslerites del‘reyensis - 036—065 30°15’—37°45’ __ X __ __. __ __ __ 105 Cutherelloidea sp. B ---_ - 003 22°00’—22°15’ __ __ __ __ __ __ __ 105F Cythe'ropteron sp. B _ — ——————————————————— — X __ —— X _— -— 106 C. dobladoensia ______________________ 001—003 21°30’—22°15’ __ __ __ __ __ _ __ 106F “Hemicythere” sp. H ———————————————————————————————————— _— —— —— —— —— —— X 107 “Radimella.” jollaensis -_ _ 021—108 26°30'—48°30' X X X X X X —~ 108 “A." schumannensis __ _ 062—108 36°45’—48°30’ __ __ __ __ __ X X 109 “Radimella” pacifica _ _ 014-073 24°34’—39°45' X X __ X _- -_ __ 110 “Aurila” Sp. D ______ _ 028—057 28°15’—35°45’ -x X x X x X x 111 Radimella sp. A _ 028‘108 28°15'—48"30’ X X X X X —— .— 112 «Ag» Sp, E ______ - 063—108 37°00’—48°30’ __ __ __ __ __ __ __ 113 “A." driveri _______ _ 028—056 28°15'-35°30' X X X X X X X 114 Ambostracon sp, P __________ _ 022—027 26°45'—28°15’ __ __ __ __ __ __ __ 115 Cytherapteron johnsonoides - - 003-006 22°00'“2_3°00' __ _.- __ __ __ __ __ 116 Ambost'racon sp. 0 _________ 023—050 27°00'—34°00’ X __ __ __ __ X __ 117 “Paijenbarchellu.” 51;). A _ 053—108 34°30'—48’°30' __ 1- __ __ __ X __ 118 Ambostracon sp_ I _______ 023—034 27°00'—30°00’ __. __ __ .. .. __ __ 119 “Paijenhorchella” sn. B 056—071 35°15’—39°15’ -_ __ _- __ __ X X 120 Amboatracon sp. F 041—051 31°30’—34°15’ X X __ X __ __ __ 121 Caudites sp. A ____________ _ 045—057 32°30’-35°45' __ X ___ X __ __ __ 122 “Hemisythere” californie sis _ 027—071 28°00’—39°15’ X X X X X __ -— 123 “Hemicythere” 51). A ________ 027—051 28°00'—34°15’ X X X X X X X 124 “H.” sp. F __________ 022—027 26°45’—28°15’ __ __ __ __ __ __ __ 125 “H." SD. E ___________________ 025—036 27°30'—30°30' ._ -_ __ __ _._ __ __ 126 Amboatrucon microreticulatum ‘ 028—056 28°15'—35°30’ X X __ X X X -— 127 Cytheromorpha sp. B _________ 051—108 34°00'—48°30’ __ X X __ X __ X 128 Neocaudites? hem‘yhowei . 026—045 27°45’—32°45’ __ __ __ __ _.. __ __ 129 “Cytheretta” sp. C _______ 041—052 31°30’—34°30’ X __ __ fl- __ __ __ 130 New genus F sp. A __ 067—108 38°00’—48°30’ __ __ __ __ __ __ X 131 “Hemicythere” 5}). C __ 014—036 24°45’—30°30’ __ __ __ __ __ __ __ 132 Hermanites sp. A _____ 026—036 27°45’—30°30’ __ __ __ ._ __ __ __ 133 “Cytheretta” sn. B ______ 025—052 27°30’—34°30’ X __ X __ __ X X 134 Orionina pseudovauuhni - 109,110, 21°30’—27°15’ __ __ __ __ __ __ .__. 001—023 135 Palacz'oaa sp. A ______________________ 036—068 30°15’—38°30’ __1 X __ __ __ X .. 136 Cytherelloz'dea californica _ 006—056 22“45’—35‘30’ X X __ X X X __ 137 Palaciosa sp. B ___________ 034—065 29°45’—37°45’ X X __ __ __ __ __ 138 Cytherella banda _ 036—051 30°15'—34°15’ __ __ __ __ __ __ __ 139 Palacz'osa sp. D __ 022—034 26°45'—30°00' __ __ -_ X __ __ __ 140 Cytherella SD. A .. 001—003 21°30’—22°15’ __ __ _.. _- __ __ _- 1401“ Hermanites sp. H ________________________ __ X __ __ X __ _ 141 Palmenella culifornica _ 025—056 27°30’—35°30’ X __ X X X _- —X 142 Paracytheridea 5p. B __ 022—051 26°45’—34°15’ X X __ __ __ X X 143 P. sp. A ____________ 022-057 26°45’—35°45’ X X __ _1 X __ .._ 144 P. sp. C ___ _ 021—050 26°30’—34°00’ __ __ __ __ __ __ __ 145 P. sp. D __ - 023—041 27°00’-31°45’ __ __ __ __ __ __ __ 146 P. granti ____ _- 041-050 31°30’—34°00’ x~' x X __ X __ __ 147 Pectoeythere 51). A _ - 053—108 34°30’-48°30' __ __ __ __ -1 X X 148 P. tomalenais ______ __ 062—108 36°45’—48°30’ __ __ __ __ __ __ __ 149 Kangarina sp. H _________ _ 048—050 33°15’—34°00’ __ __ __ __ __ __ __ 150 Ambostracon californicum _ 050—051 33°45’—34°15' __ X X X X __ __ 151 A. costutum ______________ _ 048—051 33°15'—34°15’ X X X __ X __ .. 152 A. sp. D _____ _ 050—056 33°45'—35°30’ __ __ .. __ X __ __ 153 Puriana sp. B _____ _ 006 22°45’—23°00’ __ __ __ __ __ __ __ 154 Ambost'racon sp. M _____ _ 049—108 33°30’—48°30’ X X __ __ X X __ 155 Bavssle'rites thlipsu‘roidea _ 001 21°30’—21°45’ __ __ _ __ __ __ __“ 1551’ Kangarina sp. G ___________________________ __ X __ __ __ X __ 156 Caudites Sp. C ______________________ 109,110, 22°45'—25°00' __ __ __ __ __ _._ __ 006-014 157 C. 51?. K _______.___ 109 23°15'-23°30’ __ __ __ __ __ __ __ 158 Coquzmba schenckt 028—051 28°15’—34°15’ X __ X X X X X 24 ZOOGEO-GRAPHY OF HOLO'CENE OSTRACODA OFF WESTERN NORTH AMERICA TABLE 4.—Holocene geographic ranges and fossil occurrences of ostracode species in the study area—Continued Fossil occurrence Holocene geographic range gun Petdro San orma ion Santa ' - , Palos Foxen Careaga No. Species 15-minute llzlaférfififi FDiego Lomita Tm)?“ Verdes garbara Mud- Sand- sample range 33:“. Marl 1380.1“ Sand 0131:" stone stone range (north) Member Merhber 159 Pontocythere sp. A __________________ 044—050 32°15’—34°00’ X __ X __ __ __ _.. 160 New genus D sp. A ______________ 052—068 34°15’—38"30’ __ __ __ __ __ X X 161 Cytheropteron newportense 050—052 33°45’—34°30’ __ __ __ __ __ _.- __ 162 “Hemicythere” hispida _ 048—051 33°15’—34"15’ X X _... __ X __ _. 163 “H." sp. D __________ 041—050 31”30'—34°00’ __ X __ X .. _... _.. 164 Eucythe'rura sp. A 001—003 21°30’—22°15’ __ __ __ _.. __ __ __ 164F Krithe 51). A __________________________ __ X __ __ X __ __ 165 “Hemicythere” sp. G 048—052 33°15'—34°30' __ X X —— X ___ -— 166 Hermanites sp. E ___ 036—049 30°15’—33°45’ X __ __ __ __ __ __ 167 Loxoconcha sp. E ____________________ 109, 110, 22°00’—34°00’ __ __ __ _. __ __ __ 003—050 168 Loxocorm'culum aculptoides __________ 109, 110, 22°45’—25°00’ __ __ __ __ __ __ __ 006—014 169 Loxoconcha sp. F ____________________ 109, 110, 22°45’—25°00’ __ __ __ __ __ __ .__ 006—014 170 “Loxoconcha” emaciata _______________ 109, 110, 22°45’—23°45' __ __ __ __ __ __ __ 006 171 “Bradleya” pennant 028—056 28°15’-35°30’ __ X X __ X __ -— 172 New genus A SD A ___ 028-050 28°15’—34°00' __ X __ __ _.. _.. __ 173 Paracytheridea sp. E __ 041—051 31°30'—34°15' X X X -— X __ __ 174 Bassleritea sp. C ______ 001 21°30’—21°45’ __ __ __ __ _.. _.- __ 174F Mumeyella similis‘! ___________________ __ __ __ __ __ __ X 175 Paracytheridea sp. F _________________ 109, 110, 22°45’—23°45’ __ x- __ __ __ __ __ 006 176 P. 51). G _____________________________ 041—050 31°30’—-34°00’ __ X __ X __ __ __ 177 Caudites sp. H _______________________ 109, 110, 22°45’—23°45’ __ __ __ __ __ __ __ 006 178 Bythocemtina sp. A __________________ 003 22°00’—22°15’ __ __ __ __ __ __ ._ 178F Munseyella sp. D ___- ________________ ,__ __ __ __ __ X __ __ 179 Pellucistomu scrippsi _________________ 109,110, 22°45'—34°30' __ __ __ __ __ __ __ 006—052 180 Periasocytheridea pedroensis __________ 045—050 32°30’—34°00’ __ X X X .. _.. __ 181 Cauditea sp. F _________________ _ 22°45’—25°00' __ __ -- __ ___ __ __ 182 “Radimcllu” sp. B _ ________ _ 28°15’—34°15’ __ X __ __ __ X __ 183 R. convergens ___- ______ _ 23°15’—23°30' __ __ __ __ __ __ _ 184 Kangarina sp. D 30°15’—33°45’ __ X __ __ __ __ __ 185 K. Sp. __________ 22"45’—23°00' __ __ __ __ __ __ _... 186 “Hemicytherc” sp. B _ 32°15’—-34°15’ __ X __ X __ X X 187 “Cytheretta” corrugata 31°30'—34°15' X X —— X _.- __ _... 188 Kangarina sp. _____ 22°00’—22°15’ __ __ __ __ __ __ __ 188F New genus B sp. A ____________ __ __ X __ __ __ __ 189 Pectocythere sp. B 43°15’-—48°30’ __ __ __ __ __ __ __ 190 Caudites 51). G ---_ 24°45’—25°00’ __ __ __ __ __ __ _.- 191 C. sp. I _____________________ 24°45’—25°00’ _.. __ _.. __ _.. __ .. 192 Pontocythere sp. D __________ 22°45’—23°00’ __ __ __ __ __ __ _- 193F New genus C S1). A ________ __ __ __ __ __ __ X 194F New genus E sp. A __________ _ X __ __ __ __ __ __ 195F Paijenbo'rchella sp. A ________ _ __ __ __ _- __ X X 196F Pectocythere sp. D ___________________ _.- __ __ __ __ X __ another undesoribed genus (or genera) comprising 5 Holocene species, occurs in the Surian, Californi- an, and Oregonian provinces. Cythere, a genus usually restricted to northern latitudes in other parts of the world (Hazel, 1970) , includes 3 species in the study area, 2 of which occur in the Oregonian province and probably range far- ther north. One specie-s, however (Cythere main.) , is confined to the Californian and northern Surian sub- province, and although it is very similar to one of the northern species (Cg/there sp. B), it is distinc- tive, and their ranges do not overlap. Its present dis- tribution may be the result of southward migration facilitated by the cooling of coastal waters by up- welling. H emicythere, represented by 2 Holocene species, is another cryophilic genus (Hazel, 1970), and it occurs only in the Oregonian province and most probably farther north. A summary of the provincial distribution of these and other important genera follows. Numbers in parentheses indicate, first, species occurring in the Holocene and, secondly, species occurring only in the Pliocene or Pleistocene. Panamanian—Oregonian provinces: Ambostracon (20: 1) “Aurila” (6; 2) Cytheropteron (4; 1) “Hemicythere” (9; 1), mainly Californian and south. Kangarina (8; 1), mainly Californian and south. Loxoconcha, (8; 0) Pontocythere (4: O) Radimella (4; 0) , mainly Californian and south. FAUNAL CHARACTER OF PROVINCES I 25 Panamanian—Californian provinces: Caticella Caudites Coquimba Cytherelloidea H ermanites Loxocorniculum Paracytheridea Pellucistoma Puriana (4; 0) (15; 0) (3; 1) (4:0) (8; 1) (3; 0) (8; 0) (4; 0) (3; 0) These genera do not range north of Point Piedras Blancas. Surian—Oregonian provinces: Aurila Cythere “Cytheretta” H emicytherura M unseyel la ”Rudimella” (6: 0) , mainly Californian and north. (3; 0) (4; 0) , mainly Californian and north. (12; 0) (4; 2), mainly Californian and north. (5; 0) Californian—Oregonian provinces: “Paijenborchella” Pectocythere (2; 0) , does not range south of Point Conception. (4; 1) Oregonian province: H emicythere (2; O) , does not range south of Point Piedras Blancas. This summary clearly indicates the magnitude of the fauna] discontinuity between the Californian and Oregonian provinces. Nine important genera termi- nate their northern ranges in the Californian prov- ince. The nature of other provincial boundaries, namely the Panamanian—Surian and the Surian— Californian, are not as well understood on the basis of generic distributions. From the information at hand, it seems that the Panamanian—Surian discon- tinuity may be of the magnitude of the Californian- Oregonian break. The Surian—Californian faunal boundary is probably of less magnitude, as indicated by the fact that these two provinces show more simi- larity to each other than to either the Oregonian or Panamanian provinces (fig. 8). The distributions of ostracode genera provide a broad outline of provincial configuration. The faunal character of a province can be more precisely evalu- ated on the basis of the numbers of species it con- tains and their distribution patterns within and out- side of the province (tables 4, 5). The northern limit of the study area apparently does not include the northern boundary of the Oregonian province, and therefore comparative figures on the occurrences of species in this province are incomplete. TABLE 5.—Distribution of inner sublittoral ostracode speciesAand their occurrence in faunal provinces of the west coast of North me rica [Based on 143 species which range through at least 2" of latitude; also included are 28 species which exhibit short ranges or single occurrences in the Panamanian province and adjacen t southern Surian subprovince and are considered to represent northern distributional limits of southern species] ' I Eastern Pacific faunal province PANAM‘. SURIAN CALIFORNIAN OREGONIAN I Subprovince i 3 North South N South North : Latitude north 21 2'2 23 24 2526 27 2 29 3'0 31 3'2 33 34 35 36 37 3'8 3'9 4'0 4'1 4'2 4'3 4'4 4'5 46 4'7 48i 4'9 I 12 15 34 29 22 7* j Northern range end points of ostracode species I 16 I I 27 63 29 l I 4| I 17 I 45 49 I16 8 I 4 I Southern range end points of ostracode speciesl 4 I a i 52 65 12 I 9 l N l o I 2 13 I0 0 I ? I >, Endemic species . " I g I 14 27 7 1' 25 : 4 I 8 30 I37 31 I 7 I 3 Species ranging through I ? I E 3 15 7 I .- I J _l ' I Species in common I - 21 66 39 J I i 33 I 66 100 I82 61 I 42 I Total species . 20 4| i 78 116 64 I *36 species range north to 48°24’N (limit of study area), and it is not known at this time how many range on northward. 26 ZOOGEOGRAPHY OF HOLOCENE OSTRACODA OFF WESTERN NORTH AMERICA 0f the 171 species treated in table 5, the Oregoni— an, Californian, and Surian provinces contain 64, 116, and 78 species, respectively. A similar diversity trend was noted in molluscan distributional studies (J. W. Valentine, 1966, table 2). The increase in diversity from the Oregonian to the Californian province conforms to the generally recognized lati- tudinal diversity gradient which increases from polar to equatorial regions. Ostracode as well as molluscan diversity decreases from the Californian to the Surian province, however. This reduction in the number of species may simply be the result of a lack of adequate sampling along the coasts of south- ern Baja California, or it may be partially due to lower habitat diversity in the Surian province, which is smaller areally than the Californian. The boundary between the Californian and Ore- gonian provinces is characterized by many range endpoints. This boundary is essentially the result of northern range terminations of thermophilic species and actually occurs in two steps—a southern break between the southern and northern subprovinces of the Californian province at 34° N. where, within 1° of latitude, 37 southern species end their ranges (contrasted with only 6 endpoints of northern species), and a northern break between the Cali- fornian and Oregonian province at about 35°30’ N. where 19 thermophiles end their ranges within 1° of latitude (against only 4 southern endpoints for cryo- phile-s). This major faunal discontinuity is related to the marine climatic regime (figs. 6, 7). The south- ern fauna] elements encounter the coolest water in their ranges in the area between 34° N. and 35°45’ N. where the temperature minimum is‘11° to 12° C. It is difficult to ascertain whether the temperature minima represent a survival limit for the north- ward—ranging thermophiles, or whether the maxi— mum yearly temperatures are more important. The minimum yearly temperatures of 11° to 12° C in this area also occur to the north, whereas the maximum yearly temperatures of 14° to 17° C do not, an indi- cation that maximum water temperatures may not be high enough to permit repopulation, thus limiting the northward expansion of thermophilic species. Southward ranging cryophiles are little affected be- cause the thermal maxima in this area (14° to 17 ° C), which they tolerate, extend continuously farther south within the Californian province due to upwell- ing along the coast (figs. 5, 6, 7). The boundary between the Californian and Surian provinces is characterized by the occurrence of range endpoints of northern species and is of lesser magnitude (see fig. 8, where clusters B and C indi- cate closer affinity than do A and B.) At this bound- ary 26 cryophiles end their ranges within 1° of lati- tude while only 6 thermophiles drop out. In fact, 49 southward~ranging species terminate their ranges in the southern Californian subprovince, and they drop out in a stepwise manner southward, an indication that these species are reaching their temperature limits in areas where temperature maxima are dampened by upwelling. They may be reacting to increased temperature and survival limits or to the failure of upwelling to produce temperatures low enough to allow repopulation. Temperature minima enhanced by upwelling occur both north and south of the Californian~Surian boundary at Punta Eugenia—— Bahia S. Vizcaino. Cryophilic species ranging south- ward past this boundary must pass through an ap- parent barrier of shallow, relatively warm waters of the bay to reach the cool, upwelled waters south of Punta Eugenia; they must then be able to tolerate still higher temperatures during thermal maxima in an area where upwelling occurs only seasonally. The northern and southern Surian subprovinces are similar in that their northern boundaries are marked by promontories where seasonal upwelling lowers the temperature markedly; to the south, how- ever, the coasts are embayed and little or no upwell- ing occurs. Because these waters are warmed both by solar radiation and by a warm surface countercur— rent in the summer and fall, the annual range of temperature is greater (as much as 10° C in the areas where seasonal upwelling occurs) than that (3° to 6° C) to the north in the Californian province. Many eryophilic species (62) , chiefly from the Cali- fornian province where waters are cooler, end their ranges in the Surian pro-Vince. They can survive tem- perature minima in the upwelling areas of the Surian province, but temperature maxima limit their southward movement. Perhaps they survive where they do because minimum temperatures are cool enough for repopulation to occur and maximum tem- peratures are not high enough to kill. When tempera- tures remain too high for repopulation or become high enough to be lethal, these species can no longer expand southward. Of the 27 species which end their northern expansion in the Surian province, many are endemic. It is possible that their range records will be extended southward when better collections can be made in this area and in the adjacent Pana- manian province. Formidable thermal barriers limit the northward expansion of species through the Surian province, principally the two large areas of upwelling to the south of Cabo San Lazaro and Punta Eugenia (fig. 5). Maximum temperatures in FOSSIL COLLECTIONS 27 these regions may well be tolerated by southern spe- cies, but minima due to upwelling can be more than 10° C lower than maxima and probably reach the killing temperatures of many species. The Panamanian province is here represented by only a few samples which contain a total of 20 spe- cies, 16 of which do not occur in the Surian province to the north. This suggests that the Surian—Pana- manian boundary represents a major faunal discon- tinuity in ostracode distributions, but more detailed sampling is needed to verify this assumption. Although there are few species endemic to any province, there are also few that range through an entire province, though many range through each subprovince (tables 4 and 5). The nature of their geographic distribution illustrates the response of species to changes in the marine climate within prov- inces as well as at provincial boundaries. Although faunal provinces are equated with climatic zones, a practice which lends a somewhat quantitative aspect to the provincial marine climate, such usage should not be construed to indicate that a province exhibits a particular climate throughout. If this were the case, there would be many more endemic species, and range endpoints of species would occur almost ex- clusively at provincial boundaries, a situation which of course does not occur. It is noted in table 5 that the southern Surian, northern Californian, and southern Oregonian sub- provinces do not contain any endemic species. These subprovinces occupy short segments of coastline and small areas of shelf compared to-other subprovinces, and they might be considered to represent areas of overlap of species chiefly confined to neighboring provinces. However, as shown in the cluster analysis (fig. 8), the subprovinces within each major prov- ince cluster at approximately the same level and therefore appear to constitute entities of equivalent importance. Although the importance of endemic species in defining faunal provinces has been em- phasized by some workers (Ekman, 1953; Holland, 1971), I feel thatfaunal distinctiveness rather than the presence or absence of a certain number of en- demic species should be the criterion for recognition of a faunal province. Indeed, as has just been shown, endemics make up a very small proportion of the faunas of even the larger areas considered by most workers to represent faunal provinces; this situa- tion has also been observed in molluscan studies (J. W. Valentine, 1966, table 2). MARINE PALEOCLIMATES OF SELECTED PLIOCENE AND PLEISTOCENE UNITS OF SOUTHERN CALIFORNIA A knowledge of the Holocene distributional pat- terns of the Ostracoda and determination of the ma- rine climatic conditions which govern those distribu- tions provide a basis for the initiation of paleoeco- logic analyses of fossil ostracode assemblages. Stud- ies of ostracode (paleo)zoogeography will also pro- vide a test for conclusions already drawn from more extensive investigations which have been conducted with the Mollusca and Foraminifera. It is hoped that studies of a heretofore neglected group will con- tribute to an understanding of the factors control- ling provincial biogeography in general. An effort has been made to utilize the information resulting from study of Holocene ostracode zooge— ography to interpret marine climatic conditions dur- ing the late Cenozoic at selected locations along the coast of southern California. The paucity of infor- mation on ostracode fossil assemblages on the west coast constrains the scope of the investigation. The results will therefore be of a preliminary nature. FOSSIL COLLECTIONS Collections were made from formations of Plio- cene to Pleistocene age which occur in four areas along the California coast (fig. 3). These units and their generally accepted ages include: San Diego Formation (late Pliocene age) at San Diego (32°48’ N.) ; San Pedro Formation (early Pleistocene age) and the Palos Verdes Sand (late Pleistocene age) at San Pedro (33°44’ N.); Santa Barbara Formation (early Pleistocene age) at Santa Barbara (34°25’ N.) ; and the Foxen Mudstone (middle to late Plio— cene age) and the Careaga Sandstone (late Pliocene age) of the Santa Maria district (34°54’ N.). The stratigraphy of these units has been adequate- ly described by others and will be only summarized herein. The ages of some of the formations treated here are in dispute, especially with respect to the placement of the Pliocene—Pleistocene boundary in the west coast sequence. Delineation of this boun- dary has been reviewed elsewhere (Woodring, 1952; J. W. Valentine, 1961; Bandy, 1967; Bandy and Wil- coxon, 1970). The west coast fossil ostracode assem- blages are at this time too poorly known to con- tribute to the solution of this problem. Previous paleoecologic interpretations are based mainly on fossil molluscan and foraminiferal assemblages; they are reviewed in a later section. A register of fossil localities is in table 6. 28 ZOOGEOGRAPHY 0F HOLOCENE OSTRACODA OFF WESTERN NORTH AMERICA Tm: 6.——Reyixtc'r of fossil localities M12165—2167: Pacific Beach, San Diego, Calif.; La Jolla 7.5’ quad., 1953. Locality of Hertlein and . Grant, 1944. San Diego Formation (upper Pliocene); approximately 400 ft (120 m) thick; unconformable on Eocene marine sandstones; unconformable under hori- zontal Bay Point Formation (upper Pleistocene). Mf2165: San Diego Formation (upper Pliocene); in a gully in sea cliffs off Crystal Drive where it turns from east-west to north-south; shell bed in yellow-brown marine sands; about 100 ft (30 m) above Eocene—Pliocene contact. Mf2166: San Diego Formation (upper Pliocene); 35 ft (11 m) north of the end of Law Street in sea cliff facing Pacific Ocean; white sandy marl containing Pecten and echi- noid fragments; about 100 ft (30 m) above Mf2165 and approxi- mately in the middle of the expos— ure at Pacific Beach; about 20 ft (6 m) above high tide; from the beach to the sample locality the outcrop is covered by slumped ma- terial and a thick bed of tide-de- posited cobbles. Mf2167: San Diego Formation (upper Pliocene); 4 ft (1 m) above Mf2166. Second St., between Pacific Ave. and Mesa St., San Pedro, Ca1if.; San Pedro 7.5’ quad., 1964. Lomita Marl Member of San Pedro formation (lower Pleistocene) con- sists of approximately 50 ft (15 m) of marls and calcareous sands; Lomita over— ’ lain unconformably by Timms Point Silt Member of San Pedro Formation (lower Pleistocene) which consists of about 80 ft (24 m) of fossiliferous marine silty sands. Horizontal Palos Verdes Sand (upper Pleistocene) unconformably over- lies the Timms Point. Localities 42—44 of Woodring and others, 1946. (See their p. 46 for measured sections.) Mf2168: Lomita Marl Member (lower Pleistocene); fine-grained marl; lowest sample in Lomita (base not exposed); in an alley across from a bakery on north side of Second St. Mf2169: Lomita Marl Member (lower Pleistocene); coarse marl; north side of Second St. in middle of Lomita exposure and approximate- lv 25 ft (8 In) higher than Mf2168. Mf2171: Timms Point Silt Member (lower Pleistocene); shell bed in fine-grained marine silty sands; at Lomita-Timms Point contact on Mf2168, 2169, 2171, 2172: TAIL]: 6—129ng of fossil localities—Continued south side of Second St.; contact burrowed downward into Lomita. Mf2172: Timms Point Silt Member (lower Pleistocene); shell bed 20 ft (6 m) above Mf2171 and 6 ft (2 m) below Palos Verdes Sand. Lomita Quarry, Torrance, Calif. ; Torrance 7.5’ quad., 1964, (not shown on this map; see Woodring and others, 1946, pl. 1; also Galloway and Wissler, 1927). Mf2170: Lomita Marl Member (lower Pleistocene); marl; near top of formation; “upper foram bed” of Galloway and Wissler, 1927. Vacant lot west of Pacific Ave. between Bonita St. and Miner St., San Pedro, Ca1if.; San Pedro 7.5' quad., 1964. Mf21732Palos Verdes Sand (upper Pleistocene); collected in 2—3 ft (1 m) of shelly marine sands. West side of Shoreline Drive, south of Castillo St. and 0.1 mile (163 m) south- west of the municipal pool in Santa Bar- bara, Calif.; Santa Barbara 7.5’ quad., 1952. Santa Barbara Formation (lower Pleistocene) consists of about 70 ft (21 m) of silty bryozoan marls, and uncon- formably overlies the Sespe Formation Mf2170: Mf2173: Mf2174, 2175, 2176: (Oligocene?) . Mf2174: Santa Barbara Formation (lower Pleistocene); silty bryo- zoan marl; approximately 21 ft (6 m) from base of formation. Mf2175: Santa Barbara Formation (lower Pleistocene); silty bryo- zoan marl; approximately 24 ft (7 m) above Mf2074. Mf2176: Santa Barbara Formation tion (lower Pleistocene); silty bryozoan marl; approximately 12 ft (4 m) above Mf2075 and 12 ft (4 m) below top of formation. Southern Pacific RR cut, 0.75 mile (1.2 km) northeast of Shuman, Calif. (Casmalia 7.5’ quad., 1959) where the road from Shuman to Santa Maria crosses RR, at benchmark 370; Guadalupe 7.5' quad., 1959. Locality 170 of Woodring and Bramlette, 1950. (See their p. 38 and p. 44 for measured sections.) Foxen Mudstone (middle to upper Pliocene) con- sists of about 285 ft (87 m) of fine- grained sands containing shell beds; Careaga Sandstone (upper Pliocene) con- sists of about 290 ft (90 m) of medium- to coarse-grained sands containing shell beds. Mf2177: Foxen Mudstone (middle to upper Pliocene); fine sandy shell bed, about 95 ft (29 m) above base of exposure. Mf2177, 2178, 2179: STRATIGRAPHY AND AGE OF PLIOCENE AND PLEISTOCENE UNITS 29 TABLE 6.—Register of fossil localities—Continued Mf2178: Foxen Mudstone (middle to upper Pliocene); fine sandy shell bed about 85 ft (26 m) above Mf2077. Mf2179: Careaga Sandstone, Graciosa Coarse-Grained Member (upper Pliocene); 3- to 4-ft (1-m) sandy shell bed about 170 ft (52 m) above Foxen-Careaga contact. Road cut at Fugler Point; on south side of road from Santa Maria to Garey, Calif.; Twitchell Dam 7.5’ quad, 1959 (middle of north side of sec. 35, T. 10 N., R. 33 W.). Locality 178 of Woodring and Bramlette, 1950. (See their p. 47 for measured section.) Careaga Sandstone (upper Pliocene) consists of about 50 'ft (15 In) of fine-grained marine sands con- taining shell beds, often impregnated with asphalt. Mf2180: Careaga Sandstone, Cebada Fine Grained Member (upper Plio- cene); fine-grained sandy shell bed about 6 ft (2 m) thick near base of exposure; containing many brachio- pods and impregnated with asphalt. All samples except Mf2170 collected in 1971 by Page Valentine and J. W. Valentine, Dept. Geology, Univ. of Cali- fornia, Davis, Calif; Mf2170 was provided by J. H. Lipps of the same department. Mf2180: STRATIGRAPHY AND AGE OF PLIOCENE AND PLEISTOCENE UNITS SAN DIEGO FORMATION The San Diego Formation is exposed in the mesas which extend eastward of San Diego, Calif, and southward past the International Boundary. The formation also crops out north of Mission Bay on Mount Soledad and in sea cliffs at Pacific Beach. Hertlein and Grant (1944) provide a review of earli- er stratigraphic and paleontologic studies of these beds. The gently dipping San Diego Formation lies un- conformably between Eocene marine sandstone be- low and Pliocene conglomerates and horizontal Pleis- tocene marine terrace deposits and nonmarine al- luvium above. The formation varies in thickness from over 1,000 ft (300 m) in the mesas to the east to about 400 ft (120 m) at Pacific Beach; lithologi- cally, the sediments range from fine-grained marine sands containing scattered conglomeratic lenses to a white, sandy marl. The exposures in the north-trend- ing cliffs at Pacific Beach, from which samples for the present study were collected, are the most fos- siliferous. Here yellow-brown marine sands contain- ing shell beds are overlain by white, sandy, some- what indurated marls. These beds lie unconformably on Eocene marine sandstones and are capped by horizontal marine terrace deposits of the Bay Point Formation (upper Pleistocene). According to Hertlein and Grant (1944), W. H. Dall in 1874 first referred these sediments to the Pliocene on the basis of their molluscan fauna; Hert- lein and Grant consider the San Diego Formation to be of middle Pliocene age. Woodring and others (1940, p. 112) indicate (in a correlation chart) a probable late Pliocene age for the San Diego Forma- tion at Pacific Beach, while considering other ex- posures to be early and middle Pliocene; Woodring (1957) still considered the upper part of the San Diego Formation (Pacific Beach) to be of late Plio- cene age. J. W. Valentine (1961) cites vertebrate as well as invertebrate faunal evidence in assigning a tentative age of Pliocene and Pleistocene to the San Diego Formation, and this would indicate a possible Pleistocene age for the Pacific Beach beds. Allison (1964) considers the formation to be probably Pleis- tocene on the basis of vertebrate (horse tooth) evi- dence. For the purposes of this report, the San Diego Formation at Pacific Beach is considered to be late Pliocene in age. Sample Mf2165 of this study was collected from a shell bed in marine sands about 100 ft (30 m) above the Eocene—Pliocene contact at Pacific Beach. Sam- ples Mf2166 1 and Mf2167 were collected from the coarse sandy marls about 100 ft above Mf2165 and somewhat above the middle of the exposure which is approximately 400 ft (120 m) thick at Pacific Beach. SAN PEDRO FORMATION AND PALOS VERDES SAND The San Pedro Formation and the Palos Verdes Sand are exposed in the Palos Verdes Hills, an anti- clinal, coastal fault block which rises to about 1,500 ft (460 m) in altitude on the southwest border of the Los Angeles basin. These units crop out chiefly on the north and east slopes of the Palos Verdes Hills, the best exposures occurring in and around the city of San Pedro. Woodring and others (1946) made an exhaustive study of the geology and paleontology of the area; J. W. Valentine (1961) contributed paleo- ecologic interpretations of the molluscan fauna found in these units. The following summary of the stratigraphy of these two formations has been taken in large part from these two studies. The gently folded San Pedro Formation is 350— 600 ft (105—180 m) thick and includes three mem- 1 Mfzmicrofossil. 30 ZOOGEOGRAPHY 0F HOLOCENE OSTRACODA OFF WESTERN NORTH AMERICA bers which have been assigned to the lower Pleisto- cene: the basal Lomita Marl Member, the middle Timms Point Silt Member, and an unnamed upper member (“San Pedro Sand” of Woodring and oth- ers, 1946). The Lomita Marl Member is composed of 60—70 ft (20 m) of glauconitic sands, marls, and cal- careous sands and lies unconformably on lower Pliocene rocks or Fernando Formation or on the Monterey Shale (Miocene). The age of the Lomita is in dispute. Woodring (1952) assigns it an early Pleistocene age and places the Pliocene-Pleistocene boundary at its base. Obradovich (1968) assigns it an age of 3 million years on the basis of radiometric dating of glauconite, and this would make it Pliocene according to current usage. Fanale and Schaeffer (1965) used helium-uranium ratios to date the Lomita at about 155,000 years or late Pleistocene. Bandy (1967) places it in the lower Pleistocene on the basis of foraminiferal coiling ratios; Bandy and Wilcoxon (1970), however, assign it to the upper Pleistocene on the basis of planktonic foraminiferal and paleomagnetic stratigraphy. Zinsmeister (1970) correlates the upper Pliocene deposits of Newport, Calif, with the Lomita and cites molluscan and fish fauna] evidence for a late Pliocene age of the Lomita. The San Pedro Formation is considered to be of early Pleistocene age in this report. The Timms Point Silt Member is composed of 30— 80 ft (9—24 m) of marine silts and sandy silts con- taining fossiliferous horizons and lies either uncon- formably or in gradational contact with the Lomita Marl Member below and the “San Pedro Sand” above. The Timms Point is believed to be essentially equivalent in age to the Lomita, but the succession of the two formations documents a radical change in the sedimentary environment. The “San Pedro Sand” is composed of about 175 ft (55 m) of sands, often crossbedded, which include fossiliferous horizons and beds of silt and gravel; it is truncated by upper Pleistocene terrace deposits. The Palos Verdes Sand (upper Pleistocene) is a marine terrace deposit consisting of essentially hori- zontal, fossiliferous, coarse sands and gravels, silty sands and silt. It is as much as 15 ft (5 m) thick and lies unconformably on sediments of Miocene to Pleistocene age. Samples used in this study were collected from the Lomita Marl and Timms Point Silt Members (lower Pleistocene), which crop out on 2d Street in San Pedro. The Lomita here consists of about 50 ft (15 m) of marls and calcareous sands; sample Mf2168 was taken at the base of the section, and Mf2169 in the middle. The Timms Point was sampled in a basal shell bed (Mf2171) above the burrowed contact with the Lomita and again 20 ft (6 m) above the contact (Mf2172). A sample (Mf2170) from the Lomita was taken near the top of the formation in the Lomita Quarry (“upper foram bed” of Galloway and Wissler, 1927) on the north side of the Palos Verdes Hills. The “San Pedro Sand” was not sampled. A collection was also made (Mf2173) in a sandy shell bed of the Palos Verdes Sand (upper Pleistocene) which crops out in San Pedro. SANTA BARBARA FORMATION The Santa Barbara Formation is exposed near the coast southwest of Santa Barbara, Calif, and also at Rincon Point to the east of the city. It has a maxi- mum thickness of about 2,000 ft (600 m), is com- posed of marine sand, silt, and clay, and lies uncon- formably on older Tertiary sedimentary rocks (Up- son, 1951). Woodring and others (1946) and Wood- ring (1952, 1957) consider the Santa Barbara For- mation to be correlative with the San Pedro Forma- tion (lower Pleistocene), and J. W. Valentine (1961) indicates that the Santa Barbara Formation and the Lomita Marl Member of the San Pedro For- mation are equivalent in age. Collections were made in the lower part of the Santa Barbara Formation at the Bathhouse Beach exposures which dip gently south and lie unconform- ably on the Sespe Formation (Oligocene?) . The beds are about 7 0 ft (20 m) thick and comprise highly fossiliferous, silty, bryozoan marls. Samples Mf217 4, Mf217 5, and Mf217 6 were collected in sequence from the lower to the upper part of the exposure. FOXEN MUDSTONE AND CAREAGA SANDSTONE The Foxen Mudstone and the Careaga Sandstone belong to a sequence of marine and nonmarine sedi- ments deposited in the late Tertiary Santa Maria basin. These formations are exposed around and within the Santa Maria district, a northwest-trend- ing lowland on the California coast north of Point Conception. The geography and paleontology of units found in the Santa Maria district have been described by Woodring and Bramlette (1950). The Foxen Mudstone is a sequence of fossiliferous mudstones, siltstones, and fine-grained sandstones as much as 800 ft (245 m) thick. It is conformable with the Sisquoc Formation (upper Miocene to middle Pliocene) below and the Careaga Sandstone above. On the basis of its molluscan and foraminiferal as- semblage-s, the Foxen has been assigned an age of middle to late Pliocene. The Careaga Sandstone is STRATIGRAPHY AND AGE OF PLIOCENE AND PLEISTOCENE UNITS 31 composed of two members, a lower fine-grained sandstone of 1,000 ft (300 m) maximum thickness, the Cebada Fine-Grained Member, and an upper coarse-grained sandstone and conglomerate of 425 ft (130 m) maximum thickness, the Graciosa Coarse-Grained Member. The formation lies con- formably under the nonmarine Paso Robles Forma- tion of Pliocene(?) and Pleistocene age. Molluscan fossils indicate the Careaga was deposited in late Pliocene time. Woodring and Bramlette (1950) cor- relate the Careaga with the upper part of the San Diego Formation (exposures at Pacific Beach). Collections for the present study were made in the upper two-thirds of the Foxen Mudstone and in the Careaga Sandstone which crops out in the Casmalia Hills in the western part of the Santa Maria district. Here about 285 ft (85 m) of the Foxen and 290 ft of the Careaga are exposed. Samples Mf2177 and Mf2178 were collected in the Foxen 95 and 180 ft (29 and 55 m) from the bottom of the exposure. Sample Mf217 9 was collected in the Careaga (Graci- osa Coarse-Grained Member) 170 ft (52 m) above the Foxen—Careaga contact. An additional collection (Mf2180) was made in the Careaga (Cebada Fine- Grained Member) at Fugler Point on the northeast- ern edge of the Santa Maria district. COMPARISON OF HOLOCENE AND FOSSIL ASSEMBLAGES FAUNAL COMPOSITION OF FOSSIL UNITS The Holocene fauna treated in this study com- prises 50 genera; of these, 38 also occur in the Plio- cene and Pleistocene units. Six genera occur only as fossils (tables 3, 4). - The 12 Holocene genera which are not represented in the Pliocene and Pleistocene are principally those occurring along the coast north or south of the fossil localities which are situated in the Californian prov- ince from San Diego to just north of Point Conception. Bythoceratina, Costa?, Eucytherura, Hemicythere?, “Loxoconcha,” Oriom‘na, Pterygocy- thereis ?, “Trachyleberis,” two of the three species of Puriana, and one of two species of Cytherella occur only in the Surian or Panamanian province; N eocaudites? occurs south of San Diego; and H emi- cythere occurs only in the Oregonian province. One species in particular, Puriana pacifica, does not oc- cur in fossil assemblages, although it ranges from the Panamanian province north to just south of Point Piedras Blancas at the Californian—Oregonian boundary and is a ubiquitous element of the Holo- cene assemblages in this area. Cativella sp. A is an- other common Holocene species which is not found in the fossil assemblages although it ranges from the northern Surian subprovince to just south of Point Piedras Blancas. The absence of these two species as fossils may be due to collecting bias, but the richness of the fossil deposits, especially the Lomita Marl Member of the San Pedro Formation and the Santa Barbara Formation, make this un- likely. Perhaps these species have migrated north- ward, at least into the shelf areas off San Pedro and Santa Barbara, since the early Pleistocene. The solu- tion of this problem will have to wait until fossil ostracode assemblages along the coast become better known; then it will be feasible to determine move- ments of species along the shelf through time and possibly to identify the origin of new elements of the fauna. Paleoecological interpretations of fossil units com- monly rely on comparisons of living and fossil as- semblages. One must assume, of course, that fossil representatives of a species had environmental toler- ances or limits similar to those of the modern repre- sentatives of that species. Of the 123 species occur— ring in fossil units that are included in this investi- gation, only 19 are not living today; 85 percent of the Pliocene and Pleistocene assemblages under study is extant and forms a broad data base. Table 7 indicates that practically the entire assem- blage of each unit is composed of species living to- day. Older units generally contain greater percent- ages of extinct species, except for the Careaga Sand- stone. This unit yielded the smallest fauna (27 spe- cies) and contains the most extinct species (6). Dis- covery of several of the fossil species in the Holo- cene would markedly increase the percentage of the Careaga fauna having living representatives, and it is probable that several of the “extinct” species exist today but have not yet been found in Holocene samples. CLUSTER ANALYSIS OF HOLOCENE AND FOSSIL SAMPLES The fossil and the Holocene shelf samples were subjected to a cluster analysis similar to that per- formed on the Holocene samples alone. It has been emphasized that faunal assemblages and provinces change character through time in response to evolu- tionary and migratory events and to the influences of climatic variations and therefore cannot be repli- cated, but it still is possible to compare assemblages of differing age and obtain an indication of the dis- tance in time and environment separating them. An affinity to modern faunal provinces will provide a qualitative measure of the marine climate which 32 ZOOGEOGRAPHY OF HOLOCENE OSTRACODA OFF WESTERN NORTH AMERICA TAsLE 7_.—F'aunal diversity, generic and specific, of each fos- Sil umt and the percentage of the fauna, having living rep- resentatives Fossil Per- Formation species Fossil Total centage (age) Samples Genera in Holo- only species of fauna ‘ cene extant Foxen Mudstone Mf2177, 24 35 4 39 90 (middle to late Mf2178 Pliocene). Careaga Sandstone Mf2179, 22 21 6 2'7 '77 (late Pliocene). Mf2180 San Diego Forma- Mf2165, 26 56 4 60 93 tion (late Plio- Mf2166, cene). Mf2167 Lomita Marl Mem- Mf2168, 31 68 6 73 93 her of San Pedro Mf 2169, Formation (early Mf 2170 Pleistocene) . . Timms Point Silt Mf2171, 20 28 2 30 93 Member of San Mf2172 Pedro Formation (early Pleisto- cene). Santa Barbara Mf2174, 25 48 5 53 91 Formation (early Mf2175, ’ Pleistocene). Mf217 6 Palos Verdes Sand Mf2173 20 39 0 39 100 (late Pleisto- , cene) . prevailed during deposition of the fossil unit. This analysis was based on 16 fossil samples and 105 Holocene 15-minute samples (006—110). The 15- minute samples located in the Panamanian province, and the 15 species which occur almost exclusively in these samples, were deleted from the analysis to ac- commodate 19 species that occur only as fossils. The deleted samples are clearly different from the other Holocene samples (fig. 8). The deletion of the 15 Holocene species allowed the data matrix to remain sufficiently small, and it was felt that inclusion of these species would not have altered the results of the cluster analysis, as the fossil samples are obvi- ously most similar to samples occurring north of the Panamanian province. (See table 3 for alphabetical list of ostracode species included in cluster analyses). The results of the analysis are illustrated in the dendrogram of figure 10. The fossil samples are most similar to samples from the southern Califor- nian subprovince, the Californian province, and the combined Californian—Oregonian provinces, in that order. The assemblages from south of Point Concep- tion (San Diego Formation, Lomita Marl and Timms Point Silt Members of San Pedro Formation, Palos Verdes Sand, and Santa Barbara Formation) indi- cate a similarity to those of the Californian province, whereas assemblages from the Santa Maria district, to the north, (Foxen Mudstone and Careaga Sand- stone) show a relationship to both the Californian and Oregonian provinces. This cluster pattern cor- roborates other evidence, which will be discussed later, regarding the marine temperatures prevailing during deposition of these units. It is doubtful that age plays a significant role in determining the affinities between fossil and H010- cene assemblages, as most species have living repre- sentatives. Age is a factor only insofar as it is ex- pressed in altered climatic conditions and corres- ponding changes in the composition of assemblages. PALEOTEMPERATURE DETERMINATIONS BASED ON ZOOGEOGRAPHIC EVIDENCE The determination of yearly marine paleotemper- ature ranges is based on interpretation of tempera- ture ranges of Holocene species that occur in fossil units. As a first approximation, one might simply plot these ranges and infer that the area of overlap of temperature ranges of cryophiles versus thermo- philes represented the possible maximum yearly temperature range and the endpoints the maximum and minimum yearly temperatures. This procedure is based on the assumption that the temperature tol- erances determined for Holocene species represent the maximum and minimum temperatures at which they can survive. However, Hutchins (1947) has pointed out that the distributions of marine organ- isms are limited by temperature in at least two ways. Survival temperatures (minimum and maximum) encompass the widest temperature range within which a species may live. A narrower range is bounded by the maximum and minimum tempera- tures required by the organism for repopulation, which includes reproduction and larval development. With regard to the geographic distribution of a species, the minimum survival temperature would occur poleward during the winter, and the minimum temperature required for repopulation would occur poleward in the summer (that is, the water must warm up enough to allow repopulation to occur). The maXimum survival temperature occurs equator- ward during the summer, and the maximum tem- perature required for repopulation occurs equator- ward in the winter (that is, the water must be cool enough to allow repopulation to occur). As each spe- cies may be limited at each end of its geographic range by only one of the two types of temperature limits, four types of species distribution or zonation were established by Hutchins (1947). Equatorward temperature limits summer survival winter repopulation summer survival winter repopulation Poleward temperature limits 1 winter survival 2 summer repopulation 3. summer repopulation 4 winter survival Zonal type PALEOTEMPERATURE’ DETERMINATIONS OTSUKA COEFFICIENT OF SIMILARITY 50 40 ' 15-MINUTE SAMPLE NUMBERS u: M12169 E Palos Verdes Sandxfizf 32% 3 . . San Diego { . . MI“ 166 6 Tlmms Pomt SI": Formatlon Mf 167 0 Member of Sar M: 171 “ Pedro Formation Santa Barbara m ‘74 .. «.77 m 0 IL Pt. Piedras Blancas —> 2?, loo 1) 80 70 GP I I I Cape Flattery —> 11%; 48.24, I“ 105 104 103 102 101 xs§ 97 288 93 92 '9 3 Northern subprovince .6 Salt Point —> 2% 3834' 67 ‘2 Southern subprovince 35°40’ 22 Pt Conception 23 Northern subprovince 34°30' —’ 52 N. Channei Is. -—> 51 34°00' 33 Californian B province 43 42 ' 39 Southern 37 subprovince Lomita Marl 29 Member of San { . f2168 Pedro Formation I) 76 Oregonian province A 20 Careaga Sandstone Member of San Pedro Formation Formation m 175 Foxen Mudstone / 52 Bahia swam/9% 28°l 5' 22 Punta Eugenia g}, :——‘ 27-52 19 l 18 17 1 Northern subprovince 10 Cabo San Lazaro -—> 24°48' Cabo San Lucas —> 22°52' } 15 1 4 13 12 11 10 9 B 7 6—.— 09 :l—— 10 I I I I 100 90 80 70 Surian province C Southern su bprovi nce I I I I I I 60 50 40 30 20 10 0 FIGURE 10.—Dendrogram of Holocene and fossil sample clusters. Samples are compared on the basis of their ostracode spe- cies composition using the Otsuka similarity coefficient (0/ VN1N2) X 100; clustering was by the unweighted pair-group method. 34 ZOOGEOGRAPHY OF HOLOCENE OSTRACODA OFF WESTERN NORTH AMERICA That species can be limited in their distribution by temperature tolerances other than maximum and minimum survival temperatures introduces com- plexity into the interpretation of paleotemperature. Inferences are even more uncertain if the type of zonation rep-resented by any given species cannot be determined. Hutchins (1947) shows that zonal type may be determined for a species that occurs along more than one coast by comparing winter and sum- mer temperature minima and maxima at its range endpoints. Hazel (1970) has successfully ascertained the zonal type for many amphiatlantic ostracode species. Where the types of zonation represented by spe- cies cannot be determined, one does not know wheth— er the temperatures controlling species distribution are survival limits or repopulation limits or both. Both cryophiles and thermophiles may occur in the same fossil deposit, and When the temperature ranges of living representatives of these species are compared, they usually overlap. The range of over- lap and the bounds of this range are of interest in determining the possible maximum and minimum paleotemperatures which occurred during deposition of the fossil unit. There are four kinds of tempera- ture limit involved where thermophiles and cryo- philes approach each other and overlap in tempera- ture tolerances: for the cryophile there are two tem- perature limits, summer survival and winter repopu- lation; for the thermophile there are two possible limits, winter survival and summer repopulation. Although we cannot always identify the nature of these temperature limits, it is still of interest to know the possibilities that the limits represent sur- vival maxima and minima and therefore convey the temperature limits that could have occurred at a fossil locality. In the following model, the temperature ranges of a pair of species, one thermophile and one cryophile, are allowed to approach each other and overlap. The possible interactions of the temperature limitations are illustrated in figure 11. Thirteen configurations occur, 7 of which are impossible owing to constraints applied by the survival and the repopulation tem- perature tolerances of one or both of the two species. For example, two species can occur together only if the repopulation temperature limits are equal to or lie Within the survival tolerances. Obviously, two species cannot coexist if the temperature needed for repopulation by one lies above the maximum or be- low the minimum survival temperatures of the other species. For each of the six configurations in which a cryo- phile and a thermophile can coexist, there are four subcases. The first subcase involves the situation where the cryophile is limited by a summer survival temperature and the thermophile is limited by a winter survival temperature (fig. 11, configuration 11A). This temperature overlap then incorporates a possible maximum and minimum yearly temperature which would be defined if this situation could be identified in a fossil unit. The other three subcases involve two situations in which one of the species is limited by survival temperatures and the other by repopulation temperatures, and a third situation in which both are limited by repopulation temperatures (fig. 11, configurations 11B, C, D). Of the 24 possible subcases, 4 involve temperature range disj unctions. In practice, when the temperature ranges of a cryophile and a thermophile are being compared, the overlapping temperature range endpoints represent, respectively, the maximum summer temperature and the minimum winter temperature they experience in their geographic distribution. Now, we do not know whether these temperatures actually represent sum- mer survival maxima and winter survival minima or whether the particular species are limited by re- population temperatures inside these limits. If all temperature limits in 15 theoretical subcases are counted (excluding four subcases of disjunct tem- perature ranges, and five subcases of coincident tem- perature endpoints), 10 of the 15 possible upper temperature limits are real survival maxima and 5 are repopulation limits; all these are ‘upper limits exhibited by cryophiles (that is, the real survival maxima are summer survival temperatures of cryo- philes). And 10 of the 15 possible lower temperature limits are real winter survival minima of thermo- philes. In this model, therefore, 67 percent of the maxima and minima represented by overlapping temperature ranges of pairs of cryophiles and ther- mophiles are survival temperatures and should out- line the possible yearly maxima and minima. When subcases involving disjunct temperature ranges are considered, there are four minima and four maxima, and two of each are real survival maxi- ma and minima. When temperature range endpoints of a cryophile and a thermophile are coincident (five subcases) , 40 percent always represent survival maxima of the cryophile, 40 percent represent sur- vival minima of the thermophile, and 20 percent represent repopulation boundaries. Their relation- ship to pairs of overlapping temperature ranges should indicate which they are. PALE‘OTEMPERATURE DETERMINATIONS 35 TEMPERATURE CONFIGURATION COOL WARM w 1 TEMPERATURE LIMITS * CRYOPHILES 0 Summer survival 47" 4‘) D Winter repopulation 2 THERM- . Winter survival OPHILE I Summer repopulation 3 I3 3 Survival temperature .——-——'_ limiting uJ . Repopulation temperature g a O "' limiting 4 3 g I I E w 5 H w 6 r 1 ——El—O 7 I I ——{3—0 8 t L w 9 k I ‘3 k I .——I— 12 ———El-—O D —El-------o e I emul— ———-EI-——O l 3 o I FIGURE 11.—Interactions of temperature limits (survival and repopulation) of a cryophile and a thermophile. Configurations 1—7 cannot occur because of constraints applied by temperature limits of one or both species. Configuration 11 is expanded to include four subcases. See text for further explanation. 36 ZOOGEOGRAPHY OF HOLOCENE OSTRACODA OFF WESTERN NORTH AMERICA The purpose of all this discussion is to establish a basis for the interpretation of temperature range overlaps of Holocene species occurring together in fossil deposits. (For examples, see figs. 12, 14—19). It may be noted that the species do not have to actu- ally oecur together in the Holocene——and depending on the configuration of the marine isotherms they may not—but theoretically they must be able to co- exist in some thermal regime. The temperature ranges of species occurring as fossils usually over— lap each other in a stepwise manner, each step in- cluding many species. It would seem from the fore- going discussion that about two-thirds of the tem- perature range endpoints of cryophil-es could repre— sent actual summer survival maxima, and that a similar percentage of the temperature range end- points of thermophiles could represent actual winter survival minima, thus outlining the yearly tempera- ture maxima and minima that could occur at that locality and still allow the species to coexist. If two steps of temperature range endpoints, each for cryo- philes and thermophiles, are allowed to delineate the temperature range, there would be a good chance of including actual maxima and minima. In this case, the possible yearly minimum and maximum paleo- temperatures would bound a wider temperature range than might actually have occurred. An important constraint must be placed on this technique. We have assumed until now that the oc- currence of each of the species zonal types is equally probable. This may not be a valid assumption in all situations. For example, northward-ranging thermo- philes may be limit-ed by a summer temperature which has to reach a certain value to allow repopu- lation to occur. In this case many of the geographic range endpoints of thermophiles would be attribut- able to summer repopulaticn limits and not winter survival minima. Consequently, the inferred possi- ble yearly minimum paleotemperature would be higher and the possible yearly paleotemperature range narrower than if there were random occur— rence of species zonal types. Of course, if the zona- tion type of certain species can be ascertained, the above procedure can be simplified and more confidence placed in the inferred yearly paleo- temperatures. The paleotemperatures of southern California fos- sil assemblages, which are discussed in the subse- quent section, were determined using the method outlined above. PALEOCLIMATIC IMPLICATIONS OF FOSSIL ASSEMBLAGES FROM SOUTHERN CALIFORNIA A common assumption in paleoclimatic analyses of fossil assemblages is that fossil representatives of a species had environmental tolerances similar to those of the living representatives of the species. With progressively older faunas, fewer species or even genera are living, and confidence in analyses based only on this assumption rap-idly decreases. Re- cent zoogeographic studies dealing with marine paleotemperatures of the eastern Pacific which re- view earlier work and provide new interpretations include reports by J. W. Valentine (1961), J. W. Valentine and Meade (1961), Addicott (1966), and Kern (1973). The following discussion incorporates information from the literature (mainly molluscan and foramini- feral studies) with new evidence, principally regard- ing marine temperatures, derived from investiga- tions on the Ostracoda. The temperature tolerances of Holocene ostracode species were developed by comparing their distribution to marine temperature maps (figs. 6, 7). Two sets of maps were used, one based on —10—m temperatures along the coasts of Baja California and California to San Francisco (Lynn, 1967) and the other based on surface tem- peratures from San Francisco to Cape Flattery (Robinson, 1957). The data presented by Lynn are considered to be the more accurate in reflecting bot- tom temperatures, and fortunately his maps cover the areas most critical to this report. The tempera- ture tolerances of all ostracode species ranging through at least 2° of latitude were used in the paleo- temperature determinations, and the number of spe- cies employed varied from 19 to 55 depending on the fossil unit involved. Species from all samples of a particular formation were combined in determining the paleotemperature range of that unit (determina- tions based on separate samples gave similar re- sults). Depth ranges of ostracode species in the study area are not yet known, so that depth determi- nations are based on previous interpretations of other organisms. SAN DIEGO FORMATION The San Diego Formation at Pacific Beach, of late Pliocene age, is inferred to represent a depositional environment no deeper than 300 m (Hertlein and Grant, 1944). The samples from Pacific Beach are indicative of two different environments of deposi- tion. Sample Mf2165 was collected in terrigenous sands containing scattered shell beds typical of an PALEOCLIMATIC IMPLICATIONS OF FOSSIL ASSEMBLAGES 37 open shelf environment. Samples Mf2166 and Mf2167, on the other hand, were collected in coarse sandy marls indicative of an environment receiving little inorganic debris from rivers and coastal ero- sion processes. Highly calcareous sediments are known to be presently accumulating off the eastern (leeward) shore of Santa Catalina Island (Shephard and Wrath, 1937), while the windward shelf ex- hibits chiefly terrigenous sediments. The three sam- ples from the San Diego Formation cluster together (fig. 10) ; the assemblages from the calcareous sedi- ments show more affinity for each other and may reflect habitat similarity. Together these assem- blages cluster with the combined subprovinces of the Californian province, in spite of the fact that the San Diego Formation lies on the coast well within the limits of the southern California subprovince. A comparison of the present temperature toler— ances of 46 ostracode species from the San Diego Formation (figs. 12, 13) shows that the marine tem- peratures could have ranged from 14° to 18° C, no different from the marine temperatures off that part of the coast today (figs. 6, 7). The cluster analysis shows, however, that the Pliocene and modern as- semblages at that latitude are different, probably the result of a change in the overall climatic regime not apparent in the marine temperatures at that point, but reflected in the range of species respond— ing to the change. FOXEN MUDSTONE AND CAREAGA SANDSTONE In the Santa Maria district, the Foxen Mudstone (middle to upper Pliocene) contains molluscan and foraminiferal assemblages which suggest deposition in an inner sublittoral marine environment (50 m maximum depth) ; molluscan evidence indicates the Careaga Sandstone (upper Pliocene) was deposited in shallow water (to about 30—60 m) (Woodring and Bramlette, 1950). On the basis of inferences from foraminifers, Natland (1957) also regards both units to represent inner sublittoral depths. The os- tracode collections are from fine sands and sandy shell beds suggestive of an open shelf environment. The Foxen and Careaga samples cluster together but at relatively low values, possibly indicating somewhat different habitats. These samples cluster at a low level with the Californian province; in fact they show only slightly more affinity for the Cali- fornian province than do the Californian and Ore- gonian provinces for each other (fig. 10). This low correlation is clearly due to the presence of northern elements in these fossil assemblages. The Santa Maria district lies on the coast within the modern northern Californian subprovince. The water temperatures along this part of the coast an- nually range from 12° to 15° C. Natland (1957) in- ferred, on the basis of foraminiferal assemblages, temperatures ranging from 8° to 13° C for the upper part of the Foxen Mudstone, and from 13° to 15° C for the Careaga Sandstone. Thirty-two ostracode species from the Foxen (figs. 13, 14) and 19 species from the Careaga (figs. 13, 15) indicate the yearly temperatures could have ranged from 12° to 15° C. These results agree with those of Natland (1957) for the Careaga and are identical with present marine temperatures off that coast (figs. 6, 7) . LOMITA MARL AND TIMMS POINT SILT MEMBERS OF THE SAN PEDRO FORMATION The Lomita Marl and Timms Point Silt Members of the San Pedro Formation at San Pedro are of early Pleistocene age. Though closely associated in time, they represent two radically different sedi- mentary environments which are similar to those of the San Diego Formation at Pacific Beach. Studies of the molluscan fauna by Woodring and others (1946) and J. W. Valentine (1961) indicate shoal- ing towards the top of the Lomita; the lower beds of the Lomita represent the outer sublittoral (deep- er than 100 m) whereas the upper beds generally represent depths of 50—100 m. Molluscan evidence indicates that the Lomita was deposited on the lee side of an offshore island in a sedimentary environ- ment notably low in terrigenous detritus; the water temperature at the time of deposition was similar to that in the region today, except that the outer sub- littoral was probably cooler. The mollusks of the Timms Point Silt Member represent an outer sublittoral association (100—200 m), equivalent in depth to the lower part of the Lomita. The sedimentary regime was radically dif- ferent, however, since the Timms Point, in contrast to the Lomita, is composed almost entirely of terrig- enous sediments. These sediments resemble the silts and sandy silts which are at present derived from the Coast Ranges and being deposited on large areas of the continental shelf. The molluscan fauna repre- sents a marine climate somewhat cooler than that of the region today. The ostracode genus Perissocythem'dea occurs as a minor constituent of the Lomita and Timms Point assemblages. This genus lives in shallow, brackish to normal marine environments. During this study it 38 ZOOGEOGRAPHY OF HOLOCENE OSTRACO‘DA OFF WESTERN NORTH AMERICA SPECIES ”136 -018 - 095 -109 ' 057 '103 >107 >086 -116 ~142 - 029 -143 -035 -02O I-006 -074 -113 '141 '073 -094 .123 ”099 '158 >110 -113 *093 I'126 -008 '122 I'004 I-lll - 002 -116 '146 I-173 ~129 -120 '162 '064 '066 - 062 - O70 ~137 -091 -154 '101 “C 28-29 - 27—28 - 26-27 - 25—26 - .[ ' [ 24—25 - [ t . 23—24 - t r F. 22—23 - 21—22 - ,..[ [ 20—21 - l8~19 17—18 16-17 15—16 14—15 13—14 12—13 11—12 10—11 5—6 - FIGURE 12.—Modern temperature tolerances for 46 cene); inferred possible yearly minimum and SAN DIEGO FORMATION ostracode species that occur in the San Diego Formation (upper Plio- maximum paleotemperatures, to the nearest 1°C are 14° and 18°, as indicated by the patterning. See table 4 for numerical list of ostracode species. was found to occur only in samples from shallow depths (usually 20 m or less) seaward of harbor en- trances. Although Perissocytheridea may be a trans- ported element in the fossil assemblages, it indicates proximity to a bay or estuary. Molluscan studies show that the assemblages from the Lomita and Timms Point are not equivalent. The cluster analysis (fig. 10) based on ostracode sam- ples also indicates that the assemblages are differ- ent; the Lomita samples cluster with the southern Californian subprovince, whereas the Timms Point sample shows more affinity to samples from the San- ta Barbara Formation which, in turn, cluster with the Californian province. Sample Mf2172 from the PALEOCLIMATIC IMPLICATIONS OF FOSSIL ASSEMBLAGES 39 _ . Yearly range of inner sublittoral bottom . _ Time Faunal (sub) provunce Latitude temperature to nearest 1° C Marine climate "0"“ 5 10 15 20 25 30 _________? 48°24, '_'_"_.'_'.2 _'_'L.-'_‘_' _'.L.' .. ______ N %/ ////////» . OREGONIAN - 38°30’ '— 7 Mild temperate s ////////% 35°30’ l: o , _ W/fl Warm CALIFORNIAN 34 00 7 temperate Holocene S // ///////// A 28°15’ SURIAN N 24°45' V ////////////////////// - — , Subtropical s W ////////// ///% 22°45’ _ PANAMANlAN % //////// /////22 Tropical Fossil unit Holocene y//// Late Pliocene Careaga Sandstone 34°54’ y//// Warm . / temperate Migdifcgongate Foxen Mudstone %/ Holocene 34°25, W A Warm Early Pleistocene Santa Barbara Formation W /% temperate Holocene 7//////% Late Pleistocene Palos Verdes Sand % /////A Warm . 33°44’ _ San Pedro Formation 7 temperate 53'” P'e'smcene Timms Point Silt Member // % Early Pleistocene Lomita Marl Member y/ /% Holocene 32°48’ 7 /////// warm Late Pliocene San Diego Formation 7//////, temperate | 1 I I l I I I I I I I I I I I I I I I FIGURE 13.—Year1y marine temperature to the nearest 1°C occurring in the ostracode faunal provinces of the east- tern Pacific shelf; and possible yearly temperature ranges in fossil units of southern California inferred from ostracode species distributions. Timms Point contains a very small assemblage and, consequently, shows little similarity to other fossil and Holocene samples; it is disregard-ed in paleoeco- logical interpretations. The dissimilarity between the Lomita and Timms Point samples may be due to differences in habitat represented by the two units. A determination of temperature tolerances of 19 ostracode species of the Timms Point indicates that the yearly temperatures could have ranged from 13° to 18° C (figs. 13, 16), identical to water tempera- tures determined for the Lomita (figs. 13, 17; based on 55 species) and very similar to those prevailing off the coast today (figs. 6, 7). J. W. Valentine and Meade (1961, table 12) deter- mined paleotemperatures for these two units at San Pedro based on zoogeographic as well as oxygen iso- tope temperature techniques. The zoogeographic evi- dence yields paleotemperatures ranging from 11° to 40 ZOOGEOGRAPHY OF HOLOCENE OSTRACODA OFF WESTERN NORTH AMERICA SPECIES £§8§2§§§ageageweg§§§§§§§8=ssggea °C H n T n '1‘ I 1 I n I l 1 l 7 7 T n 1 I l l n n 1 T T T T n n . A 28—29 h . ' FOXEN MUDSTONE 27—28 I . I 26-27 ' ' 25—26 ' ' 24-25 - T .r W - 23—24 - 'F 'F 22—23 - W . 21—22 - -- qr . 20—21 19—20 18—19 17—18 16-17 15-16 . 14—15 ' 13—14 12—13 11—12 10—11 FIGURE 14.—Modern temperature tolerances for 32 ostracode species that occur in the Foxen Mudstone (middle to upper Plio- cene); inferred possible yearly minimum and maximum paleotemperatures, to the nearest 1° C, are 12° and 15° C, as 1n- dicated by the patterning. See table 4 for numerical list of ostracode species. PALEOCLIMATIC IMPLICATIONS OF FOSSIL ASSEMBLAGEES 41 SPECIES '\ (I) N H M a.) as m m 0 ‘1' 0 H o 7‘ h o 05 .C a 2: z z :2 2 s 52:: : 8 :5 e 2 s: z 2 § : l l l n I l I l I I 1 1 l I l l l I 28-29 - . CAREAGA SANDSTONE 27—28 - - 26—27 - . 25-26 ' 1 24-25 .1 ll .1 23—24 - l. . 22-23 F 1 21-22 - i 1' . 20—21 — 1 .1 - d. 1 1 < 19-20 - - 18—19 F l W .. 17-18 - l l - 16-17 15—16 14-15 ' 13-14 12—13 11—12 10—11 9—10 - - 8-9 - - 7—8 - . ' . I O | I I | I I . I 6-7 - - L v . FIGURE 15.—Modern temperature tolerances for 19 ostracode species that occur in the Careaga Sandstone (upper Plio- cene); inferred possible yearly minimum and maximum paleotemperatures, to the nearest 1° C, are 12° and 15° C, as indicated by the patterning. See table 4 for numerical list of ostracode species. SPECIES Ix!) ' v-va omRSBSRSL‘S‘ZESQSQ‘REwN . HOOF‘HHo-OOHI-lHOP-IHI-‘HOOH C llllllllllllllllll 28—29 - u TIMMS POINT SILT 27-28 - MEMBER OF SAN PEDRO FORMATION d 26-27 - - 25—26 - - 24-25 - 1.1 ‘ . 23—24 - 1. . 22-23 - . 21-22 20-21 19-20 18—19 [7-18 16-17 15—16 14-15 [3—14 l2—13 = 11-12 0—11 9-10 - . h - 8-9 - - 7'8 ' I I I ' 6-7 .| ' I. 5-6L - FIGURE 16.—Modern temperature tolerances for 19 ostracode species that occur in the Timms Point Silt Member of the San Pedro Formation (lower Pleistocene); inferred possible yearly minimum and maximum paleotemperatures, to the nearest 1° C, are 13° and 18° C, as indicated by the pat- terning. See table 4 for numerical list of ostracode species 42 ZOOGEOGRAPHY OF HOLOCENE OSTRACODA OFF WESTERN NORTH AMERICA SPECIES 136 -010 ~028 ’095 '026 ~109 ~098 -057 >038 >107 '103 '029 '049 -142 -143 '035 -020 '006 '074 '083 -067 -O73 -175 h094 -123 -099 ~182 -008 -058 '069 >171 '110 '113 '126 -122 '004 '002 -111 '163 “146 '176 '120 -173 '121 “184 '070 ~062 ~135 '137 *009 '068 '090 '064 '154 '127 °C 28—29 h 27—28 - 26—27 - 25-26 ~ [ [ - [ 24—25 - [ [ 23—24 - - 22—23 - i 21-22 - [ [ 20—21 19—20 - 18—19 17—18 16—17- 15-16 14—15 - 13—14 12—133 11—12- 10—11- 5—6 - LOMITA MARL MEMBER OF SAN PEDRO FORMATION FIGURE 17.——-Modern temperature tolerances for 55 ostracode species that occur in the Lomita Marl Member of the San Pedro Formation (lower Pleistocene) nearest 1° C, are 13° and 18° 21° C for the Lomita. The molluscan temperature tolerances were determined by these authors from marine temperature data that have since been re- vised. Regarding the Lomita, their 21° C tempera- ture was believed to be the minimum temperature experienced by Panamanian forms (which occur in the Lomita) at Cabo San Lucas. According to the latest data (Lynn, 1967 ; fig. 6), this temperature should be revised to 18° C; the paleotemperature range of 11° to 18° C is more in line with the results of the present study (13° to 18° C). Isotopic tem- peratures from the Lomita fossils range from at least 132° to 19.0° C, also in agreement with paleo- temperatures based on ostracode tolerances. Paleo- ; inferred possible yearly minimum and maximum paleotemperatures, to the C, as indicated by the patterning. See table 4 for numerical list of ostracode species. temperatures for the Timms Point (J. W. Valentine and Meade, 1961) based on the molluscan distribu- tions range from 11° to 14° C and based on isotopic analyses from 5.7° to 13.8° C. These temperatures are below the range of 13° to 18° C determined from ostracode distributions and may reflect the deeper environment of deposition indicated by the mollus- can assemblages of the Timms Point. SANTA BARBARA FORMATION The Santa Barbara Formation is correlative with the Lomita Marl Member (lower Pleistocene) ; it is richly fossiliferous, especially in the lower part, and the molluscan fauna has been partially documented. DISCUSSION 43 (See references in J. W. Valentine, 1961; Woodring and others, 1946.) These molluscan assemblages are interpreted to represent inner sublittoral depths in the lower part of the unit, whereas the fauna of the upper beds is indicative of somewhat deeper condi- tions. Samples for this study were collected in the bryozoan marls in the lower part of the formation at the Bathhouse Beach locality. J. W. Valentine (1961) interpreted the molluscan assemblages here to be indicative of depths between 30 and 70 m. Bull- ivant (1969) examined the same beds, and on the basis of their faunal and lithological character inter- preted them to represent an offshore bioherm de- posited at 40—60 m in a marine climate similar to or perhaps a little cooler than that found offshore today. The Santa Barbara samples cluster with those of the Californian province at a low level but not as low as the Foxen and Careaga samples (fig. 10), which have a more northern aspect. Paleotempera— tures of the Santa Barbara Formation (figs. 13, 18; based on 36 ostracode species) could have ranged from 13° to 18° C, similar to water temperatures found off that coast today (figs. 6, 7). PALOS VERDES SAND The Palos Verdes Sand at San Pedro is of late Pleistocene age. The molluscan fauna is indicative of an inner sublittoral habitat. The ostracode as- semblage is distinctive, as can be seen in the cluster analysis (fig. 10) ; the sample clusters with no other fossil sample and shows similarity to samples of the Californian province. Temperature tolerances of 31 ostracode species of the Palos Verdes Sand indicate that the yearly temperatures could have ranged from 13° to 18° C (figs. 13, 19) , equivalent to the present annual temperature range off that coast (figs. 6, 7). J. W. Valentine and Meade (1961) have also cal- culated paleotemperatures for this unit. The mollus- can temperature tolerances indicate a range from 14° to 21° C, and the isotopic analyses give a tem— perature range from at least 12.3° to 182° C. If, as explained during discussion of the Lomita Marl Member above, the 21° C temperature is lowered to 18° C to reflect the latest marine temperature stud— ies, then Valentine and Meade’s determinations agree well with those based on ostracode tempera- ture tolerances. DISCUSSION The marine climates prevailing during deposition of the Pliocene and Pleistocene units treated here were probably not very different from those existing today off those sites. That the Holocene and fossil assemblages in those areas do not show a strong re- semblance, although most species occurring as fos- sils have living representatives, leads one to believe that the marine climate off southern California has indeed changed, even though the change is not easily detectable in the paleotemperature determinations for particular localities. This situation is under- standable, even expected, in view of the complex sys- tems involved as well as the fragmentary nature of the climatic record. The basic patterns of faunal distribution existing on the shelf can be attributed to the interaction of two very complex systems. One is the marine cli- matic system, simplistically expressed in the con- figuration of marine isotherms; the other is a system of ecological units ranging in organizational com- plexity from the individual organism to the provin- cial assemblage. Both systems alter their character very gradually. The marine climate is, in fact, a highly buffered sys- tem, and directional changes in the properties and effects of atmospheric wind systems, oceanic current patterns, solar radiation, and other major and minor environmental determinants are required to modify the climatic regime. Modifications that are of inter— est here would include alteration of annual water temperature ranges and the appearance of new iso- thermic configurations. Isotherms probably would not merely shift position along the coast in concert; movement might take place in some areas and not in others, and bunching and dispersion of isotherms would also occur, producing new thermal gradients and dissipating others. The magnitude and duration of thermal barriers might change. Climatic altera- tion would not necessarily have to be great to influ- ence considerably the distribution of marine organisms. Individual organisms, populations, communities, and provincial assemblages together form another highly complex system, each ecological unit or com- ponent of the system responding somewhat different- ly to environmental stimuli (J. W. Valentine, 1968, 1973). A change in the marine climate will elicit a different degree of response at each level of organi— zation, the less inclusive ecological units generally reacting more rapidly. Ultimately, a rearrangement or reorganization of these ecological units will occur. Although ecological systems are capable of with- standing short-term climatic changes, an enduring modification of the marine climate, however minor, should result in the appearance of new faunal as- sociations and distributions. 44 ZOOGEOGRAPHY OF HOLO‘CENE OSTRACODA O-FF WESTERN NORTH AMERICA SPECIES up In In I\ I\ as m m m to no 01 -« so 0 m N N .— v m to o N m m 0 ID 1' I\ v H I\ N 01 , messeszizeeegzszzs 28223888883328288 C I 1 l 1 1 1 1 I 1 1 l 1 I 1 1 I 1 I 1 1 I I l 1 I I I I 1 I l l 1 I I 28—29 " _ - I ' [I SANTA BARBARA FORMATION 27-28 - _ 26-27 " 4 25-26 - 'F _ 24-25 - - 'F 1r -1 23-24 - «r - . 22—23 - . 21—22 - .J . 20—21 - I I . . - . - . q . - . . M'I'IMHIM 19—20 - 18—19 17—18 16—17 15-16 14-15 13—14 12-13 11-12 10—11 - J. _ 9-10 - J. J. J. - 8—9 - - - I I I I 7—8 - ' I I I . I I I I I I I I I 6-7 - I I I I _ 5—6 - d FIGURE 18,—Modern temperature tolerances for 36 ostracode species that occur in the Santa Barbara Formation (lower Pleis- tocene); inferred possible yearly minimum and maximum paleotempera’cures, to the nearest 1° C, are 13° and 18° C, as indicated by the patterning. See table 4 for numerical list of ostracode species. REFERENCES CITED 45 SPECIES -136 ”022 >028 -010 -095 -026 '109 ~103 -107 -086 '139 >029 -020 -074 '083 ~141 '158 -123 >113 -126 -110 '093 >008 .122 b111 -163 '176 -120 L121 '070 '009 °C 28-29 ' 2%Q8- PALOS VERDES SAND 26—27 - 25-26 - 1 1 24-25 - ' 1 23-24 . l 1 . 22-23 - 1 21—22 - 20—21- 19—20 18—19 17—18 16—17- 15—16 - 14-15 13—14 12-13 11—12 10—11- 9—10 8—9 - . 7-8 ' | ‘ 5—7 ' I . 5—6 - . FIGURE 19.—Modern temperature tolerances for 31 ostracode species that occur in the Palos Verdes Sand (upper Pleisto- cene); inferred possible yearly minimum and maximum paleotemperatures, to the nearest 1" C, are 13° and 18° C, as indicated by the patterning. See table 4 for numerical list of ostracode species. It would appear fro-m this study that marine cli- mates in the study area have been fluctuating during the past several million years within a narrow spec- trum of configurations which, while influencing the distribution of organisms, remain difficult to discern from each other and from the present climate. We know, however, that at times the marine climate fluctuates far enough in one direction to produce more obvious effects in the fossil record. For ex- ample, during the glacial advances of the Pleisto- cene, thermal barriers and provincial boundaries were relocated and shifted far enough along the coast so that marked differences between present and past climates are discernible. But glacial age deposits were laid down at lower sea levels, and they are generally accessible today only through drilling or in uplifted sections of the coast. The patchiness and limited exposure of the fossil record, then, places a great constraint on climatic reconstructions. If a number of geographically dispersed fossil assem- blages of equivalent age could be studied, it would be possible, by using the present as a model, to re- construct an ancient climatic regime over a large shelf area. Studies of this nature necessarily demand many well-preserved fossil assemblages and are thus confined to the recent past. Only two such investiga- tions have been conducted in the eastern Pacific (J. W. Valentine, 1961; Addicott, 1966), and these complementary reports employ molluscan zoogeogra- phy to reconstruct the marine climate of the late Pleistocene shelf off the west coast of the United States. As a result of the present study, the Holocene zoogeography of the Ostracoda can now serve as a datum for similar paleoclimatic reconstructions. REFERENCES CITED Addicott, W. 0., 1966, Late Pleistocene marine paleoecology and zoogeography in central California: U.S. Geol. Survey Prof. Paper 523—0, 21 p. Allison, E. C., 1964, Geology of areas bordering Gulf of Cali- fornia, in Andel, P. H., van, and Shor, G. G., Jr., eds., Marine geology of the Gulf of California—A symposium: Am. Assoc. Petroleum Geologists Mem. 3, p. 3—29. Bandy, O. L., 1967, Foraminiferal definition of the boundaries of the Pleistocene in southern California, in Sears, M., ed., Progress in Oceanography—v. 4, The Quaternary his- tory of the ocean basins. Oxford, Pergamon Press, p. 27—49. Bandy, O. L., and Wilcoxon, J. A., 1970, The Pliocene—Pleisto- cene boundary, Italy and California: Geol. Soc. America Bull., v. 81, p. 2939—2948. Benson, R. H., 1959, Ecology of Recent ostracodes of the Todos Santos Bay region, Baja California, Mexico: Kansas Univ. Paleont. Contr. 23, Arthropoda, art. 1, 80 p. Benson, R. H.. and Kaesler, R. L., 1963, Recent marine and lagoonal ostracodes from the Estero de Tastiota region, Sonora, Mexico (northeastern Gulf of California) : Kansas Univ. Paleont. Contr. 32, Arthropoda, art. 3, 34 p. Bold, W. A. van den, 1963, Upper Miocene and Pliocene Ostra— coda of Trinidad: Micropaleontology, v. 9, p. 361—424. Bullivant, J. S., 1969, The Bathhouse Beach assemblage: Southern California Acad. Sci. Bull., V. 68, pt. 2, p. 86—95. Burrett, T. L., 1967, Bottom sediment data from the continental shelf ofl“ Washington, Grays Harbor to Cape Flattery: Washington Univ. Dept. Oceanography Spec. Rept. 39. Cheetham, A. H., and Hazel, J. E., 1969, Binary (presence- absence) similarity coeflicients: Jour. Paleontology, V. 43, p. 1130—1136. Coryell, H. N ., and Fields, Suzanne, 1937, A Gatun ostracode fauna from Cativa, Panama: Am. Mus. Novitates No. 956, 18 p. Crouch, R. W., 1949, Pliocene Ostracoda from southern Cali- fornia: Jour. Paleontology, v. 23, p. 594—599. Ekman, Swen, 1953, Zoogeography of the Sea: London, Sidg— wick & Jackson, Ltd., 417 p. 46 ZOOGEOGRAPHY OF HOLOCENE OSTRACODA OFF WESTERN NORTH AMERICA Fanale, F. P., and Schaeffer, O. A., 1965, Helium-uranium ratios for Pleistocene and Tertiary fossil aragoni’oes: Sci- ence, v. 149, p. 312—317. Galloway, J. J ., and Wiissler, S. G., 1927, Pleistocene Foramini- fera from the Lomita Quarry, Palos Verdes Hills, Califor— nia: Jour. Paleontology, V. 1, p. 35—87; correction of names, p. 198. Gunter, Gordon, 1957, Temperature, in Hedgpeth, J. W., ed., Ecology, v. 1 of Treatise on marine ecology and paleoecolo- gy: Geol. Soc. America Mem. 67, p. 159—184. Hanai, Tetsuro, 1957a, Studies on the Ostracoda from Japan; I, Subfamily Leptocytherinae n. subfam.: Tokyo Univ. Fac. Sci. Jour., sec. 2, V. 10, p. 431—468. 1957b, Studies on the Ostracoda from Japan; 11, Sub- family Pectocytherinae n. subfam.: Tokyo Univ. Fac. Sci. Jour., sec. 2, V. 10, p. 469—482. 1957c, Studies on the Ostracoda from Japan; 111, Sub— families Cytherurinae G. W. Muller (emend. G. O. Sars, 1925) and Cytheropterinae n. subfam.: Tokyo Univ. Fac. Sci. Jour., sec. 2, v. 11, p. 11—36. 1959a, Studies on the Ostracoda from Japan; IV, Family Cytherideidae Sars, 1925: Tokyo Univ. Fac. Sci. Jour., sec. 2, v. 11, p. 291—308. 1959b, Studies on the Ostracoda from Japan; V, Sub- family Cytherinae Dana, 1952 (emend.): Tokyo Univ. Fac. Sci. Jour., sec. 2, V. 11, p. 409—418. 1961, Spinileberis, a new genus of Ostracoda from the Pacific: Palaeont. Soc. Japan Trans. and Proc., 11. ser. no. 44, I). 167—170. 1970, Studies on the ostracod subfamily Schizocy— therinae Mandelstam: Jour. Paleontology, v. 44, p. 693— 729. Hartmann, Gerd, 1953, Iliocythere meyer-abichi nov. spec., ein Ostracode des Schlickwattes von San Salvador: Zoolog- ischer Anzeiger, V. 151, p. 310—316. 1956, Zur Kenntnis des Mangrove-Estero-Gebietes von El Salvador und seiner Ostracoden—Fauna: Kieler Mee- resforschungen, V. 12, p. 219—248. 1957, Zur Kenntnis des Mangrove-Estero-Gebietes von El Salvador und seiner Ostracoden-Fauna, II: Kieler Meeresforschungen, v. 13, p. 134—159. 1959, Zur Kenntnis der lotischen Lebensbereiche der pazifischen Kiiste von El Salvador unter besonderer Beriicksichtigung seiner Ostracodenfauna, [pt. 3 of] Beitrag zur Fauna El Salvadors Kieler Meeresforschun- gen, V. 15, p. 187—241. 1962, Ostracoden des Eulitorals, in Hartmann- Schroder, G., and Hartmann, G., Zur Kenntnis des Euli- torals der chilenischen Pazifikkfiste und der Argen- tinischen Kfiste Siidpatagoniens unter besonderer Beriick- sichtigung der Polychaeten und Ostracoden: Hamburg Zool. Mus. Inst. Mitt., pt. 3, p. 169—270. 1965, Ostracoden des Sublitorals, in Hartmann- Schroder, G., and Hartmann, G., Zur Kenntnis des Sub— litorals der chilenischen Kiiste unter besonderer Beriick- sichtigung der Polychaeten und Ostracoden: Hamburg Zool. Mus. Inst., Mitt., pt. 3, p. 307—384. Hazel, J. E., 1962, Two new hemicytherid ostracods from the lower Pleistocene of California: Jour. Paleontology, V. 36, p. 822—826. 1970, Atlantic continental shelf and slope of the United States—Ostracode zoogeography in the southern Nova Scotian and northern Virginian faunal provinces: U.S. Geol. Survey Prof. Paper 529—E, 21 p. Hertlein, L. G., and Grant, U. S., IV, 1944, Geology, part 1 of The geology and paleontology. of the marine Pliocene of San Diego, California: San Diego Soc. Nat. History Mem., V. 2, p. 1—72. Holland, C. H., 1971, Silurian faunal provinces?, in Middle- miss, F. A., Rawson, P. F., and Newall, G., eds., Faunal provinces in space and time: Geol. Jour. Spec. Issue 4, p. 61—76. Hutchins, L. W., 1947, The bases for temperature zonation in geographical distribution: Ecological Monographs, V. 17, p. 325—335. Ishizaki, Kunihiro, 1966, Miocene and Pliocene ostracodes from the Sendai area, Japan: Tohoku Univ. Sci. Repts., 2d ser. (Geol.), V. 37, p. 131—163. 1968, Ostracodes from Uranouchi Bay, Kochi Pre- fecture, Japan: Tohoku Univ. Sci. Repts., 2d ser. (Geol.), v. 40, p. 1—45. 1969, Ostracodes from Shinjiko and Nakanoumi, Shimane Prefecture, western Honshu, Japan: Tohoku Univ. Sci. Repts., 2d ser. (Geol.), v. 41, p. 197—224. 1971, Ostracodes from Aomori Bay, Aomori Prefecture, northeast Honshu Japan: Tohoku Univ. Sci. Repts., 2d ser. (Geol.), v. 43, p. 59—97. Ishizaki, Kunihiro, and Gunther. F. J., 1974, Ostracoda of the Family Cytheruridae from the Gulf of Panama: Tohoku Univ., Sci. Repts., 2d ser. (Geol.), V. 45, p. 1—50. Juday, Chauncey, 1907, Ostracoda of the San Diego region; II, Littoral forms: California Univ. Pubs. Zoology, v. 3, p. 135—156. Kern, J. P., 1973, Early Pliocene marine climate and environ- ment of the eastern Ventura basin, southern California; California Univ. Pubs. Geol. Sci., V. 96, 117 p. Kinne, 0., 1963, The effects of temperature and salinity on marine and brackish water animals; Part 1, Tempera- ture, in, Barnes, H., ed., Oceanography and marine bio- logy—An annual review, v. 1: London, George Allen & Unwin, p. 301—340. Lankford, R. R., 1962, Recent Foraminifera from the near- shore turbulent zone, western United States and north- west Mexico: Doctoral dissertation, California Univ., Scripps Inst. Oceanography, 234 p. Lankford, R. R., and Phleger, F. B, 1973, Foraminifera from the nearshore turbulent zone, western North Ameri— ca; Jour. Foraminiferal Research, V. 3, p. 101—132. LeRoy, L. W., 1943a, Pleistocene and Pliocene Ostracoda of the coastal region of southern California: Jour. Paleontology, V. 17, p. 354—373. 1943b, Name revisions of some Tertiary ostracodes: Jour. Paleontology, V. 17, p. 629. 1945, A contribution to ostracodal ontogeny: Jour: Paleontology, V. 19, p. 81—86. Lucas, V. Z., 1931, Some Ostracoda of the Vancouver Island region: Fisheries Research Board Canada, Contr. Cana- dian Biology and Fisheries, V' 6, p. 399—416. Lynn, R. J., 1967, Seasonal variation of temperature and salinity at 10 meters in the California Current: Cali- fornia cooperative oceanic fisheries investigations—Re- ports, V. 11, 1 July 1963—30 June 1966, California Dept. Fish and Game, p. 157—186. McKenzie, K. G., and Swain, F. M., 1967, Recent Ostracoda from Scammon Lagoon, Baja California: Jour. Paleon- tology, V. 41, p. 281—305. REFERENCES CITED Manar, T. A., ed., 1953, California cooperative oceanic fish- eries investigations—Progress rept. 1 July 1952—30 June 1953, California Dept. Fish and Game, 44 p. Margalef, Ramon, 1968, Perspectives in ecological theory: Chicago, Chicago Univ. Press, 111 p. Morkhoven, F. P. C. M. van, 1963, Post-Paleozoic Ostracoda, their morphology, taxonomy, and economic use: Amster- dam, Elsevier, v. 2, (generic descriptions), 478 p. Natland, M. L., 1957, Paleoecology of West Coast Tertiary sediments, in Ladd, H. 8., ed., Paleoecology, v. 2 of Treatise on marine ecology and paleoecology: Geol. Soc. America Mem. 67 , p. 543—571. Obradovich, J. D., 1968, The potential use of glauconite for late-Cenozoic geochronology, in Morrison, R. B., and Wright, H. E., J r., eds., Means of correlation of Quater- nary successions: Assoc. Quaternary Research, 7th con- gress, Sept., 1965 Proc., v. 8, p. 267—279. Ohmert, Wolf, 1968, Die Coquimbinae, eine neue Unterfamilie der Hemicytheridae (Ostracoda) aus dem Pliozan von Chile: Bayer. Staatssamml. Palaont. Hist. Geol. Mitt, v. 8, p. 127—165. 1971, Die Ostracodengattung Palaciosa und Caudites aus dem Pliozaen von Chile: Bayer. Staatssamml. Palaont. Hist. Geol. Mitt., v. 11, p. 87—116. Pokorny, Vladimir, 1968, Radimella gen. n., a new genus of the Hemicytherinae (Ostracoda, Crust.): Carolinae Univ. Acta, Geologica No. 4, p. 359—373. 1969, The genus Radimella Pokorny, 1969 (Ostracoda, Crustacea) in the Galapagos Islands: Carolinae Univ. Acta, Geologica No. 4, p. 293—334. 1970, The genus Caudites Coryell & Fields, 1937 (Ostracoda, Crust.) in the Galapagos Islands: Carolinae Univ. Acta. Geologica No. 4, p. 267—302 [1972]. Reid, J. L., Roden, G. I., and Wyllie, J. G., 1958, Studies of the California Current system: California cooperative oceanic fisheries investigations—Progress rept., 1 July 1956—1 January 1958, California Dept. Fish and Game, p. 27—56. Robinson, M. K., 1957, Sea temperature in the Gulf of Alaska and in the northeast Pacific Ocean, 1941—1952: Cali— fornia Univ., Scripps Inst. Oceanography Bull., v. 7, p. 1-98. Schmidt, R. A. M., 1963, Pleistocene marine microfauna in the Bootlegger Cove Clay, Anchorage, Alaska: Science, v. 141, p. 350—351. 1967, New generic assignments for some Pleistocene Ostracoda from Alaska: Jour. Paleontology, v. 41, p. 487— 488. Schmidt, R. A. M., and Sellmann, P. V., 1966 Mummified Pleistocene ostracods in Alaska: Science, v. 153, p. 167— 168. Shephard, F. F., and Wrath, W. F., 1937, Marine sediments around Catalina Island: Jour. Sed. Petrology, v. 7, p. 41— 50. Skogsberg, Tage, 1928, Studies on marine ostracods. Part II: California Acad. Sci. Occasional Paper 15. 155 p. 1950, Two new species of marine Ostracoda (Podo- copa) from California: California Acad. Sci. Proc., v. 26, p. 483—505. Smith. V. Z., 1952. Further Ostracoda of the Vancouver Island region: Fisheries Research Board Canada Jour., v. 9, p. 16—41. Sokal, R. R., and Michener, C. D.. 1958. A statistical method for evaluating systematic relationships: Kansas Univ. Sci. Bull., v. 38, p. 1409—1438. 47 Swain, F. M., 1963, Pleistocene Ostracoda from the Gubik Formation, Arctic Coastal Plain, Alaska: Jour. Paleon- tology, v. 37, p. 798—834. 1967, Ostracoda from the Gulf of California: Geol. Soc. America Mem. 101, 139 p. 1969, Taxonomy and ecology of near-shore Ostracoda from the Pacific coast of North and Central America, in Neale, J. W., ed., The taxonomy, morphology and ecology of Recent Ostracoda: Edinburgh, Oliver & Boyd, p. 423— 480. Swain, F. M., and Gilby, J. M., 1967, Recent Ostracoda from Corinto Bay, western Nicaragua, and their relationship to some other assemblages of the Pacific coast: Jour. Paleontology, v. 41, p. 306—334. 1974, Marine Holocene Ostracoda from the Pacific coast of North and Central America: Micropalentology, v. 20, p. 257—352. Triebel, Erich, 1957, Neue Ostracoden aus dem Pleistoz'an von Kalifornien: Senckenbergiana Lethaea, v. 38, p. 291— 309. Upson, J. E., 1951, Geology and ground-water resources of the south-coast basins of Santa Barbara County, Cali- fornia: U.S. Geol. Survey Water—Supply Paper 1108, 144 p. Valentine, J. W., 1961, Paleoecologic molluscan geography of the Californian Pleistocene: California Univ. Pubs. Geol. Sci., v. 34, p. 309-442. 1966, Numerical analysis of marine molluscan ranges on the extratropical northeastern Pacific shelf: Limnology and Oceanography, v. 11, p. 198-211. 1968, The evolution of eclogical units above the popu- lation level: J our. Paleontology, v. 42, p- 253—267. 1971, Resource supply and species diversity patterns: Lethaia, v. 4, p. 51—61. 1973, Evolutionary paleoeoology of the marine bio- sphere: Englewood Cliffs, N.J., Prentice Hall, Inc., 511 p. Valentine, J. W., and Meade, R. F., 1961, California Pleisto- cene paleotemperatures: California Univ. Pubs. Geol. Sci., v. 40, p. 1—46. Valentine, J. W., and Peddicord, R. G., 1967, Evaluation of fossil assemblages by cluster analysis: J our. Paleontology, v. 41, p. 502—507. Valentine, P. C., 1971, Climatic implication of a late Pleisto- cene ostracode assemblage from southeastern Virginia: U.S. Geol. Survey Prof. Paper 683—D, 28 p. Watling, Les, 1970, Two new species of Cytherinae (Ostra- coda) form central California: Crustaceana, v. 19, p. 251—263. Woodring, W. P., 1952, Pliocene—Pleistocene boundary in Cali- fornia Coast Ranges: Am. Jour. Sci., V. 250, p. 401—410. 1957, Marine Pleistocene of California, in Ladd, H. 8., ed., Paleoecology, v. 2, of Treatise on marine ecology and paleoecology: Geol. Soc. America Mem. 67, p. 589—597. Woodring, W. P., and Bramlette, M. N., 1950, Geology and paleontology of the Santa Maria district, California: U.S. Geol. Survey Prof. Paper 222, 185 p. Woodring, W. P., Bramlette, M. N., and Kew, W. S. W., 1946, Geology and paleontology of the Palos Verdes Hills, California: U.S. Geol. Survey Prof. Paper 207, 145 p. Woodring, W. P., Stewart, R. B., and Richards, R. W., 1940, Geology of the Kettleman Hills oil field, California: U.S. Geol. Survey Prof. Paper 195, 170 p. Zinsmeister. W. J., 1970. A late Pliocene macrofossil fauna of Newport Beach, Orange County, California: Southern California Acad. Sci. Bull., v. 69, p. 121—125 [1971]. —_—_————§—‘4.7 fiUS. GOVERNMENT PRINTING OFFICE: 1976 0—211—317/63 PLATES 1-14 Contact photographs of the plates in this report are available, at cost, from U.S. Geological Survey Library, Federal Center, Denver, Colorado 80225 FIGURES 1, 2, 4, 5. 9, 12. 10, 11, 13, 14. 15,18. 16, 17. PLATE 1 [All figures are lateral views; all X 60] Ambostmcon costatum Hazel, 1962. 1. Right valve, male. Santa Barbara Formation. USN M 207779. 2. Right valve, female. Santa Barbara Formation. USNM 207780. 4. Left valve, male. Santa Barbara Formation. USNM 207781. 5. Left valve, female. Santa Barbara Formation. USNM 207782. Ambostmcon sp. M. 3. Right valve, female. Sample 164. USNM 207783. 6. Right valve, male. Sample 164. USNM 207784. Ambostmcon sp. A. 7. Left valve, male. Sample 1. USNM 207785. 8. Left valve, female. Sample 10. USNM 207786. Ambostmcon sp. B. 9. Left valve, male. Sample 101. USNM 207787. 12. Left valve, female. Sample 101. USNM 207788. Ambostmcon sp. G. 10. Left valve, male. Sample 177. USNM 207789. 11. Left valve, female. Sample 177. USNM 207790. 13. Left valve, male. Sample 117. USNM 207791. 14. Left valve, female. Sample 117. USNM 207792. Ambostracon glaucum (Skogsberg, 1928). 15. Left valve, female. Sample 193. USNM 207793. 18. Left valve, male. Sample 193. USNM 207794. Ambostmcon sp. E. 16. Left valve. male. Sample 59. USNM 207795. 17. Left valve, female. Sample 59. USNM 207796. GEOLOGICAL SURVEY PROFESSIONAL PAPER 916 PLATE 1 AMBOSTRA CON FIGURES 1, 4. 5, 6, 8, 9. 7, 1o. 11, 14. 12, 13. 15, 16. PLATE 2 [All figures are lateral views; all X 60] Ambostracon sp. J. 1. Left valve, female. Sample 193. USNM 207797. 4. Left valve, male. Sample 193, USNM 207798. Ambostracon sp. 0. 2. Left valve, male. Sample 222. USNM 207799. 3. Left valve, female. Sample 222. USNM 207800. Ambostmcon sp. L. 5. Left valve, male. Sample 164. USNM 207801. 6. Left valve, female. Sample 164. USNM 207802. 8. Left valve, male. Sample Mf2176. USNM 207803. 9. Left valve, female. Sample Mf2176. USNM 207804. Ambostmccm diegoensis (LeRoy, 1943). 7. Left valve, female. Sample 201. USNM 207805. 10. Left valve, male. Sample 201. USNM 207806. Ambostmcon sp. K. 11. Left valve, female. Sample 244. USNM 207807. 14. Left valve, male. Sample 244. USNM 207808. Ambostmcon califomicum (Hazel. 1962). 12. Left valve, male. Sample Mf2169. USNM 207809. 13. Left valve, female. Sample Mf2169. USNM 207810. Ambostmcon sp. D. 15. Left valve, male. Sample Mf2174. USNM 207811. 16. Left valve, female. Sample Mf2174. USNM 207812. GEOLOGICAL SURVEY PROFESSIONAL PAPER 916 PLATE 2 AMB OS TBA CON FIGURES 1. 4, 7, 1o. 5, 8, 11. 6, 9. 12. 13, 14. 15, 18. 16, 17. PLATE 3 [All figures are lateral views; all X 60] Ambostmcon microreticulatum (LeRoy, 1943). Left valve, female. Sample 201. USNM 207813. Ambostmcon sp. Q. Left valve, female. Sample 242. USNM 207814. Ambostracon sp. F. Left valve, female. Sample 202. USNM 207815. Coqm'mba sp. A. 4. Left valve, male. Sample Mf2174. USNM 207816. 7. Left valve, female. Sample Mf2174 USNM 207817. 10. Right valve, female. Sample Mf2174. USNM 207818. Coqm‘mba schencki (LeRoy, 1943). 5. Left valve, male. Sample Mf2174. USNM 207819. 8. Left valve, female. Sample Mf2174. USNM 207820. 11. Right valve, female. Sample Mf2174. USNM 207821. New genus F sp. A. 6. Right valve. Sample 1. USNM 207822. 9. Left valve. Sample 15. USNM 207823. Buntom'a Sp. B. Left valve, female. Sample 154. USNM 207824. Puriana pacifica Benson, 1959. 13. Left valve, female. Sample 197. USNM 207825. 14. Right valve, female. Sample 197. USNM 207826. Buntom’a sp. A. 15. Left valve, female. Sample Mf2180. USNM 207827. 18. Right valve, female. Sample Mf2180. USNM 207828. Purimw sp. A. 16. Left valve, female. Sample 254. USNM 207829. 17. Right valve, female. Sample 254- USNM 207830. GEOLOGICAL SURVEY PROFESSIONAL PAPER 916 PLATE 3 16 AMBOSTRACON, BUNTONIA, COQUIMBA, NEW GENUS F, AND PURIANA FIGURES 1. 12, 15. 13, 16, 17. 10, 14. 11. PLATE 4 [All figures are lateral views; all X 60] Hermam'tes sp. E. Left valve, male. Sample 215. USNM 207831. Hermanites kewi (LeRoy, 1943). 2. Left valve, female. Sample Mf2174. USNM 207832;. 5. Left valve, female. Sample Mf2180. USNM 207833. Sahm'a sp. A. Left valve. Sample 197. USNM 207834. Hermam'tes sp. G. Left valve, female. Sample 203. USNM 207835. Pulmilocytheridea pseudoguardensis McKenzie and Swain, 1967. 6. Left valve, female. Sample 233. USNM 207836. 7. Right valve, female. Sample 233. USNM 207837. Hermanites sp. D. Left valve, female. Sample 215. USNM 207838. Hermanites sp. C. Left valve, female. Sample 193. USNM 207839. “Cytheretta” sp. B. 10. Right valve, male. Sample 141. USNM 207840. 14. Right valve, female. Sample 141. USNM 207841. He7~mcmites sp. A. Left valve, male. Sample 215. USNM 207842. Cytkeropteron sp. A. 12. Left valve, female. Sample 35. USNM 207843. 15. Right valve, female. Sample 35. USNM 207844. “Cytheretta” sp. A. 13. Right valve, male. Sample 61. USNM 207845. 16. Right valve, female. Sample 61. USNM 207846. 17. Left valve, female. Sample 61- USNM 207847. PROFESSIONAL PAPER 916 PLATE 4 GEOLOGICAL SURVEY 15 16 17 “CYTHERETTA,” CYTHEROPTERON, HERMANITES, PULMILOCYTHERIDEA, AND SAHNIA FIGURES 1, 2. 3, 8. 15. 16. 17. PLATE 5 [All figures are lateral views; all X 60 except where noted] Pontocythere sp. C. 1. Right valve, female. Sample 15. USNM 207848. 2. Left valve, female. Sample 15. USN M 207849. Paracytheridea granti LeRoy, 1943. 3. Right valve, female. Sample 122. USNM 207850. 8. Left valve, female. Sample 122. USNM 207851. Pontocythere sp. A. 4. Right valve, female. Sample 197. USN M 207852. 5. Left valve, female. Sample 197. USNM 207853. Pontooythere sp. B. 6. Right valve, female. Sample 197. USNM 207854. 7. Left valve, female. Sample 197. USN M 207855. Paracythem’dea sp. B. 9. Right valve, female. Sample 221. USNM 207856. 10. Left valve, female. Sample 221. USNM 207857. Paracythem'dea sp. A. 11. Left valve, female. Sample 164. USNM 207858. 14. Right valve, female. Sample 164. USNM 207859. “Kangan'na” sp. A. 12. Left valve. Sample 109. X 120. USNM 207860. 13. Right valve. Sample 2. X 120. USNM 207861. Kanga/rz'na sp. B. Right valve, female. Sample 215. X 120. USNM 207862. Kangarim sp. C. Right valve, female. Sample 99. X 120. USNM 207863. Kanga’r’lna aff. K. quellita Coryell and Fields, 1937. Right valve, female. Sample 242. x 120. USNM 207864. PROFESSIONAL PAPER 916 PLATE 5 GEOLOGICAL SURVEY KANGARINA, “KANGARINAj’ PARACYTHERIDEA, AND PONTOCYTHERE FIGURES 1, 2. 12. 13. 14. 15, 16. 10. 11. PLATE 6 [All figures are lateral views; all X 120 except where noted] Loxocomiculum sp. A. 1. Left valve, male. Sample 221. X 60. USNM 207865. 2. Left valve, female. Sample 221. x 60. USNM 207866. Loxocorm‘culum sp. B. Left valve, female. Sample 251. X 60. USNM 207867. “Loxoconcha” emaciata Swain, 1967. Left valve, female. Sample 251. X 60. USNM 207868. Hemicytheru’ra sp. D. Right valve, female. Sample 57. USNM 207869. Hemicytherum sp. F. Right valve, female. Sample 215. USNM 207870. Hemicytherum sp. G. Right valve, female. Sample 244. USNM 207871. Hemicythemm sp. L. Right valve, female. Sample 108. USNM 207872. Hemicytheru’ra sp. H. Right valve, female. Sample 108. USNM 207873. Hemicytherura sp. K. Right valve, female. Sample 215. USNM 207874. Hemicythererum sp. J. Right valve, female. Sample 202. USNM 207875. Hemicythe'rum sp. C. Right valve, female. Sample 222. USNM 207876. Hemicytherum sp. I. Right valve, female. Sample 215. USNM 207877. Hemicytherura sp. B. Right valve, female. Sample 109. USNM 207878. Hemicytherum sp. A. 15. Right valve, female. Sample 2. USNM 207879. 16. Left valve, female. Sample 2. USNM 207880. GEOLOGICAL SURVEY PROFESSIONAL PAPER 916 PLATE 6 HEMICYTHERURA, “LOXOCONCHA,” AND LOXOCORNICULUM FIGURES 1, 2. 13, 16. 14, 17. 15. PLATE 7 [All figures are lateral views; all X 60] Loxoconcha lenticulata LeRoy, 1943. 1. Left valve, female. Sample Mf2174. USNM 207881. 2. Left valve, male. Sample Mf2174. USN M 207882. “Aurila” sp. C. 3. Right valve, female. Sample 193. USNM 207883. 7. Left valve, female. Sample 193. USNM 207884. Loxoconcha sp. A. 4. Right valve, female. Sample 15. USNM 207885. 5. Left valve, female. Sample 15. USNM 207886. Loxoconcha helenae Crouch, 1949. Left valve, female. Sample 82. USNM 207887. Loxoconcha sp. B. 8. Left valve, female. Sample 193. USNM 207888. 12. Left valve, male. Sample 193. USNM 207889. Loxoconcha sp. E. Left valve, female. Sample 184. USNM 207890. Aurila sp. B. 10. Right valve, female. Sample 211. USNM 207891. 11. Left valve, female. Sample 211. USNM 207892. “Aurila” sp. E. 13. Left valve, male. Sample 40. USNM 207893. 16. Left valve, female. Sample 40. USNM 207894. “Aum’la” drive'ri (LeRoy, 1943). 14. Right valve, female, Sample Mf2176. USNM 207895. 17. Left valve, female. Sample Mf2176. USNM 207896. “Aum'la” schumannensis (LeRoy, 1943). Left valve, female. Sample 96. USNM 207897. PROFESSIONAL PAPER 916 PLATE '7 GEOLOGICAL SURVEY AURILA, “A URILA,” AND LOXOCONCHA FIGURES 1—4. 5—8. 9, 10, 12, 13, 15, 16, 18, 19. 11, 14. 17, 20. PLATE 8 [All figures are lateral views; all X 60] Aurila sp. C. 1. Left valve, male. Sample 109. USNM 207898. 2. Right valve, male. Sample 109. USNM 207899. 3. Right valve, female. Sample 109. USNM 207900. 4. Left valve, female. Sample 109. USNM 207901. Aurila lincolnensis (LeRoy, 1943). 5. Left valve, male. Sample 67. USNM 207902. 6. Right valve, male. Sample 67. USNM 207903. 7. Right valve, female. Sample 67. USNM 207904. 8. Left valve, female. Sample 67. USNM 207905. Aurila ‘sp. A. 9. Left valve, male. Sample 67. USNM 207906. 10. Right valve, male. Sample 67. USNM 207907. 12. Left valve, female. Sample 67. USNM 207908. 13. Right Valve, female. Sample 67. USNM 207909. 15. Left valve, male. Sample 3. USNM 207910. 16. Right valve, male. Sample 3. USNM 207911. 18. Left valve, female. Sample 3. USNM 207912. 19. Right valve, female. Sample 3. USNM 207913. Aurila sp. D. 11. Left valve, female. Sample 109. USNM 207914. 14. Right valve, female. Sample 109. USNM 207915. Aurila montereyensis (Skogsberg, 1928). 17. Right valve, female. Sample 108. USNM 207916. 20. Left valve, female. Sample 108. USNM 207917. PROFESSIONAL PAPER 916 PLATE 8 GEOLOGICAL SURVEY 19 AURILA FIGURES 1, 4. 2, 3. 9, 12. 13, 16. 14, 17. 15. 18. PLATE 9 [All figures are lateral views; all )< 60] Cytheromorpha sp. A. 1. Left valve, male. Sample 2. USNM 207918. 4. Left valve, female. Sample 2. USNM 207919. Cytheromorpha sp. B. 2. Left valve, male. Sample 2. USNM 207920. 3. Left valve, female. Sample 2. USNM 207921. “Radimella” aurita (Skogsberg, 1928). 5. Left valve, female. Sample 177. USNM 207922. 6. Left valve, male. Sample 177. USNM 207923. Radimella palosensis (LeRoy, 1943). 7. Left valve, female. Sample 101. USNM 207924. 10. Left valve, male. Sample 101. USNM 207925. Radimella sp. B. 8. Left valve, female. Sample 244. USNM 207926. 11. Left valve, male. Sample 244. USNM 207927. Radimella sp. A. 9. Left valve, female. Sample 202. USNM 207928. 12. Left valve, male. Sample 202. USNM 207929. “Radimella” jollaemis (LeRoy, 1943). 13. Left valve, female. Sample 177. USNM 207930. 16. Left valve, male. Sample 177. USNM 207931. “Radimella” sp. A. 14. Left valve, female. Sample 67. USNM 207932. 17. Left valve, male. Sample 67. USNM 207933. “Radimella” pacifica (Skogsberg, 1928). Left valve, female. Sample 108. USNM 207934. “Radimella” sp. B. Left valve, female. Sample 201. USNM 207935. GEOLOGICAL SURVEY PROFESSIONAL PAPER 916 PLATE 9 CYTHEROMORPHA, RADIMELLA, AND “RADIMELLA” FIGURES 1, 2. 13, 14. 15, 18. 16, 17. PLATE 10 [All figures are lateral views; all X 60 except where noted] Munseyella sp. A. 1. Right valve, female. Sample 197. USNM 207936. 2. Left valve, female. Sample 197. USNM 207937. Palmenella, oalifo’rm'ca Triebel, 1957. 3. Left valve, female. Sample 141. USNM 207938. 4. Right valve, female. Sample 141. USNM 207939. Munseyella sp. B. 5. Left valve, female. Sample 96. USNM 207940. 6. Right valve, female. Sample 96. USNM 207941. Munseyella similis? Triebel, 1957. 7. Left valve, male. Sample Mf2180. X 120. USNM 207942. 12. Left valve, female. Sample Mf2180. X 120. USNM 207943. Pectocythere sp. A. 8. Left valve, female. Sample Mf2180. USNM 207944. 9. Right valve, female. Sample Mf2180. USNM 207945. 10. Left valve, male. Sample 2. USNM 207946. 11. Right valve, male. Sample 2. USNM 207947. Pectocythere clavata (Triebel, 1957). 13. Left valve, female. Sample 117. USNM 207948. 14. Right valve, female. Sample 117. USNM 207949. Munseyella pedroensis Triebel, 1957. 15. Left valve, female. Sample Mf2174. X 120. USNM 207950. 18. Right valve, female. Sample Mf2174. X 120. USNM 207951. Pectocythere tomalensis Watling, 1970. 16. Left valve, female. Sample 68. USNM 207952. 17. Right valve, female. Sample 68. USNM 207953. ' GEOLOGICAL SURVEY PROFESSIONAL PAPER 916 PLATE 10 MUNSEYELLA, PALMENELLA, AND PECTOCYTHERE " FIGURES 1, 4. 3, 6. 7, 11. 1o, 14. 15, 16, 18, 19. 17, 20. PLATE 11 [All figures are lateral views; all X 60] Palaciosa sp. A. 1. Left valve, female. Sample 108. USNM 207954. 4. Right valve, female. Sample 108. USNM 207955. Palaciosa, sp. B. 2. Left valve, female. Sample 101. USN M 207956. 5. Left valve, male. Sample 101. USNM 207957. Oriom'na pseudovaugkm' Swain, 1967. 3. Left valve, female. Sample 233. USN M 207958. 6. Right valve, female. Sample 233. USNM 207959. Caudites sp. B. 7. Left valve, female. Sample 221. USNM 207960. 11. Right valve, female. Sample 221. USNM 207961. Caudites fragilis LeRoy, 1943. 8. Left valve, female. Sample 197. USNM 207962. 12. Right valve, female. Sample 197. USNM 207963. Caudites sp. D. 9. Left valve, female. Sample 242. USNM 207964. 13. Right valve, female. Sample 242. USNM 207965. Caudz'tes sp. E. 10. Left valve, female. Sample 241. USNM 207966. 14._ Right valve, female. Sample 241. USNM 207967. Caudites purii (McKenzie and Swain, 1967). 15. Left valve, female. Sample 242. USNM 207968. 16. Right valve, female. Sample 242. USNM 206969. 18. Left valve, male. Sample 233. USNM 207970. 19. Right valve, male. Sample 233. USNM 207971. Tm'ebelina reticulopzmctata Benson, 1959. 17. Right valve, female. Sample 221. USNM 207972. 20. Left valve, female. Sample 221. USNM 207973. GEOLOGICAL SURVEY PROFESSIONAL PAPER 916 PLATE 11 18 CA UDITES, ORIONINA, PALA CIOSA, AND TRIEBELINA FIGURES 1, 4, 7, 10. 2, 5, 8, 12. 3, 6. 9, 13. 11. 14. 15, 16. PLATE 12 [All figures are lateral views; all X 60] H emicythere Sp. A. 1. Right valve, female. Sample 32. USNM 207974. 4. Left valve, female. Sample 32. USNM 207975. 7. Right valve, male. Sample 32. USNM 207976. 10. Left valve, male. Sample 32. USNM 207977. Hemicythere sp. B. 2. Right valve, female. Sample 40. USNM 207978. 5. Left valve, female. Sample 40. USNM 207979. 8. Right valve, male. Sample 40. USNM 207980. 12. Left valve, male. Sample 40. USNM 207981. Cythere maia (Benson, 1959). 3. Right valve, female. Sample 109. USNM 207982. 6. Left valve, female. Sample 109. USNM 207983. Cythere sp. B. 9. Right valve, female. Sample 40. USNM 207984. 13. Left valve, female. Sample 40. USNM 207985. Basslerites thlipsuroidea Swain, 1967. Left valve. Sample 255. USNM 207986. Bassle’rites delreyensis LeRoy, 1943. Left valve, female. Sample 215. USNM 207987. Cythe're sp. A. 15. Right valve, female. Sample 31. USNM 207988. 16. Left valve, female. Sample 31. USNM 207989. PROFESSIONAL PAPER 916 PLATE 12 GEOLOGICAL SURVEY 15 BASSLERITES, C YTHERE, AND HEMICYTHERE FIGURES 1, 6. 10. 13, 14. PLATE 13 [All figures are lateral views; all X 60] Cativella semitmnslucens (Crouch, 1949). 1. Left valve, female. Santa Barbara Formation. USNM 207990. 6. Right valve, female. Santa Barbara Formation. USN M 207991. Cativella sp. A. Left valve, female. Sample 222. USNM 207992. Cativella sp. B. Left valve, female. Sample 255. USNM 207993. “Paijenborchella” sp. B. 4. Right valve, female. Sample 101. USNM 207994. 5. Left valve, female. Sample 101. USNM 207995. 7. Right valve, female. Sample Mf2180. USNM 207996. “Paijenborchella” sp. A. 8. Right valve, female. Sample 8. USNM 207997. 11. Left valve, female. Sample 8. USNM 207998. “Trachyleberis” sp. A. 9. Left valve, male. Sample 255. USNM 207999. 12. Left valve, female. Sample 255. USNM 208000. Neocaudites? hen'ryhowei (McKenzie and Swain, 1967). Left valve, female. Sample 215. USNM 208001. Costa? sanfelipensis Swain, 1967. 13. Left valve, male. Sample 255. USNM 208002. 14. Left valve, female. Sample 255. USNM 208003. GEOLOGICAL SURVEY PROFESSIONAL PAPER 916 PLATE 13 CA TIVELLA, COST/1?, NEOCA UDITES‘I, “PAIJENBORCHELLA,” AND “TRA CHYLEBERIS” FIGURES 1, 2, 4, 5. PLATE 14 [All figures are lateral views; all X 60] “Hemicythere” sp. A. 1. Left valve, female. Sample Mf2176. USNM 208004. 2 Left valve, male. Sample Mf2176. USNM 208005. 4. Left valve, female. Sample 164. USNM 208006. 5. Left valve, male. Sample 164. USNM 208007. “Hemicythere” sp. B. 3. Left valve, female. Sample 203- USNM 208008. 6. Left valve, male. Sample 203. USNM 208009. “Hemicythere” californiensis LeRoy, 1943. 7. Left Valve, female. Sample Mf2176. USNM 208010. 8. Left valve, male. Sample Mf2176. USNM 208011. “Hemicythere” sp. C. Left valve, male. Sample 222. USNM 20812. “H emicythere” hisp’ida LeRoy, 1943. 10. Left valve, female. Sample Mf2174. USNM 208013. 11. Left valve, male. Sample Mf2176. USNM 208014. PRC FESSIONAL PAPER 916 PLATE 14 GEOLOGICAL SURVEY “HEMICYTHERE” . ‘7 ram? F76’ 7/7 A Numerical Model of Material Transport In Salt-Wedge Estuaries GEOLOGICAL SURVEY PROFESSIONAL PAPER 917 EARTH scumcas tum Prepared in cooperation with the Municipality of Metropolitan Seattle ”CT 21 19:5 U351). A Numerical Model of Material Transport In Salt-Wedge Estuaries Part I. Description of the Model By H. B. FISCHER Part 11. Model Computation of Salinity and Salt-Wedge Dissolved Oxygen in the Duwamish River Estuary, King County, Washington By ]. D. STONER, w. L. HAUSHILD, and J. B. MCCONNELL GEOLOGICAL SURVEY PROFESSIONAL PAPER 917 Prepared in cooperation with the Municipality of Metropolitan Seattle UNITED STATES GOVERNMENT PRINTING OFFICE,WASHINGTON21975 UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress Cataloging in Publication Data A numerical model of material transport in salt-wedge estuaries. (Geological Survey Professional Paper 917) Bibliography: p. 36 CONTENTS: Fischer, H. B. Description of the model.~Stoner, J. D., Haushild, W. L., and McConnell, J. B. Model computation of salinity and salt-wedge dissolved oxygen in the lower Duwamish River estuary, King County, Washington. Supt. of Docs. No.: I 19.16:9l7 l. Sediment transport—Mathematical models. 2. Electronic data processing—Sediment transport. 3. Sediment transport— Washington (State)—Duwamish River estuary. I. Fischer, Hugo B. Description of the model. 1975. II. Stoner, J.D Model computation of salinity and salt-wedge dissolved oxygen in the lower Duwamish River estuary, King County, Washington. 1975. III. Series: United States. Geological Survey Professional Paper 917. GC56.N85 551.3’5’3 75—619036 For sale by the Superintendent of Documents, U. S. Government Printing Ofiice Washington, D. C. 20402 Stock Number 024-001-02712—7 Description of the Model By H. B. FISCHER GEOLOGICAL SURVEY PROFESSIONAL PAPER 917 Prepared in cooperation with the Municipality of Metropolitan Seattle FIGURE UIAoawi-A CONTENTS Page Abstract. ________________________________________________________________ 1 Introduction ______________________________________________________________ 1 The flow model. ___________________________________________________________ 1 Flow in the wedge ______________________________________________________ 3 Flow in the overlying layer _______________________________________________ 5 Computation of salinities ________________________________________________ 5 The material-transport modeL ________________________________________________ 6 Model operation ___________________________________________________________ 6 Summary and conclusions ___________________________________________________ 8 References _______________________________________________________________ 8 ILLUSTRATIONS Page Sketch showing distribution of flow in a stationary salt wedge ____________________________________________ 2 . Idealized longitudinal section showing division of salt-wedge estuary into parts for modeling _____________________ 2 . Graphs showing profile and longitudinal distribution of salinity in the Duwamish River estuary __________________ 3 . Diagram showing typical element and transport processes included in the model _______________________________ 4 . Simplified diagram of computer program ____________________________________________________________ 7 CONVERSION FACTORS In recognition of the worldwide use of the metric system of measurements, the following factors are provided for conversion of English values used in this report of metric values: Multiply By To Get Feet (ft) _______________________ 0.3048 __________ metre (m) Miles (mi) ______________________ 1.609. ___________ kilometres (km) Cubic feet (ft3). __________________ 0.02832 _________ cubic metres (m3) Cubic feet per second (ft3/s)_ ........ 1.699 ___________ cubic metres per minute (mein) Feet per second (ft/s). _____________ 0.3048 __________ metres per second (m/s) PART 1. DESCRIPTION OF THE MODEL By H. B. FISCHER ABSTRACT Water in a salt—wedge estuary ideally is characterized by an oscil- lating well-mixed wedge of undiluted seawater topped by a series of successively more dilute overlying layers. In the wedge the flow is back and forth, with a net landward component to replace water entrained upward into the overlying layer; in the overlying layers the flow also oscillates, but with a net seaward component because of the input of fresh river water and entrained wedge water. The flow is modeled by a computer program, and the flow is used as an input to the constituent—transport model. The computer program then is used to determine the advection and dispersion of dissolved constituents and plankton, and their concentrations throughout the system in response to given inputs. The report describes required input data and method of operation of the computer program. INTRODUCTION In some estuaries, predominantly those in which the inflow of fresh water is relatively large and the rate of mixing by tidal action is relatively small, salinity intru- sion takes the form of a wedge of nearly undiluted sea- water. The mouth of the Mississippi River, described by Stommel and Farmer (1952), and the Ishikari River in Japan, described by Fukushima, Yakuwa, and Taka- hashi (1969) are typical examples. The Duwamish River estuary at Seattle, Wash., contains a wedge of Puget Sound water even when the river discharge is relatively small. The Duwamish River estuary, for which the program described in this paper was specifically de- rived, is described by Santos and Stoner (1972) and in Part II of this paper. Part I describes the program itself, as an aid for possible application in studies of other salt-wedge estuaries. The purpose of the program is to model the transport and mixing of chemical and biological constituents in a salt-wedge estuary as a means of predicting the ecologi- cal consequences of man-induced changes in the es- tuarine system. A common ecological problem in salt- wedge estuaries is a detrimentally low level of dissolved oxygen (DO) at the toe of the wedge, caused by oxygen demand in the wedge and the lack of a mechanism for oxygen replenishment. This program therefore focuses on predicting concentrations of biochemical oxygen de- mand (BOD) and D0 in the salt wedge. The transport of other biological constituents, such as plankton, also is modeled, in part to compute its effect on DO. The hydrodynamics of a stationary salt wedge are described by Stommel and Farmer (1952) and Keulegan (1966). The distribution of flow in a stationary wedge is shown in figure 1; the flow is inward from the ocean along the channel bottom, and outward to the ocean both in the upper part of the saltwater wedge and in the overlying layer of mixed fresh and salt water. A net inward flow of saltwater in the wedge makes up for the salt water entrained from the wedge into the overlying layer. A stationary wedge is formed where a river flows into a tideless sea; if the sea is tidal the wedge will oscillate back and forth in the estuary, moving land- ward during the rising (flood) phase of the tide and seaward during the falling (ebb) phase. No complete solution for the hydrodynamics of an oscillating wedge has been found. The flow within the wedge will be turbu- lent because of the shear stress at the bottom, and en- trainment into the overlying layer may be greater than for the stationary wedge. The time-averaged flow dis- tribution may be assumed to be similar to that in the stationary wedge, although it will not be identical; the instantaneous flow distribution may deviate substan- tially from its time average during parts of the tidal cycle. In modeling constituent transport in a salt-wedge estuary, some assumptions must be made about the flow distribution and its effect. Discussed in the report are (1) the assumed flow distribution and the method of model- ing dispersion, and (2) the transport of biological and chemical constituents. THE FLOW MODEL The flow model represents an attempt to describe the flow in the salt wedge and the overlying layer in a way sufficiently simple that (1) computer modeling of con— stituent transport is possible and (2) the model approx- imately represents observed distributions of flow and salinity. It should be stressed that no attempt has been made to solve the hydrodynamic equations of motion. 1 2 NUMERICAL MODEL OF MATERIAL TRANSPORT IN SALT-WEDGE ESTUARIES V Overlying layer <———— <—— K H ‘__ <—- (—— ‘\ ) Salt wedge J I» / FIGURE 1,—Distribution of flow in a stationary salt wedge. Rather, a shape is assumed for the wedge, as shown in figure 2, and flow in the wedge is computed so that water volume is conserved. Field measurements are used to determine the thickness of the wedge at the estuary mouth, the slope of the interface of the wedge and over- lying layer, and the location of the wedge toe. Figure 3 shows a typical vertical profile and a typical longitudinal distribution of salinity, as observed in the Duwamish River estuary. The nearly constant salinity in the wedge suggests that it is a well-mixed zone of turbulent flow. The overlying layer, which carries the freshwater from the river out to the ocean, is strongly stratified because the salinity, and therefore the densi- ty, decreases almost linearly throughout the layer. Therefore, turbulence and vertical mixing may be as- sumed to be strongly suppressed in the overlying layer. These observations form the basis of the flow model. The wedge is modeled as a sectionally homogeneous tidal flow, using a technique developed by the writer in a study of Bolinas Lagoon, Calif. (Fischer, 1972). The overlying layer is assumed to float back and forth on top of the wedge. The restricted nature of vertical mixing in the overlying layer is modeled by dividing the layer into sublayers and assuming that entrainment between sub- Sublayer \ ' Upper \ layer Salt wedge Mouth of estuary l l l l ' II I l"¢o.l.os, l I as l l I (“F 0f3 + 053 ‘l 0f2 + 052 <——Of F IGL'RE 2.—ldealized longitudinal section through a salt—wedge estuary, showing divisions used for modeling. DESCRIPTION OF THE MODEL 0 l I I I 0 .— ‘_ 2 m |._ DJ In IO - — E E - - 4g E _ _ z - - - 6— I 20— "‘ I'- I? I; ‘5’ '_ _' 8g 30 -— 110 I I I I I 0 IO 20 30 SALINITY, IN PARTS PER THOUSAND RIVER KILOMETRE O I 2 4 5 6 7 8 9 IO N I2 I l | I l I II I I I || | l I | l I O .— Downstream O \ I; _ 2 (3:3 3 IO -— 4 IE E =_ 6 2 20 g :r: - 8 I" 30 — .. _ I 33 '0 I: 40 -— '2 L3 50 I I I I I4 0 I 2 5 6 7 8 4 RIVER MILE FIGURE 3.—Typica1 observed profile (upper graph) and longitudinal distribution (lower graph) of salinity in the Duwamish River estuary. Lower graph is from Dawson and Tilley (1972, fig. 5). layers and from the wedge into the lowest sublayer occurs at empirically determined rates. FLOW IN THE WEDGE The wedge is treated as a turbulent open-channel flow, except that water is removed by entrainment through the upper surface and through the vertical face of the toe. The wedge is divided into elements in the same way as was the water in the channels in Bolinas Lagoon (Fischer, 1972). A typical element is shown in figure 4. An element is defined as a homogeneous vol- ume of water occupying the entire channel cross section and extending along the channel axis a distance deter- mined by dividing the element volume by the channel cross-sectional area. An element is assumed to main- tain its coherence as it moves back and forth along the channel axis so that in the absence of entrainment or longitudinal mixing the element would always contain the same water; the effect of entrainment is to continu- ously remove water through the upper surface of each element, thereby reducing its volume. Mixing is al- lowed only with adjacent elements. Thus, the concen- tration of any constituent carried by the water is mod- ified only by mixing and by appropriate biochemical reactions. 4 NUMERICAL MODEL OF MATERIAL TRANSPORT IN SALT-WEDGE ESTUARIES Z <___ Subloyer block i, 3 .._ —— 4—+ +—1 —1— H —¢—— 4—-— Advective _ Subloyer block i 2 Advective flow ’ flow —— 4—4 4— 1~1— t—4— —¢—— 4— -~ +— Subloyer block i, l «.— l l l l i l l l 1 Top of wedge <—-> Entrainment +_. Element i+l ‘"’ Elementi «T Element i—l Mixing MBITng 77—7. FIGURE 4.—Typical element, showing transport processes included in the model; settling velocities of suspended particles also are included in To find the location of a given element the program starts with element 1, which is at the toe of the wedge, and computes its length by dividing its volume by the cross-sectional area of the wedge; then the location of the upstream end and the length of element 2 are com- puted, and so on, to the mouth of the estuary. Wedge cross-sectional areas are computed from the geometry of the given estuary and wedge, as discussed in the section entitled "Model Operation.” During tidal inflow the outermost element moves upstream from the estuary mouth, and a new element is formed from water that enters the estuary during each time step. (A 15-minute time step has been found convenient.) During tidal outflow, water leaving the estuary mouth is assumed to be fully mixed with the sea water and is removed from the computation, and the number of elements or the volume of the outermost element is reduced. Transfer between the wedge and the various upper layers is modeled by assuming entrainment velocities upward across the bounding surfaces. The entrainment velocities must be found empirically; theoretical studies and laboratory measurements such as those of Turner (1968) and Kato and Phillips (1969) do not provide an adequate basis for estimating the rate of turbulent en- trainment across a density interface in a real estuary. the model. On the other hand, the technique developed by Stoner (1972') may be used to compute entrainment velocities from detailed salinity and longitudinal-velocity mea- surements at an estuary cross section. These computed entrainment velocities are supplied to the program as observed data and can be adjusted if necessary to im- prove the program’s modeling of salinity. In the model the volume of each wedge element is reduced after each time step by an amount calculated as AV = UelWL At, (1) where U e1 is the entrainment velocity between the wedge and sublayer 1, W is the channel width at the top of the wedge, L is the length of the element, and At is the duration of the time step. Because of entrainment, the volume of each wedge element continuously decreases. To avoid having a large number of small elements, an element reduction subroutine is included. Whenever an element becomes shorter than an arbitrary length of 200 feet (60 m), the element is combined in volume with the shorter of the adjacent elements to form one new element; all the identifying numbers of the elements seaward of the combination are then reduced by one. The same element reduction, using the shortest two elements, is made DESCRIPTION OF THE MODEL 5 when the number of elements exceeds the program stor- age capability of 50 elements. A further constraint is for a special case that can occur only in element 1, the farthest upstream. Because the model provides for flow from this element out of the upstream end as well as upward into the upper layer, the flow out during a time step may exceed the volume contained in the first ele- ment. If the model determines that the volume of the first element will be less than the volume expected to flow out of it during a time step, the first element is combined with the adjacent downstream element. The way in which the flow in the wedge is modeled may be summarized from the point of view of a water particle, as follows: A water particle that enters the wedge on a floodtide becomes part of a wedge element and is carried upstream. On the ebbtide the particle moves back downstream, in most cases leaving the es- tuary not to return. A particle that enters the estuary at the beginning of the floodtide, however, may be in an element that remains in the wedge after the following ebb. It then begins a backward and forward progression, on each flood moving slightly farther landward than it returns seaward on each ebb. Occasionally longitudinal dispersion may cause it to move from one element to the adjacent one, but on the average the particle will main- tain a net landward drift. Finally, a time comes when the particle is entrained from the wedge into the first sublayer, and possibly farther into the higher sub- layers. When this happens the net drift becomes sea- ward, and the particle is carried back to sea. FLOW IN THE OVERLYING LAYER The freshwater discharge of the river mixes with the saltwater entrained from the wedge to flow as a mixed layer above the wedge. Because this layer is strongly stratified, it is modeled in the flow model as a series of sublayers, each with its own fluid velocity. The total freshwater discharge is allocated between sublayers to obtain the best possible agreement between observed and predicted salinity distributions. The sublayers be- gin directly over the toe of the wedge. Because the toe of the wedge is taken to be a vertical face (fig. 2), and because in reality the estuary upstream from the toe of the wedge is characterized by considerable vertical saltwater transport, each sublayer is given an initial flow of saltwater as well as freshwater. The total volume of saltwater put into the layers is taken from element 1 of the wedge, further reducing the volume of that ele- ment during each time step. Thus, the total discharge in sublayer j at the seaward end of element i, assuming no accumulation of water in sublayer blocks, is given as i Q/i,j:QfJ-+Q5j+‘Uej—Uej+1)haw/9L}?’ (2) where Q fjis the freshwater discharge into the upstream end of the layer, Q sj is the saltwater discharge into the upstream end of the layer, Uej is the entrainment veloc- ity through the bottom of the sublayer, Uej+1 is the entrainment velocity out of the top of the sublayer (zero for the uppermost sublayer), and Wk and L k are the top width and length of element k. Note that the term on the right side of the equation, which accounts for entrain- ment between sublayers, assumes that the channel width at each sublayer boundary is the same. This cor- responds to assuming that the estuary has vertical sides above the top of the wedge; the error introduced will depend on the geometry of the estuary. COMPUTATION OF SALINITIES The salinity of the wedge is assumed to be that of the ocean, or whatever body of water is at the estuary mouth. This corresponds to assuming that entrainment is a one-way process from the wedge to the upper layer, as verified by the observed salinity distributions shown in figure 3. The salinity of the water entering the up- stream end of each overlying sublayer is that resulting from complete mixing of the saltwater and freshwater inputs, given as sj=QsjsWAQsj+Q9> , (3) in which S w is the wedge salinity. Complete mixing is assumed within each block of a sublayer. For any block the salinity at the end of a time step is given as , - At. . .- .. S i,j‘Si,j+Vli{QLFIJ-Sl—IJ in’jSU +WiLi (UejSi,j-1—Uej+1si,j)} , (4) where S i,j is the salinity in the block at the beginning of the time step, At is the length of the time step, and VI, is the volume of each sublayer block above wedge element i. This equation expresses the conservation of salt; it states that the salt in a block at the end of the time step is equal to that at the beginning of the step, plus that brought in by the flow through the upstream end of the block, minus that carried out by flow through the downstream end, plus that entrained from the next lower sublayer, minus that entrained into the next up- per sublayer. The salinity computation for each time step begins at the landward end of the wedge and pro- gresses seaward. Salinities are determined essentially by using an explicit backward-difference—finite-differ— ence solution of the advective-transport equation. Dispersion due to flow within the sublayers is not explicitly modeled; however, the numerical scheme does generate a substantial amount of numerical dispersion of the sort discussed by Bella and Grenny (1970). 6 NUMERICAL MODEL OF MATERIAL TRANSPORT IN SALT-WEDGE ESTUARIES THE MATERIAL-TRANSPORT MODEL The model includes provisions for study of the motion, decay, and reactions of suspended and dissolved biologi- cal and chemical constituents. The transport and mix- ing of these constituents is assumed to follow that of salinity, which has already been discussed, but addi- tional programming is included for constituents which have a settling velocity, and for chemical and biological reactions between constituents. In particular, the model is equipped to simulate the motion of saltwater plankton which enter the wedge from the sea, and which are entrained into the flowing layers before possibly settling back into the wedge. The model also is equipped to study the effect of oxygen demand by plankton or other sources of BOD on DO levels in the wedge. Dis- solved oxygen in the upper layers has not been included in this model. The upper layers seldom have a low-DO problem; however, the DO deficit in the wedge often represents a severe environmental problem 'because there is no source of oxygen for the wedge, and oxygen consumed in the wedge by any source of BOD can only be replaced by mixing with aerated water from the ocean. Plankton are assumed to enter the wedge with each inflow of ocean water, and the plankton concentration in each newly formed wedge element is given as that of the ocean. Within the wedge the plankton concentration at the end of each time step, 0,, is given as C',=(1—kd)c,+%,1_{ic,l~c,) W,L,U$} l +E(C,-+1+Ci-1—2C,~), (5) in which Ci is the concentration of plankton in wedge element i, at the beginning of the time step, Ci,1 is the concentration in the first sublayer above element i, kd is a dimensionless decay coefficient for plankton in the wedge, Vi is the volume of wedge element i, U S is the sinking speed of the plankton, and E is an exchange coefficient between wedge elements due to longitudinal dispersion. On the right side of this equation the first term represents a first-order decay, the second term represents addition to the wedge by settling from above and reduction by settling to the bed, and the third term represents longitudinal dispersion due to the velocity gradient in the wedge. The coefficient E is obtained in the same way as described by Fischer (1972). Within the upper layers, the change in plankton con- centration is given by an equation similar to that for salinity, but with additional terms expressing settling and growth, as follows: C'i,j=(1+ajkg)C- +£ler Ci-1;J'_Qli,jci L] V; l—1,j ,J' U At . , +WiLi (UejCi,j*l_Uej+1Ci,j-)} +5}; “(Ci’j+1_Ci’j) , (6) where aj is a factor to account for light attenuation in layer j, kg is a growth coefficient, C i J is concentration in the jth sublayer above element i, and h i is the thickness of each sublayer. (Note that the sublayers have equal thicknesses.) Other constituents can be modeled in the same way as plankton, using appropriate growth, decay, and settling coefficients. For instance, D0 is modeled by assuming a Sink equal to the BOD in each wedge element. Although the program written for the present model includes only plankton and DO, extension to other constituents and other sources would be straightforward. MODEL OPERATION This section summarizes the inputs required by the model, its method of verification, and the results ob- tainable. A simplified flow diagram is shown in figure 5. The procedure is as follows: 1. The program is given a physical description of the estuary, in the form of tables of width ver- sus depth at a number of longitudinal stations, the tidal variation at the mouth, and the fresh- water discharge. This information is usually available from published records. The program includes an interpolation subroutine to com- pute the channel cross-sectional area versus depth at any location. 2. The thickness of the upper layer at the mouth, the slope of the interface between the upper layer and the wedge, and the number of sub- layers are specified. According to Stommel and Farmer (1952), the thickness of the upper layer at the mouth can be computed by setting the interfacial Froude number equal to unity. In practice, however, measurements of the loca- tion of the interface are desirable. 3. The location of the toe of the wedge is specified as a function of time throughout the tidal cycle. Field measurements of the location throughout the cycle are preferable, although in their ab- sence an estimate can be made based on the known location of the toe at one time and an estimate of the tidal excursion. 4. The total freshwater discharge is divided among the sublayers. An equal division can be used as an initial estimate. 5. Estimates are given of entrainment velocities Reset time index=1. Increment cycle index. Run another cycle with same input data? Run another cycle with new input data? Combine the two shortest elements. DESCRIPTION OF THE MODEL - Read estuarine geometry and control parameters. Read constituent concentrations in seawater, flow data, toe locations, and tidal elevations for one tidal cycle. Is this the first cycle? Initialize element volumes and concentrations. Compute top width and area of the wedge at each section throughout the tidal cycle. Set cycle index=1; Set time index=0, to begin a tidal cycle. Increment the time index. Is this the end of a tidal cycle? Set element index=1 (toe of wedge). Compute area, length, width, and location of an element. Is the downstream end of the element downstream from the estuary's mouth? Are there any more elements? Increment the element index. Create a new element and assign it seawater concentrations. Compute the area, volume, and length of the last element, terminating at the estuary mouth. Is any element shorter than 200 feet (60 m), or are there more than 49 elements? ntrations of wedge elements caused by Compute changes in conce n elements and biochemical reactions. diffusive transport betwee Compute the volumes of each sublayer above each wedge element. In sequence, compute the flow in each sublayer element; compute the salinity and plankton concentration in each sublayer element by equations (4) and (6); compute the plankton concentration in each wedge element by equation (5). Decrease the volume of each wedge element to account for entrainment into the overlying layer. Print results at selected time steps. FIGURE 5.—Simp1ified flow diagram of computer program. 8 NUMERICAL MODEL OF MATERIAL TRANSPORT IN SALT-WEDGE ESTUARIES between the wedge and the lowest sublayer, and across each interface between sublayers. 6. An estimate is given of the amount of saltwater transferred from the toe of the wedge into the beginning of each sublayer. A first estimate of these quantities must be a guess, to be im- proved in the verification stage. 7. The program is operated with the data described in steps 1 through 6, and a salinity distribution is generated. The model salinity distribution is compared to a distribution observed in the pro- totype, and adjustments are made to the esti- mates called for in steps 4 through 6 to achieve as close a verification as possible. If possible, prototype salinity distributions should be ob- served for a range of tidal and freshwater- discharge conditions, so that the values called for in steps 4 through 6 can be expressed as functions of tidal range and freshwater dis- charge. Assuming that an adequate verification has been ob- tained, step 7 completes the description and verification of the flow model. The material-transport model re- quires specification of biological parameters; for in- stance, to model the growth of plankton the program must be given the following parameters: 1. Settling velocity. 2. Growth rate under ideal light conditions. 3. Reduction in growth rate with reduced solar radiation. 4. Attenuation coefficient for reduction of solar in- tensity with depth beneath the water surface. 5. Decay rate in the wedge. Because none of these parameters are accurately known for any species of plankton, and because many species may be present, it is apparent that one cannot expect a high order of accuracy in the prediction of biological constituents. The verified model can be used to trace the movement and dispersion of any constituent introduced into the estuary. For example, if BOD is being introduced into the wedge from a constant source its distribution and decay can be traced, and its effect on DO can be pre- dicted. The distribution, growth, and decay of plankton also can be traced—provided the biological parameters can be estimated with sufficient accuracy—and it may be possible to predict the growth and decay of plankton blooms. A second major use of the program is prediction of the effect of changing the estuarine geometry, such as extending a dredged channel or deepening an existing one. This type of prediction requires that some of the parameters, such as the entrainment velocities, do not change significantly when the geometry is changed, and it also requires extrapolation of the wedge geometry to fit the changed estuary geometry. Within these limits, however, the program can be useful in providing an informed estimate of the effect of changes in geometry. For example, if the length of a dredged channel is in- creased, so that the length of the wedge is increased, the residence time for water in the wedge will increase and the program can be used to predict the further decrease of DO concentrations. SUMMARY AND CONCLUSIONS This paper has described a numerical program which has been used to model the transport of salt and other constituents in one salt-wedge estuary, the Duwamish River estuary at Seattle, Wash. The results of the model application are given in Part II of this report. The pro- gram can be used to predict the distributions of such constituents as BOD, DO, and plankton, and to predict the effect of changes in the estuarine geometry on these distributions. It is hoped that the program will prove useful in studies of other salt-wedge estuaries. REFERENCES CITED Bella, D. A., and Grenny, W. J ., 1970, Finite difference convection errors: Am. Soc. Civil Engineers Proc., Jour. Sanitary Eng. Div., v. 96, no. SA6, p. 1361—1375. Dawson, W. A., and Tilley, L. J., 1972, Measurements of salt-wedge excursion distance in the Duwamish River estuary, Seattle, Washington, by means of the dissolved-oxygen gradient: U.S. Geol. Survey Water-Supply Paper 1873—D, p. D1—D27. Fischer, H. B., 1972, A Lagrangian method for predicting pollutant dispersion in Bolinas Lagoon, Marin County, California: U.S. Geol. Survey Prof. Paper 582—B, p. B1—B32. Fukushima, H., Yakuwa, I., and Takahashi, S., 1969, Salinity diffu- sion at the interface of stratified flow in an estuary: Internat. Assoc. Hydraulic Research, 13th Cong, Kyoto, Japan, 1969, Proc., v. 3, p. 191—198. Kato, H., and Phillips, 0. M., 1969, On the penetration of a turbulent layer into stratified fluid: Jour. Fluid Mechanics, v. 37, p. 643. Keulegan, G. H., 1966, The mechanism of an arrested saline wedge, in Ippen, A. T., ed., Estuary and coastline hydrodynamics: New York, McGraw-Hill, chap. 11. Santos, J. F., and Stoner, J. D., 1972, Physical, chemical and biological aspects of the Duwamish River estuary, King County, Wash- ington, 1963—67: U.S. Geol. Survey Water-Supply Paper 1873—C, p. Cl—C74. Stommel, Henry, and Farmer, H. G., 1952, On the nature of estuarine circulation: Woods Hole Oceanog. Inst. Rept., reference nos. 52— 51, 52—63, and 52—88, 7 chap. Stoner, J. D., 1972, Determination of mass balance and entrainment in the stratified Duwamish River estuary, King County, Washington: U.S. Geol. Survey Water-Supply Paper 1873—F, p. F1-F17. Turner, J. S., 1968, The influence of molecular diffusivity on turbulent entrainment across a density interface: J our. Fluid Mechanics, v. 33, p. 639—656. Model Computation of Salinity and Salt-Wedge Dissolved Oxygen in the Duwamish River Estuary, King County, Washington By J. D. STONER, W. L. HAUSHILD, and J. B. MCCONNELL GEOLOGICAL SURVEY PROFESSIONAL PAPER 917 Prepared in cooperation with the Municipality of Metropolitan Seattle CONTENTS Page Page Abstract __________________________________________________ 13 Salinity __________________________________________________ 21 Introduction ______________________________________________ 13 Observed data ________________________________________ 21 Acknowledgments ________________________________________ 15 Modeling salinity ______________________________________ 21 Duwamish River estuary __________________________________ 15 Dissolved oxygen __________________________________________ 22 Tides ________________________________________________ 17 Input and output DO __________________________________ 22 Riverflow ____________________________________________ 17 Modeling DO __________________________________________ 25 Duwamish River estuary model ____________________________ 17 Predicting DO ________________________________________ 26 Estuary geometry ____________________________________ 17 Discussion ____________________________________________ 30 Wedge toe location ____________________________________ 17 Sensitivity analyses ______________________________________ 31 Wedge elements ______________________________________ 17 Summary and conclusions __________________________________ 33 Entrainment __________________________________________ 17 Definition of terms ________________________________________ 35 Upper-layer thickness __________________________________ 17 References cited __________________________________________ 36 Inflow and outflow ____________________________________ 19 Tide stages ____________________________________________ 20 ILLUSTRATIONS Page FIGURE 1. Map showing study area _____________________________________________________________________________________ 14 2. Graph showing means and extremes in monthly and annual discharges of Green River at Tukwila -_ _______________ 16 3. Graph showing mean daily discharges of Green River at Tukwila _______________________________________________ 18 4. Diagrams showing idealized longitudinal geometry of the estuary _______________________________________________ 19 5—17. Graphs showing: 5. Curves used for locating the wedge toe _________________________________________________________________ 20 6. Vertical-velocity profiles in the estuary _________________________________________________________________ 21 7. Daily extremes in dissolved-oxygen concentration at Spokane Street Bridge for J une—September, 1967—69 ________ 23 8. Daily extremes and ranges in dissolved—oxygen concentration at Spokane Street Bridge for June—September, 1969—71 ___________________________________________________________________________________________ 24 9. Frequency distributions of daily maximum dissolved—oxygen concentration at Spokane Street Bridge for August- September, 1967—69 and 1970—71 ___________________________________________________________________ 25 10. Daily extremes in dissolved—oxygen concentration at 16th Avenue South Bridge for J une—September, 1967—69 _____ 26 11. Daily extremes in dissolved-oxygen concentration at 16th Avenue South Bridge for J une—September, 1970—71 _____ 27 12. Frequency distributions of daily maximum and minimum dissolved-oxygen concentration at 16th Avenue South Bridge ___________________________________________________________________________________________ 28 13. Computed and observed dissolved-oxygen concentration at 16th Avenue South Bridge _____________________ 29 14. Predicted reduction in dissolved-oxygen concentration at 16th Avenue South Bridge _______________________ 30 15. Tide stage during the tidal cycle used in the sensitivity analyses _________________________________________ 32 16. Computed sublayer salinity during a tidal cycle for various sublayer saltwater inflows _____________________ 33 17. Computed dissolved-oxygen concentration during a tidal cycle for various thicknesses of the upper layer __________ 34 TABLES Page TABLE 1. Model oxygen-use rates in wedge, mean wedge DO concentrations, and statistics of regressing computed versus observed DO concentrations in salt wedge at 16th Avenue South Bridge during J une— September periods of 1967-71 __________ 30 2. Mean changes in sublayer salinities at 16th Avenue South Bridge during a 2-month period, relative to base values of sub- layer salinities, for selected changes in the controlling parameters used in the Duwamish River estuary model ________ 32 3. Mean changes in D0 concentration in the saltwater wedge at 16th Avenue South Bridge during a 2-month period, relative to base values of DO concentrations in the wedge, for selected changes in the controlling parameters used in the Du- wamish River estuary model _____________________________________________________________________________ 34 ll PART II. MODEL COMPUTATION OF SALINITY AND SALT- WEDGE DISSOLVED OXYGEN IN THE DUWAMISH RIVER ESTUARY, KING COUNTY, WASHINGTON By J. D. STONER, W. L. HAUSHILD, and J. B. MCCONNELL ABSTRACT Saltwater from Elliott Bay on Puget Sound forms a wedge in the lower part of the Duwamish River estuary. The numerical model described by Fischer in Part I of this report was used in computing salinity distributions in the estuary, and oxygen-use rates and dissolved—oxygen distributions in the salt wedge. Computed spatial distributions of salinity agreed well with observed distributions dur- ing about 30 slack tides in July and August 1968. Analyses of the sensitivity of computed salinity to changes in model input parameters indicate that salinity changed most in response to changes in the wedge salinity and the location of the wedge toe. The rate of use and the concentration of dissolved oxygen (D0) in the salt wedge were computed by the model for J une—August 1968 and for the J une—September periods of 1967 and 1969—71. Before 1970, the estuary received discharges of treated, partly treated, and raw indus- trial and municipal wastes; after 1970, the only major source of waste was the effluent from the Renton Treatment Plant, a secondary treatment plant. Attributable to these changes in waste disposal to the estuary were (1) observed wedge DO concentrations generally 2 mg/l greater in 1970—71 than in 1967-69, and (2) oxygen-use rates in the wedge 60 percent greater during 1967-69 than during 1970—71. Analyses of covariance indicate that computed wedge DO concentra- tions were not different (95-percent confidence level) from observed concentrations, and the standard error of estimate of the computed concentrations ranged from 10 percent (1971) to 22 percent (1967) of the observed mean concentrations. Sensitivity analyses indicate that wedge DO concentration changed proportionally with oxygen-use rate and also was sensitive to changes in the wedge toe location and in the velocity of the water entrained from the wedge. The model was used to predict the changes that would have occurred in the oxygen-use rate and DO concentrations in the wedge during J une—September 1971 if discharge of Renton Treatment Plant effluent had been increased from a 1971 average of 37 fta/s (63 m3/min) to the planned maximum of 223 ft3/s (379 mein). The predictions suggest that (1) the oxygen—use rate would have been increased by 92 percent, (2) a relatively low DO concentration (4 mg/l) would have been decreased by 45 percent, and (3) a relatively high concentration (9 mg/l) would have been decreased by 8 percent. INTRODUCTION The Duwamish River estuary, which is the lower part of the Green-Duwamish River and the important in- dustrial waterway in Seattle, Wash. (fig. 1), is undergo- ing a change in the patterns of its waste-water inputs. The estuary has been receiving industrial and munici- pal wastes since the early 1900’s, but the effects of such disposal were not considered serious until the 1940’s. Later increases in wastes resulting from population and industrial expansion degraded the quality of the es- tuarine water to the extent that endangerment of the fisheries resources and aquatic life in the estuary was of concern to local, State, and Federal agencies and private groups. In 1958 the people in the greater Seattle area voted to form the Municipality of Metropolitan Seattle, gener- ally referred to as Metro. Metro is a federation of a number of towns and cities which united to deal with the growing problems of waste-water disposal in the area. Metro’s comprehensive plan for water-pollution control resulted in construction of an extensive network of sewer trunklines and several sewage—treatment plants, including the Renton Treatment Plant (RTP) near the head of the Duwamish River estuary. In the 9 years before August 1970, discharges of raw or partly treated sewage into the Duwamish River es- tuary and into Puget Sound along the Seattle water- front (including Elliott Bay) progressively were being intercepted and pumped to primary treatment plants (not shown in fig. 1) which discharge effluent into Puget Sound. In June 1965, discharge of effluent from RTP began and increased progressively thereafter. In a re- port prepared by the Metropolitan Engineers (1971), Metro states that "the last [major] raw-sewage dis- charge to the Duwamish River and Elliott Bay was intercepted in August 1970.” Although several sources outside Metro’s jurisdiction may have continued to dis- charge minor quantities of sewage into the estuary, the principal waste discharge into the estuary after August 1970 has been the effluent from RTP. The RTP discharges its effluent into the lower part of the Green River about 1 mile (1.6 km) upstream from its confluence with the Black River. Downstream from the confluence (fig. 1), the Green River becomes the Duwamish River. Sewage received mostly from areas east of Lake Washington but also from areas south and east of RTP is given secondary treatment (activated- 1?) —<_ 14 ' NUMERICAL MODEL OF MATERIAL TRANSPORT IN SALT WEDGE ESTUARIES I22°3o' I22°I5’ El. L /077 84 Y I I (PUGET SOUND} 47° _ 35 ' 505/ Walerway West I Waferway ‘ o Spokane Street Bridge / # DRM |.2 (i.9km) Model Station 0 DRM O.75(i.2km) SEATTLE & /First Avenue South Bridge DRM 3 4(5. 5km) o (2» V47 |6ih Avenue South (9 Bridge ’5’ DRM 4.8 (7.7 km) EXPLANATION O DRM 3. 4(5 5km) Duwamish River miles(kilomefres) upstream from mouth :3 < m :3 Boeing Bridge sife / DRM 6.5(10.5km) Easi Marginal Way Bridge / DRM 7.(8 (i2.6km) 47D _ 3° WASHINGTON Black River DRM ||.9 ¥ (I9.| km) Renion x" Treatmen? Plant 1 2MILES ‘ i II | I I I ‘FJ l 2 5KILOMETRES ‘\V § (:RM l2.8 (20.6km) % Tukwila Gaging Siation\ I) GRM|3|(2|.Ikm) 1; m l m FIGURE 1 —Elhott Bay, Duwamish Rlver estuary, and downstream reach of Green Rlver MODEL COMPUTATIONS IN'THE DUWAMISH RIVER ESTUARY, WASH. sludge method) at RTP. Included in the plant effluent are the nutrients, dissolved oxygen-demanding mate- rial, and suspended solids not removed during treat- ment of the sewage. Grit, sludge, and scum removed during treatment are not discharged into the Green— Duwamish River. Grit is trucked to a landfill, and sludge and scum are pumped to the primary-treatment plant at West Point (not shown in fig. 1). By 1972, the design maximum of 37 ft3/s (63 m3/min) of discharge from the first-stage facilities at RTP had been reached and additional facilities were being constructed to dou- ble the sewage-treatment capacity. The RTP has been designed to ultimately discharge 223 ft3/s (379 m3/min) of effluent. In 1963, Metro and the US. Geological Survey to— gether began a study to evaluate the effects of changes in waste disposal on water quality of the salt-wedge Duwamish River estuary. In the estuary, the tides and freshwater inflow cause water circulation and concen- trations of dissolved and suspended material to con- tinually change in space and time. The dual inflow of water—saltwater from Elliott Bay and freshwater from the Green-Duwamish River—further complicates the transport of water and dissolved and suspended con— stituents in the estuary. Consequently, Metro and the Geological Survey decided in 1971 to proceed with the development of a numerical model for use in estimating water circulation and constituent transport in the estuary. The model described in Part I of this report was de- veloped by Fischer for application to salt-wedge es- tuaries. The purpose of Part II is to present some of the results derived from adapting and applying Fischer’s general model to the Duwamish River estuary. Part 11 includes a description of the Duwamish River estuary and information and data peculiar to the model of the estuary that supplements the description of the general model given in Part I. The modeling reported here is restricted to salinity distributions within the downstream reach of the estuary and to the rate of use and distribution of DO (dissolved oxygen) in the es- tuary’s salt wedge. From 1967 to 1971, only the June— September periods of relatively low D0 in the salt wedge were modeled. These years include (1) those dur- ing which waste entering the estuary was the discharge of treated effluent from RTP and the discharges of raw or partly treated sewage from industrial and municipal outfalls and (2) those in which waste was primarily the RTP effluent. Therefore, comparisons of results for these years indicate the relation of oxygen-use rate and DO concentration in the salt wedge to the type and quantity of waste received by the estuary. Also evaluated are the sensitivities of salinity in the estuary 15 and DO concentration in the salt wedge to changes in the controlling parameters of the model. The model was used to predict the oxygen—use rates and DO concentra- tions in the salt wedge for a future time, when RTP will be discharging an ultimate-design quantity of effluent. Finally, other probable and possible uses of the model are discussed. ACKNOWLEDGMENTS The authors are grateful for the continuing support and encouragement of Glen D. Farris, superintendent of the Water Quality and Industrial Waste Division of the Municipality of Metropolitan Seattle. In applying the model to the Duwamish River es- tuary, the authors benefited considerably from the suggestions, criticisms, and guidance provided by H. B. Fischer of the Department of Civil Engineering of the University of California at Berkeley. DUWAMISH RIVER ESTUARY The Duwamish River, which is an extension of the Green River downstream from its confluence with the Black River, generally flows northwestward and dis- charges through its East and West Waterways into El- liott Bay of Puget Sound (fig. 1). Most of the water flows in the West Waterway, because sediment deposits con- strict the channel of the East Waterway near and under the Spokane Street Bridge. The West Waterway is an extension of a waterway in the Duwamish River; ships and barges use these waterways upstream to about the First Avenue South Bridge at DRM 3.4 (Duwamish River mile 3.4; km 5.5)1 and barges go upstream to about DRM 63 (km 10.1). Regardless of tide stage, saltwater from Elliott Bay does not intrude up the Duwamish River to the East Marginal Way Bridge (DRM 7.8; km 12.6) when river discharge is more than 1,000 ft3/s (1,700 m3/min), but it intrudes at least that far upstream during most flood- tides when the river discharge is less than 625 ft3/s (1,060 m3/min) (Stoner, 1967). Salt has been observed in the river water at DRM 10.2 (km 16.4) during some periods of low discharges and high high tides. The far- thest upstream intrusion of saltwater is not known, but saltwater probably only rarely intrudes as far as the Tukwila gaging station at GRM 13.1 (Green River mile 13.1, km 21.1). For various river flows and tides, the upstream limit of tide effect on stage and flow of the Green-Duwamish River also is unknown; however, high tides usually affect stage and flow at the Tukwila gag- ing station at all discharges less than about 7,000 ft3/s ‘River miles in this report agree with those reported in previous publications; agreement was obtained, with a few exceptions, by adding 0.9 mile (1.4 kmito the river miles reported by the Pacific Northwest River Basins Commission (1969). 16 (12,000 m3/min). The farthest upstream point at which the flow may reverse directions also is unknown; re- verse flow has been observed so seldom at the Tukwila gaging station that its occurrence there is presumably rare. Observed spatial distributions of salinity indicate that saltwater from Elliott Bay intrudes as a wedge into the lower Duwamish River estuary for all river flows and tides (Dawson and Tilley, 1972; Santos and Stoner, 1972). The part of the estuary included in the model is that downstream from the wedge toe, which is defined as the farthest upstream cross section Where salinity of the wedge water is 25 ppt (parts per thousand). The river reach within which the wedge toe moves upstream and downstream with the tides varies with river discharge; graphs used in the model for determining the wedge toe location are given in the section describing the Duwamish River estuary model. A mixture of salt and fresh water from upstream of the wedge toe flows into the layer overlying the wedge. The overlying layer also receives the saltwater entrained from the wedge at the interface. Water in the overlying layer moves upstream and downstream with the tides but has a net downstream movement toward Elliott Bay. The wedge water entrained into the overlying layer or advected upstream of the wedge toe is replaced by inflowing salt- water from Elliott Bay. Therefore, although water in the wedge also moves upstream and downstream with the tides, wedge water has a net upstream movement away from Elliott Bay. TIDES The tides that originate in the Pacific Ocean affect the flow and level of water in the estuary and usually cause two high tides (high high and low high) and two low tides (high low and low low) in a day. Heights of succes- sive high or low tides normally are unequal; the largest inequality is in heights of the low tides, although the heights of two high tides differ considerably on some days. At times, the heights of the two high tides of a day differ by more than 4 ft (1.2 m) and the heights of the two low tides differ by more than 8 ft (2.4 m). Data from the National Oceanic and Atmospheric Administration (1973, table 2) indicate that the ranges and level of the tide heights at DRM 3.4 (km 5.5) are as follows: (1) the difference in height between mean high tide and mean low tide (mean range) is 7.5 ft (2.3 m); (2) the difference in height between mean higher high tide and mean lower low tide (diurnal range) is 11.1 ft (3.4 m); and (3) the elevation midway between mean low tide and mean high tide (mean tide, or halftide, level) is 6.5 ft (2.0 m). Recorded tide heights have ranged from minus 4.6 to plus 14.7 ft (—1.4 to 4.5 m). In this report, the datum from which all elevations are measured is the mean NUMERICAL MODEL OF MATERIAL TRANSPORT IN SALT-WEDGE ESTUARIES lower low tide, which is 6.6 ft (2.0 m) below mean sea level. RIVERFLOW Since September 30, 1961, the discharge of the Green River has been determined at the Tukwila gaging sta- tion (fig. 1), above which the drainage area is about 440 mi2 (1,140 km2). Downstream from the Tukwila gage, the Green-Duwamish River receives some seepage of ground water, local runoff from precipitation and (or) snowmelt, and the effluent from the RTP. Mean monthly discharges for 1961—71 (fig. 2) indicate that discharge at the Tukwila gage usually is greatest dur- ing January—February, decreases in March, increases somewhat during April—May, decreases to a minimum in August, and then increases during September— December. The maximum and minimum monthly discharges shown in figure 2 indicate that flow within a specific season varies considerably from year to year, which is further illustrated by the variation in mean daily dis- charge at the Tukwila gage for each June—September 5 Maximum ‘ Mean -8 i'" Minimum E 0 D g z 04—- _ E m cn c: in E - -6 o. “.0. i— “Ia" (Di- om INS— — (r,_ <11: __ r < IU— __ i— i—_ In: 80 ‘ 05 _ 0) Eco " —<_) 8 - _ -4°m _ —— — _ in F32 _ <7 F“- < <0 38 __ 3 2 7' i— 8 at, r r .2, § ,7? 7;‘~ ‘2 8 l —‘ O l— F—i— — 7 I l— 2 / __ “ ——__. z 7 __7—7- _ / / / O OJFMAMJJASOND MONTH FIGURE 2,—Means and extremes in monthly and annual discharges of Green River at Tukwila during 1961—71 water years. MODEL COMPUTATIONS IN THE DUWAMISH RIVER ESTUARY, WASH. period during 1967—71 (fig. 3). Although the mean daily discharges during each period generally follow the usual variation pattern for this time of the year, the magnitude and timing of changes in discharge differ from year to year. DUWAMISH RIVER ESTUARY MODEL Because Part I of this report contains a description of the numerical program used for modeling the Duwamish River estuary, and the theory and operation of the model, these will not be discussed in detail here. This section describes the estuary geometry and the various inputs used in applying the general model to the Duwamish River estuary. ESTUARY GEOMETRY The geometry of the estuary is described by 11 cross sections which were determined from maps made in 1971 by the U.S. Army Corps of Engineers of the dredged part of the Duwamish River estuary, and from US. Geological Survey measurements of cross sections upstream from the dredged part. The plan and elevation views of the estuary (fig. 4) show a general perspective of the geometry used in the modeling. Tables of width versus depth of each cross section are used in the model to generate, for any tide stage, areas at any cross section within the estuary. WEDGE TOE LOCATION Observations of temporal and spatial distributions of salinity in the upstream part of the estuary and concur- rent tide and flow data indicate that the location of the wedge toe is a function of the tide stage and the freshwa- ter inflow. The interface between the wedge and the upper layer was distinct and wedge salinity remained relatively constant downstream from the farthest up— stream cross section at which salinity was 25 ppt (fig. 4). Therefore, the location of this cross section was used as the wedge toe location. Upstream of the wedge toe, the interface between the wedge and the upper layer tends to diffuse, and salinity decreases erratically in the up- stream direction. As will be described later in the re- port, a saltwater flow out of the wedge and a saltwater flow back into the upper layer at the wedge toe has been incorporated into the model. A family of curves that define wedge toe locations for a wide range of tide stage at DRM 3.4 (km 5.5) and for various freshwater inflows was developed from ob— served data (fig. 5). The model computes wedge toe loca- tions either by using the equations for these curves or interpolating for freshwater inflows between those given for the curves. WEDGE ELEMENTS At the beginning of any period modeled, the wedge 17 was divided into a number of elements of equal volume. The elements were assigned an initial volume such that the wedge did not contain too few or too many elements; an initial volume that caused division of the wedge into 30 to 40 elements was desirable, with 50 elements being the maximum. For the Duwamish River estuary, each wedge element was assigned an initial volume of 10 million R3 (280,000 m3); thereafter, the number and volumes of elements were determined by the numerical program described in Part I of this report. ENTRAINMENT Because of strong stratification, the upper layer was divided into three sublayers, with sublayer 1 adjacent to the wedge and sublayer 3 extending to the water surface (fig. 4). Salt—wedge water is therefore entrained into sublayer 1 and water in sublayers 1 and 2 is entrained into the respective overlying sublayers. Stoner (1972) estimated that the entrainment veloc- ity between the wedge and sublayer 1 varied from 3><10—5 to 10X10‘5ft/s (9X10—4to 30x10‘4cm/s). (The entrainment velocity was computed by dividing the volume of water crossing the upper surface of the wedge per unit of time by the area of the surface.) Stoner attributed the variations in the entrainment velocity to the variations in freshwater inflow and tidal-prism thickness. Tidal-prism thickness, which is a measure of the tidal exchange, is the difference between the sum of the elevations of the two high tides and the sum of the elevations of the two low tides during a tidal cycle. The equation used in the model to compute the entrain- ment velocity between the wedge and sublayer 1 is Ue1 = —5.2>< 10-7+2.6>< 10”6(TP)+3.6>< 10-8(Qf), where Ue1 is the entrainment velocity between the wedge and sublayer 1 in feet per second, Tp is the tidal- prism thickness in feet, and Qf is the freshwater inflow in cubic feet per second. In the Duwamish River estuary, the relatively large (and usual) difference in flow velocity between the wedge and the upper layer (fig. 6) probably is the dom- inant factor affecting entrainment between the wedge and sublayer 1 even though the fluid density tends to suppress entrainment the most at their interface (fig. 6). Entrainment velocities between sublayers were not ex- pected to be as great as U61 and were expressed as pro- portions of Uel. UPPER-LAYER THICKNESS Observed spatial and temporal distributions of salin- ity and concurrent flow and tide data indicate that the average tidal-cycle thickness of the upper layer at the mouth of the Duwamish River estuary is related to tidal-prism thickness and freshwater inflow. The rela- 18 DISCHARGE, IN THOUSANDS OF CUBIC FEET PER SECOND OIO NUMERICAL MODEL OF MATERIAL TRANSPORT IN SALT-WEDGE ESTUARIES I I I I I I I I I I l I967 O O3 : 2 I970 M o — 4 : |97I 2 I I I I I I I I I I JUNE JULY AUGUST SEPTEMBER FIGURE 3.—Mean daily discharges of Green River at Tukwila during the June—September periods of 1967—71. , IN THOUSANDS OF CUBIC METRES PER MINUTE DISCHARGE MODEL COMPUTATIONS IN THE DUWAMISH RIVER ESTUARY, WASH. 19 tionship between these variables is expressed by the water flowing into the upper layer at the wedge toe regression equation results from a mixing of the freshwater of the Green- Upper-layer thickness = 2.5 + 0.30 (Tp) + 0.003 (Qf) . Duwamish River and RTP with the saltwater advected In the model, the estuary has a level water surface and upstream of the wedge toe. Saltwater inflow to the upper the interface between the salt wedge and the upper layer is modeled as equal to the outflow from the salt layer slopes downward from mouth to toe at a rate of wedge at its toe. Freshwater inflow is computed as the 0.06 foot per 1,000 feet (0.06 m per 1,000 m) (Dawson sum of the daily mean discharge of the Green River at and Tilley, 1972). The sublayers are equally thick in the the Tukwila gaging station and the mean daily outflow model. from the RTP. During some low-flow periods the outflow from the RTP was as much as 10 percent of the total freshwater inflow. The relic Black River and the drain- In the model, saltwater from Elliott Bay is advected age area downstream of RTP usually contribute only into and out of the estuary by adding or subtracting negligible quantities of water to the Green-Duwamish elements at the mouth. The mixture of fresh and salt River during summer and early fall. INFLOW AND OUTFLOW DISTANCE ALONG ESTUARY, IN THODSANDS OF METRES 4 6 8 2 600% l I I I I II I I I I? I |12 _ - I50 400 — — _ - 100 U) J: F —% ‘5 32 LIJ 200 :‘ a) 8 —_ I— ll] 0.) 5O LLJ u. : Estuary center line 2 2 z m _ / __ ac) _ z __ O > O _ .- 0) d u- :I: S J: I _ .— .. I- E, 200 — i? <2 _ 50 9 3 ‘n 3 _ - I00 400 - — _ - 150 600 I0 _ Sublayer /W01er surface 25 ————— %——————-——————————-——a O_—-——-—-l———-—--——v—-—————-——— I "‘0 /Wedge foe m _ UJ ill —I0 — '25 c: “J . _ a" ' :9 — -— - .. I— u- _ ., :‘n ’ I, 3’, - LIJ Salt wedge _, . —50 2 E _20 — ,. . , , y, ’ ’ Z " --7.5 * c2) 3 — —3O — E E ‘-I0.0 g 3 a DJ -40 — 42.5 m Maximum tidal excursion _ —5o / 400 fI3/s, (680 m3/min) — —I5.o 'r‘fi AI - -I7.5 —60 I I l I I II | I I l I O 5 IO I5 20 25 30 35 4O DISTANCE ALONG ESTUARY, IN THOUSANDS OF FEET FIGURE 4.——Idealized longitudinal geometry of Duwamish River estuary during a mean tide of 6.5 ft (2.0 m) and a Green River discharge of 400 ft3/s (680 m3/min). 20 IN THOUSANDS OF FEET Y MOUTH, UPSTREAM DISTANCE OF WEDGE TOE FROM ESTUAR Hourly values of tide stage are input to the model; they were computed by using the method and tables of -O.5 0.0 0.5 I.O I.5 2.0 2.5 3. NUMERICAL MODEL OF MATERIAL TRANSPORT IN SALT-WEDGE ESTUARIES TIDE STAGE, IN METRES 34 32 30 N 00 26 24 22 N O I I 1| III]! [I I I I I I I , IN THOUSANDS OF METRES UPSTREAM DISTANCE OF WEDGE TOE FROM ESTUARY MOUTH O 2 4 6 8 IO I2 TIDE STAGE, IN FEET FIGURE 5.—Variation of wedge toe location with tide stage and freshwater inflow of the Duwamish River estuary. TIDE STAGES data published by the Environmental Science Services Administration (1968-71). The reference station used for computing tide stages was Seattle, Wash. MODEL COMPUTATIONS IN THE DUWAMISH RIVER ESTUARY, WASH. VELOCITY, IN CENTIMETRES PER SECOND O 20 21 —40 —20 0 4O 60 20 4O 60 VELOCITY, IN FEET PER SECOND —I O O I 2 | 2 O I I \ I I I \ I I O \\ \ \ \\ \ \ Upper layer Interface \\ _ \\ _ _ \\ _ 3 IO \\ _ — \ _ _ \\\\ '— I \ Interface \ I \ \ I \ I l Interface \ I ._ I _ _ \‘ _ — \ _ 6 20 - | — — |_ r— \ — 'I Tidal time I 3°” "edge 2 hrs. IO min. before high high (I) I— |‘ _ I hr. 30 min. before high low _ _ 0 hr. 30 min. before high low- 9 I3: 3130 I l I I l I I— u. 0 IO 20 30 0 IO 20 30 0 IO 20 30 “EJ E —20 O 20 4O —20 O O 20 4O 60 Z r—W‘ r—m .-—-—-I-'—‘I——‘-‘I — I -| O I -I O '- I- 0 I \ I O i 2. O I a. \ \\ \ " LIJ \\ \ \ D— 0 ~ - ~\\ \ Lu \‘ \\ Q \\ \ \ Interface \ Interface \ _ \ IO — \ — - '\ — “ 3 3 \ \ Interface \ ‘ I \ \ I \ \ I I I - | - - I — — I — 6 20 — | - — l - I I I I l _ . . ‘ 3 hrs. 20 min. after high high _ 0 hr. BOlmIn. after hIgh Iow _ I hr. 30 |ITIII'I. after hIgh low _ / I _ 9 30 ' | I I 0 IO 20 30 0 IO 20 30 0 IO 20 30 SALINITY, IN PARTS PER THOUSAND FIGURE 6.—Velocity (solid line) and salinity (dashed line) profiles at 16th Avenue South Bridge for various tidal times during September 6—7, 1967. Negative velocity indicates upstream flow. SALINITY OBSERVED DATA Salinity is relatively easy to measure in the estuary, and many data for determining its distribution in time and space are available. Because salinity in the estuary results from mixing essentially zero-salinity freshwater from the river with essentially constant-salinity water from the wedge, its value at any time at any point indicates the percentage of water originating from both sources. Consequently, verification of salinity distribu- tion in the estuary is a check of circulation and move- ment of fresh and salt water within the estuary. The data used for model verification of salinity dis- tribution in the estuary were 31 salinity profiles (salin- ity distributions in midstream verticals) at First Av- enue South Bridge and 30 salinity profiles at 16th Av- enue South Bridge during July and August 1968. The observed data were equally distributed between high slack and low slack tides. During the salinity observa- tions, the tidal-prism thickness ranged from 9.9 to 20.0 ft (3.0 to 6.1 m) and freshwater inflow ranged from 273 to 1,220 ft3/s (464 to 2,070 m3/min); inflow at the times for the majority of the profiles was between 350 and 500 ft3/s (590 to 850 m3/min). These ranges represent those anticipated in the Duwamish River estuary during the periods of low flow in summer and early fall. MODELING SALINITY Verification of salinity distribution consisted of de- termining values of four parameters in the model so that computed distributions agreed with observed dis- tributions. The model parameters were sublayer en- trainment velocities, sublayer freshwater inflows, Sub- layer saltwater inflows, and saltwater outflow. Values of these parameters in modeling salinity were varied within the following limits: (1) Computed salinities were to agree within 10 percent with observed salin- ities, and (2) even though they could not be measured or empirically estimated from observed data, the parame- ter values should fall within ranges that were roughly estimated from gross approximations of, and some as- 22 sumptions about, the physical processes occurring in the Duwamish River estuary. Several assumptions and approximations are consi- dered in modeling salinity. Freshwater flow into the upper layer at the wedge toe might be expected to be distributed approximately equally among the three sublayers; if not, less water would be expected to flow into the lower sublayers. Saltwater flow into the upper layer at the wedge toe might be expected to be distrib- uted such that less saltwater would flow into the higher sublayers. Entrainment velocities between sublayers were constrained such that less water was entrained out of than was entrained into a sublayer. Saltwater flow out of the wedge at its toe, Qs, is approximately 200 ft3/s (3'40 m3/min), which is about the midvalue of a range (140 to 270 ft3/s, 240 to 460 mein) computed from data given by Stoner (1972). The results of using many combinations of the four variable parameters in the model indicated that com- puted salinities agreed well with observed salinities for a Qs of 200 ft3/s (340 m3/min) and the following values for the other parameters: Sublayer Item K 1 2 3 Sublayer entrainment velocity, in percent of Ue ______________ 75 60 0 Sublayer freshwater inflow, in percent of Qf ________________ 25 37.5 37.5 Sublayer saltwater inflow, in percent of QS ________________ 75 15 10 The model computes the salinities of water in the sublayers at the end of every 15 minutes. The salinities computed for the sublayers at either First Avenue South Bridge, or 16th Avenue South Bridge, were com- pared with observed salinities at these locations. The differences between the means of computed and ob- served salinities were evaluated for statistical sig- nificance by using the t test in the method of pairing observations (Dixon and Massey, 1957, p. 124—127). The results given in the following table indicate that only the means in sublayer 1 at First Avenue South Bridge were significantly different from one another at the 99-percent confidence level. Number of Values oft in sublayer Location observa- E “0“ 1 2 3 1 ,2,3 16th Avenue South Bridge __ 30 2.51 —0.16 0.61 _--_ First Avenue South Bridge _- 31 4.46 —.84 2.53 _-__ Both stations ______________ 183 ____ _-_- _-__ 2.56 By using the equation, y= bx, a regression of com- puted salinities against observed salinities indicated that (1) computed salinity is 0.92 of observed salinity, NUMERICAL MODEL OF MATERIAL TRANSPORT IN SALT-WEDGE ESTUARIES (2) the standard error of estimate is 3.1 ppt, and (3) the correlation coefficient is 0.98. The t test and the regres- sion analysis indicate that the differences between computed and observed salinities are within the 10- percent-limit criteria for model estimates of salinity in the Duwamish River estuary. DISSOLVED OXYGEN INPUT AND OUTPUT DO Because upper-layer water does not move downward into the underlying salt wedge in the Duwamish River estuary, DO concentration in the wedge is unaffected by the content of DO and dissolved BOD (biochemical oxy- gen demand) in the upper layer. Photosynthesis pro- duces only negligible quantities of oxygen in the wedge water, which is usually below the photic zone that ordi- narily extends about 13 ft (4 m) below the water surface (Welch, 1969). As postulated by Welch, the phyto- plankton that sink from the upper layer contribute to the suspended-particulate and benthic oxygen- demanding material of the wedge. Therefore, the con- centration of DO can only decrease while wedge water moves from Elliott Bay to the wedge toe. In modeling wedge DO, concentrations in Elliott Bay water entering the wedge at the model mouth were assumed to be the same as those measured by the bot- tom water-quality monitor in the West Waterway at Spokane Street Brridge. The elevation of this monitor’s intake is such that upper-layer water is sampled during as much as 2—3 hours of some low low tides. Wedge DO concentrations during those period were estimated from adjacent-in-time concentrations and from the temporal-variation pattern in wedge DO concentration at this site; the pattern was determined from concentra- tions observed frequently during several tidal cycles. Modeling of wedge DO concentration was considered verified when hourly computed concentrations at 16th Avenue South Bridge agreed with hourly observed con- centrations there; comparisons were made only for the periods when wedge water was being sampled by the lower monitor at this site. Daily extremes in observed DO concentrations of wedge water at Spokane Street Bridge are shown for each June—September of 1967—69 in figure 7 and for each J une—September of 1970—71 in figure 8. Compari- sons of the data for the two periods indicate the follow- ing major differences: (1) Daily minimum DO concen— trations during August-September periods were higher in 1970—71 (about 5—7.5 mg/l) than in 1967—69 (about 3—5.5 mg/l), and (2) ranges in daily DO concentrations (difference between maximum and minimum) were higher during 1967—69 than during 1970—71~the ranges for 1969 and 1970 given in figure 8 exemplify this difference. MODEL COMPUTATIONS IN THE DUWAMISH RIVER ESTUARY, WASH. 23 '2 I I I I I I I I Lu 0: t —— _] _ o E " Io— a o . _ 2 ° . “' é '. ° _ o o . I o _.| I I. h‘. fl 0 E 8 I' 0. ~. 0 I ‘A _ o o .- ' " A . g I00 0 '0 I . ° 0 I A o A o (S?) O O a) I A 9 fl . .. ‘- 00 OO O O ' Q.:.QA‘ ‘ ‘ 0 I. : g 0000 0 Q9300 '- ‘ ‘ A ‘A.’ l .‘ i: I 0 (Egg! 0 O o O i ‘5'.“ ...-‘ A 4 sign 3, can 623 o°o 000 a -.,.- : J45}; m a: E [In I: D O 00 o 0 AAA A ll.- ‘ A‘ uJ _ III 451E] A MA AsmA AAAAM “ 0 (gain 95? am A a U z 0 D so u 0 0 $993623 E, o IEI 1] ED L39“ % GDOOQEAAAOOO >’ I:I “9 o _ X —- O o I o 8 2 _ Year Maximum Minimum _ Z l967 A A o (D l968 I D Q — I969 o o __ o I I I I I I l I I JUNE JULY AUGUST SEPTEMBER FIGURE 7. —Daily extremes in hourly observed DO concentration in salt wedge at Spokane Street Bridge during the J une—September periods of 1967—69. The differences in observed DO concentration at Spokane Street Bridge between the two periods proba- bly were caused by lesser consumption of oxygen in the ' downstream part of the wedge and by higher DO con- centrations in Elliott Bay water that flowed into the wedge during 1970—71. The relative importance of these two causative factors may be evaluated from an analysis of the changes in extremes in daily DO con- centrations between 1967—69 and 1970—71. Daily minimums in wedge DO concentrations occur at the bridge during low low tides in water that has spent a relatively long time in the wedge. The increase of about 2 mg/l in daily minimum DO concentrations between 1967—69 and 1970—71 suggests that oxygen consump- tion in the wedge decreased between the two periods. Daily maximums in wedge DO concentrations gener— ally occur at the bridge during high high tides in water that probably has recently arrived from Elliott Bay and may have spent a relatively short time, or no time, in the wedge. The differences in daily maximum DO con- 24 NUMERICAL MODEL OF MATERIAL TRANSPORT IN SALT-WEDGE ESTUARIES '2 I I I I I I I I 3% Year Maximum Minimum I_ — ‘ A AA I970 o o — ._I A AA I97I A A A 0: A Range 0 A A _ 3f '0 . I969 ------ ‘ A A‘ ‘ l970 A g “AA 6 ‘ “ AA AA A q A Alba ‘A AA A AA A A — m AA AA A A A A A A A o A 80 1AA %A A ‘ A 3 "00% A AAAAAA“ A A A; A A A _I 8 _ 0 80A A‘A ‘ A— — 'oA AA 9 AAA A O ‘9 A 2 .o o. ' o . Am“ ‘AA 5 A ' Z ‘ 0° 0" 0 c9 0 o f ' AA A A A _ o A A A — CS)oo Q) Q1123 of A A o O o A (€160. . o -A fl ‘1. ‘ o 0 Z“ 0 “Q3 0Q 0%) o A .‘A 0 .AA 0 o. 612’ . l C_) dc? odbcp (DO do 00 % .A A z? I— 5 ‘ 00 0 63 “253° 0 . A000) ‘30me s 0 «25¢ . I._ O (1’ fi‘ 0 z _ o oo _ Lu A o 3 'l O 4 _ q :I _ co | I E ’\ 1" fl III/I O A :V‘, A h I\ P A, l g; _ I" I ‘I‘I 1" 1“ [\_I/‘I I\ [I‘llll I, ‘—' \\' I l l _ o I‘ ' II‘| I‘ I II I\l”I I IIII I_\ , I I II| I \ I: ,\ I{I [IV-I \I “I / {l \ ' I ‘I I I ,l ,' V I I ‘\ ’ Ill ' I‘ \ A I l__/ II i O 2 _ I | I \\ l‘ I L I l \ l-\ I \ I” l I LL! , - \ I I I \_I ~ \ ~ \ T > I ‘\ I l [\l V \I \l \I (D l ‘2 o O I I I l JUNE FIGURE 8,—Daily extremes and ranges in hourly observed DO concentra centrations between 1967—69 (fig. 7) and 1970~71 (fig. 8) l are not easily distinguishable. However, frequency dis- tributions of the daily maximum DO concentrations during August and September of the two periods (fig. 9) show a greater occurrence frequency of higher values of daily maximum DO concentrations in 1970—71 than in 1967—69; this suggests that DO concentrations in the . Elliott Bay water arriving at Spokane Street Bridge did increase slightly between the two JULY during the J une—September periods 0 periods. AUGUST f 1969—71. SEPTEMBER tion in salt wedge at Spokane Street Bridge C. V. Gibbs and Glen Farris (written commun., 1973) of Metro found that annual minimum DO concentra- tions in water at a depth of 3.3 ft (1 In) in Elliott Bay have increased from 4.5 to 6.5 mg/l during the period 1968—71. This increase in D0 concentration is consider- ably greater than that suggested by data reported here for the Elliott Bay water arriving at Spokane Street Bridge (fig. 9). The origin in the bay of water arriving at this bridge is unknown and determination of its origin is MODEL COMPUTATIONS IN THE DUWAMISH RIVER ESTUARY, WASH. 60 r" o 2 m D I— 3 2 40 n: m u. o 35 3 a. z m z 20 o: _ a: D o 0 '...fi”."_ 05. 0‘! 05. 0‘. 03 0? to _ % AA A CPQ) 2 — I: 6 o 15”: >6 Elihu 1; 0 lb 00 A 00 Cl) A dfihA A 0% Lu _ A _J A AAA IQAAX a A g; o I I I I I I I I I 5 JUNE JULY AUGUST SEPTEMBER FIGURE 10.—Daily extremes of hourly observed DO concentration in salt wedge at 16th Avenue South Bridge during the J une—September periods of 1967—69. mg/l in 1967 to about 6 mg/l in 1971, in mean wedge DO concentration for the J une—September periods. PREDICTING DO Several restrictions and assumptions were made in predicting the effects of a future increase in discharge of RTP effluent on wedge DO concentrations. Only the planned ultimate RTP discharge of 223 ft3/s (379 m3/ min) was used in the predictions. Because the efficiency of the RTP probably will not change with discharge, effluent quality (constituent concentrations) at the ul- timate discharge was presumed to be the same as that for the average discharge of 37 ft3/s (63 m3/min) in 1971. As discussed earlier, DO concentrations in the Elliott Bay water arriving at Spokane Street Bridge were in- ferred to be relatively unaffected by the major changes that occurred in the waste discharged to the lower Green-Duwamish River and Elliott Bay during 1967— 71. Therefore, DO concentrations of bay water entering the estuary probably would be unaffected by a change in RTP effluent discharge and were assumed to be the same as they were in 1971. Finally, the prediction was made by using in the model the river discharges and tide stages for June—September 1971 and an estimated rate of oxygen use in the wedge. As in 197 1, the oxygen-use rate in the wedge when the RTP discharges an ultimate quantity of effluent may be expected to be constant in time and with longitudinal distance. The change in oxygen-use rate between the two discharges then may be approximated by evaluat- MODEL COMPUTATIONS IN THE DUWAMISH RIVER ESTUARY, WASH. 27 10 I I l I I l I | I A Year Maximum Minimum T I . O .— z '- ‘ ‘9 I970 9 ‘ ‘ ‘ l97| A A I— A‘ ‘ AA A <(observed DO concentration), solids in the wedge may not change benthic oxygen use, because there might be no change either in the area of bed covered with particulate oxidizable material or in the oxygen demand exerted by the bed area covered by particulate oxidizable material. Assuming no change in benthic oxygen use between RTP discharges of 37 and 223 ft3/s (63 and 379 m3/min) computations of the oxy- gen use and a corresponding oxygen-uSe rate in the representative column of wedge water are as follows: Suspended- dissolved use: 7.4x 1 X 1 x 1,000X0.012 X 1.92 =170.5 mg/hr 29.6X 1.000 = 29.6 mg/hr 170.5+29.6 = 200.1 mg/hr A: 0.27 mg/l/hr 7.4><1><1><1,000 DO concentrations in the wedge during June— September 1971 were modeled using the oxygen-use rates of 0.027 and 0.031 and a RTP effluent discharge of 223 ft3/s (379 m3/min). The differences between pre- dicted wedge DO concentration, assuming present and ultimate RTP discharges, were determined from regres- sion analyses. Figure 14 shows, for a proportionate in- crease (curve A) and no increase (curve B) in benthic oxygen use, the predicted reductions in wedge DO con- centration at 16th Avenue South Bridge due to in- creased effluent, in relation to observed wedge DO con— centration there. The model projections, based on the above data and assumptions, indicate that DO concen- trations in the wedge water may be reduced by 45 per- cent given a relatively low DO concentration (4 mg/l), and by 8 percent given a relatively high DO concentra- tion (9 mg/l), when the RTP eventually is operating at full design capacity. Benthic use: Total use: Oxygen-use rate = DISCUSSION For the June—September periods of 1969—71, the ob- served and computed DO concentrations have shown that concentrations increased and oxygen consumption decreased in the wedge between 1967—69 and 1970—71. Wedge DO concentrations during June—September periods of 1970—71 still frequently were less than the Washington State standard of 5 mg/l for estuarine wa- ters such as the Duwamish River estuary (Washington Water Pollution Control Commission, 1967). Neverthe- less, DO concentrations were less often below 5 mg/l and minimum concentrations were much higher in 1970—71 than in 1967~69. The model was used to predict wedge DO concentrations during a future time, when RTP is discharging a maximum quantity of effluent. These DO concentrations may be 1—2 mg/l less than they were during periods of low DO concentrations (4—5 mg/l) in June—September 1971. The predictions assume that (1) the suspended solids that sink into the wedge from the upper layer are a principal source of oxidizable material there, (2) the concentration of suspended oxidizable par- ticulates in Green River water will be 1 mg/l, and (3) DO concentration in the Elliott Bay water entering the wedge will be relatively unaffected by the increase in RTP effluent discharge. If the quality of the Green River and (or) Elliott Bay degrade in the future, the contribu- tion of the increased discharge of RTP effluent to degra- I2 oil-u “$1352 I I I I I 3:4 m< 90:5 Q'Eml m 2 (0 ‘EE 20:: 929 '5ij I I I I I 32§3 4 5 6 7 8 9 g g DISSOLVED‘OXYGEN CONCENTRATION, IN MILLIGRAMS PER LITRE FIGURE 14.—Predicted reduction in D0 concentration in wedge (from wedge DO concentration at 16th Avenue South Bridge with 37 ft3/s (63 m3/min) effluent discharge from Renton Treatment Plant, June—September 1971), assuming the ultimate RTP discharge of 223 ft3/s (379 ms/min), and a proportionate increase (curve A) or no increase (curve B) in benthic oxygen demand. MODEL COMPUTATIONS IN THE DUWAMISH RIVER ESTUARY, WASH. dation of wedge DO concentration would be compara- tively less. The probable effect on wedge DO concentrations of not discharging RTP effluent to the estuary in June— September 1971 also may be of interest. Again using the assumptions, restrictions, and procedures that were used in predicting the effect of an increase in RTP efflu- ent, an oxygen-use rate of 0.0092 mg/l/hr in the wedge during J une—September 1971 is estimated for a zero discharge of RTP effluent. When this rate and a zero RTP effluent discharge was used in modeling 1971, the average computed DO concentration in the wedge for J une—September was 0.6 mg/l higher than that comput- ed for a RTP effluent discharge of 37 ft3/s (63 m3/ min). The reliability of predicting the effects of future changes in the variables of the Duwamish River estuary cannot be proved until the changes occur. Presently, the confidence level for the predicted changes in wedge DO concentration is low. Because a technique for predicting oxygen-use rates is necessary, a recommended use of the model is to compute oxygen—use rate in the wedge annu- ally, biennially, or after longer periods. Such modeling would provide information about changes in oxygen use in the wedge relative to changes in RTP discharge; this information could be used in improving the prediction of the effect of increases in RTP discharge on wedge DO concentrations. Present (1973) plans include (1) completion of model- ing of the upper layer of the estuary, and modeling of phytoplankton in the estuary, and (2) combining the model with a model of the adjoining upper estuary and tidal river. The model could be used for estimating transport and distribution of constituents other than wedge DO and estuary salinity. Haushild and Stoner (1973) used the model to predict effects of a proposed change in estuary geometry on residence time and DO concentrations of the wedge water. The model could be used to evaluate methods for improving DO concentra- tions in the salt wedge. For example H. B. Fischer (au— thor of Part I of this report) suggested that salt water could be pumped from the wedge into the upper layer at a location in the upstream part of the wedge to decrease residence time of water in the wedge. This probably would result in less of the D0 in the wedge water being consumed during its travel from the mouth to the toe. The effect on the upper layer DO concentrations, from the introduction of more wedge water of relatively low DO concentrations, would have to be evaluated. SENSITIVITY ANALYSES To evaluate the response (sensitivity) of estuary sa— linity and wedge DO concentration to changes in model parameters, one parameter value at a time was varied while the other parameter values were held constant. In 31 the sensitivity analyses, single-value parameters were equally increased or decreased once each, whereas dis- tribution parameters were changed for the purpose of producing one negative and one positive change in the salinity of each sublayer. The results reported here are based on one-step changes in model parameters; the degree and (or) direction of response in D0 and salinity may not be the same for further changes in the parameters. Sensitivity of variables to model parame- ters will be more completely analyzed in the model that includes the upper layer. For each change in a parame- ter, the mean increase or decrease of salinity and DO concentration during a period of approximately 2 months and the changes in salinity and DO concentra- tion during a tidal cycle were determined for the estuary only at 16th Avenue South Bridge. The variation of tide stage during the tidal cycle is shown in figure 15. The parameter values determined during the verification of estuary salinity and wedge DO concentration are called base data, and salinity and DO concentrations com- puted by using the base data are called base values. The salinity and DO concentrations computed using differ- ent data were compared with base values. In the model, salinities of sublayers change in propor- tion to changes in wedge salinity, as shown by the data in table 2. Among the seven other model parameters analyzed, the data in table 2 indicate that sublayer salinity at 16th Avenue South Bridge is most respon- sive to wedge toe 10cation and is least responsive to wedge and sublayer entrainment velocities and to outflow of wedge water at the toe. Salinities in the three sublayers generally vary dur- ing a tidal cycle (fig. 16) in accordance with the tide stage (fig. 15). A change in the distribution of saltwater inflow among the sublayers at the wedge toe causes more of a change in sublayer salinity during the period of low low tide than during the remainder of the tidal cycle (fig. 16); this was true also for a change in the outflow of saltwater at the wedge toe. This inequality of change in sublayer salinity was expected, because 16th Avenue South Bridge at low tides is near the wedge toe where sublayer salinity is specified in the model by the saltwater outflow from the wedge and by the distribu- tion of an equal saltwater inflow among the sublayers. The sensitivity analyses for the other parameters indi- cated that (1) sublayer salinity at the bridge changed less during the period of low low tide than during the other parts of the tidal cycle for a change in sublayer entrainment velocities, because most of the entrain- ment occurs downstream at low low tides; and (2) throughout the tidal cycle, sublayer salinity changed proportionately with the other five parameters. The mean changes in wedge DO concentration for changes in each of the controlling parameters in the 32 NUMERICAL MODEL OF MATERIAL TRANSPORT IN SALT-WEDGE ESTUARIES '0 I I I I I I I I I I I I 30 8 0') m E u: w I- u. tn 6 2 z — E s” — < c“; '— < m 57: DJ 4 o m I: 9 l- 2 O I I I I I I I I I I I I O O 4 8 l2 20 24 IS ELAPSED TIME, IN HOURS FIGURE 15.—Variati0n of tide stage during the tidal cycle used in the sensitivity analyses. TABLE 2.——Mean changes in sublayer salinities at 16th A venue South Bridge during a 2-month period, relative to base values ofsublayer salinities, for selected changes in the controlling parameters used in the Duwamish River estuary model [Positive and negative values indicate increases and decreases, respectively, in sublayer salinity] Change in model salinity (ppt) Parameter Change in the parameter Sublayer 1 2 3 Wedge toe location +1,000 ft (4 percent of average location) 060 086 0.36 —1,000 ft (4 percent of average location) -.68 —.94 —.36 Saltwater outflow +10 percent 24 .36 .29 —10 percent —.22 —.32 —.28 Entrainment velocity +10 percent .31 .51 ‘24 ~10 percent —.30 —.48 -.22 Upper—layer thickness +2 ft 122 percent of average thickness) .18 .51 .33 —2 it (22 percent of average thickness) —.11 —.26 —.15 Wedge salinity +2 ppt (8 percent) 145 .74 .33 —2 ppt18 percent) —l.45 -.74 —.33 Sublayer freshwater inflow Distributed as 20, 40, and 40 percent 1.15 .08 —.17 Distributed as 30, 35, and 35 percent —1.04 —.04 —.20 Sublayer saltwater inflow Distributed as 60, 25, and 15 percent —31 .44 .86 Distributed as 80, 10, and 10 percent .10 —.24 —I14 Sublayer entrainment Distributed as 80, 65, and 0 percent .05 .26 .15 veloc1ty Distributed as 70, 55, and 0 percent —.04 —.28 —.15 MODEL COMPUTATIONS IN THE DUWAMISH RIVER ESTUARY, WASH. 33 25 I I I I I I I I I I I I | 3 ‘8. ; Tides '9 E '9 \z a ‘5» _.I _I ':E o z 20 _ < V) 8 Saltwater inflow as distribution, in percent Sublayer E I 2 ‘5- 0—0—0 65 25 I5 ..I E 75 I5 I0 4 80 IO IO a t E ‘ Qw/ \‘/ >. t .— E _I < (n o m '5 f— O. _‘_—'_ 2 __ o o O I I I I I I I I I | I I O 4 8 I2 I6 20 24 ELAPSED TIME, IN HOURS FIGURE 16.—Cornputed sublayer salinity during a tidal cycle at 16th Avenue South Bridge for various distributions of saltwater inflow among the sublayers at wedge toe, Duwamish River estuary. model are given in table 3. The data suggest that, on the basis of equal percentage change, wedge DO concentra- tion probably is more sensitive to changes in wedge toe location and entrainment velocity at the top of the wedge than it is to changes in saltwater outflow at the wedge toe and upper-layer thickness. The variation of computed DO concentration in the wedge during the tidal cycle (fig. 17) agrees with the semidiurnal pattern of variation usually observed (fig. 13). For the changes in upper-layer thickness, wedge DO concentrations departed farthest from base concen- trations during the low low tide (fig. 17); this also was true for the other three model parameters used in the sensitivity analyses. SUMMARY AND CONCLUSIONS The numerical model described in Part1 of this report has been found to yield a good agreement between com- 34 TABLE 3.—Mean changes in D0 concentration in the saltwater wedge at 16th Avenue South Bridge during a 2- month period, relative to base values of D0 concentration in the wedge, for selected changes in the controlling parameters used in the Duwamish River estuary mo‘del [Positive and negative values indicate increases and decreases, respectively, in wedge DO concentration] Change in wedge Parameter Change in the parameter DO concentration (mg/1) Wedge toe location +1,000 ft (4 percent of avera e location) 0.084 —1,000 t (4 percent of average location) —.098 Saltwater outflow +10 percent .081 — 10 percent —.084 Entrainment velocity +10 percent .142 — 10 percent —.124 Upper—layer thickness +2 ft (22 percent of average thickness) .159 —2 ft (22 percent of average thickness) —.151 puted and observed salinity distributions in the Duwamish River estuary and the DO concentration in the salt wedge of the estuary. The model input data, MODEL COMPUTATION S IN THE DUWAMISH RIVER ESTUARY, WASH. determined or estimated from observed data, included estuary geometry, location of upstream end of salt wedge (wedge toe) as a function of freshwater inflow and tide stage, entrainment of water from the wedge as a function of tidal-prism thickness (tidal exchange) and freshwater inflow, upper layer thickness, slope of the wedge-upper layer interface, freshwater inflow, tide stages, DO concentration in Elliott Bay water entering the downstream end of the wedge, and saltwater outflow at the wedge toe. Computed salinities in the estuary and DO concentrations in the wedge near its upstream end were compared with the appropriate observed salinities and concentrations. Because the upper surface of the wedge becomes diffuse and unstable farther upstream, the location of the wedge toe was defined as the farthest upstream cross section where salinity is 25 ppt. The saltwater that flows out at the wedge toe is balanced by an equal flow of saltwater into the upper layer at the wedge toe. Entrainment velocities, proportioning of freshwater and saltwater inflows among the sublayers 3-5 I I I I I I I I I I I I J: g 34" TMes\\\ E 3’ S ‘ I: 3 3 g < _ 3 3 ‘f _ a: 3.3 E 1 in m Upper layer thickness _ g E 3’2 at model estuary mouth 8 3 0—0 Base value plus 2 feet 2 0: 3' Base value " In In 2 CL F—I Base value minus x a) 2 feet _ O 2 3.0 I o 3‘: "J o > — 2.9 " ‘ 5’ 3 a) a 2 - o z 2.8 _ o I2 27 _ — D . a. E 8 2.6 — _ 2.5 I I I I I I I l I I I I I ELAPSED TIME, l2 l6 IN HOURS 24 FIGURE 17 .—Computed DO concentration in salt wedge during a tidal cycle at 16th Avenue South Bridge. MODEL COMPUTATIONS IN THE DUWAMISH RIVER ESTUARY, WASH. at the upstream end of the upper layer, and oxygen-use rates in the wedge were the parameters for which values were determined by trial and error. The parameter val- ues necessary for agreement between computed and observed salinities and DO concentrations were within a credible range for the Duwamish River estuary. DO concentrations in wedge water increased between 1967-69 and 1970—7 1. Raw and partly treated indus- trial and municipal wastes as well as effluent from the . RTP, a secondary treatment plant, were discharged into the estuary during 1967—69, whereas nearly all the wastes discharged into the estuary during 1970—71 were those given secondary treatment at RTP. In the wedge at a downstream station near the estuary mouth, observed daily minimum concentrations during August-September periods were about 2 mg/l higher in 1970—71 (about 5—7.5 mg/l) than in 1967-69 (about 3—5 mg/l). In the wedge at an upstream station near the wedge toe, observed daily minimum and maximum con- centrations during August-September periods in- creased by 2—3 mg/l and 2 mg/l, respectively, between 1967-69 and 1970—71. The sources of wedge water near the estuary mouth in relation to times of daily extremes in D0 concentration suggested that a lower oxygen-use rate in the wedge in 1970—71 than in 1967—69 was the principal contributor, and the slightly higher DO con- centration in Elliott Bay water entering the wedge in 1970—71 was a minor contributor to the increase in wedge DO concentrations between the two periods. Wedge DO concentrations computed by the model agreed well with observed concentrations during the June—September periods of 1967—71. The modeling re- sults indicated that the oxygen-use rate did not vary within any one J une—September period, that rates were different (range from 0.023 to 0.028 mg/l/hr) each year in 1967—69, and that rates were constant (0.016 mg/l/hr) in 1970 and 1971. The average oxygen-use rate in the wedge during 1967—69 was 60 percent higher than it was during 1970—71. The model program was used to predict wedge DO concentrations by assuming that parameters and input data, except freshwater inflow and oxygen-use rate, were the same as they were during June—September 1971. Freshwater inflow was changed to include the ultimate designed discharge of RTP effluent into the estuary (233 ft3/s or 379 m3/min) instead of the 1971 discharge (37 ft3/s or 63 m3/min). A model oxygen-use rate for the higher discharge was computed by assum— ing that (1) the rate would increase in proportion to the increase in concentration of suspended solids in.the river and, consequently, in the upper layer; (2) the suspended-solids concentration in the river water was 1 mg/l for both 1971 and the time when RTP discharge will be the planned ultimate rate; and (3) the 35 suspended-solids concentration in plant effluent did not change with discharge. An increase from 37 to 223 ft3/s (63 to 379 m3/min) in plant—effluent discharge may de- crease DO concentrations in the wedge water by 45 percent given a relatively low concentration (4 mg/l) and by 8 percent given a relatively high concentration (9 mg/l). Also, it was estimated that average wedge DO concentrations might have been 0.6 mg/l higher in J une—September 1971 if RTP effluent had not been dis- charged to the estuary. Sensitivity analyses in which only one positive and one negative change in model parameters was made indicate that sublayer salinity is most responsive to changes in wedge toe location and in salinity of the wedge water and is least responsive to changes in sub- layer entrainment velocities and in saltwater flow out of the wedge at its toe. Wedge DO concentration is most responsive to changes in wedge toe location and en- trainment between the wedge and the upper layer. The upper layer is presently (1973) being modeled. Using the model to compute oxygen-use rates in the wedge at the end of future periods is recommended to provide information about response of the rate to changes in RTP discharge and quality of freshwater and saltwater inflows. The model could be used to estimate transport of constituents other than wedge DO and es- tuary salinity. The model has been used to predict ef- fects of a proposed change in estuary geometry on resi- dence time and DO concentrations of the wedge water. Also, the model could be used to evaluate suggested methods for improving DO concentrations in the salt wedge. DEFINITION OF TERMS Entrainment velocity—Velocity at which water moves from the wedge into the overlying layer, in feet per second (centimetres per second). Freshwater inflow—The total flow of freshwater into the upper layer at the wedge toe, in cubic feet per sec- ond (cubic metres per minute). Initial volume—Initial volume of the wedge elements, in cubic feet (cubic metres). Saltwater outflow—The total flow of saltwater out of the wedge at its toe, in cubic feet per second (cubic metres per minute); it equals the total flow of salt- water into the upper layer at the wedge toe. S ublayer entrainment velocity—An entrainment veloc- ity at the top of a specific sublayer, in feet per second (centimetres per second). S ublayer freshwater inflow—The flow of freshwater into a specific sublayer, in cubic feet per second (cubic metres per minute). S ublayer saltwater inflow—The flow of saltwater into a specific sublayer, in cubic feet per second (cubic metres per minute). 36 Upper layer thickness—The total thickness of the upper layer at the model-estuary mouth, in feet (metres). Wedge salinity—Salinity of the saltwater wedge, in parts per thousand. Wedge toe location—Location of the model-wedge toe, in feet (metres) upstream from model-estuary mouth. REFERENCES CITED Dawson, W. A., and Tilley, L. J ., 1972, Measurement of salt-wedge excursion distance in the Duwamish River estuary, Seattle, Washington, by means of the dissolved-oxygen gradient: U.S. Geol. Survey Water-Supply Paper 1873—D, p. D1—D27. Dixon, W. J., and Massey, F. J ., Jr., 1957, Introduction to statistical analysis: New York, McGraw—Hill, 488 p. Haushild, W. L., and Stoner, J. D., 1973, Predicted effects of proposed navigation improvements on residence time and dissolved oxygen of the salt wedge in the Duwamish River estuary, King County, Washington: US. Geol. Survey open-file report, 13 p. Metropolitan Engineers, 1971, Report on operations, Municipality of Metropolitan Seattle, January 1, 1970—December 31, 1970: Seat- tle, Wash., 48 p. Pacific Northwest River Basins Commission, Hydrology and Hy- draulics Committee, 1969, River mile index—Deschutes, Nis- qually, Puyallup, and Green Rivers, Lake Washington, and Snohomish River, Puget Sound Basin, Washington: Vancouver, Wash., p. 21—29. NUMERICAL MODEL OF MATERIAL TRANSPORT IN SALT-WEDGE ESTUARIES Santos, J. F., and Stoner, J. D., 1972, Physical, chemical and biological aspects of the Duwamish River estuary, King County, Washington 1963—67: US. Geol. Survey Water-Supply Paper 1873—C, p. C1—C74. Stoner, J. D., 1967, Prediction of salt-water intrusion in the Duwamish River estuary, King County, Washington, in Geologi- cal Survey research, 1967: U.S. Geol. Survey Prof. Paper 575—D, p. D253—D255. 1972, Determination of mass balance and entrainment in the stratified Duwamish River estuary, King County, Washington: US. Geol. Survey Water-Supply Paper 1873—F, p. F1—F17. [US] Environmental Science Services Administration, 1968—71, Tide tables, high and low water predictions, west coast of North and South America including the Hawaiian Islands: Washington, pub. ann. [U.S.] National Oceanic and Atmospheric Administration, 1973, Tide tables, high and low water predictions, west coast of North and South America including the Hawaiian Islands: Washington, pub. ann. Washington Water Pollution Control Commission, 1967, A regulation relating to water quality standards for interstate and coastal waters of the State of Washington and a plan for implementation and enforcement of such standards: Wash. Water Pollution Con- trol Comm. Publ., 23 p. Welch, E. B., 1969, Factors initiating phytoplankton blooms and re- sulting effects on dissolved oxygen in Duwamish River estuary, Seattle, Washington: US. Geol. Survey Water-Supply Paper 1873—A, p. A1—A62. l'J’U.S. GOVERNMENT PRINTING OFFICE: 1975-0-690-036/12 7 DAY EARTH SCIENCES WY ,3“ Lithium in Unconsolidated W Sediments and Plants of the Basin and Range Province, Southern California and Nevada ! 918 GEOLOGICAL SURVEY ‘PROFESSIONAL PAPER {8‘3 TY or 0,41 § ”0.9 xx“ OCT .16 1975 (247/9 59 SCIENgE ma“ OCT 1 5 1975 l 3 LIBRARY 5 firm/man OF‘VEfiUanm ‘r‘ 5 ) 5 : Sf? 2 1975 U.S.S.D Lithium in Unconsolidated Sediments and Plants of the Basin and Range Province, Southern California and Nevada By HELEN L. CANNON. THELMA F. HARMS, and J. C. HAMILTON GEOLOGICAL SURVEY PROFESSIONAL PAPER 918 A study of the accumulations of lithium in the evaporz'tes, clays, freshwater limestones, waters, and vegetation of open and closed basins UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON ; 1975 UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress Cataloging in Publication Data Cannon, Helen E. Leighton, 1911— Lithium in unconsolidated sediments and plants of the Basin and Range province, southern California and Nevada. (Geological Survey Professional Paper 918) Bibliography: p. Supt. of Docs. No.1 119.162918 1. Lithium. 2. Sediments (Geology)—California. 3. Sediments (Geology)fiNevada. 4. Plants—Chemical analysis. 5. Water, Underground—California. 6. Water, Underground—Nevada. I. Harms, Thelma F. 11. Hamilton, J. C. III. Title: Lithium in unconsolidated sediments and plants of the Basin and Range province . . . IV. Series: United States Geological Survey Professional Paper 918. QE516.L5C36 553'.499 75—619074 For sale by the Superintendent of Documents, US. Government Printing Office Washington, DC. 20402 Stock Number 024—001-02213—3 CONTENTS Page Page Abstract ........................................................................................... 1 Lithium in the Basin and Range province—Con. Introduction 1 Lithium uptake by desert plants ............................................. 6 Collection and treatment of samples ...................................... 1 Open basins discharged by underflow .................................... 7 Geography and ecology of the Basin and Range province 3 Hydrologically closed basins ................................................... 11 Geochemistry of lithium .......................................................... 3 Death Valley ..................................................................... 12 Rocks ............................................................. 3 Amargosa Desert ............................................................... 14 Soils and sediments ...................................... 4 Other closed basins ........................................ 16 Water ....................................... 4 Plant prospecting .............................................. 19 Vegetation .................................................... 4 Health aspects of lithium ...................... 21 Lithium in the Basin and Range prov1nce ................. 5 Summary ................................................. 21 Hydrological relationships ......................................... 5 References Cith ............................................................................... 22 ILLUSTRATIONS Page FIGURE 1. Map showing concentration of lithium in basin sediments, California and Nevada .................................... 2 2. Graph showing relationship between lithium in plants and that in soil ........................................................................................... 9 3—7. Photographs: 3. Crown of rush killed by hot-spring precipitate, Clayton Valley ............................................................................................. 12 4. Precipitate of biogenetic origin occurring in place, Amargosa Desert ......................... 12 5. CaCOg precipitate on organic detritus, Amargosa Desert .................... 13 6. Cast of former root or stem structure, Amargosa Desert ................... 13 7. Pickleweed (Allemolfea) on salt mounds in Death Valley ..................................................................................................... 14 TABLES Page TABLE 1. Lithium, strontium, boron, and fluorine content in some waters of Nye County, Nev., and Inyo County, Calif, compared with 100 municipal water supplies in the United States ...................... 6 2. Maximum lithium concentrations in some basins of Nevada and California... 7 3. Plant species sampled in Nevada and California .............................................. 7 4. Lithium in plant groups of the Great Basin, California-Nevada ........................................................................................... 8 5. Maximum contents of lithium and sodium in some plants and basin soils ..................................................................................... 9 6. Percentage of lithium and associated ions in sediments and the ash of plants from open basins in Nevada and California ......... 10 7. Percentage of lithium and associated ions in basin sediments, collected within 6 inches of the ground surface ............................ 11 8. Percentage of lithium and associated ions in sediments and the ash of plants from Death Valley, Calif ........................................ 15 9. Chemical constituents of some springs and wells in Amargosa Desert ............................................................................................. 16 10—12. Percentage of lithium and associated ions in sediments and the ash of plants from: 10. Amargosa Desert, Nevada and California .............................................................................................................................. 17 11. Three closed basins in which economic quantities of lithium occur .................................................. 18 12. Some other closed basins in Nevada and California ............................................................................................................. 20 III LITHIUM IN UNCONSOLIDATED SEDIMENTS AND PLANTS OF THE BASIN AND RANGE PROVINCE, SOUTHERN CALIFORNIA AND NEVADA By HELEN L. CANNON, THELMA F. HARMS, and J. C. HAMILTON ABSTRACT Geochemical and geobotanical sampling of 20 hydrologically closed and open basins in the southern part of the Basin and Range province shows contrasting lithium contents of their sediments, vegetation, and natural waters. These findings have significant economic and environ mental implications. In hydrologically closed basins, where ground- water discharge is mainly by evapotranspiration, economic concentra— tions of lithium are of three types—as lithium chloride in evaporites and brines, in the crystal lattice of montmorillonitic clays, and in carbonate precipitates in hot-spring and lacustrine deposits. Along with potas- sium, boron, and halite, lithium compounds are precipitated at a late stage of evaporation. Sediments collected from low areas of closed basins contain 304,500 ppm lithium, whereas sediments from open, drained basins contain 20—150 ppm lithium. Recycled concentrations occur in association with hot springs at the alluvium-playa contact, where lithium dissolved at depth is precipitated around hydrophytic plants and organic debris. The average lithium con- tent of native vegetation sampled in closed basins is 100 times the con- tent reported for average plants. The lithium content of halophytic vege- tation is closely related to that of the soil, anomalous accumulations oc- curring in plants that also take up large concentrations of sodium chlor- ide. Ground water contains 50 times more lithium than average municipal water supplies in the United States. Lithium is produced com- mercially from volcanic clays and, also, by brine evaporation in several closed basins; analysis of vegetation may furnish a means of prospecting for deep-seated lithium brines. Contents of as much as 3,000 ppm lithi- um were found in the ash of pickleweed (A llenrolfea) and in rushes root- ed in hot springs. An average of 150 ppm lithium in the ash and 25.8 ppm in the dry weight of all plants that were collected in both closed and open arid basins is considerably higher than the average of 1.3 ppm in dry weight reported for plants growing in a nonarid climate. The high lithi- um content in well waters and vegetation of closed basins is probably beneficial, rather than harmful, to animals and man. INTRODUCTION The geochemistry of lithium in sediments and its ab- sorption by plants in the arid basins of the Basin and Range physiographic province were studied after a broad geochemical study had found large amounts of lithium in the native plants of the basins. The importance of con- centrations of lithium in plants is twofold: first, the plants may be of value as indicators of lithium-rich brines at depth; and second, the uptake of lithium by concentrator species of forage or produce may be of significance nutri- tionally to livestock and man. Therefore, we studied the plant-soil relationships in closed basins that are dis- charged by evapotranspiration and in open basins that are discharged by underflow through permeable beds at considerable depths to lower basins. Plants and sediments were sampled in 20 basins of California and Nevada (fig. 1), as time permitted, during 1963—67. Plants and the sedi- ments in which the plants grew were collected in each basin from the lowest and wettest part of the basin, and, 10cally, suites of samples were collected from several chemical zones in the basin. Generally, the plants were salt-tolerant species—phreatophytes in the closed basins and xerophytes in the open basins. The flora and deposits of springs were of particular interest. Analytical data on eight elements are reported here, together with available water analyses. COLLECTION AND TREATMENT OF SAMPLES Branch tips and leaves were collected from the larger shrubs, and the above-ground parts were sampled from herbaceous plants. The samples were not washed but were dried in paper sacks and analyzed in the US. Geological Survey laboratories in Denver, Colo. Soil samples were air-dried and then sifted through an 80-mesh sieve. Plant samples were dried at 50°C, ground in a Wiley mill, and ashed in a muffle furnace at 500°C. Of the elements re- ported, magnesium and carbonate were analyzed by rapid rock analysis; calcium, sodium, and potassium by atomic absorption; sulfate by a colorimetric method; fluorine by selective electrode; and boron by emission spectrograph. The lithium content of each sample was determined by two methods—semiquantitative spectrographic analysis (Myers and other, 1961) and atomic absorption method (Ward and others, 1969). Determinations by atomic ab- sorption for lithium in the plants were made by leaching 100 mg of ash in 2 ml of 6 N HCl and diluting the leachate to 25 ml before atomizing. Lithium detection limits are 10 ppm for the atomic absorption method and 100 ppm for the 6-step spectrographic procedure, which appears to be more reproducible for higher values. Therefore, the reported low values were determined by atomic absorp- tion and the high values by emission spectrograph. A 1 LITHIUM IN UNCONSOLIDATED SEDIMENTS AND PLANTS, CALIFORNIA AND NEVADA 1 19 ° "L 5 F haFlt‘ recrnn a F065? - mm _ , ., _ , , e - * rindggaandanc 09?» ' Ao oL25 O 50 MILES L17Boo L17A 0 50 KILOMETRES FIGURE l.—Concemration of lithium in basin sediments, California and Nevada. Geology and base modified from Stose (1932). Map explanation is on facing page. GEOCHEMISTRY OF LITHIUM 3 EXPLANATION [:1 Sedimentary rocks — Largely Paleozoic limestones Igneous rocks — Largely volcanic tuffs 5:3 Closed basin discharged by evapotranspiration Playa —+ Direction of ground-water movement Lithium in sediment sample — In parts per million. Plants were collected at all sample localities 0 < 100 0 100—300 0 > 300 AM Ash Meadows CS Station number spread in lithium values was noted in high-sodium plants that were analyzed by both methods. Probably sodium ’masks lithium in determinations by atomic absorption to some degree but enhances the lithium values on the spectrograph. Also presented herein are analyses of waters sampled by US. Geological Survey hydrologists; the samples were filtered, acidified, and analyzed chemically and by emis- sion spectrograph in the Denver laboratories. The senior author collected the samples, assisted at var- ious times by Thelma F. Harms, Margaret E. Hinkle, and Mary Strobell. Analyses by emission spectrograph and atomic absorption were made in US. Geological Survey laboratories, Denver, Colo., and by rapid—rock analysis in Washington, DC. Analysts were as follows: Chemical analysis of plants and soils: M. E. Hinkle, J. B. McHugh, K. W. Leong, J. H. Turner, T. F. Harms, Clara Papp Spectrographic analysis of plants and soils: J. C. Hamil- ton, Barbara Tobin, J. L. Finley, H. G. Neiman, B. N. Lanthorn, E. L. Mosier, L. D. Forshey Rapid-rock analysis: Paul Elmore, H. Smith, S. D. Botts, Lowell Artis, J. L. Glenn, Gillison Chloe, D. Taylor. Isaac Winograd, US. Geological Survey, provided help- ful information concerning hydrological relationships. GEOGRAPHY AND ECOLOGY OF THE BASIN AND RANGE PROVINCE The Basin and Range province is characterized by a series of tilted fault blocks forming longitudinal, asym- metric ridges or mountains and broad intervening valleys. Weathering and erosion have lowered the mountain peaks and have filled the valleys with alluvium and products of volcanism to a depth of several thousand feet. The rocks range in age from Precambrian to Quaternary, but we are concerned here only with Pleistocene and Holocene basin- fill sediments interbedded with volcanic tuffs and basaltic flows, and with the gravel fans that surround the basins. During the several pluvial periods of historic time, those basins in which the alluvium overlies relatively imper- vious rock formations were filled with lake water to a con- siderably higher elevation than .exists today. Basins that are not hydrologically closed drain through the under- lying Paleozoic sedimentary rocks to nearby lower basins. The lowest closed basin is Death Valley, whose surface is 280 feet (85 m) below sea level. Large concentrations of salts have built up in the closed basins by processes of evaporation and transpiration; the composition of the salts varies from basin to basin depending on the source of the water and the geology of the recharge area. The climate is generally arid, and so surface waters are apt to be ephemeral; the depth to ground water in open basins may be several hundred feet. Which plant species can live in a particular basin is controlled by altitude, by water condi- tions, the total salt content, and the tolerance of various species for particular ions that may occur in very high con— centrations. Hunt (1966) described the salt zones of Death Valley as a central area of chlorides ringed successively by sulfide and carbonate zones. Each zone has a character- istic flora. The plants of greatest significance as indica- tors of soil conditions and as accumulators of ions are woody shrubs of the Upper and Lower Sonoran plant zones. Open basins are generally populated by xerophytic plants except at springs along the mountain-basin con- tact. Closed basins are populated by phreatophytes (deep- rooted plants that depend on ground water for their water supply) on the basin proper and by xerophytes (plants adapted to dry conditions) on the peripheral gravel fans. Bare areas completely devoid of vegetation occur in the salt pan of Death Valley, playa muds of drier valleys, and on the steeper gravel fans. We sampled sediments of the bare areas in several basins as well as the soils in which plants were able to grow. GEOCHEMISTRY OF LITHIUM Lithium, a light metal, third in the periodic table, is widely distributed throughout the Earth’s crust. It is used as a pottery glaze, as an absorbent in air conditioners, and as an ingredient of waterproof and high-temperature lubricants. The use of lithium in hydrogen bombs and of lithium flares accounts for some of the lithium that occurs at the Atomic Energy Commission’s Nevada Test Site. ROCKS Lithium is concentrated in the silicates and aluminosilicates of acidic igneous rocks where it replaces magnesium, ferrous iron or aluminum; it is also present in pegmatite dikes where it forms independent lithium minerals, such as spodumene (LiAlSizOs), and the lithium 4 LITHIUM IN UNCONSOLIDATED SEDIMENTS AND PLANTS, CALIFORNIA AND NEVADA micas, such as lepidolite, which are the main ores of the element. Averages for lithium in igneous rocks have ranged from 22 to 65 ppm (Parker, 1967 ). Averages for sedi- ments, calculated by Rankama and Sahama (1950) from Strock (1936), are 17 ppm in sandstones, 46 ppm in shales, and 26 ppm in limestones; contents of more than 100 ppm are reported for iron ores. In a Russian study of the geochemistry of lithium in the sedimentary cycle (Ronov and others, 1970), 21,000 samples were analyzed. Lithium was found to concentrate in clays in which it correlates most strongly with alumi- num. The results of the Russian study show that the behavior of lithium is the weathering zone is not depend- ent on the petrographic type of the initial rock, on the initial lithium content, or on the geologic age of the crust but is determined by the sequence of stages or the physio- chemical conditions of weathering: the metal is usually concentrated finally in kaolinite. Lithium entering clay minerals differs from sodium, potassium, rubidium, cesium, magnesium and calcium by being firmly bound during the formation of the clay minerals but by ulti- mately being released during the disintegration of clays to form laterite. Tardy and others (1972) showed that the retention of lithium in clay minerals of weathering is weak and on the same order as the initial content of the rock (<60 ppm); that the fixation of lithium is moderate in sedimentary clay minerals of detrital origin (< 300 ppm); and that lithium is concentrated in new magnesium clay minerals formed under arid conditions in a carbonate or sulfate environ- ment (as much as 6,000 ppm). The carbonates and sulfates that accompany the silicates, however, contain scant lithium (<30 ppm). Accordingly, the behavior of lithium can serve to distinguish in sediments the inherited clay minerals from newly formed clay minerals. Lithium follows the pattern of magnesium and, in one mode of accumulation, is absorbed into the crystal struc- ture of montmorillonite and illite (Mason, 1952). Lithium is not in an exchangeable site but is probably in the octa- hedral layer, substituting for magnesium (R. A. Sheppard and H. C. Starkey, written commun., 1972). A second mode of lithium accumulation is in salt lagoons or basins where contents of lithium correlate negatively with those of aluminum but positively with those of magnesium, sulfate, and chloride; large concen- trations of lithium occur in these evaporative basins (Ronov and others, 1970). Lithium not only is very soluble but is most easily displaced by ion exchange of all the cations and, hence, is a very mobile element that moves readily into and out of waters and sediments (Bear, 1964). In an evaporative basin, lithium may remain in solution until a late stage and then be precipitated along with sodium, potassium, and boron in the chloride and sulfate zones (Stewart, 1963). In several basins in the Western United States, lithium and potassium are commercially produced by evaporating the brines. SOILS AND SEDIMENTS Swaine (1955) reported 8 to 400 ppm lithium in soils. Steinkoenig (1915) reported 10 to 100 ppm in soil and 20 to 70 ppm in subsoil from 19 soils in 6 different areas of the United States. Recent US. Geological Survey studies in Missouri show a geometric mean of 21 ppm lithium for that State. In the arid Basin and Range province, lithium is concentrated in hydrologically closed basins but not in open basins. In closed basins, soluble salts move upward to the ground surface, where they are further concentrated by evaporation and plant transpiration; in open basins the salts move downward to the ground-water table at depth and then laterally to a lower basin. Surface sediments (ex- clusive of alluvium) collected in our study contained 30 to 2,000 ppm lithium (median, 150 ppm) in 12 closed basins and 30 to 150 ppm (median, 50 ppm) in 8 open basins. WATER The lithium content of fresh water is considerably less than that of sea water (0.1 ppm), because lithium chloride is highly soluble and remains in solution in sea water more or less indefinitely. A median of 2 mg/l (0.002 ppm) lithium was stated by Durfor and Becker (1964) for muni- cipal water supplies; some of this lithium may be sorbed by suspended clay. Thermal waters commonly contain large quantities of lithium leached as lithium chloride from rocks penetrated by the waters (Rankama and Sahama, 1950). VEGETATION Lithium is not known to be an essential plant nutrient, but its presence in all plants and its effect on plant growth suggests that lithium is necessary for plant growth and development. It exerts both stimulating and retarding effects on plant growth, depending on the concentration and on the plant species. According to Evans and Sorger (1966) at least one enzyme from a halophile is activated by sodium and lithium but not by potassium or rubidium. Those enzymes that are activated by potassium are not activated by lithium. After analyzing 680 plants from 68 families, Bertrand (1952; 1959) reported an average of 0.85 ppm in the dry weight of Monocotyledons and 1.3 ppm in Dicotyledons. Averages by Bertrand for some families are given in table 4. Robinson and others (1917) found an aver- age of 0.3-0.42 ppm in dry weight of various classes of vegetation. Borovik-Romanova (1965) analyzed 138 plants from 8 soil types by emission spectrography, using potassium chloride as buffer. She found that plants generally have an LITHIUM IN BASIN AND RANGE PROVINCE 5 average content of 0.15 to 0.3 ppm, but plants rooted in saline soils contain 2 to 10 times that level. Lithium accumulates first in the root and then moves into the older leaves where it becomes immobilized (Kent, 1941). High values were found in the dry weight of crops irrigated with water containing 0.1 to 0.2 ppm lithium; more lithium was concentrated in the leaves than in the roots (Bradford, 1966). Vlasiuk and Okhrimenko (1967) experimented with various increments of lithium to determine effects on growth, development, and productivity of tomatoes and potatoes. The lithium was introduced as lithium sulfate in amounts of 0.1, 1.0, 2.0, 3.0, 10.0, and 30.0 ppm, and the ex- periments were repeated 6 times. Curves plotted on doses versus product weight are bimodal, with two optimal values—for tomatoes, 0.1 and 2.0 ppm; and for potatoes, 0.1 and 5.0 ppm. The reason for these double optima is unknown. In 1969 the same authors reported that lithium stimulates the photochemical activity of chloroplasts and chlorophyll content in tomato and pepper leaves, with a consequent increase in the yield. According to Borovik- Romanova (1965), lithium forms part of the protein frac- tion of the leaf, and lithium ions substantially stimulate the metabolism of protein and carbohydrate in the plant. The amounts of lithium that stimulate growth and amounts that produce toxicity symptoms depend upon the tolerance of the particular species and the form of salt administered. Puccini (1957) reported that several lithium salts improved the growth of carnations, whereas lithium carbonate was toxic. Voelcker (1912) found any lithium salt to be stimulating to wheat in amounts of less than 20 ppm but to be toxic in amounts greater than 30 ppm in the soil; lithium nitrate was found to be the most toxic salt but nevertheless the most stimulating salt to growth when present in amounts less than 10 ppm. Plant species vary widely in their tolerance. Citrus trees are sensitive to lithium, and toxicity symptoms develop in citrus growing in California in areas where the lithium content of the soil is only 12 ppm (Bradford, 1966). Lithium contents in the tree reach 40 ppm dry weight or perhaps 400 ppm in the ash. Toxicity in avocado seedlings in pot culture was pro- duced by the addition of 16 ppm lithium. Symptoms in- clude necrotic spots in the interveinal leaf tissue and even- tual browning and curling of the leaf margins. Meristem damage occurred in root tips of corn exposed to high lithium concentrations (Edwards, 1941). Some plants, on the other hand, tolerate large amounts of lithium, and growth is stimulated when the soil con- tains considerable concentrations. The best-known such plant is tobacco, a member of the Solanaceae. Headden (1921 ), in plot experiments, found 496 ppm lithium oxide in the ash (or :40 ppm in the dry weight) of Nicotiana affinis and 108 ppm in N icotiana tobacum but only a trace in alfalfa grown in the same soil. A maximum value of 4,400 ppm (or :t 750 ppm in the dry weight) was reported by Strock (1936) in tobacco ash. Borovik-Romanova (1965) found species of Thalictrum, Adonis, C irsium, N ico- tiana, Salsola, Solanum, Ranunculus, Lycium, Atropa belladonna, and Datum stramonium to accumulate lithi- um to an unusually large degree. She noted that the lithi- um content is highest in samples with high alkaloid con- tents, which also varies with locality and growing season. In recent years, lithium compounds have often been used in organic synthesis, and they may participate in the bio- synthesis of alkaloids. Vinogradov (1952) described a lithi- um flora which includes Thalictrum (Ranunculaceae). Linstow (1929) reported Lycium barbarum of the Solanaceae to be a lithium indicator. Lycium, which is used in making jelly, was sampled on lands inhabited by the Pima Indians in Arizona; it contained 1,120 ppm lithium in the dry weight of the plant (Sievers and Cannon, 1974). LITHIUM IN BASIN AND RANGE PROVINCE HYDROLOGICAL RELATIONSHIPS Two distinct types of basins occur in the Basin and Range province: hydrologically closed basins, in which the ground water is only discharged by evapotranspira— tion, and hydrologically open basins, in which the ground water is discharged by subsurface drainage through under- lying sedimentary formations into a lower basin. Topog- raphy and surface drainage patterns in open basins do not necessarily coincide with the directions of ground-water movement. The Ash Meadows basin in the Amargosa Desert, for instance, drains several intermontane basins, all of which are hydrologically connected. The connec- tion is effected by movement of ground water through the Paleozoic carbonate rocks that underlie the valleys at con- siderable depth (Winograd, 1962). The thousands of feet of fill in a closed basin act as an immense underground reser- voir, constantly becoming more concentrated with salts as water is lost through evaporation at the surface and through transpiration by plants. In a closed basin the composition of solutes depends largely on the reaction of natural waters with the particu- lar lithologic units underlying that basin and occurring in its area of recharge. The composition is dependent on the simple solution of readily soluble compounds, as might occur in preexisting salt beds, and on hydrolysis of various silicate minerals, depending on their relative solubility (Jones, 1966). The end product of weathering in the Basin and Range province is generally montmorillonite and carbonates except that in basins fed by waters draining ex- tensive mineralized areas sulfates predominate. Sodium bicarbonate waters are common in volcanic areas, and cal- cium-magnesium carbonate waters are common in areas 6 LITHIUM IN UNCONSOLIDATED SEDIMENTS AND PLANTS, CALIFORNIA AND NEVADA of limestones. Mixed waters may have originated in tuffs and then been altered by traveling through limestones, or the sequence may be the reverse (Eakin and others, 1963). Secondary influences on the solute composition are con- centration by evaporation and concentration by organic respiration and decay (Jones, 1966). The evaporative process involves a “continuous precipitation of salts in concentric zones from the edge inward according to their relative solubility: alkali-earth carbonates, then sulfates, then chlorides at the center.” The chemical zonation of sediments and the typical plant species that are able to grow on each zone (generally identified by anion as car- bonate, sulfate, and chloride zones) were studied by C. B. Hunt (1966; Hunt and others, 1966) in Death Valley, where he made on-site analyses of sediments and plants. Lithium is reported to remain in solution until a late stage and to be precipitated with potassium and boron in the chloride zone (Stewart, 1963). Our sampling in Death Valley shows high values for lithium in both the sulfate and chloride zones; and Lombardi’s (1963) analyses of springs in Saline Valley (west of Death Valley) show a similar concentration of lithium in high-sulfate water. Lombardi noted that in transit lithium is rapidly ab- sorbed by clay in suspension. For example, the water of Vega Springs, Saline Valley, loses 98 percent of its lithi- um in flowing 2 miles (3.2 km) from spring to playa. Lithium also is concentrated to a considerable degree in lacustrine montmorillonite clays of closed basins whose recharge is from volcanic rocks. The amount of lithium concentrated in the brines from a large recharge area is considerable; a maximum of 1.62 percent lithium was found in the dried residue of Owens Lake water (Gale, 1915). The extraction of lithium, as well as potassium, from these large closed-basin reservoirs of brine is economically feasible; lithium chloride is cur- rently being extracted from brines in several basins in Nevada and California. Lithium concentrations in wells and springs are greater in chloride and sulfate waters than in bicarbonate waters and are significantly greater in closed basins than in open basins. Contents in springs and wells in southern Nevada, reported by B. P. Robinson and W. A. Beetam (written commun., 1965), Pistrang and Kunkel (1958), and Eakin and others (1963), range from 0.00 ppm lithium to 0.3 ppm, or 300 Jug/1 (table 1). Eighteen water samples from Amargosa Desert, a discharge area for several basins to the north, had a median lithium value of 105 ug/ l, 50 times the median of 100 major city water supplies in the United States. The contents of lithium and associated ions in basin sediments that we sampled are given in table 2. We con- clude, as did the Russian scientists, that lithium is con- centrated in residual chloride-sulfate brines of evapora- tive basins and also in evaporative lake clays close to volcanics. LITHIUM UPTAKE BY DESERT PLANTS Plants and soils were collected in relatively low areas of 20 basins in Nevada and California to study the uptake of lithium by salt-tolerant species (table 3). Our analyses indicate (table 4) that native plants in the arid basins of the Western United States absorb much more lithium than do native plants in humid areas (Bertrand, 1952, 1959; Robinson and others, (1917). For the arid-basin samples, the median lithium content for 68 samples of dicotyledons is 150 ppm in the ash or 22.8 ppm in the dry weight, as compared with an average of 1.3 ppm in dicotyledons collected by Bertrand. A median content of 27 ppm lithium in dry weight of 47 samples of Chenopodiaceae from our desert collections is markedly higher than the average of 3.8 ppm found in Chenopodia- ceae by Bertrand; except for Atriplex polycarpa, the median is higher (in lithium) than for plants sampled at the Nevada Test Site and Mohave Desert by Wallace, Romney, and Hale (1973). They found as much as 111 ppm lithium in the dry weight of Atriplex polycarpa and 176 ppm in Lycium andersoni in Rock Valley, which is hydrologically an open valley but which contains a TABLE l.—Lithium, strontium, boron, and fluorine content (At/l, ppb) in some waters of Nye County, Neu, and Inyo County, Calif., compared with 100 municipal water supplies in the United States [Data on municipal water supplies from Durfor and Becker (1964); data on Nevada Test Site and Amargosa Desert from B. P. Robinson and W. A, Beetam (written commun., 1965). ND, not detected] Water supply, Open basins, Closed basins, 100 major U.S. Nevada Test Site Amargosa Desert cities (8 wells) (18 springs, wells) Median Range Median Range Median Range 2 ND— 170 20 13— 46 105 0— 300 110 2.2— 1,200 40 32— 160 800 500- 7,700 31 2.5— 590 180 l 10— 280 420 90— 2,800 400 0 — 7,000 1,000 500— 2,400 2,100 700— 7,000 Hardness as CaCOa (Ca+Mg) ............................... 90,000 0 —738,000 45,000 8,000-116,000 119,000 26000-482000 LITHIUM IN BASIN AND RANGE PROVINCE 7 TABLE 2.—Maximum lithium concentrations (percent) in some basins of Nevada and California TABLE 3.—Plant species sampled in Nevada and California [Localities are shown in fig 1] Family Genus and species Common name A Maximurgilithium Chlorophytaceae ........ Phovmidium ambiguum Gorn Algae. rea m se iment or Ephedraceae .. .. Ephedra viiidis Cov. .............. Green mormontea. 8‘ aporite Gramineae. Distichlis strictu (Torn) Beetle Saltgrass. Closed basins, discharged by evapotranspiration Sporobolus airodes (’l‘ort.) Torr Dropseed. Scirpus olneyi Gray .................................... Cyperaceae ,,,,,,,,,,,,,,,,,, Olney bulrush. Amargosa Desert, Nye County, Nev., and lnyo County, Cali 0.15 Big Smoky Valley, Nye County, Nev ....................... .01 Juncaceae .......... Juncus cooperi Engelm .............................. Cooper rush. Bristol Lake, San Bernardino County, Calif. .03 Capparidaceae .. Cleomella obtusifolia Torr. 8c Frem. Stinkweed. Clayton Valley, Esmeralda County, Nev .................................... .05 Chenopodiaceae ......... Allenrolfea oecidentalis (Wats.) ........ Pickleweed. Kuntze. Columbus Salt Marsh, Esmeralda County, Nev .023 Atriplex caneseens (Pursh.) Nutt ...... Saltbush. Death Valley, Inyo County, Calif ........... .03 Fourmile Flat, Churchill County, Nev .. .007 con/ertifolia (Torr. 8c Frem.) iiiiiiiiiiiiiii Shadscale. Hector, Mohave Desert, San Bernardino County, Calif. .20 Wats. hymznelytra (Torn) Wats. . Desertholly. Oasis Valley, Nye County, Nev ,,,,,, .015 lenti/ormis (Torn) Wats ..... Big saltbush. Owens Lake, lnyo County, Calif .08 linearis (Wats.) Hall 8c Clem. Narrow leaved saltbush. Railroad Valley, Nye County, Nev. .015 Sarcobatus Flat, Nye County, Nev ..................... .03 torreyi (Wats.) Wats. Torrey saltbush. Median ............................. 0.03 PM?" Wa‘S-u Pam! Sallbum palycarpa (Torn) Wats. .. Cattle spinach. , . s ini era Macbr. .......... Spiny shadscale. Open basms, discharged by underflow Eurot’ila lafnata (Pursh.) Moq. Winterfat. Cactus Flat, NYE County, Nev. 0-004 Kochia amen'cana Wats.. Greenmolly. Dry lake Valley, Clark County, Nev. '005 Salsola kali L. ................ Russian thistle. Frenchman Flat, Nye County, Nev '015 Sarcobatus vermiculatus (Hook. .. Greasewood. Gold Flat, Nye County, Nev .008 Torr. Suaeda torreyana ramossissima .................. Bush seepweed. Kawich Valley, Nye County, Nev .............. .008 (Standl.) Munz. Silver Lake, San Bernardino County, Cali .008 Stonewall Flat, Nye County, Nev .007 [offeyana Wats. Torrey seepweed. Yucca Flat, Nye County, Nev ........................... fl oecidentalis Wats. Western seepweed. Median .......................................................................... 0.0075 Composi‘ae " L is 3' Peri Seepwillow Bebbia juncea aspen: Greene Rush bebbia. anseria dumosa Gray ......... White bursage. Hymenoclea salsola T. 8: G.. White burrobush. sodium-calcium sulfate type of water unusual to the c: . “”43?" "Tm (NP'L) 0“" Amwmd' . . . uciferae L r , ‘ Wats. . Fremont peppergrass. reglon. The source of the Water 15 a sequence of volcanics Leguminosae .............. Prosopis glandulosa (Sw.) DC ................... Honey mesquite. (SChOff and Moore’ 1964) Saururaceae.. Anemopsis ealifomica Hook. .................... Yerbamansa. The highest lithium contents in plant ash were found in Solanaceae-m Lyn'um andersom' Gray. Anderson wolfbmy. ‘ . - L. allidum Miers ...... Pale wolfberry. Juncus 60017611, a ruSh’ and A llenTOZ-fea OCCldentalzs’ Tamaricaceae .. Titian): pentandia Pall. or gallica. Tamarisk. pickleweed, rooted in a hot spring on the east side of Clayton Valley. Most species that contained more than 400 ppm lithium in the ash are sodium chloride accumulators in which sodium may be absorbed and used in place of potassium. For certain species of Atriplex, sodium is an es- sential element (Brownell, 1965). The sodium plus potas- sium content in the plant appears to remain constant for the genus, but the sodiumzpotassium ratio varies with the species. Except for the hydrophytes juneus and Anemopsis, the high-lithium plants are all halophytes from the Chenopodiaceae family (table 5). The relation- ship between the uptake of sodium and lithium in all the plants collected is as follows: Percentage of sodium Concentration ratio of in plants lithium in plants:lithium in soils 0-10 ............................................................. 1.6 10—25.... 25+ ....... The lithium content of native plants correlates well with the lithium content of the soils in which the plants were rooted (fig. 2) The correlation is better than might have been expected, inasmuch as only the top 5 inches of soil or evaporite were sampled, and the roots of the larger phreatophytic shrubs may extend to considerable depths and actually reach a deeper zone of chemical precipitate. The concentration ratio of lithium in plant to lithium in soil is greatly increased for sodium accumulator plants. A maximum concentration ratio of 10 occurred in samples of pickleweed and desertholly. Commonly, the ratio is also increased where the soils contain less than 100 ppm lithium. A plot of lithium in plants against calcium in soils shows no pattern that confirms an interference in up— take of lithium by the presence of high calcium as Epstein had reported (1960). A more detailed discussion of the dis- tribution of lithium in soils and plants, by area, follows. OPEN BASINS DISCHARGED BY UNDERFLOW Many basins of the Basin and Range province are discharged by underflow and have been filled by clastic LITHIUM IN UNCONSOLIDATED SEDIMENTS AND PLANTS, CALIFORNIA AND NEVADA TABLE 4.—Lithz'um content (parts per million) in plant groups of the Great Basin, California-Nevada [Comparative dam on analyses of plants growing in Franu‘ from Bertrand (19:39). I’, phl’l‘alophylc; Ha, halophylv; Hy, hydrophyle; X, xerophyle; ND, not dt'lt'rll'd; < , less than; N , approximalcly] France Great Basin, California—Nevada order and family L1 3::rage Genus and species 2:. 11‘; P::}Cle:rt1 Lithium in ash dry wt. samples dry wt. dry wt. Median Median Median Range Lower plants Blue—green algae—- ———— Phomidium ambiguum (P) ------- 2 170.0 85.4 200 110- 300 Monocotyledons -------- 0.85 Group average-----—--—-----——— 6 11 15.8 70 60-3,000 Cyperaceae ———————— .4 Scirgus o_1neL1 (Hy) ----------- 1 54 15.5 350 ---- Gramineae ———————— .75 Distichlis stricta (P) -------- 3 9.5 15.8 60 30-1,000 SEorcbolis airoides (Ha) ------ 1 12.6 15.8 80 ---- Juncaceae --------- .89 w coogeri (Hy) ----------- 1 186 6.2 3,000 —--- Dicotyledons ———————— — 1. 3 Group average ------------------ 68 22.8 15.2 150 <20-3,000 Capparidaceae ----- -—-— Cleomella obtusifolia (Ha)--—— 1 135 9. 150 ---- Chenopodiaceae-——- 3. 8 Family average ----------------- 47 27 23. 120 <20-3,000 Allenrolfea occidentalis (Ha)- 6 258 28. 7 900 300-3,000 Atriglex canescens (P) -------- 2 18 20. 90 20- 150 A. confertifolia (X) ---------- 6 6.7 16.8 40 30- 500 A. hzmenelxtra (X) ------------ 2 124 31.2 400 300- 500 A. lentiformis (P) ------------ 1 28.5 19. 150 —--— A. linearis (P) --------------- 2 14.6 18.3 80 20- 150 A. REY}. (X) ----------------- 4 22.8 21.1 108 30- 200 A. Eolxeaga (X) -------------- 4 25.5 17. 150 20- ”0 A. sginifera (X) -------------- 3 14.2 12.9 110 50- 150 A. torrexi (P)-------—~---—-- 1 102 20.4 500 ---- Eurotia 1an*at§ (X) ————————— 1 16.5 11. 150 ---- mg americana (P) ---------- 1 6 24.8 20 --—- Salsola Lall (X) -------------- 1 ND 22.8 ND ---- Sarcobatus vermiculatus (P)-—— 7 13. 7 12. 7 60 <20— 700 Suaeda occidentalis (P) ------- 1 43.5 29. 150 -—-- g. ramssissima (P) ----------- 2 67 22.3 300 100- 500 A. torrexana (P) -------------- 3 99.3 33.1 300 50- 700 Compositae -------- 1.33 Family average ------------------ 8 15.1 9.6 155 40— 175 Baccharis glutinosa (P) ------- 1 15 10. 150 ---- M M m (x) —————— 2 13.9 9.25 150 150- 150 Franseria dumosa (X) ---------- 1 3.9 9.7 40 --—- Hymenoclea salsola (P) -------- 2 26.6 15.2 175 150- 200 Pluchea sericea (P) ----------- 2 15.2 9.5 160 35- 300 Cruciferae -------- 1.7 LeEidium fremonti (X) --------- 2 4.45 7.8 57 40- 75 Leguminosae ——————— .81 M W (P) ——————— 1 .39 3.9 10 ---- Saururaceae ------- ---- Anemogsis californica (Hy)--—- 1 25 16.7 150 —--— Solanaceae -------- 2.12 Lycium andersoni (X)—-———-—-— 1 25.5 8.5 300 ----- A. Eallidum (X) -------------- 2 “44.2 7.7 N55 <25- 100 Tamaricaceae ------ -—-- Tamarix Eentandra (P) --------- 5 16 11.4 140 30- 300 LITHIUM IN BASIN AND RANGE PROVINCE TABLE 5,—Maximum contents of lithium and sodium in some plants and basin soils [Analysts: Thelma F. Harms, B. N. Lanthorn] Plant Soil Na Li Na Li Plant ash 111:? (percent) (ppm) :2). (percent) (PPm) 1131:10th r3110 ASh wgrgyht ASh with: wag/m wzght Allenrolfea occidentalis (pickleweed) ............ D413057 29 10.1 3,000 1050 D413102 2.1 300 10 luncus cooperi (rush) ......................... D408136 22 1.4 3,000 185 D408181W 3.0 500 6 Distichlis spicata (saltgrass)... D413091 18 2.7 1,000 150 D413098 2.0 120 8.3 Suaeda torre’yana (seepweed) .............. D413054 26 6.2 700 168 D413098 2.0 120 5.8 Sarcobatus vermiculatus (greasewood) D408134 23 2.9 700 90 D408177W 4.7 300 2.3 Atriplex torreyi (Torrey’s saltbush).... D411534 26 5.3 500 100 D413125 1.4 70 5.0 confertifolia (shadscale) ............ D405826 20 3.5 500 90 D405847 3.0 70 7.1 hymenelytm (desertholly) ....................... D412065 23 6.9 500 150 D412058 1.3 50 10 0.5 Yucca Flat, for instance, there is no caliche in the top 18 feet (5.5 m). ' ' ‘ Alluvial basins discharged by underflow are not large reservoirs of water that are constantly being evaporated to brines; rather, they are undergoing a flushing action that ' ° ' prevents the buildup of salts. These basins can support w OJ lakes only if the underflow is considerably less than the E recharge or if the sedimentation is such as to form a g ' ' perched water table. Some such situation must have 3 0-05 . existed in the pluvial period in Gold Flat and Dry Lake 3 , , Valley, as evidenced by old beaches, although the present 3:: - ° ' 0 - -. ground-water table is several hundred feet deep (Eakin, E . 1963). Perched water appears to be close to the surface on 8 ' .. the southern part of the Gold Flat playa, and many 3 z': ' ' ' '" phreatophytes grow there. The soil contains 5.4 percent L5” 001 water-soluble sulfate, and a white efflorescence of salts E . . . . coats the surface of the ground. a 0 005 ‘ ' ' Analytical data for eight open basins are given in table 6; ' , the sample localities are shown In figure 1. . ; . The lithium content of sediments collected from 0 - ' hydrologically open basins was generally low, ranging “ ' from not detected to 150 ppm, with a median of 50 ppm and a mean of 54 ppm (table 6). The playas of open basins are largely composed of calcium-magnesium carbonate 001510.01 0.005 0.01 0.05 0.1 and clay; the surrounding sediments are sandy alluvium PERCENTAGE OF LITHIUM IN SOILS containing considerable caliche. The magne- FIGURE 2 —Relationship between lithium in plants and that in soil. Slumillthlurn ratlo ls hlgh. Boron contents are 51m11ar t0 ' those of 11th1um and range from 20 to 70 ppm. Probably materials of variable texture deposited from intermittent no lithium or boron deposits 0f economic value can be streams during torrential storms. The coarser materials are expected in basins drained by underflow. deposited on the higher parts of alluvial fans, and the fine A maximum of 500 ppm lithium was found in a sample Silt and clay eventually reach the basin, where they are of Atriplex confertifolia from Stonewall Flat. This high deposited in a thin layer over the playa as the water evapo- lithium content, however, may be caused by contamina- rates. Caliche, which is distributed in this type of basin tion because neither the soil nor Lepidium collected at the through hundreds of feet of alluvium above the water same station contained much lithium; all three materials table, probably is flushed out of the nearosurface soils and were collected near ground zero of a surface plutonium re— is precipitated from surface water as it percolates down- lease which may have been accompanied by lithium flares. ward through the beds; this contrasts with the upward The 20 plant samples collected in open basins had a medi- capillary flow and near-surface deposition of caliche in an content of 30 ppm lithium in the ash or 4.8 ppm lithi- closed basins (Johnson and Hibbard, 1957). In northern um in the dry weight. 10 TABLE 6.—Percentage of lithium and associated ions in sediment [Stations shown in fig. 1. Analysts art- named in introduction. N, not llt‘lflk‘tl; Ma, (’lt'mC LITHIUM IN UNCONSOLIDATED SEDIMENTS AND PLANTS, CALIFORNIA AND NEVADA The sail in which the sampled plant grew is listed directly below each plant] s and the ash of plants from open basins in Nevada and California nl present in major quantity; X, xerophytc; P, phreatophyle; M, mesophyte; WSS, water-soluble salts Sta- Lab. Carbo- Sulfate tion Sample No. Ash WSS Lithium Magnesium Calcium Sodium Potassium Boron nate (water— No. (D-) (as 002) soluble) (percent) Plant Soil Plant Soil Plant Soil Plant Soil Plant Soil Plant Soil 5011 Plant Soil Frenchman F1at——Nal-IC03 ground water at depth 623—1,103 ft (190-336 m) F6 Lycium paZZidwrt ( ———————————— 403492 5.8 ——-- 0.01 ———- 5.0 ———— 28 ——-— 1.6 12 —-—— ——————————— -——— 0.19 ———- Sandy clay west of playa ——————— 403201 ———— 0.30 ————— N ————— 1.4 ————— 4.0 ————— 1.7 ————— 2.9 0.01 —————— 2.4 ----- <0.01 Lyaiztm‘pallidwn (x)1-'-— 412496 9.5 ———- <.002 ————— 1.5 ———— 33 —--- .6 -——— 5.1 ———— ————— <0.002 -—-— Playa clay- 412501 ————————— 0.005 ————— 2.0 ————— 9.9 ————— .95 ————— 2.3 .015 .003 --—- F10 Playa clay an east side (no vegetation)——-- 411562 ———— ----- .003 ————— 3.2 ————— 10 ————— .85 ————— 2.0 ————— <.002 ———- Atriplex linearis (P) ------------ 413078 16 —--- 002 ————— 7.0 9.0 —-—— .28 25 .015 A. Spiflifera (X)- 413079 21 —--— .002 ----- 5.0 12 -——— 10 17 .02 Kochia amzricana (P) ———————————— 413080 16 --—— .003 ————— 7.0 11 ———— 11 ——-- 16 ———— .07 ------ ———- .33 ---— Playa clay with plants ————————— 413115 ————————————— .004 ————— 3.0 ————— 10 ————— .8 ————— 1.9 ————— .005 —-—— ————— .03 F21 Playa clay (no vegetation)-——- 406182 ———— .13 .015 ————— 3.1 ————— 7.8 ————— .53 ————— 2.7 ----- 002 8.1 ————— —--— Yucca F1at-—NaI-IC()3 ground water at depth 670—1,539 ft (204—470 In) Y40 Tamarix pentandra P) — 406532 14.1 ---— 0.003 ————— >10 --—— 17.6 ——-— 4.0 —--- 8.2 ——-- 0.07 ------ ——-— 1.7 ---- Atm'plex canes— czns (P) ------- 411573 23.0 -——— .002 ————— .5 -——— 6.6 ———- 3.6 --—- 32.0 --—-- .01 —————— ———- ----- ---- North playa soil in cattle reservoir ----- 411564 ————————————— 0.004 1.0 ————— 2.0 ————— 1.2 ----- 3.0 ————— 0.007 1.4 ----- —--- Y5 South playa clay at crack O—l.5 (no vegetation) 403218 .007 ————— 1.9 74 .005 3.1 <.01 <0.01 at 1.5-15 cm--— 403219 .005 1.9 .74 .003 2.9 ----- .23 at 1 m --------- 403220 N 1.2 1.3 .003 .15 ----- <.01 Atriplez confer— tifoZia (x)—-- 403522 17.3 -——— .003 —————— 2.0 ———- 13.1 ———— 10.6 —--- 14.4 —--— ------ ———- 55 ---- Sandy alluviuIn-— 403221 ——-- .15 ————— .005 ————— 1.2 ----- 2.8 ————— 1.5 ————— 3.2 .003 1.1 ----- .02 Yl4 Lycium andersani ( ) ——————————— 408648 8.5 ———- .03 —————— 2.0 -——— 25.2 .20 15.2 ——-- .02 —————— .52 —--- SaZSoZa kali (X) 408649 22.8 ——-- N 0035 1.5 ———- 8.6 .35 27.2 ---- .01 .003 .34 ---- Alluvium ———————— 118531 ——-— .08 ———————————————— .54 ————— --—— 3.0 --—— —————— ---— .01 Kawich Valley——NaHCO; ground water at depth 660 ft (200 111) K12 Atriplex confer- tifOZia (X)-——— 403585 ——-- 0.0045 —————— 2.0 — -— 11.2 —--— 11.8 -——— 13.2 ---— 0.01 —————— —--- 0.36 --—— Playa Clay —————— 403255 0.05 —————— 0.005 ————— 1.3 ----- 1.8 ----- 1.5 ---— 3.3 ---- 0.005 0.55 0.17 K44 Playa clay (no Vegetation)—-- 405840 ———- .76 —————— .008 ————— 1.8 3.2 ————— .89 ———- 3.2 ——-- .005 2.2 ——-- 1.3 Cold F1at——NaHCOg‘ ground water at depth 230 ft (70 m) GF33 Kochia mericana (P) ——————————— 405816 24.8 -——— 0.002 2 --—— 8.8 ———— 24 ———- 15 ———— 0.02 —————— ---— 0.49 ———- Suaeda occiden— talis (P) ————— 405817 29.0 ——-- .015 ————— 1.5 ———- 3.8 —--- 31 9.4 ---- .015 —————— ———- .68 ---— silty sulfate zone —————— 405844 —--— 2.1 ————— 0.008 ————— 1.5 ----- 4.5 ----- 1.3 ---— 3.2 ----- 0.005 1.6 ---— 5.4 Cactus Flat—-NaI-lCOa ground water at depth 110—443 Er. (35—135 m) C3 Atriplex confer- tifolia (x)-—- 405822 10.5 ————— 0.0035 ————— 2.0 —--— 7.0 -——— 17 ——-— 14 0.01 —————— --—— ———- ——-- Eurotia lanata (x) ----------- 405823 11.0 15 ----- 2 --—— 7.0 ———— 1.1 ———— 16 ——-- .02 —————— ———— ---- Alluvium ———————— 405845 ——-- —————— 0.004 ————— 1.0 ————— 2.0 ————— 2.0 ———- 3.0 <0.002 ——-- ---— Stonewall F1at—-altered NaHCOa ground water at depth 100-275 it (30-85 in) C5 Atriplzx confer— 405826 17.5 ————— 0.05 ————— 1.5 ——-— 5.0 ---— 20 -——— 9.4 ———— 0.02 —————— --—— —--- ---— 405827 5.3 ————— 004 ————— 2.0 —--— Ma --—— 2.4 -——— 21 ———— .03 ---— ---— Alluvium ——————— 405847 ——-— 0.1 ————— 0.007 ----- 1.0 ----- 3.0 3.0 ———— 5.0 ———- 0.003 ---— -—-— Dry Lake Valley-~ground water at depth >300 ft (>90 in) L22 Atriplex Zineflwis (P) ———————————— 413074 15 ----- 0.002 ----- 15 -—-— 0.26 ——-- 21 ———— 0.02 ------ —--- 0.65 ———— A. antiformtis (P) —— 413075 19 .013 ————— 10.0 ———— 9.2 ———- 12 —-—— 17 .02 —————— ——-- .77 --—— Playa _sands ————— 413112 ————— 0.005 ----- 2.0 9.4 ----- 0.70 ---- 2.0 --—— 0.007 ---- ---- 0.15 Silver Lake L21 Atriplm.‘ poly- carpa (P) —————— 413090 13 ————— 7.0 ———— 6.0 ———— 16 __-_ 16 ———— 0.05 —————— ———— 0.46 ——-- Sandy soil at playa edge ————— 413124 —————————————— 0 007 ————— 2.0 ----- 3.4 ----- 2.2 ———— 2.3 --—— 0.002 ———— ——-— 0.02 Playa mud (no vegetation)——-— 415475 ———— 2.0 ————— .008 ————— 2.0 ————— 3.8 ————— 2.5 -~—— 1.9 —-—— .007 --—— .54 lFluorine content of plant: 0.0063 percent. LITHIUM IN BASIN AND RANGE PROVINCE 11 TABLE 7.—Percentage of lithium and associated ions in basin sediments, collected within 6 inches (15 cm) of the ground surface [N, not detected; number of samples shown in parentheses] Li thium Magnes ium Mg /Li P otass ium Boron Median Range Median Range Ratio Median Range Median Range Closed basins Lake clay (5) associated with voleanics-— 0.07 0.03 -0.2 3.0 0.3 —11.0 43. 1.7 0.24— 2.4 0.005 0.002—0.015 Spring—deposited tufa and lake marls (8)— .045 .003 — .20 2.85 .33— 5.0 63. .73 1 —18.0 .003 <.002— .03 Chloride zone (13) ---------------------- .015 .0025- .03 2.0 .03- 4.0 133. 1 9 8 - 2.4 03 .015- .07 Sulfate-chloride mixed (9)-----—-—--- .015 .003 - .03 1.7 .07- 5.0 113. 2.9 .24- 3.7 .05 N - .15 Black oozes in hot springs (6) ----------- .01 .004 - .08 1.0 .70— 2.0 100. 1.12 34— 3 0 .005 3002- .02 Carbonate zone (11) ---------------------- .015 N - .03 1.0 .07- 4.8 66. 2.3 24- 3 6 .015 N - .03 Playa clays, calcareous (7) -------------- .015 .006 — .023 1.3 7 — 3 2 86. 2 45 2 3 - 2.6 .019 .005— .07 Alluvium (4) ----------------------------- .003 N - .005 1.0 1 0 - 1.1 333. 3.0 2.8 — 3.5 .003 <.002- .005 Basin sands (6) -------------------------- .004 N — .017 1.5 48— 3 0 375. 2.4 2.0 - 3.4 .005 .003— .2 Open basins Sulfate zone (1) ------------------------- 0.008 ——————————— 1.5 ————————— 187. 3.2 ———————— 0.005 ———————— -— Playa clays (12) ------------------------- .005 N -0.015 2 0 1.0 - 3 2 400. 2.7 1.9 - 3.3 .005 <0.002-0.007 Playa sands (2)------------------------- .006 0.005 - .007 2.0 --------- 333. 2.2 2.0 - 2.3 .004 .002- .007 Alluvium (4) ---------------------------- .005 .004 - .007 1.0 54- 1 2 200. 3.1 3.0 - 5.0 .003 (.002- .003 HYDROLOGICALLY CLOSED BASINS Lithium is concentrated particularly in the evaporites, spring deposits, and lacustrine clays of closed basins. The maximum lithium concentration found in sediment and clay samples that were collected in 12 closed basins in Nevada and California ranged from 30 to 2,000 ppm. Samples collected from lacustrine clay deposits have a median concentration of 700 ppm lithium; spring- deposited tufa and lake marls, 450 ppm; those from the sul- fate, chloride, and carbonate zones and, also, those from playa clays, which commonly contain considerable detrital calcium carbonate, contain 150 ppm; samples of organic black oozes in hot springs, 100 ppm; alluvium and basin sands from the fans that surround the basins, 30 and 40 ppm (table 7). Apparently lithium and boron differ in their geochemical behavior. Lithium becomes concentrat- ed in lake clays, particularly those of volcanic origin and in spring deposits, whereas boron does not. A general increase for both lithium and boron content from the fans to the center of the basin is probably real, as shown by samples collected in Death Valley (fig. 1), although the content of a particular zone is not consistently high or con— sistently low. A close correlation of lithium, boron, and potassium might be expected in view of their common volcanic origin and their precipitation at a late stage in brine evaporation. There is little zonal variation in the potas- sium concentrations of basin sediments, although the brines are highly enriched in potassium. However, the boron contents increase from 30 ppm in alluvium on the fans to 500 ppm in the sulfate-chloride zone in the central parts of the basins. Neither element is precipitated in quantity in the spring deposits. In evaporites, the boron content exceeded the lithium content, but in detrital sedi— ments the reverse was found. The magnesium content of the basin sediments varies widely because it occurs in residual dolomite that has been deposited in the carbonate zone and in the playas. Magnesium also occurs in sulfates and chlorides that have been precipitated from evaporite brines. The magnesiumflithium ratio varies from basin to basin, which affects the economic value of the deposits, a low magnesiumzlithium ratio being most desirable. The ratio is lowest in collections from Owens Valley and is also relatively low in spring deposits and lake clays. The amount of lithium found in the spring-deposited tufas and the marls of springs that occur at the alluvium- evaporite contact in these closed basins is much greater than the average content of lithium in marine limestones. The phenomenon probably reflects a secondary cycling in the basins, as lithium trapped in clays and other sedi- ments may be easily returned to solution by thermal waters. Hot springs are highly charged with carbon dioxide, and so alkali carbonates precipitate around the orifice of a spring of low discharge due to evapotranspira— tion, respiration, and decay of marsh vegetation (Jones, l2 LITHIUM IN UNCONSOLIDATED SEDIMENTS AND PLANTS, CALIFORNIA AND NEVADA FIGURE 4.—Precipitate of biogenous origin occurring in place, Amargosa Desert. Sample from Station A5B. FIGURE 3,—Crown of rush killed by hot-spring precipitate, Clayton Valley, Sample from station L4B. 1966). Such concentrations around the roots of rushes (Juncus cooperz') in a hot spring of Clayton Valley gradu- ally kills the plants. These “fossilized” root crowns con- tain more than 1,000 ppm lithium (fig. 3). Lithium-rich lacustrine deposits of marl and clay commonly show a definite root-cast structure that suggests precipitation around plant debris and in old root casts in marls of Amargosa Desert (figs. 4, 5, 6). Thus, the lithium content of many types of sediment is highest in closed basins discharged by evapotranspira- tion, particularly basins that receive recharge from several adjacent basins. Within a given basin, lithium is particu- larly concentrated in lacustrine deposits of volcanic origin and in biogenous deposits precipitated from thermal waters associated with these closed basins. DEATH VALLEY Death Valley is a closed basin from which no water escapes, as the floor of the valley extends 280 feet below sea level. Old shorelines mark the lowering by evaporation of Pleistocene Lake Manley, which was originally 600 feet (180 m) deep. Mineralized water has also been contributed yearly by freshwater rivers and by underflow from nearby LITHIUM IN BASIN AND RANGE PROVINCE l3 FIGURE 5.—Calcium carbonate precipitate on organic detritus, Amargosa Desert. Sample from station A5. FIGURE 6.——Cast of former root or stem structure, Amargosa Desert. Sample from station ASA. basins (Hunt and Robinson, 1960). Thus, lithium has been concentrated as a residue in the mother liquor in this evaporite basin from an enormous quantity of water and has been redistributed laterally through seasonal flooding and hot-spring action. Discharge by evapotranspiration of this large quantity of water has resulted in a lateral zonation of salts charac- terized by chlorides in the center of the basins ringed suc- cessively by sulfates and carbonates. The carbonate and sulfate zones are commonly contaminated with salt (sodium chloride) during seasonal flooding. Ground- water conditions are rarely homogeneous to the depth penetrated by plant roots because permeable layers alter- nate with impermeable layers. The carbonate zone may be underlain by a caliche of sulfates, and so forth (Hunt and others, 1966). Only bacteria grow in the chloride zone of the salt pan and fungi and algae in the sulfate zone (Hunt, 1966). Pickleweed (Allemolfea) grows on the sand facies of the carbonate zone where the sands are impregnated with salt. Sixty percent of the ash of this plant is composed of sodium chloride, and mounds of exudated salt build up as high as 2 feet (0.6 m) around the base of the plant (fig. 7). Saltgrass (Distichlis), which grows on the same sand facies, contains a lesser percentage of salt in the ash, but the true amount of salt in the plant is masked by the large amount of silica in the ash which affects all the per- centage values. Arrowweed (Pluchea) absorbs large amounts of sulfate but little sodium chloride. Arrowweed is restricted to soils of less than 5 percent water—soluble salts and mesquite to less than 0.5 percent. Thus, mesquite (Prosopis) grows only around springs of relatively fresh water that issue along the contact between the alluvial fans and the salt pan. Xerophytes, such as desertholly (A triplex hymenelytm), cattle spinach (A. polycarpa), and creosote- bush (Larrea) cover the alluvial fans (Hunt, 1966). Hunt (1966) reported the following percentage con- l4 LITHIUM IN UNCONSOLIDATED SEDIMENTS AND PLANTS, CALIFORNIA AND NEVADA FIGURE 7.—Pickleweed (Allemolfea) on salt mounds in Death Valley. Mound beneath near bush is approximately 2 feet (0.6 m) high.- centration of major ions in the ash of some Death Valley plants: C05 or P0< Plant name and number Ca Mg K Na SO. Cl needed to SiO2 of samples balance cations Xerophytcs Atriplex hymenelytra (2) ..... 6 2 7 26 4 25 30 Atriplex polycarpa (1) ......... 11 6 8 14 4 14 40 < 5 Phrcatophytes Pluchea (6) .............. ll 4 10 12 23 8 32 Distichlis (6).. 2 l 2 5 5 9 5 70 [uncus (2)... ..... 4 2 19 8 l3 9 20 25 Suaeda (l) 3 5 9 20 8 20 35 Allenrolfea ( ............ 2 2 5 32 7 27 25 Atriplex linearis (l) ............. 9 4 13 17 5 l2 The relatively constant ratio of sodium to chloride in the plants suggests that the two ions are absorbed together. The correlation between lithium and sodium shown in our analyses suggests that lithium is adsorbed as a chloride. Our samples in Death Valley came chiefly from the car- bonate zone, the lower parts of the alluvial fans, and near springs so that we could study the uptake of ions by plants (table 8). The content of water-soluble salts in the soils that were collected at plant sample localities was higher in sev— eral instances than the tolerance limits reported for these species because the samples were of surface soils and in- cluded an efflorescent crust. The soils actually in contact with the plant roots (for example, mesquite) may be as much as 50 feet (15 m) below the surface. In general, lithium contents in surface sediments of this evaporite basin decrease laterally from the center out- ward: Chloride and sulfate zones ..... 150—300 ppm Carbonate zone ......................... 30—150 ppm Alluvium and dune sand ..... 35— 50 ppm Anomalous amounts of lithium (300 ppm) were detected in the travertine deposits of Travertine Spring in Furnace Creek Wash, the sulfate sediments at Badwater, and in the sediment (as opposed to the halite pinnacles) of the salt pan or chloride zone (DlA). This concentration was found in the brown soft sediment below the polyg- onal hard crust of the streambed which meanders across the salt pan. The surface polygons of halite with up- turned edges are hard enough to support a man’s weight. Lithium in the ash of plants that were collected in Death Valley ranges from 30 to 500 ppm in the ash, or 7.3 to 160 ppm dry weight. Plants containing high lithium corres- pond with high lithium in the substrate. The highest con- centrations of lithium are in Allenrolfea (pickleweed), in Atriplex hymenelytra (desertholly) and Suaeda ramossis— sima (seepweed) collected near the edge of the salt zone where the roots presumably are periodically contaminat- ed with salt during floods. The blue-green algae at Bad- water (sta. D8) in the sulfate zone (which is also high in sodium chloride) contained only 110 ppm lithium, a sur- prising low value considering the metal-accumulative ability of most algae. AMARGOSA DESERT The Amargosa Desert is not topographically closed; the Amargosa River flows into it from Oasis Valley in the northwest corner and out into Death Valley at the south- east corner. Nevertheless, little water is actually lost from the valley because the streambed is completely dry in the summer, when the temperatures and evaporation rates are high. Hydrologically, therefore, the basin is presumed to be closed and is characterized by a noticeable upwelling of ground water in many large springs; even that flow is con- sumed in the summer by evapotranspiration. The chem- istry of Death Valley waters shows (Hunt and Robinson, 1960) that a certain amount of water from the Amargosa Desert is discharging by underflow into Death Valley. According to Winograd (1962), who studied the altitude of the piezometric surface of the semiperched Cenozoic aqui- fers in Yucca, Frenchman, and Jackass Flats, the ground water of these valleys moves downward into the under- lying Paleozoic carbonates and discharges into the Amar- gosa Desert. Analyses of water from springs and wells (Walker and Eakin, 1963; B. P. Robinson and W. A. Beetam, written commun, 1965) show high calcium and magnesium in the ground water discharged largely from limestones in the northeastern part of the valley and high bicarbonate, sulfate, chloride, sodium, fluorine, potassium, and boron in waters recharged largely from tuffs in the western part of the valley (table 9). Water from Jackass Flats, which is unusually high in sulfate attributed by Schoff and Moore (1964) to hydrothermal alteration of volcanic rocks, con- tributes sulfate and probably lithium to the western part of Amargosa Desert. Water from a shallow well southeast of Allens Well and 2 miles (1.2 km) south of station A4 con- tains 0.30 ppm lithium and 2.8 ppm boron—the highest percentages recorded by Walker and Eakin for their 18 Amargosa water samples. According to Walker and Ea- kin, the boron content should be toxic to any but the most LITHIUM IN BASIN AND RANGE PROVINCE 15 TABLE 8,—Percentage of lithium and associated ions in sediments and the ash of plants from Death Valley, Calif. [Stations shown in fig. l. Analysts are named in introduction. N nm dou-ru-d; X, xerophylt; P, phrealophyle; Ha, halophyte; Hy, hydrophyte; WSS, water-soluble salts. The soil in which lht‘ sampled plant grew is listed directly below each plant] Carbo— Sta- Sample Lab. Ash WSS Lithium Magnesium Calcium Sodium Potassium Boron nate Sulfate Chlorine tion No. (as (water- No. (D—) C02) soluble) (percent) Plant Soil Plant Soil Plant 5011 Plan Soil Plant Soil Plant Soil Soil Plant Soil Soil At edge of salt pan, west of Devil's Golf Course MA AZZarzro lfea occi- dentalis (Ha)——- 403640 27.5 ——-— 0.035 ————— 2.0 ———— 2.7 —-—— 33.2 —-—— 6.4 -—-— 0.05 -——- —--— 0.50 -—-— -——— Sediment mixed with halite ----- 403338 —--— 2.8 ----- 0.015 ----- 4.0 --—- 7.8 ----- 8.3 ————— 1.7 ----- 0.03 6.2 ----- 1.72 28.5 Salt pinnacle (no plants) ————— 405463 -——— ——-— .02 ----- .07 —-—— .50 ————— 35 ————— .8 ————— .07 --—- ————— 2.0 --—- Brown sediment be— tween pinnacles— 405464 -——- ——-— ————— .03 ————— 1.0 —--— 3.0 ————— 20 ————— 1.7 ————— .15 —-—— ————— 4.5 ——-— Streambed to salt zone ———————————— 405465 ——-— ——-— ————— .03 2.0 —--— 5.1 ————— 15 ----- 1.9 ————— .10 —--— ————— 9.4 ——-— DlB Atriplex hymene— Zytra (X) ------- 403642 32.3 ——-— .03 ————— 1.5 ——-— 5.4 ——-— 31.3 ——-- 5.4 --—- .02 -——- ——-— .71 -—-- -——- Suaeda ramassis- sima (P)—- —-- 403641 22.5 ——-- .05 ————— 1.5 -——- 3. 2 -——- 31.5 --—- 6.4 —-—— .02 -——- ——-— 1 . 2 -——- —-—— Carbonate zone plus salt ------- 403339 ——-— .85 ————— .007 ————— 3.2 -——- 8.8 ————— 3.1 ————— 1.9 ————— .01 12.2 ——-— .44 —-—— DlC Atriplex hymene— lytra (x)1—- - 412065 30. 0 -—-- .05 ----- 7 . 0 —-—— 9.6 -——- 23 --—- 7 .0 -—-— .03 ———— -—-- ——-— - --—- Alluvium —————— 412058 -——- --—- ----- .005 ————— 1.0 ——-— 1.8 ————— 1.3 ————— 3.5 ----- .003 —-—- ---- —-—- —--— D5 Atriplex Zinearis (P) ------------- 403649 16.8 —-—- 0.015 ----- 3.0 --—- 8.8 -—-— 18.0 -——- 14.6 -—-— 0.05 -——— ——-— 0.23 -——- -——- Distichlis stricta (P) ----- - 403650 24.4 .003 ----- 1.5 —--— 3.1 —-—- 5.8 —-—— 2.6 -——- .03 —-—- --—- .13 —--— ——-— 413095 —-—— —-—— ----- 0.0035 ——-- 2.0 -—-- 3.8 ————— 1.1 ————— 2.4 ----- 0.015 —--— ————— 0.03 —-—- Near Tule Spring D2 Pluchea sericea (P) 403643 10.7 ——-— 0.0035 ————— 3.0 -——- 15.5 -—-—- 10.2 ———— 6.2 ———— 0.1 ————— ——-— 0.76 -—-- -——— Prosopis glandu— Zosa (P) ———————— 403644 3. 9 —-—— .015 ————— 5.0 -——- 20.8 -——- 1.0 —-—— 19. 2 —-—— .05 ----- ——-— .08 ——-- --—- Atriplex polycarpa (X) ------------- 403645 13.5 -——- .015 ————— 7.0 ——-— 9.4 ——-— 23.2 ——-— 8.2 --—- .03 ————— —-—- .19 —--— ——-- Sandy facies of carbonate zone-— 403340 --—— 1.05 ----- N --—- 1.8 -——- 2.8 ————— 4.2 ——-— 2.5 ————— 0»015 2.0 ————— 0.91 ——-— DZA Atrip Zea: p0 Zycarpa (x)2 ———————————— 412066 16.0 --—— .015 ————— 7.0 ——-- 7.3 ——-— 23 ———— 6.5 -——- .05 ————— —-—— ————— ——-- -——- Sandy carbonate zone—- 412059 -—-- -——- ————— 0.003 —-—— .7 ——-— 7.0 ————— .75 —-—— 2.0 ————— .005 —-—- ——-— -——- of Badwater, and 8 mi (13 km) S. of Badwater D7 Allem‘alfea occi- dentalis (Ha)--— 408644 25.5 ——-— 0.03 ————— 1.5 —-—— 2.2 —-—— 30.2 ———— 8.2 -——- 0.05 ----- —-—— 4.2 ——-- --—- Sulfate zone plus salt ------------ 118528 ---- 38.8 ----- 0.03 --—- 3.2 -—-- 5.1 ----- 9.7 -——- 2.7 ————— 0.3 3.2 ————— 5.9 16.7 DB Phomidi'wn ambiguwn ———————— 408645 87.0 ———— .011 ————— 5.0 --—- 2.3 -——- 2.0 —-—— 1.4 .02 -——- —-—— .22 ——-— --—- Crust of gypsum——— 118529 ———- 5.7 ————— .007 ——-— 1.4 -—-- 18 ————— 1.6 -—-- .41 ————— .02 1.0 ----- 2.2 —--— D9 AZZenrtoea occi— dentalis (Ha)——- 408646 30 ———- .03 ————— 2.0 -——- 2.0 -——— 33.2 -——- 5.6 ——-— .03 ——-— -——- 1.9 —--— ——-— Spongy sulfate one plus salt—— 118530 18.0 ----- .015 -——- 3.4 -——— 3.4 --—- 4.4 -—-— 3.4 ————— .05 1.2 -—-— 2.6 7.9 Furnace Creek Wash. Travertine Spring D6 Anemopsis cali— f‘ornica (Hy)——-— 403651 16.7 ——-— 0.015 ————— 5.0 -——— 5.6 —-—— 12.1 -——— 27.0 -——- 0.02 ————— —-—— 0.16 -——- —-—— Travertine deposit ————————— 403342 ——-— 0.30 ————— 0.03 ——-- 3.3 ———- 17.7 ————— 0.74 -——- 1.2 ————— 0.003 20.4 ————— 0.02 —-—— D3 Franseria dumosa (X) ————————————— 403647 9.7 —-—- 0.004 ——-- 2.0 ———— 20.2 —-—— 1.4 ——-- 20 ——-- 0.05 ----- ———— 0.22 -——- ——-— Hymenoc Zea salsola (P) ————————————— 403646 9.1 —-—— .015 --—- 3.0 —-—— 12.7 ———— 1.1 ——-- 28.4 -——- .07 ————— —-—— .14 -——- —--— Alluvium at 2,000 ft elev. (610 m) 403341 -——- ———— ————— N -——- 1.1 -——— 2.6 ————— 8.1 —-—— 2.8 ————— 0.005 2 . 0 ————— 0.01 ——-- D4 Eabbia juncea aspera (X) —————— 403648 9.25 -——- .015 ——-— 6.0 —-—— 14 —-—— .5 ———- 16 ——-— .3 ————— ——-— .13 —-—— -——- Alluvium at 1,000 t elev. (305 111) 413096 -—-— -—-— ————— 0.0035 -——- 1.5 -——- 3.8 ————— 1.1 —-—— 2.4 ————— .005 --—- ————— .03 —-—- Ji’r/t/g/ DlO Soft sediment (NaCl) —————————— 415466 —-—— 2.0 ————— 0.016 -——- 3.0 -——- 4.9 ————— 2.0 --—— 2.3 ————— 0.03 -——- ———- 0.30 -—-— 1Fluorine content in plant 0.0033 percent. 2Fluorine content in plant 0.0112 percent. 16 LITHIUM IN UNCONSOLIDATED SEDIMENTS AND PLANTS, CALIFORNIA AND NEVADA TABLE 9.—Chemical constituents (parts per million [Analyses from Walker and Eakin (1963, p. 33) and B. P. Robinson and W. A. B ) of some springs and wells in Amargosa Desert eetam (written Commt'rn., 1965). Surfact- samples unless depth figure shown] Name of spring, sample-collection De th date, station No. (feet) p(metres) C3 Mg Dissolved K Li HCO; 80‘ C1 F B 50]de Hardness Recharge largely from limestones Fairbanks Spring 10—27—64 (A7) ................................ 51 18 8 0.11 304 80 22 2.2 0.51 422 201 Point of Rocks (small spring) 10—‘26—64 (A9) ................................ 52 19 66 7.9 .1 306 70 21 1.2 .66 419 208 Point of Rocks (King Spring) 10—‘26—64 (A8) ................................ 42 19 7.9 .09 308 70 22 1.6 .45 416 208 Recharge largely from altered tuffs Southeast of Allens Well 8-18-52 (2 miles south of A4) ...... 27 8.2 4.8 3.3 370 16 0.30 542 256 102 3.2 2.8 1,119 26 Death Valley Junction 8—‘29—52 (A10) ................................ I46 45 1.9 1.9 325 12 556 149 49 7.9 1.3 874 12 Ash Tree Spring 5—8—52 (A5) ................................... 16 4.8 7.9 *.45 160 37 7.2 2.8 .29 293 60 ‘USGS analysis of sample collected by H. L. Cannon. boron-tolerant crops, and the fluorine content of 7.9 ppm at Death Valley Junction is too high to meet US. Public Health Service drinking-water standards. Lithium con- tent was not reported for Death Valley Junction but is probably also high. Two phreatophytes collected near the well contained 200 ppm lithium and 116 and 140 ppm fluorine in the ash. Although correlations are apparent for lithium and sodium, lithium and chloride, and lithium and boron in the waters, no consistent relationship between lithum and any other element can be demonstrated in the various types of sediments sampled (table 10). Plant and soil studies were made in relation to a playa northeast of Ash Mea- dows, springs issuing from limestones in the same part of the valley, a playa northeast of Allens Well, the well at Death Valley Junction, and the clayand carbonate depos- its north of Death Valley Junction. The unnamed playa is a relatively wet playa with ground water at a depth of 5-6 feet (1.5—1.8 m) discharging from Paleozoic limestones. Several A triplex species were sampled in sandy soil 600 feet (180 m) northeast of the playa (station Al). Also growing at the same station were the phreatophytes Prosopis (mes- quite), A triplex canescens (saltbush), and Suaeda mmos— sissima (seepweed) and the sulfur indicator Lepz'dium fre- monti. N 0 plants grew in the center of the playa, at station A2. Seepweed and shadscale were collected at station A3 where ground water stands at 6 feet (1.8 m) in depth in the clay pits. These clay sediments contained appreciable car- bonate and sulfate and 200 ppm lithium. At four springs issuing from Paleozoic limestones (stations A6, A7, A8, A9) only a small amount of lithium was found in tufa deposits but as much as 150 ppm was found in organic mud in McGilivrays Spring. At the unvegetated wet playa northeast of Allens Well, the capillary fringe extends to the surface and is fed by water from the Paleozoic limestones; a maximum of 150 ppm lithium was found in the outcrop- ping sandy carbonate facies. Of greater interest were the Pleistocene lacustrine clay and travertine deposits in the northwest part of the valley where the waters are of vol- canic origin. There, clays contained as much as 1,500 ppm lithium; travertine or lacustrine limestones, as much as 700 ppm; and plants, a maximum of 300 ppm lithium. The area appears to be favorable for the commercial pro- duction of lithium salts. The carbonates with the highest concentrations of lithi- um are clearly of biochemical origin (figs. 3, 4, 5) and probably reflect selective transpiration of particular ele- ments by halophytes. Lithium has been shown to sub- stitute for magnesium in the crystal lattice of the clays. Clays deposited or formed in saline lacustrine deposits commonly have relatively high lithium contents. Studies by P. D. Blackmon and H. C. Starkey (written commun., 1974) at the Pleistocene Lake Tecopa in the southern part of the Amargosa Desert showed the zeolitic rocks to con- tain 50—100 ppm lithium and montmorillonitic clay- stones to contain 100—800 ppm. A nearly pure separate of saponite from a claystone contained 0—34 percent lithium oxide. OTHER CLOSED BASINS Lithium and potassium have been commercially pro- duced in Owens Lake and are being produced in Clayton Valley; halite, calcium chloride, and gypsum are being mined in Bristol Lake, and halite is being recovered at Fourmile Flat. In all these evaporite basins the brines are pumped from depth and evaporated to crystallize the salts. Lithium chloride and potassium are or have been pro- duced from the remaining mother liquor in Owens Lake and Clayton Valley. Hectorite, a lithium clay, is being LITHIUM IN BASIN AND RANGE PROVINCE TABLE 10.—Pe1centage of lithium and associated ions in sediments and ash of plants from Amargosa Desert, Nevada and California [Stations shown in fig. 1. Analysts are named in introduction. N, not determined; X, xerop‘hylc; P, phreatophyte; Ha, halophyte; WSS, water-soluble salts. The soil in which the sampled plant grew is listed directly below each plant] Carbo- Sta- Sample Lab. Ash WSS Lithium Magnesium Calcium Sodium Potassium Boron nate Sulfate tion No. (as (water— No. (13-) 002) soluble) ____: f __.__ ______ ,_____ (percent) Plant Soil Plant Soil Plant Soil Plant Soil Plant Soil Plant Soil Soil Plant Soil Clay pits N of Death Valley Junction, from Ash Tree Spring to ESE, NazHC03 ground water A5 Lake clay with root casts ------------- 403349 —-—— 0.25 ------ 0.15 —————— 11.0 ————— 0.92 ————— 1.1 ————— 1.7 ————— 0.01 0.04 ——-— 0.04 Limestone of bio- genlc origin —————— 412759 -—-- .48 —————— .07 —————— 3.0 ----- 2 ----- .1 ----- N -—-- .09 ASA Atm'plex spinifem 408130 10 7 0.015 ------ 7 0 ----- 25 ----- 4 3 3.7 0.12 ------ 03 __-__- 1.9 —---- 32 6 --——- -——-- --—— Lepidium fremonti - 0075 ------ 3.0 ————— l4 ————— 3 7 19.4 1.64 Casts on roots--- 408172 —-—- 1.02 ------ .03 ------ 2.7 ----- 12.4 ---------- ——-— Brown sulfate zone with crust -------- 408173 -—-- 11.0 —————— .02 ------ 2.1 ----- 2.9 6.4 ----- ---- 2.58 Cleamella obtusi folia (Ha)- 413089 ----- .015 ---------- 18 1.2 ——-— Calcareous 5011- 413122 02 ------ .015 3.0 .11 ----- -—-— .02 Sulfate 5011—- 413123 2 3 —————— .028 5.0 2.2 ----- ——-— 2.3 A513 Distichlis stricta (P) --------------- 408640 15.8 ----- .006 ————— .7 ----- 1.3 ————— 7.8 ————— 2.2 ————— .03 ---------- .14 ———— Sporobolus airoides (Ha) 408641 15.8 ----- 008 ————— 1 0 2 3 ————— 6.8 ----- 1 7 ————— 02 ————— .36 ~—-— Tamarisc pentandra (P) --------------- 408642 14.5 ----- .03 ----- >10 ————— 14.8 ————— 10.3 ————— 5.6 ----- .05 —————————— 2.3 --—- Calcareous lake c1ay1 ————————————— 418525 —-—— .18 ----- .039 ------ .96 ————— 9.9 ————— 1.7 ----- 2.2 ————— .015 21.2 —--— .03 Limestone of bio- genic origin ------ 118526 --—- .28 ----- .05 33 ----- 38.3 ————— 06 ----- 15 35.4 —-—— .08 Death Valley Junction, near spring A10 Tamar-ix pentandra (9)2 ------- 412756 6.9 ----- 0.02 ----- 7.0 ————— 8.4 ----- 12 ————— 19 ----- 0.07 ----- 1.5 -—-- Hymenoclea salsola (r)a -------------- 412757 21.4 .02 1.5 ————— 1.4 ————— 8 015 ————— 1.4 Atm’pleac paw-142' 00- 413088 25.0 .02 1.5 ————— 4.6 ----- 3.4 20 ----- 54 Sandy soil ---------- 413121 — ----------- 3 o ————— 6.0 —————————— 0.20 --—- Playa NE of Allens Well (no vegetation) A4 Brown carbonate layer with NaCl — 403347 2.0 0 015 ------ 3.4 8.7 2.2 2.7 ————— 0.015 10.2 - - 0.41 Salt crust ——————— 403348 80.6 N ------ .4 78 29 3.2 ----- .05 9.4 — 19.5 Playa NE of Ash Meadows; ground water at 5—6 ft (1.5—1.8 m), CaMgHCOa) A1 Atriplex parryi 00— 403652 21.6 ----- 0.005 ----- 0.005 ----- 5.0 ————— . ————— . ————— . ---------- A. confertifolia (X) --------------- 403653 16.3 ----- .01 ————— .01 ————— 7.0 ————— 19.2 ----- 12.4 ————— .007 —————————— .24 —--— Sandy soil at sta. NE of playa—- 403343 ————— 0.005 —————— 1 1 4.5 ---------- 2.3 ----- 0.01 Atriplex parryi 00- 413087 015 ----- 3 0 —————————— 15 8 4 ----- 02 ——-- Sandy soil-—-- 413120 ————— 1.5 4.3 1.3 12.4 (.01 A2 Playa crust--- - 403344 ----- 3.2 7.0 2.3 2.6 1.13 Playa at 6 inches (15 cm) depth ----- 403345 -——- 1.5 ----- .015 ------ 3.2 ————— 4.8 ----- 7.4 ————— 2.5 ————— .03 10.4 ---- .95 A3 Suaeda rwnossissima (P) --------------- 403654 22.0 ----- .01 ----- 2.0 ————— 4.1 ----- 27.9 ————— 8.3 ----- .015 —————————— .72 --~— Atm‘plex confert falia (X)-- 411477 38.6 ————— .0035 ----- 1.5 6 5 ----- 21 ————— 17.0 ----- 01 ----- —-—- --—- Carbonate clay with 50.. —————————— 403346 .02 .02 —————— 4.8 6.9 2 2 ————— .03 15 2 3.65 Efflorescence—— 411455 .01 .01 ------ 3.0 16.0 1 7 ----- .05 ----- ---— Fairbanks Spring, mixed CaMgHCO; and Nail-1003 A6 Tamariac pentaadra (P) ------------ - 403655 9.5 ----- 0.0065 ————— 5.0 ————— 13.3 1.6 ————— 16.8 ------ 0.03 —————————— 1.4 ~—-— Hard white Calcare— ous clay —————————— 403350 --—- 0.45 ————— 0.005 0.96 ----- 9.9 ----- 1.7 ----- 2.2 ----- 0.003 9.6 ---- 0.10 Brown silty carbo- nate lnterbedded—— 403351 —--- .15 ————— N ------ 3.3 ----- 38.3 ----- .06 ----- .15 ————— (.002 42.5 ---— .01 M McGilivrays Spring, CaMgI-l003, recharge in limestone /— ___/_’__/ A7 Bacharis glutiaosa (P) ——————————————— 411478 10.0 >10 ————— 23 ————— 0.91 ————— 6 6 ————— 0.15 ---------- Black organic mud 411456 -——- ————— . ------ 2.0 ————— 10 ————— 1.1 ————— 1.5 ----- 0.005 6 ___'_—_____4——— A8 Tamarix pentranda (P) ---------------- 411480 11.4 ————— 0.01 ————— . . Black organic mud—-- 411457 ——-- ---------- 0.0045 ------ 1.5 ————— l9 ————— 2.2 ----- 1.5 ----- 0.002 Point of Rocks Spring, CaMgHCOg, recharge in limestone r/"— f A9 Pluohea serviced (P)- 411479 8.3 Calcareous tufa ————— 411458 —--— ____________4———#_(_______. 1Chlorine content of soil 0.09 percent. 2Fluorine content of plant 0.0140 percent. 3Fluorine content of plant 0.0116 percent. 17 18 LITHIUM IN UNCONSOLIDATED SEDIMENTS AND PLANTS, CALIFORNIA AND NEVADA TABLE ll.—Percentage of lithium and associated ions in sediments and the ash of plants from three closed basins‘in which economic quantities of lithium occur [Stations shown in fig. 1. Analysts are named in introduction. P, phyteatuphytc; Ha, halophyte; X, xerophyte; Hy, hydrophyte; WSS, water-soluble salts. The soil in which the sampled plant grew is listed directly below each plant] Carbonate Sulfate Sta- Lab. Ash WSS Lithium Magnesium Calcium Sodium Potassium Boron (as CO ) (water— tion Sample No. 2 soluble] No. (D—) Plant Soil Plant Soil Plant Soil Plant Soil Plant Soil Plant Soil Plant Soil Plant Soil Clayton Valley (Lithium and potassium produced commercially from chloride brines) L3 Sarcobatus vermiculatus (P) ------------------- 408134 12.7 ~—-- 0.07 —-—- 0.5 -——— 6.8 --—~ 23.2 ———- 14.6 —~-— 0.015 -——— --—— ——-- 0.38 ---— Sulfate soil under plant 408177 ———- 13.3 ~--- 0.03 --—— 1.4 —--- 6.8 --- 4.7 —-—- 2.57 -—-- 0.015 ---— 1.6 -——— 2.33 Spongy chloride 5011 408178 —--- 13.8 —--— .03 -—-— 1.4 -——- 3.0 ——-— 6.1 -—-- 2.24 ---- .02 ——-— 1.6 ---- 1.22 Efflorescent crust—- 408179 —-~- 88.6 —-—— .03 —--— .24 ———- 1.8 ———- 35.0 -—-— 1.3 —--- (.02 ---— .18 -——— 3.02 L4A Soft brown lake clay (no plants) ------------ 415459 ---- 2.8 ———- .03 ———— 1.0 ———— 3.3 -——- 2.9 ———- 3.2 ———- .015 ---- ——-— - -— .10 White efflorescent crust 415460 ———- 3.2 ---— .05 —-—— 1.5 -—-- 3.3 ———- 2.8 —--- 2.8 -——- .015 —--- —--— —-—- .24 L4B Allenrolfea occidentalis (Ha) ------------------ 408135 25.8 .15 __-_ 1.0 ——-~ 3.6 -——- 31.5 4.5 ——-— .05 —--- --—- ——-— 2.08 -——- Juncus cooperi (Hy) ----- 408136 6.2 .30 --—— 1.0 --—- 4.6 -—-— 21.6 4.8 --—— .2 ——~- —--- ——-- .4 —--- Algae in hot spring (Hy) 408148 83.8 ---- .03 --—- .3 -—-— 33.4 —--- 3.6 -—-- 3.6 ---— .007 —--- ---- ~--- -------- Spring deposit between Juncus ---------------- 408180 ———— 7.45 —--- .03 -__- .7 ---— 30. --—- 3.0 ——-— 3.0 —--- .01 ---- 34.1 ---— .50 Spring deposit under Jgncus ————— 408181 —--— 9.51 ——-- .05 —--— 1.0 ~--— 25. ---— 3.0 ———— 3.0 --—- .02 ---- 25.6 ---- 2.51 Allenrolfea occidentalis (Ha) ------------------ 413056 44. —--- .3 ---- 2.0 -——- 1.4 ——-— 31. ---- 3.4 -——- .07 —--- ---- 3.4 -—-— Hot spring mud ---------- 413101 ---- ---- —-—- .01 -——- 1.0 -—-- 23. ---- 1.0 ---- .85 -—-- .007 ---- ---- .18 L18 Allenrolfea occidentalis (Ha) ————————————————— 413057 33. ~-—- .3 ———- 3.0 —--— 1.0 -—-— 29. ---— 3.0 ~--- .05 ---- ---- ---— 1.8 ---- Brown soft soil with NaCl ------------------ 413102 ——-- -~—- ——-- .03 ---- 2.0 ~—-— 3.9 ~--- 2.1 —--— 2.4 ---— .015 —-—- ---— Owens Lake (Lithium and potassium formerly produced from chloride brines) L12 Scirgus olneyi (Hy)— —- 411481 0.035 ---— 1.5 -——— 4.2 ——-- 15. -—-— 13. ———- 0.7 —_—— —__— —_—_ __.- __-- Mud in spring at playa edge —————————————————— 411459 ——-- --—- —-—— 0.01 -—-— 1.5 --—— 7.8 -—-— 1.45 -——— 2.5 ---- 0.007 —--- -—-- —--— 0.65 Atriplex torreyi (X)~-—— 411534 20.4 —-—- .05 ~--- .7 ———- 3.2 ———- 26. --—~ 11. —--— .03 —-—— ———— ———— 2.2 -4“ Sand at edge of playa--- 126550 v-~— ———— -—-— .007 --—- 1.5 —--- 7,4 —--- 3. ———- 2.0 -——— .03 -—-— ———— -_-_ -___ Suaeda torrezana (P)—~—— 413054 24.0 —--— .07 ———- 3.0 ——-— 1,4 -——- 25. —_—- 3.2 —__— .03 -___ ___- -_-.. 2,2 -__- 119 Distichlis spicata (P)-- 413091 15.0 -~—- .10 —--— 1.5 -——- 2.0 -——- 13. -_—_ 3.0 __-- .15 --_- ____ _-_- .19 -_-_ Soil under evaporites in evaporating pan—-—- 413098 ---- ———- —-—— .012 ——-- 3.0 ——-- 9.2 —--— 2. ——-- 1.9 ——-— .015 ---- -—-— —-—— .12 Evaporites being shipped for BI ———————— 413099 --~— --—— ---- .07 -——- 3.0 —--— 6.1 ———— 17. ——-— .85 _—-- .05 _--_ -_-- _--- .76 L193 Soft spring mud with salts ————————————————— 415469 -—-- 1.8 —--- .03 -—-- 3.0 -——- 11.0 —--- 2.5 ---- 1.9 ——-- .03 -—-— --—- ——-- .15 L19A Crystals in Distichlis plants ———————————————— 415468 —-—— 44. ~—-— .01 -——— 1.5 --—— 3.5 ——-- 10. -—-- 1.7 -——- .1 —--— -——- -—_—15. Spring mud with algal scum --------- 415467 --—— .7 --—— .004 --—— .7 -—-— 2.8 --—— 2.8 ——-- 3.1 ---- .1 -——— —--~ ---- .17 Hector (Lithium clays being mined (hectorite)) L24 white carbonate lithium clay being shipped---- 415473 ---— 1.0 ---- 0.20 -—-- 7.0 -—-- 22. -—-— 1.1 ~--— 0.24 ---- 0.003 ---- --—- ---- 0.01 Darker clay surrounding white nodules--- ----— 415474 ---- 1.6 ---— .07 ———— 3.0 -—-~ 20. —--— 1.5 —-—- .85 —-~- .002 —-—- ——-- ---— .03 Shale underlying volcanic flow --------- 415476 --—- .3 —--- .005 --—- .3 ——-- .80 --—- 4.2 ---- 2.4 ---- .01 ——-- —--- --—- .01 1Chlorine content of soil 1.4 percent. mined in the Mohave Desert. Analyses of samples from In Clayton Valley playa sediments consisting of 5—20 three closed basins from which lithium has been pro- feet(].5—6 m) of clay impregnated with haliteand gypsum duced are given in table 11. overlie beds of crystallized halite and saturated brine LITHIUM IN BASIN AND RANGE PROVINCE l9 (Meinzer, 1917). Brines that contain 12 percent sodium and 3 percent potassium are currently being pumped by the Foote Mineral Co. from a depth of 500 ft (150 m). Lithium contents of the brine were not ascertained from company geologists, but a magnesium:lithium ratio of l was men- tioned. The percentage of lithium in the brine can then be assumed from the available analyses for magnesium in the brines to be at least an order of magnitude higher than that of sediments sampled at the surface. The highest lithium value in a surface soil analyzed by the Foote Mineral Co. was 1,200 ppm. Our surface sediment samples averaged 300 ppm lithium. Lithium-rich precipitate around the roots of Juncus in the hot spring on the east side of Clay- ton Valley contained 500 ppm lithium; a Juncus root crown (fig. 3) contained 3,000 ppm. Water—soluble sulfate has been concentrated from 0.50 percent in the spring mud to 2.51 percent in mud around the roots of a Juncus plant (sta. L4H). Hunt (1966, p. 34) reported 13 percent total SO4 in the ash of juncus; the plant described above contained 0.4 percent water-soluble 80,. The water of Owens Lake is a dense brine containing halite, soda, borax, and other salts. The dried residue analyzed by Stone and Eaton in 1905 contained 38.09 per- cent potassium, 1.62 percent lithium, and only 0.02 per- cent magnesium (Gale, 1915). The mother liquor still con- stitutes a large proportion of the mass, permeating the en- tire thickness of crystallizing salt. Mud in the bottom of the old evaporating pans at Owens Lake contained only 120 ppm lithium; evaporite, which is dug and trucked only for boron payments, contained 500 ppm boron and 700 ppm lithium. Plants collected at sev- eral stations contained 350 to 1,000 ppm lithium. A high-lithium clay, hectorite, is mined at Hector, east of Barstow from lake beds close to volcanic flows from Mount Pisgah (Foshag and Woodford, 1936). The pure mineral contains 1.1-1.2 percent L120. The material be- ing shipped is a white talclike clay that occurs in nodules and contained, in our sample, 0.2 percent lithium. The surrounding impure clay matrix contained only 0.07 per- cent. The hectorite resembled structurally the clay being mined in Amargosa Desert (A5) and shown by our anal- yses to contain a considerable amount of lithium. Analyses of samples from seven other closed basins are given in table 12. On the Bristol Lake playa, rock salt (sodium chloride) is mined by the Leslie Salt Co. from a 6- to 7-foot-thick bed, which is covered by a similar thick- ness of mud over an area of about 4 square miles. In the recovery method, calcium chloride is precipitated from brines believed to be the mother liquor of the evaporative basin. The abundance of calcium chloride in this basin is unique for basins of the Basin and Range province. The calcium chloride is believed by Gale (1951) to have orig- inated in the basaltic flows from Bagdad crater, which have in Holocene times separated Bristol Lake from Bag- dad Lake to the northwest. Lithium is not mentioned in Gale’s report. Polygonal salt crust in evaporating pans sampled during our study (sta. L25) contained only 20 ppm lithium. A sample of the playa carbonate mud over- lying the salt contained 300 ppm lithium, and Atriplex linearis rooted in the carbonate near the toe of the vol- canic flow contained 150 ppm. Gypsum and oolitic sur- face material 350 feet (100 m) onto the barren playa from the south edge also contained scant lithium. The surface alluvium at the south edge of the lake was low in lithium, but desertholly and cattle spinach collected there con- tained 300 ppm lithium. On Sarcobatus Flat, the sediments of the soft brown carbonate zone that are impregnated with halite con- tained 300 ppm lithium, but greasewood ash contained only 150 ppm. Lesser amounts in both plants and soils were found in samples from Fourmile Flat (where halite is mined), Big Smoky Valley, Railroad Valley, and Oasis Valley. Samples of gypsiferous sediments collected near an old borax works in Columbus Salt Marsh contained only 30 ppm lithium but contained 10,000 ppm boron. A triplex spinifera growing on them contained 150 ppm lithium and 1,500 ppm boron. The barren playa muds 500 feet (150 m) beyond the borax works contained 280 ppm lithium and 1,500 ppm boron. Seemingly, lithium is not associat- ed directly with halite and borates but has a rather spotty distribution pattern in the basins, owing to sources of vol- canic activity, to thermal hot springs, to adsorption on clay, or to subsequent flooding and subsurface movement of the solutions. PLANT PROSPECTING The analysis of plants for assessing lithium contents in brines in the Great Basin might be a useful prospecting tool in those areas where brines occur at depth and surface soils are relatively low in lithium. The lithium contents of plants growing in two plant societies in closed basins may indicate lithium concentrations at depth. First, hot springs at the contact of alluvium with basin sediments are apparently dissolving lithium at depth and the' hydro- phytes, such as Scirpus olneyi and [uncus cooperi, are ab- sorbing and concentrating large quantities of lithium in plant tissues. Second, phreatophytes, such asAllenrolfea, Distichlis, and Suaeda, that are tolerant of and absorb large quantities of sodium chloride and grow in the chloride and sulfate zones also are tolerant of and absorb large quantites of lithium chloride. Xerophytic species of Atriplex, even though they may be lithium accumulators, have shallow root systems adapted to collecting surface water rather than ground water and, therefore, do not re- flect lithium deposits at depth. No symptoms of lithium toxicity were identified in plants, nor were useful indicators identified. Lycium is definitely an accumulator of lithium and may even be an 2O LITHIUM IN UNCONSOLIDATED SEDIM NTS AND PLANTS, CALIFORNIA AND NEVADA TABLE 12.—Percentage of lithium and associated ions in sediments and the ash of plants from other closed basins in Nevada and California [Stations shown in fig. 1. Analysts are named in introduction. N, not detected; P, phrealophyic; X, xerophyle; WSS, water-soluble salts. The soil in which the sampled plant grew is directly below each plant] Carbo— Sta- Sample Lab. Ash WSS Lithium Magnesium Calcium Sodium Potassium Boron nate Sulfate tion No. (as (water— No. (D—) C02) soluble) (percent) Plant Soil Plant Soil Plant Soil Plant Soil Plant Soil Plant Soil Soil Plant Soil Bristol Lake (salt (NaCl) produced commercially) L16 Atriplex Zinearis 411575 21.5 ----- 0.015 ------ 0.5 ————- 8.6 —--- 21.0 12 ———— 0.015 ————— -——— --—— 411565 ———— ————— — 0.01 —--- 2.0 -——— 5.6 ---— ——-- 2.3 ————— 0.005 1.5 ---- L16a Carbonate over— lying Chlorides 413104 —-—— —————————— .03 --—— 1.5 ———- 16.0 2.1 ——-- 1.6 ----- .007 ---- ————— ———- Limestone under— lying flow ----- 415470 -——— 4.4 ————— .004 ———— .Z —--- 15 ———— 1.5 —--— 1.1 ————— <.002 2.1 L17 Atriplex hymene- Zytra ---------- 413060 27 ————— .03 5.0 15 -——— 11 .05 ---- 1.5 A. polycarpa (X)— 411576 19.4 ————— .03 1.0 18 ———— 14 .03 ———- ————— Alluvial fan, south Side ----- 411566 --—— —————————— <.0025 l 0 ———— 3 4 ———— 2 0 ———- 3 0 ————— < 02 1 2 ----- --—— L25 Oolitic gypsum (no plantS)--—- 415471 ———— 54 ----- .005 --—— .07 ———- 25 -——— .95 ———- .24 N 2.1 ————— 14 Polygonal evapo— rite crust (no p1ants)-——- 415472 -——— 9O ————— .002 -——— .03 -——— 1.4 ---— 34 --—— .33 N 2.0 ————— 15 Columbus Salt Marsh (borax previously produced commercially) L5 Atriplex spini— fora (x) ——————— 408137 7.2 ----- 0.015 —————— 5.0 -—-— 12.8 12.2 ———— 15.6 -——- 0.15 ————— ---— 0.24 ---- Soft sand with sulfate and borates ———————— 408182 -——— 12.5 ————— 0.003 ——-- 1.4 '--- 5.2 6.0 ---— 2.6 —--- 1.0 3.8 ----- 4.3 Soft brown gyp— siferous 5011 (no p1ants)-——— 415455 —--- 7.0 ----- .015 ---— .7 -——— 3.5 -——— 3.8 ———— 2.9 -——— .2 --—- ----- 2.2 Playa sediment (no plants)-——— 415456 ——-- 8.4 ————— .028 --—— .7 ———— 3.2 ———— 5.0 ———— 3.0 -——— .15 ---— 2.0 Railroad Valley (NaHCOa water) 1.7 Sarcobatus ver— miculatus (P)—— 408139 13.7 ————— 0.0025 —————— 1.0 —--— 5.6 ---— 28.2 ---- 11.8 ---- 0.01 ----- ---- 0.30 ---- Suaeda torre— yam (P) ------- 408140 9.2 .005 ------ 1.0 —--- 4.8 —--- 25.2 —--- 13.6 —--- .02 ----- ---- .55 ---- Sandy soil, brown —————————— 408184 ——-- 0.37 —————— N ——-- 0.48 v—-- 2.1 ---— 1.8 --—- 3.4 —-—- 0.005 1.2 —--- 0.02 Playa mud with sulfates ------- 408185 -——-' 7.5 —————— 0.015 -——— 1.7 ———— 6.1 -——— 4.4 --—— 2.4 ---- .007 3.5 2.1 Big Smoky Valley (NaHC03 water) L14 Sarcobatus ver— miculzztus (P)--- 408643 16.3 ————— 0.007 —————— 0.7 -——— 8.9 --—— 20.2 ---— 12.2 --—— 0.03 ----- -——— 0.28 ---— Playa mud (water at 43 ft (13 m)) 118527 -——— 4.4 ------ 0.008 ---— 1.6 -——- 3.8 -——— 1.2 ———— 2.6 -——— 0.007 1.6 --—— 0.15 L23 Sarcobatus ver— miculatus (P)——— 413059 16.0 ————— (.002 ------ 1.5 3.2 ———- 25 ———— 9.0 ———- .05 ----- —--— .75 ----- Atriplex parryi (X) ------------- 413058 11.0 ----- .003 ------ 2.0 —--- 5.2 —--— 20 —--- 8.2 —--- .05 ----- ---- .31 ----- Sandy soil with NaCl ———————————— 413103 ----------- .01 2.0 2.0 ---— 2.9 —--— 3.6 --—— 2.6 —--- .07 ---- ---- .54 Hard playa mud (no plants) ————— 415457 ——-- 3.3 ------ .006 ---- 1.0 —--- 3.6 ——-- 2.5 ---- 2.6 ——-- .007 ---- ---- .01 Sandy clay (no plants) ————— 415458 ---- .4 ------ .009 ---- 1.0 ---- 4.0 —--— 3.5 ---- 2.4 —--- .07 ---- .21 Fourmile Flat (halite mined) L6 Sarcobatus ver— miculatus (P)——- 408138 9.7 ----- 0.0025 ------ 0.7 ---- 7.4 ---- 19.8 ---— 15.8 ---- 0.015 ----- ---— 0.28 ----- Chloride 500 ft (150 m) onto playa ----------- 408183 ---— 5.9 ------ 0.007 ---— 2.2 --—— 2.9 ----— 4.9 2.2 ----- 0.05 2.6 ---— 0.40 1-4 in. chlorides 4,400 ft (1,320 m) onto p1aya--- 415453 ---— 1.8 ------ .0035 ---—— .7 ---— 4.5 —--- 3.6 2.4 ————— .07 —--- —--- .08 Sediments in slough near edge 415454 8.8 —————— .0025 -——— .5 ---— 3.2 -——— 6.5 -——— 2.4 ————— .10 ——-- -—-- .54 Sarcobatus Flat (altered Nal‘COg water at depth 0—5 ft (0-1.5 m)) L2 Sarcobatus ver— miculatus (P)——— 408133 5.0 ----- 0.015 ------ 1.5 ——-- 23.2 —--- 5.5 —--- 5.9 —-—- 0.03 ----- ---- 0.06 ----- Brown carbonate Zone ------------ 411479 ---— 1.6 ————— 0.03 ---— 1.0 --—— 6.2 --—— 2.0 --—— 3.6 ————— 0.02 5.7 ---- 0.07 Soft dembo in playa edge —————— 415461 -——- 3.2 ----- .02 ———- 1.0 6.0 ———- 3.1 ———- 3.2 ————— .03 ---- ---- .55 Hard playa mud (no plants) ————— 415462 ——-- 2.2 ----- .015 ——-- 1.5 ——-- 5.5 ———— 2.4 ———- 3.1 .03 ---- .18 Oasis Valley (drains into Sarcobatus Flat; NaHC03 water at depth 0-5 ft (0-1.5 m)) L1 Sarcobatus ver- miculatus (P)——— 408132 7.5 ————— 0.006 —————— 1.5 --—— 8.0 22.4 ---— 8.8 ---— 0.015 ————— ---- 0.16 ----- Brown carbonate zone with NaC1—— 408175 ———— 2.3 ----- 0.015 ——-- 0.96 ---— 2.6 —--— 3.0 ——-- 3.0 ————— 0.01 2.3 ---- 0.18 Efflorescent crust of NaCl ————————— 408174 ——-— 46 ----- .01 .30 —--— 1.3 ———— >35 ———— 3.0 ————— .02 5.9 ---- .47 SUMMARY indicator of lithium but it cannot grow in soils of high salt (sodium chloride) content. Plants would not be useful in prospecting for high- potassium brines, because nutrient elements, such as potassium, that are required by plants in large amounts in normal physiological processes are not generally accu- mulated to any great extent. The plant absorbs only the amount needed regardless of the amount in the soil unless the quantities available are extremely high. As the par- ticular species of plants that grow in the chloride and sulfate zones use sodium in place of potassium even when the latter is available, no correlation between plant con- tents and brines would be expected. HEALTH ASPECTS OF LITHIUM Little is known concerning the long-term effects of lithi- um deficiency or excess on health or disease of domestic animals or man. Claims for the efficacy of lithium in the treatment of gout, diabetes, and epilepsy have not been substantiated. Probably the most promising use for lithi- um in medical practice arises from its profoundly tran- quilizing effect on manicdepressive patients (Cade, 1949; Schou, 1959). In Texas, a strong negative correlation exists between lithiumintake, as indexed by concentrations in drinking water, and admission rates to State mental insti- tutions (Dawson and others, 1970). According to Law— rence Razavi (written commun., 1970), lithium stabilizes chromosomes by shortening the amplitude of helical poly- mers, including DNA. Although the concentrations cannot be compared with those used in medical treat- ment, one might conclude that relatively high levels (20—100 ug/l) of lithium in drinking water as might be en- countered in the Basin and Range province would be bene- ficial in improving chromosomal quality and cell mem- brane strength. The well-known negative correlation between water hardness and cardiovascular mortality can be explained statistically by lithium concentrations concomitant with water hardness (Voors, 1971). Biological evidence that low levels of lithium provide a protective effect against athero- sclerosis is accumulating. No cases of toxicity from naturally occurring lithium have been reported; indeed, increasing evidence suggests several beneficial effects. SUMMARY Lithium contents average 22-655 ppm in igneous rocks, 17 ppm in sandstone, 46 ppm in shale, and 26 ppm in lime- stone. During weathering, much of the released lithium is trapped in montmorillonitic clays; but lithium that reach- es the sea tends to remain in solution as the highly soluble compound lithium chloride (average, 0.1 ppm). From 8 to 400 ppm lithium have previously been reported in soils. 21 Both open and closed basins occur in the Basin and Range province; hydrologically closed basins are dis- charged by evapotranspiration, and open basins are dis- charged by underflow to basins of lower hydraulic head. The median lithium content for 100 major municipal water supplies of the United States is 2 ug/l; that of 18 springs and wells of the Amargosa Desert, a closed basin, is 105 pg/ 1. Thermal hot springs are a common source of lithium in geochemical processes, and more lithium is contained in sodium bicarbonate water originating in tuffs than in calcium-magnesium bicarbonate water orig- inating in the Paleozoic limestones. A further concentra- tion of lithium takes place by the evapotranspiration of brines contained in closed basins that drain a large re- charge area and that were occupied by lakes in Pleistocene time. The percentages of lithium in the brines along with potassium, bromine, boron, and halite are of economic value in several areas. Further investigation of lacustrine clays in closed basins, particularly in the Amargosa Des- ert, may reveal lithium concentrations of economic value. In an evaporative basin, lithium may remain in solu- tion until a late stage and then be precipitated along with sodium, potassium, and boron in the chloride and sulfate zones, although lithium is readily adsorbed by clay miner- als in lacustrine deposits and also precipitated in marls of biogenous origin. Sediments collected for our project con- tained maximum values of 70—2,000 ppm lithium in 12 closed basins and maximum values of 40—150 ppm in 8 open basins. Median contents of lithium in the type of sediments in closed basins were: lacustrine clays 700 ppm lithium; hot- spring travertines 450 ppm; sulfate-chloride, and carbon- ate zones, 150 ppm; playa clay, 150 ppm; and alluvium, 30 ppm. Lithium may be concentrated in amounts greater than 1,000 ppm in hot-spring deposits where lithium is dissolved at depth, pumped to the surface, and precipi- tated around the roots of hydrophytes. Lithium is reported to be toxic to most cultivated plants at concentrations as low as 30 ppm in the soil, and for citrus plants, as low as 12 ppm. Vegetation generally con- tains about 1 ppm lithium in the dry weight; a maximum concentration of 4,400 ppm has been reported in the ash (or £750 ppm in dry weight) of healthy tobacco. In the Great Basin, pickleweed and rush containing 3,000 ppm in the ash grow without apparent damage in soils con- taining 300-500 ppm lithium. A median of 150 ppm in the ash or 22.8 ppm in dry weight of higher plants collected from these basins is considerably greater than the average of 1.3 ppm in dry weight found by Bertrand. The lithium content in the vegetation correlates well with that of the soil (as shown in fig. 2) except at levels of less than 100 ppm; at lower levels, the greater relative con- tent in plants suggests that lithium is essential to plants. The species of plants that accumulate the greatest 22 LITHIUM IN UNCONSOLIDATED SEDIMENTS AND PLANTS, CALIFORNIA AND NEVADA amounts of lithium are largely from the Chenopodiaceae family, and most lithium accumulators also take up large amounts of sodium chloride. The analysis of hydrophytes in springs at the alluvium-basin sediment contact and of halophytes growing in the chloride and sulfate zones of closed basins may be useful in evaluating the economic possibility of lithium extraction from brines. The high lithium concentrations in water and vege- tation of hydrologically closed basins may have a beneficial effect on the health of both lower animals and people living in the Basin and Range province. REFERENCES CITED Bear, F. E., ed., 1964, Chemistry of the soil [2d ed.]: Am. Chem. Soc. Mon. 160, 515 p. Bertrand, Didier, 1952, Sur la répartition du lithium chez les Phanéro- games [The distribution of lithium in the phanerogams]: Acad. Sci. Comptes Rendus, v. 234, p. 2102—2104. 1959, Nouvelles recherches sur la répartition du lithium chez les Phanerogames [New investigations on the distribution of lithium in the phanérogams]: Acad. Sci. Comptes Rendus, v. 249, pt. 1, p. 787-788. Borovik-Romanova, T. F., 1965, The content of lithium in plants, in Khitarov, N. 1., ed., Problems of geochemistry [in Russian]: Akad. Nauk SSSR Inst. Geokhimii i Analiticheskoi Khimii; trans- lated into English in Jerusalem, Israel Program for Sci. Trans- lations, 1969. Bradford, G. B., 1966, Lithium, Chap. 17, in Chapman, H. D., ed., Diagnostic criteria for plants and soils: Berkeley, California Univ., Div. Agr. Sci., Agr. Expt. Sta., p. 218—223. Brownell, P. F., 1965, Sodium as an essential micronutrient element for a higher plant (Atriplex vesican'a): Plant Physiology, v. 40, p. 460—468. Cade, J. F. J., 1949, Lithium salts in the treatment of psychotic excite- ment: Med. Jour. Australia, v. 2, p. 349—352. Dawson, E. B., Moore, T. D., and McGanity, W. J., 1970, The mathe- matical relationship of drinking water, lithium, and rainfall to mental hospital admission: Diseases of the Nervous System, v. 31, p. 811—820. Durfor, C. N., and Becker, Edith, 1964, Public water supplies of the 100 largest cities in the United States, 1962: U.S. Geol. Survey Water-Supply Paper 1812, 364 p. Eakin, T. E., 1963, Ground-water appraisal of Dry Lake and Delamar Valleys, Lincoln County, Nevada: Nevada Dept. Conserv. and Nat. Resources Ground-Water Resources—Reconn. Ser. Rept. 16, 26 p. Eakin, T. E., Schoff, S. L., and Cohen, Philip, 1963, Regional hydrol- ogy of a part of southern Nevada—A reconnaissance: U.S. Geol. Survey open-file report, 40 p. Edwards, J. D., 1941, Cytological studies of toxicity in meristem cells of the roots of Zea mays.‘ South Dakota Acad. Sci. Proc., v. 21, p. 65—67. Epstein, Emanuel, 1960, Calcium-lithium competition in absorption by plant roots: Nature, v. 185, no. 4714, p. 705—706. Evans, H. J., and Sorger, G. L., 1966, Role of mineral elements with emphasis on the univalent cations: Ann. Rev. Plant Physiology, v. 17, p. 47—76. , Foshag, W. F., and Woodford, A. 0., 1936, Bentonitic magnesian clay mineral from California: Am. Mineralogist, v. 21, no. 4, p. 238—244. Gale, H. S., 1915, Salines in the Owens, Searles, and Panamint basins, southeastern California: U.S. Geol. Survey Bull. 580—L, p. 251-323. 1951, Geology of the saline deposits, Bristol Dry Lake, San 113§rr2itlirdino County, California: California Div. Mines Spec. Rept. , p. Headden, W. P., 1921, Titanium, barium, strontium and lithium in certain plants: Colorado Agr. Expt. Sta. Bull. 267, 20 p. Hunt, C. B., 1966, Plant ecology of Death Valley, California, with a section on Distribution of fungi and algae, by C. B. Hunt and L. W. Durrell: U.S. Geol. Survey Prof. Paper 509, 68 p. Hunt, C. B., and Robinson, T. W., 1969, Possible interbasin circulation of ground water in the southern part of the Great Basin, in Geo- logical Survey research 1960: U.S. Geol. Survey Prof. Paper 400—3, p. B273—B274. Hunt, C. B., Robinson, T. W., Bowles, W. A., and Washburn, A. L., 1966, Hydrologic basin, Death Valley, California: U.S. Geol. Sur- vey Prof. Paper 494—B, 138 p. Johnson, M. S., and Hibbard, D. E., 1957, Geology of the Atomic Energy Commission Nevada proving grounds area, Nevada: U.S. Geol. Survey Bull. lO2l‘—K, p. 333—384. Jones, B. F., 1966, Geochemical evolution of closed basin water in the western Great Basin, in Symposium on salt, 2d, Geology, geo- chemistry, mining, v. 1: Cleveland, Ohio, Northern Ohio Geol. Soc., p. 181—200. Kent, N. L., 1941, Absorption, translocation, and ultimate fate of lithium in the wheat plant: New Phytologist, v. 40, p. 291—298. Linstow, 0. von, 1929, Bodenanzeigende Pflanzen [Soil-indicating plants] [2d ed.]: Preussischen Geol. Landesanstalt, Abh., new ser. 114, 246 p. Lombardi, O. W., 1963, Observations on the distribution of chemical elements in the terrestrial saline deposits of Saline Valley, Cali- fornia: U.S. Naval Ordnance Test Station NOTS TP 2916, 41 p. Mason, B. H., 1958, Principles of geochemistry [2d ed.]: New York, John Wiley 8c Sons, 310 p. Meinzer, O. E., 1917, Geology and water resources of Big Smoky, Clayton; and Alkali Spring Valleys, Nevada: US, Geol. Survey Water-Supply Paper 423, 167 p. Myers, A. T., Havens, R. G., and Dunton, P. J., 1961, A spectrochemical method for the semiquantitative analysis of rocks, minerals, and ores: U.S. Geol. Survey Bull. 1084—1, p. 207—229. Parker, R. L., 1967, Composition of the Earth’s crust, Chap. D, in Data of geochemistry [6th ed.]: U.S. Geol. Survey Prof. Paper 440—D, 19 p. Pistrang, M. A., and Kunkel, Fred, 1958, A brief geologic and hydro- logic reconnaissance of the Furnace Creek Wash area, Death Valley National Monument, California: U.S. Geol. Survey open-file report, 63 p. Puccini, Giuliano, 1957, Stimulant action of lithium salts on the flower production of the perpetual carnation of the Riviera [in Italian, with English summary]: Ann. Sper. Agrar. (Rome), v. 11, p. 41-63. Rankama, Kalervo, and Sahama, T. G., 1950, Geochemistry: Chicago Univ. Press, 912 p. Robinson, W. 0., Steinkoenig, L. A., and Miller, C. F., 1917, The relation of some of the rarer elements in soils and plants: U.S. Dept. Agriculture Bull. 600, 27 p. Ronov, A. B., Migdisov, A. A., Voskresenskaya, N. T., and Korzina, G. A., 1970, Geochemistry of lithium in the sedimentary cycle [in Russian]: Geokhimiya, 1970, no. 2, p. 131—161; translated in Geochemistry Internat., v. 7, no. 2, p. 75-102, 1970. Schoff, S. L., and Moore, J. E., 1964, Chemistry and movement of ground water, Nevada Test Site: U.S. Geol. Survey open-file report. Schou, M., 1959, Lithium in psychiatric therapy—Stock-taking after ten years: Psychopharmacologia, v. 1, p. 65—78. Shawe, D. R., Mountjoy, Wayne, and Duke, Walter, 1964, Lithium associated with beryllium in rhyolitic tuff at Spor Mountain, wes- tern Juab County, Utah, in Geological Survey research 1964: U.S. Geol. Survey Prof. Paper 501—C, p. C86—C87. Sievers, M. L., and Cannon, H. L., 1974, Disease patterns of Pima REFERENCES CITED 23 Indians of the Gila Indian Reservation of Arizona in relation to the geochemical environment, in Hemphill, D. D., ed., Univ. Missouri Symposium on Trace Elements in Environmental Health, v. 7, p. 57-61. Steinkoenig, L. A., 1915, Lithium in soils: Indus. and Eng. Chemistry Jour., v. 7, no. 5, p. 425-426. Stewart, F. H., 1963, Marine evaporites, Chap. Y, in Data of geo- chemistry [6th ed.]: U.S. Geol. Survey Prof. Paper 440-Y, 52 p. Stose, G. W., assisted by Ljungstedt, O. A., 1932, Geologic map of the United States: US. Geol. Survey, scale 1:2,500,000 [reprinted 1960]. Strock, L. W., 1936, Zur Geochemie des Lithiums: Nachr. Ges. Wiss. Gottingen Math-Phys. Kl., IV, N. F. 1, no. 15, p. 171—204. Swaine, D. 1., 1955, The trace-element content of soils: [England] Commonwealth Bur. Soil Sci. Teeh. Comm. 48. Tardy, Yves, Krempp, GErard, and Trauth, Norbert, 1972, Le lithium dans les mineraux argileaux des sediments et des sols [Lithium in clay minerals of sediments and soils] [in French]: Geochim. et Cosmochim. Acta, v. 86, no. 4, p. 397—412. Vinogradov, A. P., 1952, Fundamental laws of the distribution of trace elements among plants and environment, in Mikroelementy v zhizni Rastennii i Zhivotnykh-—-Mikroelement Konf., 1950 [in Russian]: Akad. Nauk SSSR Trudy, p. 7—20. Vlasiuk, P. A., and Okhrimenko, M. F., 1967, On biological para- doxical phenomena [in Russian]: Acad. Sci. Ukranian, RSR Kiev, Ser. Geology, Geophysics, Chemistry, and Biology, 1969, no. 4, UDK 581.1:581.143; English translation in Acad. Agr. Sci. Rept. 10, V. I. Lenin All-Union (Order of Lenin), Kolas Publisher. 1969, The effect of lithium on photochemical activity of chloro- plasts of tomato and pepper leaves [in Russian]: Acad. Sci. Ukran- ian, RSR Kiev, Ser. Geology, Geophysics, Chemistry, and Biology, no. 4, UDK 581132254634. Voelcker, J. A., 1912, Pot-culture experiments: Royal Agr. Soc. joum., England, v. 73, p. 314—338. Voors, A. W., 1971, Minerals in the municipal water and atherosclerotic heart death: Am. Jour. Epidemiology, v. 93, p. 259—266. Walker, G. E., and Eakin, T. E., 1963, Geology and ground water of Amargosa Desert, Nevada-California: Nevada Dept. Conserv. and Nat. Resources, Ground-Water Resources—Reconn. Ser. Rept. 14, 58 p. Wallace, A., Romney, E. M., and Hale, V. (2., 1973, Sodium relations in desert plants—[PL] 1. Cation contents of some plant species from the Mojave and Great Basin deserts: Soil Sci., v. 115, no. 4, p. 284—287. Ward, F. N., Nakagawa, H. M., Harms, T. F., and VanSickle, G. H., 1969, Atomic-absorption methods of analysis useful in geo— chemical exploration: U.S. Geol. Survey Bull. 1289, 45 p. Winograd, I. J., 1962, Interbasin movement of ground water at the Nevada Test Site: US. Geol. Survey open-file report, 11 p. fiUS. GOVERNMENT PRINTING OFFICE: 1975—677-340/2 0575' 7 DAYS P6; 1/. 9/ 9 Volcanic Suites and Related Cauldrons of Timber Mounta1n—Oas1s Valley Caldera Complex Southern Nevada 1 1 GEOLOGICAL SURVEY PROFESSIONAL PAPER 919 Prepared in cooperation with the U. S. Atomzc Energy Commisszon SEP 281,976 \\ ’7’)». _' (‘5 /’ \ @flfiflmg‘é‘y SEP 1 6 1976 U.&S.D. —4— VOLCANIC SUITES AND RELATED CAULDRONS OF TIMBER MOUNTAIN—OASIS VALLEY CALDERA COMPLEX, SOUTHERN NEVADA 4 K WEST CALDERA RIM WESTERN CALDERA MOAT T ff fC t ff Road 3"“ If u 0 :0 .5 V f “Vv‘ -._ .~_,.\ Mountain Thirsty Canyon Tuff and basalt/ M~ , M WKWV/flN/j , w ;_ , w; ;-_ ‘flwyr —— ,2 x? N ---—-~~~ wwfi MW“ ’ “ " ’* ‘\'-\—/<~- ‘ ~.. ' ‘ ‘ 43‘ R “JN ?\r W” \«q‘ “\5\ ~t’x, \\ (3A ,a-“N .- AmmoniabTanks/ \ ‘ \ AémRSK: afinmggfsf‘xfl 4:} \, :2;\\‘. \fl‘ em ”MM ‘1“ ' v“\:;," 4;”?t"(' \,,\~ /\’*‘ ’NM ~ a... mw’g/ M w- [“7“ an /l:>34 /V// '\‘ Jfi‘a/ <\C A’\,§\\< i gRamler Mesa a? "‘" x ‘5 P’ ’2 \K“\ dfiu ‘~ ‘ M» , , , w/ / mm?” \ «5}; 1‘: ~;> r s ~i ember x» )/ /, . ins? ,3“ .fl‘b‘n" gAmmonia Tanks Member,.~.'*’ " _, ”N" ( ~— " \“4/ //"' éfi'fii “ a,“ A 5K , M Beatty Wash \ w W / Q, Ammonia Tanks Membe ,'>\ ”'7 //r’-:;:E‘W j W’fi" ,_,, /7// / :‘x‘v “‘3\ A o / / flag, is; "—1 - 1/ T' / ,1, ., Panoramlc view of Timber Mountain caldera from point southwest of Beauy Wash. Photograph by P. W. Lipman, 1964. q ‘ ._ A4 .‘ M < .( ‘4 N -< y" 14 F ’ SOUTH RIM ¥ $53332?“ 3ENTRAL DOME SOUTH CALDERA MOAT //§¢®\ Mountalr I” be Mo t ' Mr,” ‘ [ILL/.31" in“ Enhgjggafif‘ fl‘A/y—I‘ My)!“ w .1 , A/ :/ w «4/, iii?“ iv Agmonia Tankg MJDQGI’ u“ \ vddera— —fiH deposits V m "2» - ”Ammonia T nks - W‘K 'r I: «‘1. / 1.0 / - ."/ I / ",// /_1~'.' j I: g f. : ’:" 1 f3. ’f/ Volcanic Suites and * Related Cauldrons of Timber Mountain-Oasis Valley Caldera Complex, Southern Nevada By F. M. BYERS, JR., W. J. CARR, PAUL P. ORKILD, W. D. QUINLIVAN, and K. A. SARGENT GEOLOGICAL SURVEY PROFESSIONAL PAPER 919 Prepared in cooperation with the U.S. Atomic Energy Commission Stratigraphic, petrochemical, and structural relations of igneous rocks of Timber Mountain—Oasis Valley caldera complex, including calc-alkalic rocks from Silent Canyon caldera, southwestern Nevada volcanic field UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON: 1976 UNITED STATES DEPARTMENT OF THE INTERIOR THOMAS S. KLEPPE, Secretary GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress Cataloging in Publication Data Main entry under title? Volcanic suites and related cauldrons of Timber Mountain—Oasis Valley caldera complex, southern Nevada. (Geological Survey Professional Paper 919) “Prepared in cooperation with the US. Atomic Energy Commission.” Bibliography: p. Supt. of Docs. no.1 119.162919 1. Volcanic ash, tuff, etc.—Nevada—Nye Co. 2. Calderas—Nevada—Nye Co. I. Byers, Frank M., 1916- 11. United States. Atomic Energy Commission. 111. Series: United States Geological Survey Professional Paper 919, QE461.V63 552’.2'0979334 76-608057 For sale by the Superintendent of Documents, US. Government Printing Office Washington, DC. 20402 Stock Number 024—001—02868—9 CONTENTS Page Metric-English equivalents ......................................................................................................... IX Abstract ............................................ . . 1 [Introduction ......................................... 2 Acknowledgments ................................................................................. 2 Nomenclature ................................................................................. 2 General geologic relations ............................................................................... 5 Tuffs and lavas related to Sleeping Butte caldera .................................. 7 Redrock Valley Tuff ..................................................................... 7 Crater Flat Tuff and intercalated lava .. ................................................. 10 Bullfrog Member ............... . .............................................. 11 Intercalated lava flow. ................................................... .. l4 Prow Pass Member ..................................................................................... l4 Tuffs of Sleeping Butte and their relation to caldera wall ....................... l5 Tuff of Tolicha Peak .............................................................................................. 15 Biotite-hornblende rhyolite lavas west of Split Ridge ......................... l5 Peralkaline rocks of Silent Canyon caldera ............................................................................. l6 Stockade Wash Tuff and related calc-alkalic rocks associated with Silent Canyon caldera ....... l6 Paintbrush Tuff and rocks related to Claim Canyon cauldron ......................................... 21 General features ............................................................................................ 21 Original and redefinition of the Paintbrush Tuff ............................ 21 Bedded tuff ..................................................................................................................... 22 Geologic relations between welded ash-flow tuffs, lavas, and Claim Canyon cauldron ................................................................................................................. 24 Lithology and field recognition ............ 24 Topopah Spring Member ................................... 25 Pah Canyon Member and related pre-Pah Canyon lavas ................................................. 25 Lavas and ash-flow tuff between Pah Canyon and Yucca Mountain Members ................... 25 Yucca Mountain Member ............................................ 31 Tiva Canyon Member .................................................. 31 Tuff breccias within Claim Canyon cauldron segment ..................... 33 Tuff of Pinyon Pass .................................................................................................. 35 Post—Tiva Canyon rhyolite lavas ........................................................................... 35 Cauldron subsidences related to eruption of ash—flow tuff members ..................... 36 Possible magmatic resurgence of Claim Canyon cauldron .................................... 37 Timber Mountain Tuff and rocks related to Timber Mountain caldera ...... 38 Original definition and redefinition of Timber Mountain Tuff ........ 38 Rainier Mesa Member .......................................................................... 39 Ammonia Tanks Member . ............................................................ 43 Tuff of Buttonhook Wash ...................................................................... 47 Tuffs of Crooked Canyon ........................................... 49 Debris, flows ............................................................................................................ 50 Lavas petrologically related to Timber Mountain Tuff ..................... 51 Pre-Rainier Mesa lavas ...................................................................................... 51 Pre«Ammonia Tanks lavas in Timber Mountain caldera moat. ...................... 52 Intrusive rocks and relation to resurgent doming ................................................................. 52 Caldera collapses related to eruptions of Rainier Mesa and Ammonia Tanks Members 56 Younger intracaldera rocks in Timber Mountain and Oasis Valley calderas .............................. 59 Rhyodacitic and mafic lavas ..................................................................................... 59 Tuffs of Fleur-de-lis Ranch and related rhyolite lavas... .................... 61 Tuff of Cutoff Road and related rhyolite lavas ....................................... 51 Rliyolite lavas of Fortymile Canyon .............................................. 62 Summary of geologic history .......................................................... 63 References cited ........................................................................................................................... 67 VIII CONTENTS ILLUSTRATIONS FRONTISPIECE. Panoramic view of Timber Mountain caldera from point southwest of Beatty Wash. FIGURE TABLE 1. Map of Southwestern Nevada volcanic field showing location of Timber Mountain caldera area ......................................... 2. Index to original sources of geologic data and other maps for the Timber Mountain caldera art-a, Nevada Test Site, and adjacent region ....................................................................................... 3. Generalized schematic diagram through southwestern Nevada volcanic field.... ........................... 4. Map showing areal distribution of Redrock Valley Tuff, members of Crater Flat Tuff, and other volcanic rocks are probably related to Sleeping Butte caldera ........................................................................ related lavas ................................................... 7. Chart showing stratigraphic relations and revisions of Stockade Wash, Paintbrush, and Timber Mountain Tuffs, Timber Mountain caldera and vicinity 8- Map showing areal extent of Topopah Spring and Tiva Canyon Members of Paintbrush Tuff and related lavas 9. Graph showing modal and silica ranges of units of Paintbrush Tuff and petrologically related lavas 10. Map showing areal extent of Pah Canyon and Yucca Mountain Members of Paintbrush Tuff ....... ll. Generalized section A—A’ through drill holes UE19f and UElQi, Silent Canyon caldera ............................ l2. Generalized geologic map of Claim Canyon cauldron segment and vicinity 13. Geologic sections B—B’ and C—C’ through Claim Canyon cauldron segment ........................................... 14. Generalized isopach map of Rainier Mesa Member of Timber Mountain Tuff ......................................................... l5. Graph showing modal and silica ranges of units of Timber Mountain Tuff and petrologically related lavas l6. Generalized isopach map of Ammonia Tanks Member of Timber Mountain Tuff ...... 17. Photograph showing structural unconformity between uppermost high-silica rhyolite and underlying quartz latite of upper part of Ammonia Tanks Member 18. Map showing combined thicknesses of tuffs of Buttonhook Wash and Crooked Canyon and intrusive rocks on Timber Mountain resurgent dome 19. Graphs showing phenocryst mineralogy of rhyolite lava of Windy Wash on opposite flanks of Timber Mountain caldera .................................................................................................................................. 20. Photograph of rhyolite tuff dike that cuts basal quartz latite of Ammonia Tanks Member and dips inwar 2]. Graph showing modal and silica ranges of intrusive rocks on Timber Mountain compared with ranges of high-silica rhyolite and quartz latite subunits of upper part of Ammonia Tanks Member 22. Generalized interpretive section across Timber Mountain resurgent dome 23. Section across Oasis Valley caldera segment 24. Map showing distribution of younger intracaldera rocks of Timber Mountain~Oasis Valley caldera complex ........................... 25. Graph showing modal and silica ranges of younger intracaldera volcanic rocks of Timber Mountain—Oasis Valley Caldera complex 26. Sequence of north-south interpretive diagrams through Timber Mountain caldera ............................................................... TABLES l. Generalized stratigraphic nomenclature of tuff formations and members, and their indicated volcanic center ......... 2. Nonmagnetic heavy minerals of Paintbrush and Stockade Wash Tuffs and related tuffs and lavas ......................................... 3. Thicknesses and K-Ar ages of Paintbrush Tuff units and genetically related lavas within Claim Canyon cauldron segment and outside on cauldron rim ............................................................................. 4. Combined thin-section modes and computed accessory mineral percentages of local ash-flow tuff of Paintbrush Tuff and overlying pre-Tiva Canyon rhyolite lava that 5. Graph showing modal and silica ranges of Redrock Valley, Crater Flat, and Stockade Wash Tuffs and petrologically 6. Map showing isopachs of Stockade Wash Tuff, Stockade Wash(?) Tuff, and tuff of Blacktop Buttes ................................... ........................... , Page 12 17 18 22 26 28 30 33 34 40 42 44 47 48 51 53 54 55 58 60 62 65 Page 5 21 24 31 CONTENTS METRIC-ENGLISH EQUIVALENTS IX Metric unit English equivalent Metric unit English equivalent Length Specific combinations—Continued millimetre (mm) = 0.03937 inch (in) litre per second (l/s) : .0353 cubic foot per second metre (m) : 3.28 feet (ft) cubic metre per second kilometre (km) : .62 mile (mi) per square kilometre [(m3/s)/km2] : 91.47 cubic feet per second per Area square mile [(ft3/s)/mi2] metre per day (m/d) 3.28 feet per day (hydraulic square metre (m2) : 10.76 square feet (ft?) COHdllCtiVltY) (ft/d) square kilometre (km?) 2 .386 square mile (mi?) metre per kilometre _ ‘ hectare (ha) : 2.47 acres m/km) = ".25 feet per mile (ft/ml) kilometre per hour Volume (km/h) 2 .9113 foot per second (ft/s) metre per seccimd (an/s) : 3.28 feet per second . 3 ___ . i h 3 me re square per ay fig? (0193ntimtre (cm ) = 62.831 33315 igghesun ) (mZ/d) = 10-764 feet squared per day (fwd) cubic metre (m3) : 35.31 cubic feet (fta) (transmissivity) cubic metre = .00081 acre-foot (acre-ft) cubicsmetre per second . cubic hectometre (hm3) 2810.7 acre-feet m /S) = 22-826 million gallons per day litre : 2.113 pints (pt) . (Meal/d) litre : 1.06 quarts (qt) cublc metre per minute litre : .26 gallon (gal) (ms/min) =264.2 gallons per minute (gal/min) cubic metre = .00026 million gallons (Mgal 0r litre per second (I/S) = 15-85 gallons per minute 106 gal) litre per second per cubic metre = 6.290 barrels (be) (1 bbl=42 gal) metre [(I/S)/m] = 83 gallons per minute per foot [(gal/mln) /ftl Weight kilometre per hour (tkm/h) d ( / ) = 2.3?” mile per hglur (mi/h) gram (g) = 0.035 ounce, avoirdupois (oz uvdp) me re per secon m S = ' m es per our = gram per cubic 55332 (t) = 122022 igfisfds’hfi‘iiidfggésugb mp) centimetre (g/cmS) = 62.43 pounds per cubic foot (lb/ft“) tonne = .98 ton, long (2,240 lb) gram per square centimetre (g/cm”) : 2.048 pounds per square foot (lb/ft?) ' ' ' gram per square SpeCIfic combinatlons centimetre .0142 pound per square inch (lb/in”) kilogram per square centimetre (kg/cm?) : 0.96 atmosphere (atm) Temperature kilogram per square centimetre : .98 bar (0.9869 atm) degree Celsius (°C) : 1.8 degrees Fahrenheit (T) cubic metre per second _ degrees Celsius (m3/s) : 30.3 cubic feet per second (ft3/s) (temperature) =[(1.8X°C) +32] degrees Fahrenheit VOLCANIC SUITES AND RELATED CAULDRONS OF TIMBER MOUNTAIN—OASIS VALLEY CALDERA COMPLEX, SOUTHERN NEVADA By F. M. BYERS, JR., W. J. CARR, PAUL P. ORKILD, W. D. QUINLIVAN, and K. A. SARGENT ABSTRACT The Timber Mountain—Oasis Valley caldera complex occupies a slightly elliptical area about 40 km (25 mi) in maximum diameter in southern Nye County, Nev., and is a major part of the southwestern Nevada volcanic field, which includes peralkaline, alkali-calcic, calc- alkalic, and calcic centers. Upper Miocene and lower Pliocene calc- alkalic and alkali-calcic ash-flow sheets and petrologically related igneous rocks of the Timber Mountain—Oasis Valley caldera complex were erupted 16—9 m.y. (million years) ago and were associated with multiple cauldron subsidences. The newly named calc-alkalic Redrock Valley and Crater Flat Tuffs were associated with collapses 16—14 m.y. ago, possibly within the Sleeping Butte caldera, only a small part of which is still exposed. The Redrock Valley consists of only one known cooling unit, whereas the Crater Flat consists of two newly named members, the Bullfrog and the Prow Pass. The quartz-bearing Stockade Wash Tuff, formerly the Stockade Wash Member of the quartz-poor Paintbrush Tuff, is here raised to separate formational rank. It is a calc-alkalic tuff, erupted between 13.8 and 13.2 m.y. ago as a late effusive from the dominantly peralkaline Silent Canyon caldera. The Stockade Wash is unrelated to the Paintbrush, in both its petrologic features and volcanic source, but is closely similar petro- graphically to two other late ash—flow sheets of limited extent within and around Silent Canyon caldera and also to 1,500 m (5,000 ft) of calc-alkalic tuffs and rhyolite lavas of Area 20, which fill Silent Canyon caldera. The Paintbrush Tuff is here restricted to include the quartz-free to quartz-poor bedded and intercalated welded ash-flow tuffs that occur between the underlying redefined Stockade Wash Tuff, or the tuffs and rhyolites of Area 20, and the overlying Timber Mountain Tuff. The Paintbrush is a major alkali-calcic volcanic sequence of genetically related bedded tuff and ash-flow tuff sheets that were erupted 13.2—12.5 m.y. ago from the Claim Canyon cauldron center, the southernmost part of which is exposed in an arcuate segment at the south side of the complex. The ash-flow sheets consistof (1) the Topopah Spring Member near the base, (2) the Pah Canyon Member, (3) a local subsurface unit that may grade upward into a lava flow, (4) the Yucca Mountain Member, (5) the Tiva Canyon Member, which includes the intracauldron tuff of Chocolate Mountain, and (6) the intracauldron tuff of Pinyon Pass. The Claim Canyon cauldron segment exposes greatly thickened intra- cauldron facies of Paintbrush welded tuffs, intercalated lavas, and tuff breccia. The estimated volume of the Topopah Spring Member is about 250 km3 (60 mi’), but may be much greater, as the unit is buried by younger welded tuffs and lavas within the exposed segment of the cauldron. The post-Topopah Spring ash-flow sheets of the Paintbrush Tuff and intercalated lavas are more than 1,500 m thick in the Claim ’ Canyon cauldron segment. This thick post-Topopah Spring intra- cauldron filling implies considerable structural depression inside the cauldron. It seems unlikely that this much structural depression could have occurred within the cauldron following the eruption of only the Topopah Spring Member. We interpret the exposed Claim Canyon cauldron segment as probably recording continued episodic subsidence during or immediately following eruption of each ash-flow sheet of the Paintbrush. The maximum subsidence of Claim Canyon cauldron probably occurred during the late stages of eruption of 1,000 km3 (250 mi3) of the Tiva Canyon Member. An alternative interpretation infers that subsidence at the Oasis Valley caldera segment was related to erup- tion of the Tiva Canyon. A thick tuff breccia along and near the wall of the Claim Canyon cauldron segment, however, grades laterally and vertically into welded masses of the Tiva Canyon and may mark a locus of Tiva Canyon volcanic vents. Cauldron resurgence possibly affected the Claim Canyon cauldron, as the tuffs filling the exposed Claim Canyon cauldron segment now stand topographically and structurally high; the buried larger part of the cauldron to the north has been dropped to great depth by collapses of the younger Timber Mountain caldera. Alternatively, the segment was raised with a much larger Yucca Mountain block during early broad doming of Timber Mountain caldera prior to eruption of the Timber Mountain Tuff. The lower Pliocene alkali-calcic Timber Mountain Tuff is redefined to include all quartz-bearing ash-flow tuff sheets and minor interbedded ash-fall tuff erupted about 11 m.y. ago from the Timber Mountain caldera center. The tuff includes in ascending order, the Rainier Mesa and Ammonia Tanks Members, which are the two widespread ash-flow sheets of the original definition, and the newly added intracaldera tuffs of Buttonhook Wash and Crooked Canyon. The Ammonia Tanks Member has been further redefined to include the local tuffs of Cat Canyon and Transvaal, which are now known to be equivalent to part or all of the Ammonia Tanks. Two major subsidences of Timber Mountain caldera were associated with eruption of the Rainier Mesa Member, having a volume of more than 1,200 km5 (300 mil), and with the eruption of the overlying Ammonia Tanks Member, having a volume of about 900 km5 (230 mi’). The tuff exposed on the central resurgent dome of Timber Mountain caldera is intracaldera Ammonia Tanks, more than 900 m (3,000 ft) thick, and was formerly mapped as tuff of Cat Canyon. Thicknesses in excess of 450 m (1,500 ft), granophyric texture, and fluidal flow banding of both the Rainier Mesa and Ammonia Tanks Members within the Oasis Valley caldera segment suggest also partial collapse of that segment during eruption of the members, forming a large volcano-tectonic depression. The area that collapsed because of the eruption of the Rainier Mesa is somewhat larger than that which collapsed because of the Ammonia Tanks, as might be expected from the greater volume of the Rainier Mesa. The Timber Mountain resurgent dome in the central area of Timber Mountain caldera contains both inward-dipping high-silica rhyolite tuff dikes, possibly cone sheets, and an outward-dipping microgranitic ring 1 2 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA dike approachingquartz latite composition (68 percent silica). These two rock types are petrologically similar to limiting compositions of the Ammonia Tanks Member and may have come from different levels of a zoned silicic magma chamber high in the crust. Post-Timber Mountain alkali-calcic tuffs and related lavas are confined to the Oasis Valley-Timber Mountain caldera complex and probably were erupted ll.0-9.5 m. y. ago from within the Oasis Valley caldera segment. Rhyolite lavas extruded from the rim and moat areas of Timber Moun- tain caldera concluded the activity of the caldera complex about 9 m. y. ago. INTRODUCTION The ash-flow sheets discussed in this report constitute the major stratigraphic units shown on the U.S. Geological Survey map of the Timber Mountain caldera area (Byers and others, 1976) in the Nevada Test Site region of southern Nevada. An attempt is made to relate the tuff sheets and petrologically similar lavas and intrusives to major volcanic centers of the Timber Mountain—Oasis Valley caldera complex (fig. 1). A closely related paper (Christiansen and others, 1976) emphasizes the structural setting and associated Tertiary volcanism in the region around the caldera complex. Present knowledge of the relations and distribution of volcanic rocks in the Nevada Test Site region of south- western Nevada is based on 34 geologic quadrangle maps (fig. 2), about 30 exploratory drill holes, several Bouguer gravity (Don L. Healey, written commun., 1968) and aero- magnetic maps (G. D. Bath, written commun., 1968), more than 500 polarity determinations of natural remanent magnetism of rocks (G. D. Bath, written commun., 1963—68), radiometric age determinations of 51 volcanic rocks (Kistler, 1968; Marvin and others, 1970), and more than 400 modal analyses of thin sections. During the concentrated effort of geologic mapping from 1960 to 1964, the volcanic rocks at the Nevada Test Site were grouped into formations because of the need to define stratigraphic units for the geologic maps (fig. 2). Ash-flow tuff sheets of similar mineralogy and chemistry were grouped into formations, the individual sheets being members of the formations (Smith, 1960a, p. 812—813; R. L. Smith, oral commun., 1960—68; Christiansen, and others, 1968), but some stratigraphic divisions were based on limited knowledge regarding areal distributions of the units and their associations with major volcanic centers. In this paper the geologic and petrologic relations of the tuff sequences to lavas and intrusives at caldera centers are further emphasized in arriving at a logical basis for redefining some of the units. ACKNOWLEDGMENTS Several geologists have contributed materially to this report by supplying us with basic data and interpretation. R. L. Smith helped us define and solve many of the caldera problems through field conferences. D. L. Healey and G. D. Bath contributed gravity and aeromagnetic inter- pretation, which greatly aided in delineating the outlines of the buried caldera structures in figure 1. D. C. Noble contributed to the interpretation of volcanic history and location of the caldera boundaries. J. T. O’Connor supplied about 50 thin-section modes of the tuffs. Our colleagues R. E. Anderson, R. L. Christiansen, E. B. Ekren, and P. W. Lipman provided geologic information and stimulating discussion. We, however,- assume full responsibility for any errors or misinterpretations of the data. This work was fully supported by the U.S. Atomic Energy Commission. NOMENCLATURE In general, we follow guidelines of the ash-flow tuff nomenclature of Smith (1960a, p. 800—801; 1960b) and Ross and Smith (1961). For brevity in this report and on the geologic map of the Timber Mountain area (Byers and others, 1976) ash-flow tuff (Ross and Smith, 1961, p. 3) or simply the word “tuff” in the proper context carries the genetic connotation of ash-flow cooling unit of Smith (1960a, p. 801 ), as well as that of a lithologic term. A tuff or ash-flow cooling unit, in this connotation, was emplaced in an instant of geologic time and is therefore a time- marker as well as a lithologic unit (compare with Smith, 1960b, p. 150). The plural term “tuffs,” analogous to “lavas” or “lava flows,” is used for two or more ash-flow cooling units mapped as a unit (for example, Christiansen and Lipman, 1965; Sargent, 1969; Rogers and others, 1967). The singular term “tuff” is also used in formal stratigraphic nomenclature—for example, Paintgrush Tuff—in which it carries the dual genetic and lithologic implication just noted, inasmuch as the component tuff members and the informal tuffs are single ash-flow cooling units. “Ash-flow sheet” is also used descriptively in discussing the extensive cooling units of large volume associated with cauldron subsidence. Other types of tuff are indicated in this report and on the Timber Mountain map (Byers and others, 1976) by appro- priate modifiers (compare with Ross and Smith, 1961, p. 3), such as bedded tuff (descriptive) and ash-fall tuff (genetic). In Smith’s (1960b, p. 157—158) summary discussion of the cooling unit and composite sheet, he implied the existence of complete cooling breaks between cooling units and partial cooling breaks within compound cooling units by giving criteria for recognizing types of hiatuses between ash-flow units. Recognition of the nature of these hiatuses, herein referred to as either a partial or complete cooling break, is necessary in order to define an ash-flow cooling unit as the basic stratigraphic unit with time-marker implications (Smith, 1960a, pl. 1; Christiansen and others, 1968; Noble, Bath, and others 1968, p. C61). NOMENCLATURE 1174/s U \« T. “\A Mr») E, Obsidian ‘ BLACKM 7 ‘ Buttex ) H ”"oliéha , Peakx 8 x )j/ / >\/ ( ;' ‘ s/LEEPIN x l } c'ALUEgA \ ' 1 1,' to. OASI‘S'l/‘ALLEY‘ l ,/ CALDEBA SEGMENT- Fieundedis Ranch ‘ / ’ \\/\’ 11/ Q‘ X 6002 N -A ’17,,_,,.:7?:.w,."7 .Hnwul.N,,..é.,..,_,_.,. my, l . BULLFROG Sawtooth Mtn , k‘. g , Lt, ~Lc‘ l 639 N ’\ a... T w I l a \ / ;//\) ’\ / :_ “L SILENf\CANYON " 1’,” V—"f‘f‘w: CALD‘ERA-xgao /' ,/ ./ 1/ ‘Kv/ \\ I 1/00 {/1 ) //*~J(63l< }\ ,.»—\ ‘~ Mtn e ‘ \A . <5" \\ . <4) \_.__ no ( \. ‘\% } ”\W Tippipah ”age/fl f) . PointM" \‘ L '37? X6612 \ \“V ' 'r's‘o'" >< ’M’i‘ae’nm W i W :3” K.“ t ,, 4 Q \\ agate 00‘?" ’ .>< 7058 :9, {Hg/y gvofi‘v {f \gtlmshone \‘m The Bench (43mm {SAN you .\. Peek y X 4925 12\\ j / .1 camanonsesnem L093“ \“Ww x ' , ’ , ea \\ ; \“T'\»\ /' fl». H>CRATEF§ FLAT c! u )1 x5651 -~7 \— W/ ‘ . l ‘ / ,/ I ~. 3mm MS: 1 ,. n 0‘ ’12 ,»——7// 2 Black’Cohe l ( , ,r’“ [\1 T, “3’ ' , t : / . \\ Mt Salve: \1 3 c,» 30 (7/ a 1&3700 / E / 8" FLA/TS/ Wahmome-Salfir “/ x4682 3'. 431$? \x 12: P0465 f“ (I volcanic center \f/Nw R 46 E_ R 49E 10 20 MILES l ' l J 10 20 KILOMETRES EXPLANATION —l—_|——l- Approximate outer limit of Timber Mountain— ._L_l_.l_.L_1. Periphery of Timber Mountain resurgent dome Oasis Valley caldera complex. Includes Sleeping Butte and Claim Canyon segments. Dashed where indefinite FIGURE l.—Southwestern Nevada volcanic field, Nye County, Nev., showing location of Timber Mountain caldera and other major volcanic centers. Area shown on the geologic map by Byers and others (1976) is shaded. A compositional subdivision of an ash-flow tuff sheet, (Lipman, Christiansen, and O’Connor, 1966, p. F5), characterized by a unique and laterally correlatable although a topographic term, conveniently describes the phenocryst assemblage, is called a subunit, following more crystal-rich, commonly moremafic upper subunitof usage of Smith and Bailey (1966, p. 11) and Lipman, a compositionally zoned cooling unit. The caprock is so Christiansen, and O’Connor (1966). The term “caprock” named because it commonly forms a resistant capping TIMBER MOUNTAIN -OASIS VALLEY CALDERA COMPLEX, NEVADA 116°30’ 116°00' I I I I I NORTHERN NELLIS AIR FORCE BOMBING AND GUNNERY RANGE I (Ekren, Anderson, Rogers, and Noble, 1971) I I I OUARTZITE MOUNTAIN BELTED PEAK I (Rogers, Anderson, (Ekren, Rogers, I I Ekren and O'Connor, Anderson and Botinellv l 1967) 1967) I I I 37°3o’ I I NORTHERN HALF (Otaldugiggnfisgnd WHEELBARROW PEAK I BLACK MOUNTAIN Snyder 1989) RAINIER MESA AREA | I (Rogers, Ekren, Noble, ' (Sargent and Orkild, I _ (Blankennagel and 1973) | and War, 1968) Weir, 1973) I I I , ', — 7 I OAK SPRING I BLACK BUTTE I MOUNTAIN (Rogers and I GROOSMWM'NE I sw Noble 1969) I ' I \\ (Noble and CLIMAX srocxl (0°th;de \ Christiansen, ("WWW I 1967) I \ 1968) Poole, 1960) I \\ , _ : ,—' - ~—___ —. \‘ ' THIRSTY I Immense A L I I I \ l a . \\ l CANYON I I - , N315 RIDGEI \ I (O'Connor, I _, ,, l_ | \\ Anderson, 1 ' , :I‘ (I I \\ I and Lipman I I GOLDF/ELD, 45 mi (73 km) \ I I I K \\ I 1963) I I | \ I | TIMBER M0UNTAIN~T‘\ I CALDERA l THIRSTY (Byers, Carr, | CANYON SE TIMfiER a Christiansen, I (Lipman. E MOBN‘I'AIN , , Lipman, OrkiId, I OuinIivan. g In." am: __ - ,9 g and Ouinlivan, | Carr, and > a“ :1 1976) I Anderson, Fl I (I) I 1966) _, I 37000, (D I E | I m TBPOPAH I I an. I I I (Rinmhs | I _ and I , I _ mm. , I Beattyg I ‘ , 195??) | BULLFRDG L ——————— - ‘ _' (Cornwall and BARE MOUNTAIN Kleinhampl, (Cornwall and Kleinhampl, 1964) 1961) \ \ (37(gob wove 4; ‘7 4,, “I EXPLANATION I-_ Area of geologic map quadrangle Nevada Test Site FIGURE 2.—Index showing original sources of for the Timber Mountain caldera area, larger than 71/2-minute Lathrop Wells LATH ROP WELLS (Burchfiel, 1968) 15 I I I | Nevada Test Site, and date and are listed in the references at the end of the text. ‘ LAS VEGAS, 72 mi (115 km) 20 KILOMETRES 20 MILES | T CLARK—CO geologic data and other maps published by the U S. Geological Survey and adjacent region. Mapped areas are identified by author GENERAL GEOLOGIC RELATIONS 5 ledge, which is the result of a densely welded zone over- lying a less welded zone in a compound cooling unit. In many places the contact between compositional subunits is gradational within a foot or two and commonly nearly coincides with the contact between welding and crystallinity zones. The Paintbrush Tuff and younger silicic volcanic rocks of the Timber Mountain-Oasis Valley caldera complex are alkali-calcic in the Peacock (1931) classification and do not readily fit some of the commonly used classifications (Rittmann, 1952; Nockolds, 1954; O’Connor, 1965). In order to emphasize the differences in chemical composi- tion among these younger rocks of the Timber Mountain area (W. D. Quinlivan and P. W. Lipman, written commun., 1974), tuffs and lavas ranging from about 65 to 72 percent SiO2 are called quartz latites; rocks ranging from about 72 to 76 percent silica are called rhyolites or low-silica rhyolites to emphasize compositional range; and rocks in the range from about 76 to 78 percent silica are called high-silica rhyolites. The pre-Paintbrush volcanic rocks are normal calc-alkalic rhyolites, which are lower in total alkalis, particularly potassium, and slightly higher in lime than younger rocks of similar silica content. Peralkaline rocks (Shand, 1947, p. 229) are represented by the early effusives of Silent Canyon caldera and contain sodium in pyroxene and amphibole, as well as in feldspar. These rocks are considered briefly in this report. The terms “caldera” and “cauldron” are used in this paper to emphasize topographic expression. Caldera is used in the sense proposed by Williams (1941, p. 242) for a large, roughly circular or oval topographic depression in the central area of a volcano or volcanic complex. We also recognized filled and buried calderas. The Silent Canyon caldera no longer has the topographic form of a caldera, but it is filled with low-density volcanic rocks that make a large gravity anomaly having the shape of a caldera (D. L. Healey and C. H. Miller, written commun., 1967; Orkild and others, 1968, fig. 2). The term “cauldron” is used for a generally circular or oval, structural block which has sub- sided in the area of a volcanic or intrusive center but which has lost through subsequent erosion any topographic features of the former caldera. Our usage is virtually synonymous with “cauldron subsidence” as defined in the study of Glen Coe cauldron by Clough, Maufe, and Bailey (1909). The term “segment” is applied to a portion of a caldera or cauldron that does not have a circular form either because a part of it has been truncated by a later adjacent subsidence or because its original form has been severely modified by later structural movement. The term “volcanic center” is occasionally used herein for the entire volcanic edifice including the underlying magma chamber, the subsided cauldron, the surficial caldera, if preserved, and the extrusive vents that are preserved as dikes and vent breccias. The intimate associa- tion of central stocks, ring dikes, and cone sheets with cauldron structures has been emphasized many times in the literature, notably in review papers by Richey (1932), Richey, MacGregor, and Anderson (1961), Anderson (1936), Billings (1943), Buddington (1959, p. 680—685), Smith, Bailey, and Ross (1961), Branch (1966), Hamilton and Myers (1967, p. C6—C9), and Eggler (1968, p. 1555—1557). GENERAL GEOLOGIC RELATIONS The southwestern Nevada volcanic field (fig. 1) comprises upper Tertiary effusive rocks from the Timber Mountain-Oasis Valley caldera complex, rocks from per- alkaline calderas in the northern part, and several minor satellite lava piles ranging from calc-alkalic to calcic (Noble and others, 1965; Christiansen and others, 1976). The ash—flow tuffs and related rocks discussed in this report constitute most of a thick upper Miocene and lower Pliocene calc-alkalic and alkali-calcic volcanic sequence shown on the geologic map of the Timber Mountain caldera area (Byers and others, 1976) and include all the known silicic eruptive products of the Timber Mountain- Oasis Valley caldera complex. Peralkaline volcanic rocks of the Silent Canyon and Black Mountain calderas to the north (fig. 1) intertongue and postdate, respectively, the tuffs and lavas of this report and have been discussed else- where (Noble and others, 1963, 1965; Sargent and others, 1965; Christiansen and Noble, 1965; Noble, Sargent, and others, 1968; Christiansen and others, 1976). The newly named upper Miocene Redrock Valley and Crater Flat Tuffs (table 1) are calc—alkalic effusives of probably the oldest recognizable caldera within the TABLE 1.—Generalized stratigraphic nomenclature of tall formations and members as herein redefined and their indicated volcanic center [Minor intracaldera informal units are omitted] Age Volcanic Formation Member center or llnll Pliocene Timber Timber Tuffs of Crooked Canyonl Mountain Mountain Tuff of Buttonhook Washl Caldera Tuff1 Ammonia Tanks Member1 Rainier Mesa Member Claim Paintbrush Tuff of Pinyon Pass Canyon Tuff1 Tiva Canyon Member Cauldron Yucca Mountain Member Pah Canyon Member Topopah Spring Member Miocene Silent Stockade Canyon Wash Caldera Tuffl Sleeping Crater Flat Prow Pass Member2 Butte Tuff2 Bullfrog Member2 Caldera Redrock Valley Tuff2 lGeologic name redefined. 2New geologic name. TIMBER MOUNTAIN -OASIS VALLEY CALDERA COMPLEX, NEVADA EXTRACALDERA CLAIM CANYON AREAS CAULDRON SEGMENT Ammonia Tanks Member Tuff of Buttonhook Wash Tuff of A Pinyon Pass QTr Upper part Mambar ' Canyon _ Twa am Member Yucca MW“ ea“ Canvon Member Topopah Spring Member hTuff stockade was Lithic-rich tuffS Grouse Canyon Memtfier of Belted Range Tu Crater Flat Tuff Redrock Valley Tuff Pre-R edrock Valley rocks EXPLA NATION Post-Timber Mountain Tuff rock of late Tertiary and Quaternary age Paintbrush Tuff Genetically related lavas Bedded tuff undivided m Tuff breccia Timber Mountain Tuff Debris flows and breccia ‘ ' I ’ A » VV W/A Genetically related lavas Timber Mountain—Oasis Valley caldera complex. Only a defined, consists of several quartz-poor densely welded small part of this caldera is now exposed and is herein ash-flow tuff cooling units that are lithologically and called the Sleeping Butte caldera segment (fig. 1). The petrologically distinct from those of the formations above redefined Stockade Wash Tuff (table 1) is not a product of and below. Most of the units are exposed within the Claim the Timber Mountain caldera complex but is a late calc- Canyon cauldron segment (fig. 1), where they are very alkalic ash-flow sheet of the dominantly peralkaline Silent thick and intertongue with petrologically similar rhyolite Canyon caldera. The overlying Paintbrush Tuff, as herein lavas. The uppermost formation, the Timber Mountain ———— Partial cooling break T UFFS AND LAVAS RELATED TO SLEEPING BUTTE CALDERA TIMBER MOUNTAIN CALDERA RESURGENT DOME Tuff of Tuff of Crooked Canyon Buttonhook Drill Wash hole Tuff of UE18’ Crooked Canyon / QTr Tuffs and g lavas LIJ D Tuff and " h - 4 f,“— < r yolltes *_’ O of area 20 Z Lithic-rich E tuff and , i’ g lavas ’ \ 1‘ 1’1/— 0 L ‘ IQC" , 1: - , k , . 7(1) Pre- I E ~ \, I \‘ I» ’4 / ’\\_‘ \\ - _. \‘z: _ 5:1,“ |\\,, 1:, ’}\L\,J,\/: (1:3: "’7‘,” T’s/31‘”. / ’ 2,7,) Paintbrush ‘\ ’ ‘ \—,/, ll ‘\I~\"’ ”4 \/-_\‘\/r\\ r- “1““): ,w' ' \ “\\’.‘ _‘ k »-\\\,‘/\\/g \\,,/l/\‘\,, ,_’l, - \,\ \_\,'/\L/l/\/["\/‘I "NI 5 \l‘\/,\ 71:, .{I Il:l_\ll ”(’7 “at“: \,/\,L /\/\7 ,\, 47W" mgr. /__l: 33 Local tuffs and e _’ II’/../"—/‘—/“ e" ‘~ ‘ 1’1??? ’xlig‘lt’fif’: ‘7’» '57» ’ I T"? affix”) C’C‘b‘xa‘F—qal‘,‘ ./' r" CD lavas of peralkallne \,’ 'I —~/‘,- /w\- ,’/ \\,\ ,, l"._ ':'|~/\"I r’ __ /\1\:\\’\”\’Q Ark“, ’.’\I‘r\7‘/:\’,‘~:\/~l\; 5,2/‘1-919313573h" E composmon e a» , . /\(\’\ ~"‘/I,— ,\,l""IfQ’fi-l\I,“—\,\\/I—\F,‘_’\\/‘,‘[_\‘/Lvr_\l:i\’—\ ’— LV,‘,\\'\’er/l,\/I‘\/\/‘\\l\\i -l\”/\,- 11"!!\—\/\\’\;(\~H~. " ~/ . " (eff I:/\‘\5 5(1) tr (,5, Second calldera collapse assomated malnly ,w, P I . \,/-- , - g , _ ’ ...:~__\ -,\\.- [\x (L ...\ ,- 4, r _\‘.."’ \\ ~\ ‘I\’, (x ‘/ _\;/‘,\w1therupt:on\of”Amm'gnIa Tanks Member ‘1’ '7 ,9, 7 Grouse Canyon Member ’ ’ \‘z, \ \II A \’I " -—’\\‘\,’1~\ w — \—f‘l, - \\ ,rl— , / f/lI\ ’1’ /,_\ \ \' / \le\ \ .. Lax t‘7\/T\/'/‘/~l\\3'LENT_CANY,ON. CA1. EBA/FA L I" of Belted Range Tuff Tanks Member SILENT CANYON CALDERA Ammonia Tanks Member: Upper part Lower part Ammonia Upper part Tuff of Blacktop Buttes FIGURE 3 (left and above).—Generalized schematic diagram through southwestern Nevada volcanic field, showing geologic relations between ash-flow tuff sheets and related rocks discussed in this report. A few minor units omitted. Length of diagram about 50 km (30 mi). Tuff, contains common to abundant quartz phenocrysts and consists of two widespread coextensive members, the Rainier Mesa and the overlying Ammonia Tanks. Within parts of the Timber Mountain caldera these two members thicken, intertongue with rhyolite lavas, and are overlain by intracaldera informal units of the Timber Mountain Tuff, as herein redefined. Post-Timber Mountain tuffs and lavas are of relatively small volume and are largely confined to the Timber Mountain—Oasis Valley caldera complex. The stratigraphic and geologic relations of the ash-flow tuffs and intercalated rhyolite lavas in the south- western Nevada volcanic field are shown in figure 3. TUFFS AND LAVAS RELATED TO SLEEPING BUTTE CALDERA REDROCK VALLEY TUFF The Redrock Valley Tuff, the oldest ash-flow sheet related to the Timber Mountain—Oasis Valley caldera complex, is here named after exposures in Redrock Valley (fig. 4). The least altered and most complete exposure, however, is on the west side of Yucca Flat at hill 5504, 2.4 km (1.5 mi) east of the head of Redrock Valley (fig. 4) in the northern part of the Tippipah Spring quadrangle (Orkild, 1963) and is designated the type locality. The unit, as now known, consists of only one ash-flow tuff cooling unit and has been informally called “tuff of Redrock Valley” (Marvin and others, 1970, p. 2659, 2667). The Redrock Valley Tuff is the oldest known volcanic rock that TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA EXPLANATION Biotite-hornblende rhyolite west of Split Ridge Tuffs of Sleeping Butte Prow Pass Member (underlain by Bullfrog Member) Rliyolite lava flow between members Bullfrog Member Bullfrog Member (underlain by Red- rock Valley Tuff) Redrock Valley Tuff Zero isopach of ash-flow tuff Limit of information Drill holes used for control ~ Number mentioned in text O—r—O 10 15 I17°I5‘ ”7° 31? so | : l ‘3 STONEWALL "L MOUNT 7;: JACKSON MOUNTAIN Gold Point wan”, TOLICHA PEAK 41mm“ 379 NéOLo MOUNTAIN Is‘T‘ = >- 5 l— l , z , _ 8 t E u I <1 I: c _l 7. < \c 1: g :I 5 3L 0 1‘ 0v g » I 06‘ | 37 . I : / ’«7 I i // J GRAPEVINE \A PEAK /.,§ \ / ": ‘.m\“"”‘wu“‘" , /, : 2“ Amy," ”,Hfi \\ / g "a,,,.,,u,,w$ 4 ’44"? , :3 A6; ’0000 \\ 5‘ >“ S N""u..§ , 4v 5 ‘ >3“ ‘3 ' ,4?“ °o\;\p\\ g Type/47% A " a: locality Original i 3“";ng Bullfrog Mamba-r mine 15 20 MILES I | I 20 KILOMETRES FIGURE 4.-—Areal distribution of Redrock Valley Tuff, members of Crater Flat originated from the Timber Mountain—Oasis Valley caldera complex. The known extent of the Redrock Valley Tuff is shown in figure 4. At hill 5504 the unit is approximately 125 m TUFFS AND LAVAS RELATED TO SLEEPING BUTTE CALDERA 45' HS" 30‘ 6' “5° \‘SMOUNT HELEN ‘7» Q" "r'mo‘ ‘v 4» 60 ’9 COUNTY mum-1.,“ M E 54 NYE CQ_UNTY LIN60LN BLACK MOUNTAACIKN SILENT CANYON 5’32"," QUARTZ m- NTAIN MOUNTAIN CALDERA (/ CALDERA M E S A l L _‘ 37.0 ‘ V 15 éSLEEPING EE-é BUTTE 4 . CALDERAI 7%,?” SEGMENT TIMBER\MOUNTAIN , ., 5‘ CAL-pERA , , , », swarm; l Cw cAN‘oN Den R v9 5 A M8 9 : CALDERA ., SEGMENT I m} U) -37., (u . la, Kg u . .u, 3 RANGER MOUNTAINS —367 "m, 45 , ‘c u an» Type locality ‘ i , ' , , ' 3"?“ \ Crater Flat Tuff TE NW, , il,,.,..........m- \ . _._._.-._§,_ . . \ nun-mu“ m Lathrop Wells sum“ SPECTER \ W W,.um“‘\ in, H RANGE _ \ ‘ I \ 1 f, 3 \, Tuff, and other volcanic rocks that are probably related to Sleeping Butte caldera. (400 ft) thick. It reaches a maximum thickness of 418 m ranges in thickness from about 30 to 210 m (100—700 ft). (1,370 ft) where penetrated in test well Sand is about 300m The tuff is exposed at the surface in three widely (1,000 ft) thick in two wells west and south of Rainier scattered localities besides the one northwestof Yucca Flat. Mesa. Where penetrated in other drill holes (fig. 4), it The most significant of these is a small outcrop 5 km (3 mi) 10 TIMBER MOUNTAIN -OASIS VALLEY CALDERA COMPLEX, NEVADA south of Sleeping Butte near the intersection of Sleeping Butte and Oasis Valley caldera segments (fig. 4); this exposure is the only one known on the west side of the caldera complex. The northernmost exposure at the south end of the Belted Range was locally mapped in Quartet Dome quadrangle (fig. 2; Sargent and others, 1966) as “ash-flow tuff of Kawich Valley.” North of Frenchman Flat the nonwelded distal edge of the Redrock Valley Tuff is exposed in a few fault blocks along the border between Nye and Lincoln Counties. Little is known of the original total extent of the Redrock Valley Tuff, but if the unit was as widespread as suggested by the scattered outcrops and drill holes, as much as 360 km3 (90 mi?) may be present outside the Timber Mountain—Oasis Valley caldera complex, assuming an average thickness of 120 m (400 ft) over 3,000 km2 (1,200 miz). The volume of tuff buried beneath the Timber Mountain—Oasis Valley caldera complex would not increase this figure significantly in view of the inferred wide areal extent and the uncertainties involved. The Redrock Valley Tuff rests on Paleozoic rocks in the Eleana Range or on a thin, bedded unnamed tuff sequence immediately overlying the Fraction Tuff (Rogers and others, 1967; Marvin and others, 1970) in drill holes of the Rainier Mesa and Yucca Flat areas shown in figure 4. The tuff is overlain with local unconformity by bedded tuff, as much as 60 m (200 ft) thick, which is in turn overlain by the Bullfrog Member of the Crater Flat Tuff. The tuff at hill 5504 consists of a light-yellowish-gray nonwelded basal zone of shards, several metres thick, grading upward to a brown densely welded zone, which locally includes a thin dark~gray vitrophyre less than 5 In (15 ft) thick. The upper 100 m (330 ft) becomes less welded with increasing crystallinity and is light purplish gray, mottled with red—the red coloration giving the name to Redrock Valley, which in turn is here reapplied to the tuff. Small light-colored pumice lenticles 1-2 cm long in the devitrified zone contain abundant tiny (0.2—1.0 mm) biotite and hornblende phenocrysts that aid in identi- fication of the unit. Quartz is absent to very sparse and, where present, commonly cannot be seen with a hand lens. The uppermost part of the tuff locally reflects vapor-phase type crystallization (Smith, 1960b). The tuff has the zona- tion of a simple cooling unit. Petrographically, the rock can be distinguished from all other ash-flow tuffs at the Nevada Test Site by the criteria shown in figure 5. It typically contains hornblende, like the other calc-alkalic tuffs and lavas of the Sleeping Butte center. The lower part has a plagioclase to alkali feldspar phenocryst ratio of about 3:1 and very little or no quartz. The upper part, in contrast, has about equal feldspar phenocrysts and sparse quartz. Phenocrysts seemingly increase slightly in the upper part, despite less compac- tion and welding. Small euhedral quartz micropheno— crysts, commonly less than 0.5 mm and never exceeding 1.0 mm, are diagnostic of the unit; two thin sections examined from the lowermost part of the tuff were quartz-free, but all sections cut from the upper part contain quartz. The division between the upper and lower parts is based on petrography and has not been recognized in the field. The twofold division can be recognized in surface samples and also in drill core from the area northwest of Yucca Flat. The K-Ar age of sanidine in the local basal vitrophyre of the Redrock Valley Tuff from the type locality at hill 5504 is 15.7 m.y., which is in good agreement with K-Ar ages of the overlying and underlying units (Marvin and others, 1970). This age is close to the onset of volcanism at the Timber Mountain—Oasis Valley caldera complex and postdates nearly all the volcanic activity in the Northern Nellis Bombing and Gunnery Range to the north (Ekren and others, 1971 ). The thermal remanent magnetization is reverse polarity (G. D. Bath, written commun., 1968). Two chemical analyses of the tuff, one of the lower part and the other of the upper part, were made (W. D. Quinlivan and P. W. Lipman, written commun., 1974); the silica contents of these rocks are shown graphically in figure 5. CRATER FLAT TUFF AND INTERCALATED LAVA The Crater Flat Tuff is here named from exposures around the edges of Crater Flat (fig. 4), which is designated the type area of the formation. The tuff consists of a lower member, the Bullfrog Member (new), and an upper member, the Prow Pass Member (new), and local, unnamed intercalated breccia, bedded tuff, and ash-flow tuff. Both members and the intercalated breccia are well exposed at the southeast end of an unnamed hogback at south side of Crater Flat which is designated the type locality of the Crater Flat Tuff (fig. 4). This is the only known section in which both members are largely glassy and virtually unaltered and, therefore, amendable to K-Ar age dating. The thickness at the type locality totals about 190 m (620 ft), of which the Bullfrog Member is 130 m (430 ft), intercalated breccia, 10 m (30 ft), and the Prow Pass Member, 50 m (160 ft). The Crater Flat Tuff has been shown as “tuff of Crater Flat” on several U.S. Geological Survey 7%-minute quad- rangle maps (fig. 2) covering the southern part of the Nevada Test Site (Christiansen and Lipman, 1965; Orkild and O’Connor, 1970; Poole, Elston, and Carr, 1965; Sargent and others, 1970). On the east side of the Wahmonie volcanic center (Poole, Elston, and Carr, 1965; Poole, Carr, and Elston, 1965) a local middle ash-flow tuff, similar to the underlying Bullfrog Member, is included with the formation. A local intercalated rhyolite lava in the south wall of the Oasis Valley caldera segment (fig. 4) is not included in the Crater Flat Tuff, in accordance with prior practice by authors of reports on Nevada Test Site geology. This lava TUFFS AND LAVAS RELATED TO SLEEPING BUTrE CALDERA 11 is described here because it has about the same phenocryst ratios as the underlying Bullfrog Member—suggestive of a common origin. The lava has been altered in most places along with the underlying member and cannot con- veniently be mapped separately. It has been included with the Crater Flat Tuff on the map of the Timber Mountain caldera area (Byers and others, 1976). Duplicate K-Ar ages on biotites from vitrophyre in the lower part of the Bullfrog Member are 14.0 and 13.0 m.y. For reasons previously discussed (Marvin and others, 1970, p. 2667), the 14.0-my age is believed more reliable. The natural thermal remanent magnetism of both members of the Crater Flat Tuff is normal (G. D. Bath, written commun., 1965). BULLFROG MEMBER The Bullfrog Member, whose known distribution is shown in figure 4, is here named for the lower ash-flow sheet of the Crater Flat Tuff from exposures at Bullfrog Mountain (fig. 4). The type locality of the member is exposed on the mountain just north of the original Bull- frog mine, shown on the Bullfrog quadrangle (fig. 2) mapped by Cornwall and Kleinhampl (1964, pl. 4). These authors (1964, pl. 5, p. J10) described the member as cooling unit 2 of their Bullfrog Hills caldera and showed a composite thickness of 280 m (930 ft) in the Bullfrog Hills. The Bullfrog Member is 130 m (430 ft) thick in the type locality of the Crater Flat Tuff at the south end of Crater Flat and is locally as much as 180 m (600 ft) thick in the northern part. Exposures of the member either are incomplete or were not mapped separately east of Crater Flat; therefore, maximum thicknesses are not known, but are probably as much as 150 m (500 ft) in paleovalleys. Thicknesses in Yucca Flat and in the Rainier Mesa area to the northwest do not exceed 120 m (400 ft). As the distal edges are approached, such as near the southwest corner of Lincoln County, the thickness of the unit ranges from 15 to 30 m (50 to 100 ft). The unit is incompletely exposed south of Sleeping Butte (fig. 4), but probably the thickness does not exceed 60 m (200 ft). From drill hole UE20j to UE20f, just outside and inside the Silent Canyon caldera, respectively, the unit decreases slightly from 170 to 130 m (560-430 ft) in thickness (Orkild and others, 1969, section A—A’). The unit probably predates the formation of Silent Canyon caldera, for there is no great difference in thickness between the two holes inside and outside the caldera, and the unit has been downdropped 2,660 m (8,720 ft) to the east along later caldera faults (Orkild and others, 1969, section A—A’). The Bullfrog Member probably originally covered a larger area than that indicated on figure 4 and may have extended into the Death Valley area west of the Grapevine Mountains and southwesterly under the Amargosa Desert. The northwestern extent under Sarcobatus Flat is also unknown. If an original area of 6,500 km2 (2,500 mi2), twice that of the Redrock Valley Tuff, were covered, a maximum of 1,000 km?’ (250 mis) may have originally been present. The Bullfrog Member is commonly conformable on underlying bedded ash-fall tuff, which in turn separates the Bullfrog from the underlying Redrock 'Valley Tuff. In the Bullfrog Hills and on the eastern flank of the Grape- vine Mountains (fig. 4), the member rests concordantly on bedded tuffaceous sandstone that overlies the Oligocene Titus Canyon Formation (Stock and Bode, 1935). At the type locality at Bullfrog Mountain, the member rests on cooling unit 1 of Cornwall and Kleinhampl (1964). To the east along the southern border of its known extent, the member rests on the rocks of Pavits Springs (Poole, Elston, and Carr, 1965; Poole, Carr, and Elston, 1965). The member is overlain at the type locality of the Crater Flat Tuff by a monolithologic breccia composed of fragments of the member. It is overlain by the lava flow that is petro- graphically similar to the member in the south wall of the Oasis Valley caldera segment (fig. 4). Elsewhere the Bull- frog Member is overlain by bedded tuff of the Crater Flat, which, in turn, is overlain by the Prow Pass Member. Locally on the east flank of the Wahmonie volcanic center, a middle ash-flow petrographically similar to the under- lying Bullfrog Member intervenes between the members. In drill hole UE20j on Pahute Mesa, the Bullfrog Member is separated from the overlying peralkaline Tub Spring Member of the Belted Range Tuff (Noble, Sargent, and others, 1968; Orkild and others, 1969) by 51.8 m (170 ft) of bedded ash-fall tuff. In drill hole UE20f, lavas of per- alkaline composition occur below the member, and the tuff of Tolicha Peak occurs above the member. In exposures just north of Redrock Valley and in drill cores in the Rainier Mesa area the Bullfrog Member occurs as a local nonwelded tuff at or near the top of tunnel bed 1 of the Indian Trail Formation (D. L. Hoover, oral commun., 1974). In lithology and petrography the calc-alkalic Bullfrog Member of the Crater Flat Tuff is generally similar to the Redrock Valley Tuff but differs in containing significant euhedral quartz phenocrysts (fig. 5) as long as 2.5 mm. The member has a feldspar compositional zonation with plagioclase significantly in excess of alkali feldspar in the lower part and feldspars subequal in the upper part. The common biotite flakes in small (1—3 cm) white pumice lenticles are an identifying criterion which, along with readily visible quartz phenocrysts, serve to distinguish the member from the Redrock Valley Tuff. The ash-flow sheet is brown and glassy in the lower part at the type locality of the Crater Flat Formation at the south side of Crater Flat, but elsewhere the Bullfrog Member is light yellowish gray, slightly mottled with yellow and pale reddish brown and microcrystalline from base to top. The sheet, where the top l2 TIMBER MOUNTAIN -OASIS VALLEY CALDERA COMPLEX, NEVADA MAJOR STRATIGRAPHIC UNIT < SUBUNIT PHENOCRYSTS FELSIC PHENOCRYSTS AS PERCENTAGE VOLCANIC Number of thin sections (N) in parentheses AS PERCENTAGE OF TOTAL PHENOCRYSTS CENTER OF TOTAL F(OCK Quartz m Alkali feldspar m Plagioclase ‘1 fl Tuff of Blacktop Buttes (Hinricks and others, 1967) (3) Stockade Wash(.7) Tuff ' a ‘ ' ' 1 (Stockade Wash(?) Member - , q)f Sargent and, others, 196;?) (2) . I I‘ I I I I I < . . Up'per lavas . i . i . , fl tr (20 thin sections, 3 5 drill holes‘) 2| Tuffs and rhyolites Upper part 0 Of Area ?0 . (6 sections, - g (2) (Sargent 1969, Orklld Lower 3dri|l holes‘) >. and others, 1969) lavas 2 Lower m < 0 part (7) - E 7 ? ? .7 .7 l l i l _l Stockade Wash Tuff — a) of this paper (7) I ) I I Tuffs and rhyolites of Area I 1 fl 20—Lithic-rich ash-flow tuffs — (Orkild and others, 1969) (9) : H l l l l *4 Peralkaline sequence of soda rhyolite lava, Grouse Canyon Member of Belted Range Tuff, and tuffs and lava and tuff of Dead horse Flat (Noble, Sargent, and others, 1968, Orkild and others, 1968, 1969) l l l l l l fi‘ ,1 Upper I I Biotite-hornblende lava (5) V////////////l |'\. E ) rhyolite west : “l' of Split Ridge Lower E “' lava (4) I 8 ? ? .7 l 4 < E Tuffs of Sleeping Butte (9) — 3 g .7 .7 ? l A 5 ”DJ Prow Pass >- 3:1 Member (10) LU _, O " E 3‘ l— Lava flow (5) a) a Crater Flat Tuff g (D and intercalated 4 9 Z lava Upper 2 E part (12) <7: 3 Bullfrog E m Member D Lower E K part (10) W l *l l l l l *l a: 3‘3. Upper part (5) - W E Redrock < ’- Valley Tuff 3 Lower art (5) - ' § p I m I | l l | | ‘ Thin-section modes of at least 2 cores from each drill hole 10 20 0 20 40 60 80 100 were averaged to reduce scatter, because too few phenocrysts PERCENT were counted per thin section. FIGURE 5.——-Modal and silica ranges of Redrock Valley, Crater Flat, and Stockade Wash Tuffs and petrologically related lavas in general MAFIC PHENOCRYSTS AS PERCENTAGE OF TOTAL PHENOCRYSTS - Total mafics m Biotite I I w w @- ®' I l I l l r Peralkaline sequence of soda and tuffs and lava and tuff of Dead her I l l order of stratigraphic succession. Units with queried boundaries may b _l_ #I Hornblende Pyroxene TUFFS AND LAVAS RELATED TO SLEEPING BUTTE CALDERA 13 SILICA PERCENTAGE RANGE Number of analyses (N) in parentheses (H20 and CO2 free) PERCENT I | I r l | I (Hornblende present in heavy mineral suite; see table 2 (0) not present in thin section) I I 9I fi‘ I I I I J. (Hornblende present in heavy mineral suite; (0) not present in thin section) I I I I I I ' I A Glassy (2) (4) (Hornblende present In heavy Glassy mineral suite; see table 2 not present In thin section) (1) (2) (2) I I 4I If I I I I 4I (1) | I 4I l I I I I I I I | I I I I I | (O) I I 4‘. I7‘ I I I I 4I rhyolite lava, Grouse Canyon Member of Belted Range Tuff, se Flat (Noble, Sargent, and others, 1968, Orkild and others, 1968,1969) I I I I | I I I I I I If I I I I I (0) (1) I I a .fi I I I I a Clinopyroxene in uppermost unit )2) I I a. P I I I I I 3_\ Ill/A Orthopyroxene I2) Glassy (1) (1) (2) Glassy l (1) (1) I I I I I I I I J. (1) (1) | I I | | | 0 2.5 5.0 7.5 70 72 74 76 78 80 e in reverse stratigraphic order from that shown or may intertongue. l4 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA is not eroded, has an upper part that contains crystals of the vapor-phase zone in pumice. In the Bullfrog Hills, the member is silicified and locally mineralized with sulfides and gold (Cornwall and Kleinhampl, 1964, p. J22) and is moderately welded. In the Yucca Flat area, northern Frenchman Flat, and in drill hole UE20j (fig. 4), the member is nonwelded to slightly welded and the original glass is now altered to zeolites. In drill hole UE20f (fig. 4) where the tuff member was penetrated between depths of 3,500 and 3,650 m (11,500 and 12,000 ft), the biotite has been completely replaced by chlorite, and the feldspars have been partly replaced by sericite. All these differing facies and degrees of alteration of the ash-flow sheet might cause great difficulty in correlation were it not for the lateral persistence of a unique phenocryst assortment (fig. 5), general strati- graphic position, and certain lithologic features, such as small, flattened biotite-bearing pumice lenticles not destroyed by alteration. The petrochemistry (W. D. Quinlivan and P. W. Lipman, written commun., 1974) of the Bullfrog Member is similar to that of the underlying Redrock Valley Tuff (fig. 5) in that the lower part of the ash-flow sheet is slightly more mafic and less silicic than the upper part. Although sampling has been limited, field observations suggest no sharp break between the upper and lower parts; the subdivision is arbitrary and based on petrography. INTERCALATED LAVA FLOW A brownish-gray fluidal lava flow overlying the Bull- frog Member but not part of the Crater Flat Tuff is exposed in fault blocks in the south wall of the Oasis Valley caldera segment (fig. 4). The lava pinches out a short distance south of the wall, for exposures of both members of the Crater Flat Tuff at Prow Pass and at the north end of Crater Flat include only intervening bedded tuff. A down- faulted portion of the lava is doubtless buried beneath younger volcanic rocks in the Oasis Valley caldera segment. The exposed part of the lava flow does not exceed 60 m (200 ft) in thickness. The Prow Pass Member rests directly on the lava, but as much as 15 m (50 ft) of bedded tuff separates the lava from the underlying Bullfrog Member. The lava superficially resembles the underlying Bull frog Member, but its fluidal flow banding identifies it as a lava flow. The lower part locally includes a vitrophyre, whereas the brownish-gray uppermost part contains spheroidal white pumice with abundant fine biotite phenocrysts resembling the tuff. The thin-section modes of three specimens of the lava are well within the range of 22 modes of the Bullfrog Member (fig. 5). Two chemical analyses of the lava (W. D. Quinlivan and P. W. Lipman, written commun., 1974) indicate that the devitrified specimen is several percent higher in Si02 content than is the specimen of the basal vitrophyre (fig. 5). PROW PASS MEMBER The Prow Pass Member of the Crater Flat Tuff is here named after Prow Pass, the type locality, at the north end of Yucca Mountain (fig. 4). At the type locality at Prow Pass, 15.2 m (50 ft) of devitrified welded tuff constitutes the upper cooling unit of the Crater Flat. The member is slightly thicker and more densely welded in the northern part of the Crater Flat area than it is in the southern part. At the type locality of the Crater Flat Tuff (fig. 4), the member is 50 m (160 ft) thick, but is partly glassy and only slightly welded. Near the eastern and southeastern edges of its extent, the Prow Pass Member is about 15 m (50 ft) thick and, like the Bullfrog Member, it is nonwelded and zeolitized. The member has not been recognized and probably is not present outside the Timber Mountain—Oasis Valley caldera complex north of the area shown in figure 4. The estimated extracaldera complex volume of the unit is only about 30 km3 (8 mil), but the member may be several times thicker within the complex where an equal or greater volume could be present. The Prow Pass Member underlies all the construc- tional lavas of the calcic Wahmonie-Salyer volcanic center (Poole, Elston, and Carr, 1965; Poole, Carr, and Elston, 1965; Noble and others, 1965) and the rhyolite lavas of the Calico Hills (Christiansen and Lipman, 1965). The age relations to the post-Crater Flat tuffs on the west rim of Sleeping Butte caldera segment are unknown from strati- graphic relations but the member may be older, on the basis of an inferred upward increase of phenocrystic alkali feldspar (fig. 5), typical of the later volcanic suites. The age relations to the peralkaline Tub Spring Member of the Belted Range Tuff (Sargent and others, 1965) are uncertain, but the member underlies a phenocryst-poor green peralkaline ash-fall tuff that underlies the Grouse Canyon Member of the Belted Range Tuff (fig. 5). The calc-alkalic Prow Pass Member has some general similarities in lithology to the Bullfrog Member, but it is thinner, generally less welded, and practically devoid of biotite. The unit is generally light gray, grayish pink, or moderate orange pink where devitrified, but is light brown where glassy at the type locality of the Crater Flat. Petrographically, it is a unique unit of all the silicic tuffs of the region in that it contains orthopyroxene as the dominant mafic phenocryst (fig. 5) and, in the extreme, almost lacy resorption of the quartz phenocrysts. No other ash-flow sheet in the region could be mistaken for this unit. However, orthopyroxene is preserved only in the glassy upper and lower parts of the cooling unit at the type locality of the Crater Flat Tuff; elsewhere the unit is devitrified and only smudgy opaque oxide pseudomorphs K remain. The silica contents of two devitrified specimens ‘ " are shown graphically in figure 5. TUFFS OF SLEEPING BUTTE AND THEIR RELATION TO CALDERA WALL ’ a Rocks under this heading comprise a sequence of three _ .(similar ash-flow tuff cooling units occupying the west rim of the Sleeping Butte caldera in the vicinity of Sleeping ;’ Butte (fig. 4). The caldera rim is delineated for several ,. miles by the axis of an easterly dipping monocline in the tuff. The tuffs of Sleeping Butte are slightly more than 300 ’ m (1,000 ft) thick at Sleeping Butte, where the upper .cooling unit is well exposed. Along the west rim of the - caldera the monocline exposes 150—240 m (500—800 ft) of E“ the tuffs. The outcrop area shown in figure 4 is delimited hby the onlap of younger rocks. 1 The monocline dips easterly under the peralkaline ‘ Grouse Canyon Member of the Belted Range Tuff and ,younger rocks, which fill the Sleeping Butte caldera segment. Near the southern end of the exposure, relatively unfractured but locally vertically dipping tuffs of Sleeping LButte probably onlap the highly fractured caldera wall consisting of the Redrock Valley Tuff and the Bullfrog ,Member of the Crater Flat Tuff, although the contact is not exposed. The local vertical attitude in the onlapping tuffs of Sleeping Butte strikes north, parallel to the inferred caldera wall, and is typical of the most easterly “exposure. The steep initial dip of the tuffs probably occurred as they were compacted and plastered against the Sleeping Butte caldera wall. The inferred younger age relation of the tuffs of ‘Sleeping Butte to the Prow Pass Member of the Crater Flat VTuff is suggested partly by the close stratigraphic succession of the units of the Crater Flat Tuff, and partly by the increase of alkali feldspar upward. The unfractured uature of the tuffs with respect to the highly fractured Bullfrog Member also indicates their younger age, ‘probably as postcaldera tuffs draped over a wall composed ‘of the Bullfrog Member and older rocks. The three ash-flow cooling units of the tuffs include a lower nonwelded to partly welded, phenocryst-poor shard tuff (not shown in figure 5), a middle purple welded tuff . that contains fluidal flow banding where it dips inward ’(easterly) into the caldera, and an upper tuff, mainly >exposed around Sleeping Butte. All the tuffs of Sleeping Butte are mostly light purplish gray, light brown, and fnicrocrystalline; basal vitrophyres are practically non- existent and the rocks are commonly mildly silicified and sericitized to the extent that former plagioclase pheno- ‘crysts are completely gone. Xenoliths are common and consist mainly of altered silicic volcanic rock and less ‘ common pilotaxitic intermediate and mafic lava. The fluidal flow banding of the middle unit, locally vertical, gccurs only at its eastern margin where it dips into the inferred Sleeping Butte caldera. ‘ TUFFS AND LAVAS RELATED TO SLEEPING BUTTE CALDERA 15 The thin-section modes of the middle and upper tuffs of Sleeping Butte are pooled graphically in figure 5. The lower shard-rich tuff is phenocryst-poor (2 to 3 percent phenocrysts) somewhat altered, and therefore not pooled with the overlying units. Its phenocryst percentages, based on only two sections, are roughly similar to those of the middle and upper unit; the chemical analysis (W. D. Quinlivan and P. W. Lipman, written commun., 1974) indicates a more silicic rock at 75.1 percent SiOz (water- free) than the two phenocryst-rich specimens of the middle and upper units (fig. 5). The middle and upper units are similar in modal petrography except for sparse pigeonitic(?) clinopyroxene (low 2V, slightly pleochroic in pale green and yellow) in addition to hornblende in the upper unit. This is the oldest unit containing significant clinopyroxene phenocrysts. The presence of pigeonitic (low Ca) clinopyroxene in the highest stratigraphic unit of the tuffs of Sleeping Butte is analogous to the occurrence of orthopyroxene (low Ca) in the Prow Pass Member, the uppermost cooling unit of the Crater Flat Tuff. High potassium (W. D. Quinlivan and P. W. Lipman, written commun., 1974) in all three tuffs is due partly to very minor sericite replacing plagioclase, but the dominantly alkali feldspar phenocrysts indicate that most of the potassium is primary. Therefore, these youngest rocks of the Sleeping Butte caldera center approach the alkali-calcic composition of the overlying Paintbrush and Timber Mountains Tuffs and may be products of a late alkali enrichment in the upper part of the Sleeping Butte magma chamber. TUFI" 0F TOLICHA PEAK The tuff of Tolicha Peak (Noble and Christiansen, 1968; Rogers and others, 1968) is a local phenocryst—poor densely welded shard tuff, about 60 m (200 ft) thick, that occurs above the Bullfrog Member of the Crater Flat Tuff in drill hole UE20f (fig. 4; Orkild and others, 1969, cross section A—A’). The unit crops out in the north half of the Black Mountain 15-minute quadrangle (fig. 2) and in the southwest quarter of the quadrangle in the vicinity of Quartz Mountain and Tolicha Peak (fig. 1); it was penetrated in a drill hole in eastern Yucca Flat. The unit is distinctive lithologically and also petrographically in that it contains sparse quartz and plagioclase dominant over alkali feldspar. The tuff may be related to the Sleeping Butte caldera (Christiansen and others, 1976); however, E. B. Ekren (oral commun., 1974; Ekren and others, 1971, p. 60) has evidence suggesting a source from the north in the Mount Helen area (fig. 4). BIOTITE-HORNBLENDE RHYOLITE LAVAS WEST OF SPLIT RIDGE Two calc-alkalic rhyolite lava flows and underlying petrographically similar ash-flow and ash-fall tuffs inter- tongue with peralkaline members of the Belted Range 16 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA Tuff and related lavas (Sargent and others, 1965; Noble, Sargent, and others, 1968) in the triangular salient between Silent Canyon caldera, Timber Mountain caldera, and the peralkaline lava pile of Split Ridge (fig. 4). These rocks occupy approximately the same strati- graphic position as the tuffs of Sleeping Butte; they also contain hornblende but are petrographically more mafic like the Redrock Valley Tuff (fig. 5). They are therefore generally similar to other calc-alkalic rocks grouped under Sleeping Butte caldera and are included here for convenience. Magnetic polarity determinations by G. D. Bath (written commun., 1965) indicate a reverse orienta- tion in these rocks in contrast to normal remanent magnetization of the enclosing peralkaline rocks. The two lavas and underlying tuffs were called quartz latite flows and quartz latite tuff on the geologic map of the Ammonia Tanks quadrangle (fig. 2; Hinrichs and others, 1967); however, the one analyzed vitrophyre of the lower, more mafic, plagioclase-rich lava is chemically rather silicic at 73.9 percent SiO2 (fig. 5), and total alkalis are 7 percent (W. D. Quinlivan and P. W. Lipman, written commun., 1974); hence these rocks are called rhyolite in this report and on the geologic map of the Timber Moun- tain area (Byers and others, 1976). Hinrichs, Krushensky, and Luft (1967) show a maximum thickness of 364 m (1,195 ft), comprising 114 m (375 ft) of underlying tuff, 174 m (570 ft) of lower lava, and 76 m (250 ft) of upper lava. The lower flow is a thick lenticular exposure in a short canyon tributary to Timber Mountain caldera; the over- lying and underlying units are more stratiform and do not pinch out within the exposed area. The biotite-hornblende tuff-lava sequence west of Split Ridge is transected by green, peralkaline feeder dikes and vent breccias of the rhyolite complex of Split Ridge and is also overlain by greenish-gray peralkaline bedded tuff— breccia under the rhyolite lava of Split Ridge, which in turn underlies the Grouse Canyon, the upper member of the Belted Range Tuff. The sequence overlies greenish- yellow peralkaline bedded tuff that in turn overlies the Tub Spring, the lower member of the Belted Range Tuff. The basal tuffs include two, possibly three, nonwelded ash-flow tuffs as much as 23 m (75 ft) thick, and inter- calated bedded tuff. They are yellowish gray, weather lighter gray, and are typical of calc-alkalic zeolitized tuff stratigraphically between the Grouse Canyon Member and the Crater Flat Tuff in the area between Yucca Flat and Timber Mountain caldera (fig. 4). The phenocryst mineralogy of these basal tuffs is generally similar to that of the overlying biotite-hornblende rhyolite lavas. The lavas superficially resemble other calc-alkalic rhyolite lavas associated with the Timber Mountain-Oasis Valley caldera complex. Both lavas have a glassy breccia envelope, a dark basal vitrophyre, and a purplish-gray fluidal flow-banded middle part with conspicuous small phenocrysts,less than 2 mm, of feldspar, biotite, and horn- blende (fig. 5). The occurrence of a few phenocrysts of sphene in every thin section of these lavas contrasts with the sporadic occurrence of sphene in other volcanic rocks of the Sleeping Butte caldera. PERALKALINE ROCKS OF SILENT CANYON CALDERA A complete description of the peralkaline rocks related to the Silent Canyon caldera center is beyond the scope of this paper; these rocks are described elsewhere (Sargent and others, 1965; Noble, Sargent, and others, 1968; Orkild and others, 1968; Orkild and others, 1969; Sargent, 1969; Noble and others, 1969; Noble, 1970; and Christiansen and others, 1976). These unique rocks, however, provide an extremely helpful datum for stratigraphic assignments of the rocks of the Timber Mountain—Oasis Valley caldera complex and, therefore, are mentioned briefly here. The comenditic Tub Spring and Grouse Canyon Members constitute, respectively, the lower and upper ash- flow sheets of the Belted Range Tuff (Sargent and others, 1965). Many peralkaline lava flows, including the rhyolite lava of Split Ridge, are genetically related to these sheets and occur within and around the Silent Canyon caldera center (Noble, Sargent, and others, 1968). The Tub Spring Member overlies the Crater Flat Tuff in drill hole UE20j (fig. 4), and the exposed uppermost part of the Tub Spring underlies the calc-alkalic biotite-hornblende rhyolitic rocks west of Split Ridge. The Grouse Canyon Member is much more widespread, covering about 8,000 km2 (3,000 miz) (Noble, Sargent, and others, 1968, p. 67), and represents the peak of peralkaline igneous activity. The maximum extracaldera thickness is about 90 m (300 ft), and the original volume of the member is about 200 km3 (50 mi”) The Grouse Canyon and related peralkaline rocks (fig. 5) separate the underlying calc-alkalic rocks of the Sleeping Butte caldera from overlying similar calc- alkalic effusives that were erupted from the Area 20 center late in the development of Silent Canyon caldera. STOCKADE WASH TUF F AND RELATED CALC-ALKALIC ROCKS ASSOCIATED WITH SILENT CANYON CALDERA The Stockade Wash Member was first named by Hinrichs and Orkild (1961, p. D98) as a unit of the Oak Spring Formation from the thick exposure in Stockade Wash south of Rainier Mesa (fig. 6). The member was mapped in Rainier Mesa (Gibbons and others, 1963) and Tippipah Spring quadrangles (fig. 2; Orkild, 1963). Poole and McKeown (1962, p. C61) raised the rank of the Oak Spring to group status and divided it into the Indian Trail and Piapi Canyon Formations. Their Piapi Canyon included, in ascending order, the Survey Butte, Stockade Wash, Topopah Spring, Tiva Canyon, and Rainier Mesa STOCKADE WASH TUFF AND ROCKS ASSOCIATED WITH SILENT CANYON CALDERA 17 1 1 6° ’ a X ': l ,, Tuff of Btlacktopi ;\ S EXPLANATION 9‘: I, o ”W”: Z; . Locality ~ Approximate thickness known 0 Drill hole I/IuII ——0—_7— lsopachs, in metres —Queried where control is poor \\\\'lllll\‘\lé u, [UH/”HUM IHIIIIIIu III‘HH'“ 9‘ Blacktop/ ¢ \le Buttes ”“‘ a. - m we :, .3 II tockade VVash '\ TIMBER MOUNTAIN CALDERA —-——-—-"rrn ‘ NE'VAD_A TEST VSITE‘ / «M ”Hum, I I NYE COUNTY fl 0eTim er V 5 A?“ 5‘ veM oun ain \‘t 3‘ Q Resufient ‘ 4 us“ 0 V D LINCOLN COUNTV \II Imuu“ 5) ’0’,“ \“’ wow.“ s. $(OCLAIM ”i A? \ 37u_ CANYON 0 411533 CAULDRON "SEGMENT I: t ’/ \\\|l( \\ \ \\\ L c \ \ \‘I III\ 1 v 1,, I ’IIIIIIIm, E 9 E: ’2 : is 2 659 “""t “3 :3 ‘2; ‘3‘ E: E ’3 Boundaiy of TimbepMourgtain- 2W3 f: g 5:0 OasisValley calderatcorhplex 2‘7 2 ’I(/ - ’1 I carp , I,,,H I; g “Isms o: E _ " I Z WAHMONlE-SALYER e i I CO e E 2 <1 9 - 194 z 1, a VOLCANIC CENTER Q? éi A ) E 3 <<’ c9 5 i JACKASS <" 1 ' ’\ E FLATS | 5 \u F3 N": 0‘“ \n‘ | i i \\\ O 5 10 15 MILES i | l 4i If I i i O 5 10 15 KILOMETRES FIGURE 6 —Isopachs of Stockade Wash Tuff, Stockade Wash(?) Tuff, and tuff of Blacktop Buttes in relation to Silent Canyon caldera Members Later Orkild (1965) abandoned the Oak Spring division of these formations, which included the Stockade Wash Member at the base of the Paintbrush, is shown in Group and raised the Piapi Canyon to group, consisting of the Paintbrush and Timber Mountain Tuffs. His sub- the left column of figure 7. l8 TIMBER MOUNTAIN ~OASIS VALLEY CALDERA COMPLEX, NEVADA Topopah Spring Member Stockade Wash Member Local informal units BELTED BELTED RANGE Grouse Canyon Member RANGE TUFF TUFF T opopah Spring Member (peralkline composition) ORKILD (1965)l THIS REPORT NEVADA TEST SITE _ NEVADA TEST SITE AND VICINITY CLAIM CANYON CAULDRON WEST AND NORTH AREAS OUTSIDE KNOWN CALDERA SEGMENT AND PAINTBRUSH AREAS INCLUDING YUCCA FLAT CANYON TUFF TYPE SECTION 2' Ammonia Tanks Member4 Z| LL Ammonia Upper part 3 t T ff fT I“ 8 5' Tanks 0 u o ransvaa Member 5 )2 Tuff of Cat Canyon4 5 E Lower part I E i (1 I LLl - LIJ < m < s r 2 r _ Rainier Mesa Member I: l— / Tuff of Pinyon Pass Tuff of Chocolate Tiva Canyon Member LL Tiva Canyon Member & Mou taln la. la Tiva Canyon Member )— . v 3 U) ,_ ,, .13 l- 3 l: I I no 5’ i3 5 _ D: E _ V , . E :5 ., . g ////////////////4 o. . Topopah Spring Member Stock d WashT ff ‘Geometry of Orkild's (1965) diagram slightly altered in an attempt to make it fit new information presented here and still retain the stratigraphic relations in his diagram. 2Noble, Krushensky, McKay, and Ege (1967). 3Hinrichs, Krushensky, and Luft (1967). The Pah Canyon Member shown in their section A-A' is a local unit of the Paintbrush Tuff that seemingly grades upward into the overlying pyroxene-bearing rhyolite lava. See text. ‘Ammonia Tanks Member as mapped by Orkild, Sargent, and Snyder(1969)in most places around the rim of Timber Mountain caldera is the upper part of the Ammonia Tanks of this report and the main part as mapped by Noble, Krushensky, McKay, and Ege (11967). The tuff of Transvaal of 0rkildr(1965) is equivalent to all the lower part of the Ammonia Tanks Member as mapped by Noble. Krushensky, McKay, and Ege (1967). The intracaldera ruff of Cat Canyon as mapped by Carr and Ctuinlivan (1966) is equivalent to the entire Ammonia Tanks of this report and is a single compound cooling unit. See text. 5The former Stockade Wash Member of the Paintbrush Tuff was placed above the tufts and rhyolites of Area 20 by Orkild, Sargent, and Snyder (1969) but is now known to be stratigraphically equivalent to some part, possibly near the middle, of the tuffs and rhyolites of Area 20. FIGURE 7 (above and right).—Stratigraphic relations and revisions of Stockade Wash, Paintbrush, and Timber Mountain Tufts, Timber Mountain caldera and vicinity. Dark-shaded units intertongue with these named tuffs but are not part of them. Diagonal ruling indicates unit is not present; light shading, bedded tuff. The Stockade Wash Member is herein removed as the basal unit of the Paintbrush Tuff and raised to forma- tional rank (second column, fig. 7) because it is litho- logically and petrologically unlike the ash-flow tuff units of the Paintbrush and appears more closely related to the calc-alkalic tuffs. and rhyolite lavas of Area 20 in Silent Canyon caldera. The areal distribution of the Stockade Wash Tuff and closely related units is shown in figure 6. The tuff reaches its maximum thickness of slightly more than 120 m (400 ft) along the west side of Rainier Mesa and also in the low hills north of Frenchman Flat (fig. 6) where Hinrichs and McKay (1965) mapped the unit as the Stockade Wash(?) Member. Subsequent field and petro- graphic work have established the identity of their unit in the northern Frenchman Flat area with that at the type section. The volume of the Stockade Wash Tuff is STOCKADE WASH TUFF AND ROCKS ASSOCIATED WITH SILENT CANYON CALDERA 19 THIS REPORT ORKILD, SARGENT AND SNYDER (1969) AND OTHERS 23 TIMBER MOUNTAIN CALDERA TYPE AREA AND OASIS VALLEY CALDERA SEGMENT SILENT CANYON CALDERA (ONLY) PAHUTE MESA COMPOSITE (INCLUDES SILENT CANYON CALDERA AN ADJACENT AREAS) Tuff of Rhyolite lavas of Cutoff Road BeattY Wash Tuffs of Rhyolite lavas of FIeur-de-lis Ranch West Cat Canyon Tuffs of Crooked Canyon and Buttonhook Wash _ “I | 2 U er art I U t ' ‘ 2" D t Ammonia Tanks pp p 2 LL Ammonia Tanks é u. A$mima Main part 0 3 Member D H- Member 3 LL an s E I- Lower part 0 D Lower part 0 3 Member Lower part2" cc 2 E l— 2 ,_ L|.l - II 2 [I 2 m < LLI — I.I.I _ g l- in '<_( m E F ITuffc of " a on an on I— Rainier Mesa Member E Rainier Mesa Member E c y I— I— Rainier Mesa Member Quartz-rich rhyolite lava % _ Quartz-bearing rhyolite lava E Biotite rhyolite lava -‘6 g o- ”:5 u. Tiva Canyon Tiva Canyon g u. Member Member 5 :> E " 3’ I LL ‘- LL 0 g 3 2 n: *- 0 m (H . . l— a : Pre-Timber Mountain 2 3 -' volcanic rocks 2 Local subsurface tuff n: D. CD (not exposed or penetrated l— in drill hole) E < Topopah Spring Member Topopah Spring Member Sto k d h M 5 Tufts and rhyolite lavas of Area 20 c a e Was ember (Stockade Wash Tuff not recognizable _ in Silent Canyon Caldera) Tuffs and rhyolites of Area 20 Lava and tuff of Deadhorse Flat (peralkaline composition) Lava and tuff of Deadhorse Flat BRELJED Grouse Canyon Member BELTED TAU FLEE (peralkaline composition) RTAUIiiE Grouse Canyon Member probably between 20 and 40 km3 (5 and 10 mi3), depending on its largely unknown thickness and extent under Yucca and Frenchman Flats. The areal distribution pattern of the Stockade Wash, trending generally southeast from Silent Canyon caldera, is similar to that of the peralkaline ash fall of the Belted Range Tuff (Noble and others, 1968, fig. 2), which came from Silent Canyon caldera, and is unlike the distribution patterns of members of the Paintbrush Tuff (figs. 8, 9) which originated in the southern or southwestern part of the Timber Mountain-Oasis Valley caldera complex. The closest exposure of the Stockade Wash Tuff to the type section of the Paintbrush is about 15 miles. The Stockade Wash Tuff, where present in the Rainier Mesa and Yucca Flat areas, rests on zeolitic tuffs of the tuffs and rhyolite lavas of Area 20 but has not been recognized in drill holes inside Silent Canyon caldera (figs. 6 and 7). The lithic-rich ash-flow tuff in the lower part of the tuffs and rhyolite lavas of Area 20 within Silent Canyon caldera (Orkild and others, 1969) has also been recognized in several drill holes in Yucca Flat where it reaches a maximum thickness of 64 m (210 ft). The lithic-rich ash- flow tuff is separated from the overlying Stockade Wash by less than 10 m (35 ft) of zeolitic lithic-rich bedded tuff and rests on the Grouse Canyon Member of the Belted Range Tuff. The Stockade Wash Tuff is overlain by quartz-free friable glassy bedded tuff of the Paintbrush Tuff (Orkild, 1965), although, locally, a few metres (10 ft) of zeolitic lithic~rich bedded tuff occurs between the Paintbrush and the Stockade Wash. The best correlation of the Stockade Wash, based on petrography, is with the lower part of the lower rhyolite lava of the tuffs and rhyolites of Area 20 (see fig. 5), analogous to the relation of the Bullfrog Member of the Crater Flat Tuff with the petrographically similar lava (fig. 5). The StOckade Wash Tuff was probably erupted from the same magma as the lower lavas just before, in between, or just after the sequence of several lower rhyolite lavas of the tuffs and rhyolites of Area 20, 20 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA which underlie the Paintbrush Tuff (fig. 7; Orkild and others, 1969). Therefore, where zeolitic quartz-bearing lithic—rich tuffs are present beneath or just above the Stockade Wash, they are assigned to the tuffs and rhyolite lavas of Area 20, rather than to the Paintbrush Tuff (fig. 7). The Stockade Wash Tuff is a simple cooling unit of massive nonwelded to partly welded ash-flow tuff having 4—9 percent of small phenocrysts generally less than 1 mm in length. It is characterized in the field by randomly oriented orange pumice fragments that average half an inch in length in a pale-yellowish-gray to light-yellowish- brown matrix. Locally the tuff is zeolitized at the base (Hinrichs and Orkild, 1961, p. D98; Gibbons and others, 1963). Despite its partly welded character, the Stockade Wash Tuff is a cliff-former and locally south of Stockade Wash contains rosettes of closely spaced polygonal joints possibly localized by fumaroles during cooling. The Stockade Wash Tuff lithologically and miner- alogically resembles two local massive nonwelded to very slightly welded simple cooling units, one of which is found outside the Silent Canyon caldera on the east and the other within the caldera (fig. 6). The one east of Silent Canyon caldera was mapped as Stockade Wash(?) Member of the Paintbrush Tuff in Quartet Dome quadrangle (fig. 2; Sargent and others, 1966). This cooling unit is likewise raised in rank to Stockade Wash(?) Tuff. The other cooling unit that resembles the Stockade Wash is the tuff of Blacktop Buttes (cols. 5 and 6, fig. 7; Hinrichs and others, 1967; Orkild and others, 1969), which is entirely confined within the Silent Canyon caldera (fig. 6). This tuff inter— tongues with the Paintbrush Tuff (as restricted in this paper) and was mapped with the Paintbrush on two geologic maps (Hinrichs and others, 1967; Orkild and others, 1969). Because of its lithologic similarity to the Stockade Wash, however, this tuff is herein regarded as local informal unit separate from the Paintbrush. The Stockade Wash and Stockade Wash(?) Tuffs and the tuff of Blacktop Buttes are all characterized by sparse to abundant dark-greenish-gray dense devitrified xenoliths of peralkaline volcanic rock. These xenoliths closely resemble the peralkaline rocks, including the devitrified, microcrystalline facies of the rhyolite of Split Ridge and the Grouse Canyon Member of the Belted Range Tuff from the floor and the rim of Silent Canyon caldera. We therefore infer that these units were extruded from vents inside or on the rim of Silent Canyon caldera. Average modal analyses of the Stockade Wash Tuff, Stockade Wash(?) Tuff, and tuff of Blacktop Buttes are compared in figure 5 with the lavas of the tuffs and rhyolites of Area 20; the intracaldera lavas are arbitrarily placed together. The modes suggest that the Stockade Wash and Stockade Wash (P) Tuffs each may generally correlate with the lower and the upper lavas, respectively, of the tuffs and rhyolites of Area 20. The Stockade Wash(?) Tuff, therefore, probably slightly postdates the type Stockade Wash and is a separate cooling unit. The tuff of Blacktop Buttes is known to be the youngest, for it post— dates all the lavas of Area 20 (Orkild and others, 1969) and intertongues with two lavas related to the Paintbrush Tuff (cols. 5 and 6, fig. 7). In contrast with thin-section modes of Paintbrush Tuff units, the Stockade Wash and related tuffs contain common to abundant quartz and no clinopyroxene, although clinopyroxene has been found as a minor constituent of heavy-mineral separates. Percentages of nonmagnetic heavy mineral separates from the Stockade Wash Tuff and probable related calc- alkalic units associated with the Silent Canyon caldera are compared with those separated from some units of the Paintbrush Tuff in table 2. The tuffs and rhyolite lavas of Area 20, the Stockade Wash Tuff, the Stockade Wash(?) Tuff, and the tuff of Blacktop Buttes are all characterized by a high allanite/sphene ratio and very low content of augite in contrast to alkali-calcic tuff units of the Paint- brush. In the Topopah Spring Member, however, the paucity of both allanite and sphene makes the ratio insignificant for comparison. None of the three Paint- brush units contain orthopyroxene in the heavy mineral assemblage, but the Stockade Wash Tuff and tuff of Blacktop Buttes each contain 1 percent. The trace amounts of aegirine—augite in the tuffs and rhyolite lavas of Area 20, as well as in the Stockade Wash (P) Tuff and in the tuff of Blacktop Buttes, may have been derived from abundant peralkaline xenoliths. Chemical analyses of the lavas of Area 20 and the Stockade Wash Tuff were made (W. D. Quinlivan and P. W. Lipman, written commun., 1974); silica percentages of glass and divitrified rocks are plotted in figure 5. The lavas of the tuffs and rhyolites of Area 20 become progressively more silicic upward. The devitrified specimens are relatively enriched in silica with respect to the glassy specimens. In summary, the Stockade Wash and the tuff of Blacktop Buttes are removed from the Paintbrush Tuff and from the Piapi Canyon Group. The Stockade Wash is raised in rank to a separate tuff formation, because it is lithologically and petrographically unlike any units in the Paintbrush Tuff but does show lithologic and petrographic affinities to late effusives occurring entirely within the Silent Canyon caldera. The Stockade Wash(?) Member of Sargent, Luft, Gibbons, and Hoover (1966), which occurs justeast of and outside the Silent Canyon caldera, is also raised in rank but is retained as a queried unit because its petrographic features suggest that it correlates with a slightly higher part of the intracaldera sequence. The tuff of Blacktop Buttes (Hinrichs and others, 1967; Orkild and others, 1968), the youngest ash flow entirely confined to the Silent Canyon caldera, is regarded as an informal unit not assigned to a formally named formation. It u» PAINTBRUSH TUFF AND ROCKS RELATED TO CLAIM CANYON CAULDRON 21 TABLE 2.—Nonmagnetic heavy minerals of Paintbrush and Stockade Wash Tuffs and related tuffs and lavas in order of stratigraphic. succession [Stockade Wash Tuff and related units in italic type; r, mineral rare] Unit and subunit Volume percent of total heavy minerals (number of specimens) E” u q, '5 A ._.>‘ w 2 a E e E 3 3 t .. e a - v .“=' E 2 a“ 2 .43 o a 2 2 5 m 3; 2 s 3 6 a Paintbrush Tuff: Tiva Canyon Member: Quartz latite (2) ....... 19 31 O 0 9 18 12 9 1 3 O 0.7 Tuff of Blacktop Buttes (l) .................................. 76 2.6 r 1.0 0.4 1.2 12 6 0 0 10 Paintbrush Tuff: Local unit in _ hole UE19f (l) ......................... 57 6 0 0 0 24 8 6 0.8 0 0.3 Paintbrush Tuff: Topopah Spring Member: Quartz latite (1) .................... 80 18 0 0 0 0 0.8 0.4 0 0 (1) Stockade Wash(?) Tuff (l) .................................... 229 1.6 r 0 59 0.6 6.0 2.2 1.0 r 10 Stockade Wash Tuff (1) .............. 72 0.4 0 1.0 18 0.2 3.4 6 2.2 0 17 Rhyolite lavas of Area 20 (14) .............................. 79 0.1 r r 5.2 0.2 12 3.6 0.2 0 60 Lithic-rich tuffs of Area 20 (8) ........................... 42 0.1 0.3 0 r 'r 33 21 1.2 1.8 +99 ILow content of allanite and probable limit of detection of sphene in Topopah Spring not significant for this comparison. PAINTBRUSH TUFF AND ROCKS RELATED TO CLAIM CANYON CAULDRON GENERAL FEATURES ORIGINAL AND REDEFINITION OF THE PAINTBRUSH TUFF The Paintbrush Tuff was defined by Orkild (1965, p. A49) from exposures in Paintbrush Canyon, a small gulch 6.4 km (4 mi) northeast of Yucca Mountain (figs. 1, 8). The most complete stratigraphic section, however, is on the north end of Yucca Mountain where four members are well exposed. The Topopah Spring and Tiva Canyon Members are widespread welded ash-flow sheets that were first defined as members of the former Oak Spring Formation (Hinrichs and Orkild, 1961, p. D97) in the Yucca Flat area (figs. 1, 2). These members were sub- sequently included in the Piapi Canyon Formation of the Oak Spring Group (Poole and McKeown, 1962, p. C61). The Yucca Mountain Member of the Piapi Canyon Formation was described by Lipman and Christiansen (1964) and occurs between the Topopah Spring and Tiva Canyon Members (fig. 7) in the Yucca Mountain area (fig. 8). Orkild (1965) included the Stockade Wash Member (Stockade Wash Tuff of this paper), the Topopah Spring Member, a newly defined Pah Canyon Member, and the Yucca Mountain and Tiva Canyon Members in his Paint- brush Tuff (fig. 7) of the Piapi Canyon Group. The Paintbrush Tuff is here redefined to include, in ascending stratigraphic order, the Topopah Spring, Pah Canyon, Yucca Mountain, and Tiva Canyon Members and similar, quartz-poor, informal ash-flow and ash-fall 2Biotite percentage disproportionately low with respect to other mafic minerals owing to oxidation to magnetite, tuffs that occur between the top of the underlying Stockade Wash Tuff, as herein defined, and the base of the over- lying Rainier Mesa Member of the Timber Mountain Tuff, except for thin local bedded lithic-rich tuff that occurs just above the Stockade Wash (fig. 7). Excluded are the Stockade Wash Member, which was part of the original definition (Orkild, 1965), and also quartz- bearing, lithic-rich tuffs of the tuffs and rhyolites of Area 20, which occur between the Belted Range and Stockade Wash Tuffs and locally just above the Stockade Wash Tuff. In the Yucca Flat-Rainier Mesa area, these lithic-rich tuffs were included in the Paintbrush by Orkild (1965) as part of the now-abandoned Survey Butte Member (Poole and McKeown, 1962; Orkild, 1965) before the strati- graphic pOSition of the lithic-rich tuffs in the subsurface of Area 20 was known (Orkild and others, 1968 and 1969). Where the Stockade Wash Tuff is missing, the lower contact of the Paintbrush can almost always be recognized within a few metres (several feet) as glassy quartz-poor friable tuff lying on zeolitic quartz-bearing lithic-rich tuff. In northeastern Yucca Flat and locally elsewhere in and around Yucca Flat (fig. 8), bedded tuff of the Paintbrush rests directly on the Belted Range Tuff or on a local thin peralkaline bedded sequence just above the Belted Range (Sargent and others, 1965; Noble and others, 1968). BEDDED TUFF The bedded tuff of the redefined Paintbrush, is generally recognized by the glassy state of contained shards and pumice lapilli and by its somewhat lesser induration than that of the underlying bedded tuff. The distinguishing 22 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA l 37? u7°ls' ”7° , 9 so i f STONEWALL I l I' MOUNTAIN ‘1 A p In 4 5/0 r5 k, l. EXPLANATION b POST-TIVA CANYON RHYOLITE ‘ , LAVAS ‘ W Quartz-bearing rhyolite lava TOLICHA \\ Bronte—bearing rhyolite lava THICKNESS OF TIVA CANYON MEMBER O—ISOm 7 150—300 In Includes tuff of Chocolate Mountain in Claim Canyon cauldron segment >300 m -——— Limit of Topopah Spring Member 7—— Limit of information other than caldera walls UE19' i O I Drill holes — Numbered holes dis- cussed in text A A’ Line of geologic cross section O—E‘O FIGURE 8.—Areal extent of Topopah Spring and Tiva Canyon Members of Paintbrush Tuff and related lavas. Isopachs of Tiva Canyon ‘ petrographic feature ofall except the uppermostpart is the dominant in the overlying Rainier Mesa Member of the' near absence of quartz, by comparison with the under- Timber Mountain Tuff. The lower part of the bedded tuf‘t‘, p lying and overlying rocks. In the uppermost 6—12 m (20—40 as observed in cores of some drill holes in Yucca Flat (fi i ft) of the bedded tuff, quartz increases upward and becomes 8), has been locally zeolitized and indurated but, neverthe- ‘ PEAK E A 37? 15 20 MILES l l l 20 KILOMETRES cross sections shown » PAINTBRUSH TUFF AND ROCKS RELATED TO CLAIM CANYON CAULDRON 23 US" l i 45' ll6°30' é ~QWNIOUNT s HELEN ,\ . <3 E M E S 4 SI gNT CANYON BLACK 0 $ALDEORA O MOUNTAAIN QUARTZ M NTAIN BMOSCTAIN CALDERA 5 SLEEPING / BUTTE CALDERA ....... TIMBER 6x MOUNTAIN ."o A x / . ‘ l 9) "CALDERA OASIS -' . NYQN MOUNTAI ()5 CA I. A ’v \/ ,1 OASIS I AVV :. CALDERA “.‘SEGMENT an We“ ‘ ....... @051 & c9) / \ / NEVADA TEST SITE \ ___.____ I Lulhrop Walls AW“ SPECTER _ . \ I 77 RANGE \_ \ l \ '9 Y NILE c_UNT _ LINCOLN COUNTY Gloom /“ Lake f— /\~/ \‘ OAK SPRING BUTTE ~ 37.0 ‘5 SSOOd Vd |NC_0LN _couN1;Ir_ CLARK COUNTY /6AHMONlE-SALYER VOLCANIC CENTER ounmfl” MOUNTAINS as? 0mm.“ — ‘t ”u,” 45 Member show great thickness of its uppermost subunit ( in figures 11, 12, and 13. less, contains fewer lithic fragments and much less quartz than the underlying lithic-rich tuffs of the tuffs and rhyolites of Area 20. At a few places along the east wall of Timber Mountain caldera (fig. 8) where the lithic-rich tuff of Chocolate Mountain) within Claim Canyon cauldron segment. Geologic tuffs are glassy, the base of the bedded tuff of the Paint- brush is at the top of the Stockade Wash Tuff. Although induration by zeolitization does, in places, cross the lower stratigraphic contact of the bedded tuff, the relative 24 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA abundance of quartz and lithic fragments is the best criterion for separating the bedded tuff of the Paintbrush from the underlying lithic-rich tuffs of the tuffs and rhyolites of Area 20. On the 7%-minute quadrangle maps (fig. 2), bedded tuff of the Paintbrush was not distinguished consistently from the underlying lithic—rich tuffs and from intertonguing bedded tuff of the Wahmonie Salyer volcanic center to the south (fig. 1; Poole, Carr, and Elston, 1965). Moreover, on the north and south rim of Timber Mountain caldera, some beds within the Paintbrush Tuff coarsen into tuff breccia toward lava vents and intertongue with rhyolite lava. In some places the tuff breccia was included with the lava flows; in others, it was mapped as bedded tuff. On the geologic map of the Timber Mountain caldera area (Byers and others, 1976), the bedded tuff map unit includes bedded tuff of the Paintbrush, lithic-rich bedded and nonwelded ash-flow tuffs of the tuffs and rhyolites of Area 20, tuff-breccia around rhyolite lava vents, and other minor local nonwelded ash flows, such as the tuff of Blacktop Buttes. The grouping together of all this light- colored porous tuff sequence that occurs between the underlying Belted Range Tuff and the overlying Timber Mountain Tuff was necessitated by the difficulty in separating these various units in the field within the time allotted for the completion of maps at the Nevada Test Site. Drill cores have provided the means for positive separation and identification of these units in the sub- surface, in contrast to their generally low profile in outcrop. GEOLOGIC RELATIONS BETWEEN WELDED ASH-FLOW TUFFS, LAVAS, AND CLAIM CANYON CAULDRON Welded ash-flow tuff members of the Paintbrush Tuff intertongue with genetically related, alkali-calcic lavas that are not part of the Paintbrush and are within the Claim Canyon cauldron segment, which is the exposed high-standing southern remnant of the now largely buried Claim Canyon cauldron (fig. 8). Eruption of the members of the Paintbrush Tuff resulted in cauldron subsidence in overlapping areas within the inferred limits of the Claim Canyon cauldron shown in figure 8. All four tuff members are probably present within the buried part of Claim Canyon cauldron. The lowermost member, the Topopah Spring, is not exposed within the cauldron segment, but a fragment of intracauldron facies that was included in the Timber Mountain caldera debris flow was cored at a depth of 1,402 m (4,600 ft) in drill hole UE18r (fig. 8); hence, the site of UE18r was probably within the cauldron. North of Timber Mountain caldera, lavas intercalated with ash- flow tuff units are exposed in the wall of the caldera and were penetrated in drill holes north of the wall. Flow directions in the surface exposures indicate a southerly source (Cummings, 1964), probably from the rim area of the now-buried Claim Canyon cauldron within a few miles to the south. Welded ash-flow tuffs of the Paintbrush Tuff and inter- calated lavas are several times thicker (table 3) inside the Claim Canyon cauldron segment than where they are exposed on the cauldron rim (fig. 8). A total thickness of more than 3,000 m (6,500 ft) of Paintbrush Tuff and inter- calated lavas is present inside the Claim Canyon cauldron segment, but the Topopah Spring Member is not exposed. The tuff of Chocolate Mountain and the tuff of Pinyon Pass are informal map units of the Paintbrush Tuff entirely within the Claim Canyon cauldron segment (see fig. 7). The tuff of Chocolate Mountain (table 3) is an extremely thick intracauldron welded quartz latite caprock of the Tiva Canyon Member and, at one critical exposure, grades downward within a stratigraphic interval of about 50 feet into the underlying main part of the Tiva Canyon Member. The tuff of Pinyon Pass is a separate overlying cooling unit which closely resembles the main part of the Tiva Canyon but is not considered part of it because of the complete cooling break between the units. TABLE 3.—Thicknesses and K-Ar ages of Paintbrush Tuff units and genetically related lavas within Claim Canyon cauldron segment and outside on cauldron rim [Modified from Byers and others ( 1969). K-Ar ages from Marvin and others (1970)] Cauldron rim Cauldron segment Units, in stratigraphic Thickness K-Ar age Thickness K-Ar age order (m) (x106 yr) (m) (x106 yr) Tuff of Pinyon Pass ................................ 0 150 Tiva Canyon Member Tuff of Chocolate Mountain........... 0 600+ 12.5, 12.6 Main sheet 120 12.4, 12.4 500+ 12.6 Yucca Mountain Member ....................... 60 335 Lavas between Pah Canyon and Yucca Mountain Members ..................... 30 300 Pah Canyon Member 90 200+ Pre-Pah Canyon lavas.... 0 150+ Topopah Spring Member ....................... 120 13.2 Not exposed Composite thickness .................... 420 2,000+ (1,400 ft) (6,500+ ft) LITHOLOGY AND FIELD RECOGNITION There are few unique criteria that can be used to dis- tinguish individual welded tuff units of the Paintbrush Tuff in the field. Locally strong crystallization owing to slow cooling has destroyed primary textures, and so in places it is difficult to distinguish tuffs from intercalated lavas. The field geologist, following a long period of familiarization, eventually learns to distinguish, on the basis of very subtle differences, individual cooling units in a restricted area. Part of the problem of field identification arises from the fact that lithologic variations within individual welded tuff units tend to be similar for all units and to be much more conspicuous than variations between equivalent parts of separate units. All the welded ash—flow units of the Paintbrush Tuff and even the intercalated lavas have crystallization zones PAINTBRUSH TUFF AND ROCKS RELATED TO CLAIM CANYON CAULDRON 25 similar to those described by Smith (1960b), and most of the lithologic contrasts observed in outcrop are related to these zones. Color, for example, is related to crystalliza- tion zones in the following way: dark gray to black, glassy; brownish gray, cryptocrystalline; light brownish gray to purplish gray, microcrystalline; and light gray, coarsely microcrystalline with granophyric texture in pumice lenticles. The light-gray granophyric tuffs are typical of the thick intracauldron sheets. The degree of welding is also similar for each ash-flow cooling unit and tends to be moderate to dense except at the very distal edges of the units. Notwithstanding these similarities there are a few features, such as lithic content, pumice content and size, and compositional zonations, that the experienced geologist can use in making field identifications. Generally modal analyses are necessary to identify a unit. TOPOPAH SPRING MEMBER The Topopah Spring Member of the Paintbrush Tuff has been rather thoroughly described by Lipman, Christiansen, and O’Connor (1966); therefore, only the salient features are summarized here. The Topopah Spring, whose distribution is shown in figure 8, extends generally east of the Tiva Canyon Member but overlaps the member by more than 50 percent. Variations in thick- ness, with a maximum thickness of nearly 275 m (900 ft) on the west flank of the CP Hills (fig. 8), are mainly the result of paleotopography. The member is not exposed inside the Claim Canyon cauldron segment but an unknown volume of it is buried there. The total extracauldron volume is about 160 km3 (40 mi3) (Lipman, Christiansen, and O’Connor, 1966); probably more than 250 km3 (60 mi?) was erupted. The unit has a high-silica rhyolitic lower zone and a quartz lati tic upper zone or caprock. Near the south rim of the Claim Canyon cauldron segment (fig. 8) a middle xenolithic subunit containing granitic fragments separates the lower and upper parts (Lipman, Christiansen, and O’Connor, 1966, fig. 18). Densely welded quartz latitic bedded tuff, which occurs at Prospector Pass (fig. 8) just west of the cauldron segment, overlies the high-silica rhyolite zone and is probably very near the source area of the Topopah Spring area (Lipman, Christiansen, and O’Connor, 1966, p. F26). The modal compositions of the lower and upper compositional zones are shown in figure 9; the Topopah Spring is distinguished modally from all other members of the Paintbrush Tuff by the lack of sphene. The rhyolitic lower part of the Topopah Spring is also distinguished readily by a high plagioclase-to-alkali feldspar ratio, but the quartz latitic caprock of the Topopah Spring is dis- tinguished from younger caprocks of the Paintbrush mainly by the absence of sphene. Chemical analyses of the different ash-flow units of the Topopah Spring Member were published by Lipman, Christiansen, and O’Connor (1966, table 2); the silica ranges of the high-silica rhyolite and quartz latite are shown in figure 9. The Topopah Spring Member can also be identified by its normal thermal remanent magnetization (G. D. Bath, written commun., 1965), in contrast to reverse polarities of younger units of the Paintbrush Tuff. Its K-Ar age is 13.2 m.y., in contrast to an average 12.5 m.y. for the Tiva Canyon Member (table 2). PAH CANYON MEMBER AND RELATED PRE—PAH LAVAS The Pah Canyon Member is lithologically similar to the quartz latitic caprock of the Topopah Spring Member but is a simple ash-flow cooling unit without compositional zoning. Also, its silica content—74 percent (fig. 9)—falls in the rhyolite range in contrast to that of the Topopah Spring caprock, which is in the quartz latite range of 69—72 percent. Flattened pumice lenticles, less than 2 cm in length, contrast with larger ones in the Topopah Spring. Sparse to common small cognate inclusions, mostly less than 0.5 cm, are distributed throughout the Pah Canyon. The extracauldron extent of the Pah Canyon is only about 200 km2 (75 miz) (fig. 10). The extracauldron volume of the Pah Canyon does not exceed 20 km3 (5 mif‘), but an equal or greater volume of the tuff must be contained within the Claim Canyon cauldron, for the average thickness of the tuff inside the cauldron is at least three times that of the tuff outside. The buried areal extent inside the cauldron is probably greater than that outside the cauldron. Two lava flows underlie the Pah Canyon within the Claim Canyon cauldron segment (figs. 3, 7). Their modes are pooled and are similar to but slightly more mafic than that of the Pah Canyon (fig. 9). These pre-Pah Canyon lavas are the lowest stratigraphically in the Claim Canyon cauldron segment and are also closest to the cauldron wall, near which they are brecciated and silicified. LAVAS AND ASH-FLOW TUFF BETWEEN PAH CANYON YUCCA MOUNTAIN MEMBERS One or more lava flows of similar composition were extruded on the Pah Canyon Member before eruption of the Yucca Mountain Member within and just outside the Claim Canyon cauldron segment (fig. 10). At about the same time a local ash-flow tuff and two overlying lava flows were poured northward into Silent Canyon caldera (Cummings, 1964; David Cummings, written commun., 1964) from the now-buried Claim Canyon cauldron. The local ash—flow tuff to the north was penetrated only in drill holes UE19f and UEl9i (fig. 10). On the north side of Timber Mountain caldera, the Pah Canyon and Yucca Mountain Members are not exposed or penetrated in drill holes, and so the local lavas and underlying tuff occur between the Topopah Spring and Tiva Canyon Members. The petrography (fig. 9) and cooling history of these local intercalated units on the north side, however, indicate that 26 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA STRATIGRAPHIC UNlT {SUBUNIT Number of thin sect ions (N) in parentheses PHENOCRYSTS AS PERCENTAGE OF TOTAL ROCK FELSIC PHENOCRYSTS AS PERCEN OF TOTAL PHENOCRYSTS Quartz m Plagioclase TAGE m Alkali feldspar l | I | | Quartz-bearing - m rhyolite lava (4) Post-Tiva Canyon < rhyolite lavas B‘ - b . Iotlte- earlng rhyolite lava (5) I W ? ? l l l l l r _ El Phenocryst-rlch - . m rhyolite caprock (5) W/A Tuff 0f High-silica ] , Pinyon Pass j rhyolite (4) - 7/4 w “Mme ‘3’ - r///////////////// m 3: W Tuff of Chocolate I 1 i I a D - b |_ Mountain (8) — W 5 (lntracauldron) a < 3 .D 5E .E Quartz latite (13) — ' m {E _ (extracauldron) m Tiva Canyon < Member 3 Rhyolite (13) - W m A High-silica - m L rhyolite (10) % Yucca Mountain Member (10) I 7 :§ (essentially phenocryst-free) 2 _ ? a l l l . l l Hornblende- - , m Rhyolite lavas between bearing lavas (4) M Yucca Mountain and ‘ Pah Canyon Members Pyroxene- Z) V m t bearing lavas (12) - V/////////////A 3 l l l I l l I : Pah Canyon Member (17) I ‘ I Y Y fl ‘3 - m E W E l l : i l l fi‘ 0. Pre-Pah Canyon rhyolite lavas (5) \V A - fl ? ? l l l l l l l w- . ] _._ Quartz latlte (11) E l — W m g% Topopah Spring 5 Member _ . _ a ngh-S'hca - s\\\\\\\\\,\\\\\\\\\\\\\\ ti? L L rhyolite (19) 7///////////////////////////////% | l l l l l l 0 1O 20 0 20 40 60 80 100 PERCENT FIGURE 9.—Modal and silica ranges of significant units of Paintbrush Tuff and petrologically related lavas in order of stratigraphic sequences indicated by modal trends. Blank area in phenocryst column indicates absence of mineral. Modes of Topopah Spring and W. D. Quinlivan and P. W. Lipman (written commun., 1974). PAINTBRUSH TUFF AND ROCKS RELATED TO CLAIM CANYON CAULDRON MAFIC PH ENOCRYSTS AS PERCENTAGE OF TOTAL PHENOCRYSTS - Total mafics Amphibole Biotite Clinopyroxene '———‘I_”—_—I - ——4——H ———I ‘ - 51553 (R) Resorbed c/ Km I — ———I————( -——I——4 ————I S — w (RlResorbed ———I———I ——I {% ———I———I ——a 1% ~———I———I ———I m— 3 W I ____l 0 5 10 O 2.5 PERCENT succession. Queried successions between a few units inferre Member from Lipman, Christiansen, and O’Connor (1966). NUMBER OF SPH ENE PH ENOCRYSTS PER THIN SECTION ____I—J 0 20 40 27 SILICA PERCENTAGE RANGE (H20 and CO2 free) Number of analyses (N) in parentheses (’7 I I I I (2) (4) l7 I I I l q (1) (1) (1) If I I I I 4I (3) (4) (0) (6) If I I I I 4 (3) If I I I I #I (2) (2) I7 I I I I A (4) If I I I I I (0) )— I I I I (6) (8) I I I I I 68 70 72 74 76 78 PERCENT d in part from structural relations and in part from general eruptive Silica percentages from Lipman, Christiansen, and O’Connor (1966) 28 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA 37? 30 MOUNT JACKSON EXPLANATION Yucca Mountain Member RHYOLITE LAVAS BETWEEN YUCCA MOUNTAIN AND PAH CANYON MEMBER Hornblende-bearing lavas ”57‘A Pyroxene-bearing lavas Overlap of Yucca Mountain and Pah Canyon Members — Not shown at overlap with pyroxene—bearing lavas [:1 Pah Canyon Member Drill hole used for control — Num- bered holes discussed in text A O UE19i A A I Line of geologic cross section 37 V, O 5 10 15 I_ I I I I I I I I O 5 10 15 2O KILOMETRES 31° \_«“§0LD MOUNTAIN Wu”, II7°I5' “10 I I STONEWALL I l '22, MOUNTAIN TOLICHA PEAK 3 A COUNTY ESMERALDII l v” : (‘0 f <9 ““““ x I I? A? ‘7» o 4’ '2 . VA“? ' s Q ’3 40‘} 04 g :: '1 6N I '3 BULLFROG / : : 4 J : : GRAPEVINE g \A PEAK ,, HILLS In, 20 MILES 1 FIGURE lO.—Areal extent of Pah Canyon and Yucca Mountain Members of Paintbrush Tuff they were probably extruded between the times of eruption of the Pah Canyon and Yucca Mountain Members. Where both members are exposed just outside the Claim Canyon cauldron segment, coarse-bedded ash-fall tuff, as much as 60 m (200 ft) thick, related to lava venting occurs between the members, except where two erosional remnants of lava fronts are preserved: one at the north end of Yucca Moun— tain and the other 3.2 km (2 mi) south of the head of Pah PAINTBRUSH TUFF AND ROCKS RELATED TO CLAIM CANYON CAULDRON 29 ”6° 30' Is' He“ o‘SMOUNT I I 3. HELEN 5 : 2 : > g _5 ,\ 3 i 1‘} 5 5 V E E 4 EIIIIIHHH‘: :E ‘9’ : E g : 9‘" V9 E E 5 v6 ., Go _5 " J: _5 Q‘ g 2 S : g : q z Hm MI )‘l'; M E s E 8 4 I, 8 ° “WWI U z — V — _l Eé BLACK 3 Gloom+ / MOUNTACIN taIe+ , I OUARTZ M NTAIN MOUNTAIN I \ I I I CALDERA 37° I5‘ asood Va" I E I I I (”I RIDG SLEEPING / BUTTE I ______________ CALDERA ” Claim. ‘ TIMBER °/\ MOUNTAIN °a‘\ CALDERA O OASIS 9“ “map? Q, VALLEY t 0? MOUN AIN CALDERA D°NIE “_SEGMENT \’\ ,, b5 g _ o __I_ E; 37 . ”‘5 CP YUCCA’EASS I I gl WAHMONlE-SALYER \AI é'L COLN _CMDL p _— . V“ LLARK COUNTY ,, 1; ~ CLAIM CANYON VOLCANIC CENTER ‘6} | E' ”1.3 CAULDRON SEGMENT p 2| ' JACKASS q» I ' FLATS I I I I \‘IRLINGER E_‘ MOUNTAINS I II —36.° I v ”mt “5 0 60¢ SKULL I I I I» \\ NEVADA TEST J“”” 7 \ :jSPOTTEoI " Lnthrop Wells S‘IIII‘N‘ SPECTER \ Merit—u rIly‘ II " SWI W o“ \ Em» w RANGE \ '3 RANGE I ”m” \ Ems“ '3 ' ‘ . 9pm I \ | fr \' I =_ I and extent of lavas between the members. Geologic cross sections shown in figures ll, 12, and 13. field (G. D. Bath, written commun., 1965), and is typical of post-Topopah Spring units of the Paintbrush. The lavas between the Pah Canyon and Yucca Moun- tain Members in Claim Canyon cauldron segment are Canyon (fig. 10). The local ash—flow tuff and all the lavas at this stratigraphic position on both north and south sides of Timber Mountain caldera have a thermal remanent magnetism that is reverse to the Earth’s present 30 pyroxene bearing and are almost identical in modal petro— graphy to the lower lava flow on the north side in Silent Canyon caldera. These lavas differ from the Pah Canyon in that sphene is more common and alkali feldspar pheno- crysts are dominant over plagioclase; the modal ranges shown in figure 9 represent seven thin-section modes from the south side pooled with five from the north side of the Timber Mountain-Oasis Valley caldera complex. The 74—percent silica content of the lower lavas indicates a rhyolitic composition similar to that of the Pah Canyon. The local ash-flow tuff penetrated in drill holes UE19i and UE19f (fig. 10) is closely associated with the overlying pyroxene-bearing lava flow. This ash flow is locally fused beneath the lava flow, as penetrated in drill hole UE19f (figs. 10, 11). In drill hole UElQi, 4 km (2% mi) farther away from the inferred volcanic source, the two units are separated by nonwelded tuff, and a basal vitrophyre is present in the lava. Although the tuff resembles the Pah Canyon Member lithologically and was shown as Pah Canyon in a cross section by Hinrichs, Krushensky, and Luft (1967, section A—A’), both the tuff and the overlying pyroxene-bearing rhyolite lava (Byers and Cummings, A S.S.W. 7000’ — PAH UTE Drill hole Dense welded tuff General direction of flowage from inferred source vent ———> TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA 1967) have similar phenocryst ratios (table 4). The sub- surface tuff, moreover, differs modally from the Pah Canyon Member of the Paintbrush Tuff south of Timber Mountain caldera mainly in the sanidine/plagioclase phenocryst ratio (2: 1). The Pah Canyon Member has about equal sanidine and plagioclase phenocrysts in the mode and contains other petrographic features that relate it more closely to the underlying Topopah Spring Member. The petrographic evidence and close association with the overlying pyroxene-bearing lava indicate that the local subsurface tuff is probably slightly later than the Pah Canyon Member and therefore is considered a local informal unit of the Paintbrush Tuff. The close spatial and temporal association of the tuff and the lava, combined with their close petrographic similarity also suggest that they were probably erupted from the same Claim Canyon cauldron vent, probably just inside the present north wall of the Timber Mountain caldera. The upper, hornblende-bearing lava flows are exposed in the north wall of Timber Mountain caldera (fig. 10). They originally flowed only a short distance northward (Cummings, 1964; David Cummings, written commun., MESA Drill hole Timber Mountain Tuff ‘. :; as, we,” \9 - , n... ,>,, , 2... x, .‘x, , .. . w w. y.” . k u. MM r (31,5 W). 9”}; .‘ . “’ “ ‘ “ “)I“E<~.‘:s‘£&3.fib‘“¢:~i;l’t " -- Tiva Canyon Member of Paintbrush Tuff Pyroxene-bearing rhyolite lava flow des into lava (no ‘ ————————— ~xu 5000’ C gra ________ B ‘‘‘‘‘‘‘‘‘‘‘‘ /vitrophyre in lava) ::::::r_:—,‘f——__fiaLfltioEEylt: .................. Subsurface tuff of Paintbrush Tuff ’ _ 4000 Tuffs and rhyolites of Area 20 _"_,' filling Silent Canyon caldera D < LL 2000 | l 500 1000 O——O 4000 6000 FEET l l | 1500 METRES Vertical exaggeration X 2 FIGURE ll.—Generalized section along A—A’ through drill holes UEle and UE19i, Silent Canyon caldera, showing relations between subsurface ash-flow tuff of Paintbrush Tuff and overlying, mineralogically similar pre-Tiva Canyon rhyolite lava. Vitrophyres shown in black; nonwelded to partially welded zones, stippled. Inferred transitional contact shown by dot-dash line. Line of section shown in figures 8, 10. PAINTBRUSH TUFF AND ROCKS RELATED TO CLAIM CANYON CAULDRON 31 TABLE 4.—Combined thin-section modes and computed accessory mineral percentages of local ash-flow tuff of Paintbrush Tuff and over- lying pre-Tiva Canyon rhyolite lava where penetrated in drill hole UE19f, Pahute Mesa (fig. 11) [Location of drill hole shown in fig. 11. N.d., not determined] Phenocrysts and accessory minerals as volume U A‘ 5 § percent of total crystals 2 E a E - g g? E 3 E _ J. "7: em 6; s ~ .. a s u N a Unit (m) 3‘. 8. 5t 2 5 13 6 m“ 2 :5 3 t3 Lava ............................. 429 10 0 62 29 5.0 (3) 0.9 N.d. N.d. N.d. 8.8 524 9 0 64 29 3.5 (3) 1.0 0.4 0.2 0.03 7.1 607 ll 0 61 32 5.1 0.6 .3 .l .l .03 7.1 ..................................................................................... Insignificant or partial cooling break? Tuff ............................. 655 14 1.9 64 29 .5 .2 l l 3 2 .03 5 7 lAccessory mineral percentages computed from counts of heavy mineral separates using sphene as control to combine thin-section mode and heavy-mineral count. 1964), as shown by the fact that they have not been penetrated by the extensive drilling in Silent Canyon caldera. Two lava flows of slightly different modal compositions are exposed. The volumetrically larger of the two lava flows is as much as 125 m (400 ft) thick and is separated from the underlying pyroxene—bearing rhyolite lava by 75 m (250 ft) of coarse, near-vent tuff breccia. The overlying Tiva Canyon Member pinches out over the thickest part of the larger hornblende-bearing lava flow. There are two exposures of this flow separated by a thick paleovalley filling or source-vent filling of post-Tiva Canyon biotite-bearing lava (Byers and Cummings, 1967; Byers and others, 1976). Two thin-section modes from each of the two exposures are very similar and are pooled on figure 9. Two chemical analyses (W. D. Quinlivan and P. W. Lipman, written commun., 1974) indicate that the hornblende-bearing lava approaches high-silica rhyolite in composition (fig. 9). One thin section from the smaller flow at the same stratigraphic position slightly westof the larger flow is petrographically similar but contains a sub- ordinate amount of quartz phenocrysts. YUCCA MOUNTAIN MEMBER The Yucca Mountain Member (Lipman and Christiansen, 1964) is a simple cooling unit of uniform, nearly phenocryst-free, high—silica rhyolite (fig. 9) even where it is 335 m (1,100 ft) thick within the Claim Canyon cauldron segment (table 3). Outside the Claim Canyon cauldron segment, a small thin lobe is present east and south of Pah Canyon (fig. 10) and a larger lobe with a maximum thickness of about 75 In (250 ft) occupies about 100 km2 (40 mi?) around northern Yucca Mountain and northern Crater Flat. Only about 8 km3 (2 mi3) is present outside the cauldron and roughly an equal volume can be inferred within the Claim Canyon cauldron segment. The member differs from all other units of the Paint- brush Tuff, in having less than 1 percent phenocrysts, largely sodic sanidine. Sparse small pumice lenticles and grayish—red xenoliths less than 1 cm in length are strewn 2Includes opaque oxides not tallied separately. 3C1inopyroxene altered in strongly devitrified interior of lava flow. through a matrix of shards or devitrified groundmass. The unit has distinct crystallization zones with brownish-gray to gray glassy upper and lower zones and a pale-purplish- gray middle zone, which contains irregular white lithophysae where it is thick. The field and petrochemical evidence indicate that the Yucca Mountain Member postdates not only the clinopyroxene—bearing lavas but probably also the hornblende-bearing lavas exposed in the north wall of Timber Mountain caldera where the Yucca Mountain is absent. Between the Yucca Mountain and overlying Tiva Canyon Member, only a foot or two of fine-grained ash- fall tuff, typical of the initial phase of the Tiva Canyon eruption, is nearly everywhere present; no coarse tuff breccia that would indicate nearby lava vents has been found between the members. This close stratigraphic association and the petrochemical trend toward increasing phenocrysts and decreasing silica content strati- graphically upward (fig. 9) indicate that the Yucca Moun- tain Member is an early eruptive phase of the same magma chamber that supplied the Tiva Canyon. Moreover, very little time, probably measurable in tens of years, elapsed between eruptions of the two members, for little if any erosion has been observed on the Yucca Mountain where it is overlain by the Tiva Canyon. The eruption of the Yucca Mountain, therefore, was probably very close to that of the Tiva Canyon Member in time and was almost certainly later than the extrusion of the hornblende-bearing rhyolite lava. TIVA CANYON MEMBER The Tiva Canyon Member consists of a main part—a compositionally zoned extensive ash—flow sheet—and an upper thick quartz latite, the tuff of Chocolate Mountain (Christiansen and Lipman, 1965; Byers and others, 1976), which is herein included as part of the redefined Tiva Canyon Member. Outside the Claim Canyon cauldron segment (fig. 8), the main part of the Tiva Canyon, like the Topopah Spring Member, is compositionally zoned (fig. 32 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA 9); inside, the tuff of Chocolate Mountain overliesrhyolite of the lower main part of the Tiva Canyon above a thin transition zone characterized by crystal-rich quartz latitic pumice lenticles in a matrix of welded crystal-poor rhyolite. In its type section at Tiva Canyon (fig. 8), the member is 107 m (350 ft) thick, and the maximum thickness, west of Beatty, is about 200 m (700 ft). The unit extends for about 115 km (10 mi) along a broad arcuate belt outside the southern wall of the Claim Canyon cauldron segment. The distribution of the member is similar to that of the Topopah Spring Member, except that the Tiva Canyon extends not as far east and somewhat farther west—just barely into California. The Tiva Canyon was mapped by Cornwall and Kleinhampl (1964) as cooling unit 3 of their Bullfrog Hills caldera. In Silent Canyon caldera, two lobes of the Tiva Canyon each about 105 m (350 ft) thick in separate areas were penetrated in five drill holes (fig. 8). Inside the Claim Canyon cauldron segment the lower main part of the Tiva Canyon and the upper part, the tuff of Chocolate Mountain, aggregate more than 900 m (3,000 ft) in thickness (table 2). The total area covered by the Tiva Canyon outside the cauldron is probably in excess of 2,600 km2 (1,000 mi?) if the member extends northwest under Sarcobatus Flat (fig. 8). Inside the Claim Canyon cauldron, possibly as much as 520 km2 (200 mi?) of tuff having a thickness averaging more than 0.8 km (0.5 mi), is now largely buried under younger rocks of the Timber Mountain-Oasis Valley caldera complex. The total volume of the Tiva Canyon Member, therefore, may be in excess of 1,000 km3 (250 mi3). The Tiva Canyon Member is a compound cooling unit, predominantly gray to reddish-brown devitrified densely welded ash-flow tuff with minor light-brown and gray nonwelded to partially welded tuff at its base and top. Generally a few feet of genetically related light-gray ash- fall tuff occurs at the base of the unit. In the lowermost part the Tiva Canyon is crystal-poor sanidine- and hornblende-bearing high silica rhyolite tuff; it grades upward into a middle crystal-poor rhyolite with biotite, which is in turn overlain by an upper (caprock) crystal- rich quartz—latitic tuff (fig. 9). These three distinct compositional zones are laterally persistent over much of the sheet outside the Claim Canyon cauldron segment and, therefore, provide an intrasheet stratigraphy. The Tiva Canyon Member inside the cauldron segment is locally flow-laminated and contains flow folds. Upward, it grades into normal welded tuff. This secondary flowage may have occurred where steep preemplacement topography marked the site of the cauldron wall. Welded caprock vitrophyre of the member is very thin in fault slices and blocks where it overlies tuff breccia near the cauldron wall, but locally the caprock vitrophyre grades downward into massive tuff breccia, possibly indicative of vent breccia related to the Tiva Canyon. The welded rhyolitic lower part thickens from a wedge-edge to nearly 300 m (1,000 ft) downdip, 3.2 km (2 mi) away from the cauldron wall (fig. 8). A 15-m (50-ft) stratigraphic transition zone between the lower flow-laminated, phenocryst-poor rhyolite of the main part of the Tiva Canyon and the overlying tuff of Chocolate Mountain is marked within the Claim Canyon cauldron segment by an increase of dark-gray phenocryst-rich collapsed pumice lenticles of quartz latite, similar to those in the overlying tuff of Chocolate Mountain. No xenoliths of the main rhyolitic part of the Tiva Canyon have yet been found in the tuff of Chocolate Mountain, although sparse xenoliths of the Yucca Mountain Member and older rocks are present. Because of these gradational relations and the lack of Tiva Canyon xenoliths, the tuff of Chocolate Moun- tain must be the upper partof one compound cooling unit that includes the widespread main sheet of the Tiva Canyon. The tuff of Chocolate Mountain contains three com- positionally similar subunits that were mapped by Christiansen and Lipman (1965); near the cauldron walls the subunits are separated by beds of yellow tuff breccia that thin and pinch out within a few kilometres northward (figs. 12, 13; Byers and others, 1976). The lower subunit is slightly more than 300 m (1,000 ft) thick and is largely gray devitrified tuff with granophyric pumice. The middle and upper subunits are brown and purple devitrified tuff locally underlain by basal vitrophyres indicative of partial cooling breaks. Petrographically, the tuff of Chocolate Mountain differs from the quartz latite caprock of the main part of the Tiva Canyon in containing abundant resorbed feldspar phenocrysts, a higher ratio of plagio- clase to total feldspar phenocrysts (fig. 9), and sparse rather than very sparse hornblende phenocrysts. Although there is an overlap in content of silica (fig. 9) and other consti- tuents (W. D. Quinlivan and P. W. Lipman, written commun., 1974) the petrographic differences indicate that the tuff of Chocolate Mountain is slightly more mafic than the main part of the Tiva Canyon outside the cauldron segment. At one locality within the cauldron segment, where a thick complete section is exposed in the south wall of Timber Mountain caldera (adjacent to the dikes between sections B—B’ and C—C’, fig. 12), rhyolitic Tiva Canyon grades upward through about 15 m (50 ft) into typical extracauldron quartz latitic caprock of the Tiva Canyon. The transition upward to the typical tuff of Chocolate Mountain, however, is faulted out. A tabular mass of fluidal flow—banded tuff of Chocolate Mountain 6 m (20 ft) thick dips moderately and is inter- calated with coarse tuff-breccia. Near the head of Claim Canyon (fig. 12), this mass shows chilling above and below and may be a dike. PAINTBRUSH TUFF AND ROCKS RELATED TO CLAIM CANYON CAULDRON OASIS VALLEY CALDERA SEGMENT V x \ s§~ s \ $3 @344 . 5“ 5‘ ‘\ [/1 . '\ EXPLANATION E Post-Rainier Mesa rocks Rainier Mesa Member and under- lying lava Lava feeder dike 36°55' Intracauldron tuffs of Chocolate Mountain and Pinyon Pass, and tuff breccia Pre-tuff of Chocolate Mountain, Paintbrush Tuff, and intercalated lava(A 0) 33 116° 30' I TIMBER MOUNTAIN CALDERA [/r ‘0 V, M. 1;;éféjéw \VZW’ ‘1 \‘ {'3 d?" \§ Ill/I \ , I’ll/11,, I“ Ill/III // [III] II II II /‘ ' //// w A 41/11/77,)!” “RE ’// // 'i‘,\ . /////'/;’ i, intracauldron e, extracauldron Pre-Paintbrush rocks Fault , Dotted where concealed B—B’ Geologic cross section FIGURE 12.—Generalized geologic map of Claim Canyon cauldron segment and vicinity, om geologic map of Timber Mountain caldera (Byers and others, 1976). cauldron than outside. Generalized fr TUI-‘F BRECCIAS WITHIN CLAIM CANYON CAULDRON SEGMENT Tuff breccias intertongue with and overlie the tuff of Chocolate Mountain (Christiansen and Lipman, 1966; Byers and others, 1976) but are not included with the Paintbrush Tuff. These breccias are more than 150 m (500 ft) thick along the south wall of the Claim Canyon cauldron segment but wedge out within a mile or two O 1 2 3 MILES l | | I l O I 2 3 KILOMETRES showing greater intensity of faulting of tuffs within north of the wall. Only the larger outcrops are shown in figure 12 (see Byers and others, 1976); the breccias shown in cross sections B—B’ and C-C’ in figure 13 are interpreteted partly as vent breccias near the south wall of the cauldron. Within about 150 m (500 ft) of the cauldron wall, however, faulting and, locally, silicification have complicated inter- pretation of the origin of the breccia; there it is probably a landslide breccia consisting of rocks derived from the cauldron wall. 34 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA The breccia from an area about 150 m (500 ft) to 0.8 km (0.5 mi) from the cauldron wall contains, in order of decreasing abundance, large blocks of intracauldron Yucca Mountain and Pah Canyon Member, pyroxene- bearing sanidine-rich lava like that between the members, and sparse amounts of both members of Crater Flat Tuff. North of Crater Flat (fig. 12) this breccia locally includes a few giant blocks of welded rhyolitic Tiva Canyon about 30 m (100 ft) thick and 90 m (300 ft) long. These blocks dip steeply and are intricately crackled. The nonwelded B 6000’ Post = RIM Paintbrush m,” 400' mm III 'Pre~P6_intbrush' _‘7 ' i ' O' 200 rocks SEA LEVEL ' . ‘ Possuble source vent of Tiva Canyon Member C Yucca Mountain CAULDRON Ill i2 \‘ I 6000I SEA LEVEL 2000' . matrix of the breccia is also modally rhyolitic Tiva Canyon and, moreover, grades upward in a few places to unaltered moderately to steeply dipping densely welded Tiva Canyon quartz latite or tuff of Chocolate Mountain. The nonwelded matrix is not crackled or highly fractured and the densely welded quartz latite locally occurs as steeply dipping vitrophyre lenses with contacts grada- tional into the matrix. This part of the tuff-breccia may be an explosion breccia genetically related to eruptive vents of the Tiva Canyon Member. Lipman’s (1976) model of CLAIM CANYON CAULDFION SEGMENT TIMBER B’ MOUNTAIN CALDERA Tpcm II: \ e \ L ‘l\ 6000' 4000' SEA LEVEL TIMBER MOUNTAIN CALDERA v 6000' 4000’ Paintbrush 2000' EXPLANATION IIIIII Tuff of Pinyon Pass \ . .. '- Tiva Canyon Member and related tuff breccia Tpcm, tuff of Chocolate Mountain Tpc, Tiva Canyon Member > v , tuff breccia '7 fl. Yucca Mountain Member Paintbrush Tuff A Pah Canyon Member Intercalated lava flows f SEA LEVE L 2000' ———::_: Fault — Showing relative move- ‘— f 1 2 ment during cauldron subsi- dence (1) and inferred later movement during cauldron resurgence (2). Faults dashed where inferred 0 2000 4000 6000 8000 FEET 0 1000 2000 METRES Bedded tuff of Paintbrush Tuff FIGURE 13.—Geologic sections through Claim Canyon cauldron segment, with critical units and faults projected above ground surface to indicate possible cauldron resurgence. Lines of sections shown in figure 12. Only major faults shown. PAINTBRUSH TUFF AND ROCKS RELATED TO CLAIM CANYON CAULDRON 35 caldera collapse breccias in tuff associated with a caldera— forming eruption would also fit the above observations—except for the presence of intracauldron blocks. Within the l50-m wide zone disturbed by faulting along the cauldron wall, a block of vertically dipping Topopah Spring Member was identified by thin-section examina- tion. The contact of this breccia with the cauldron wall is locally exposed in Claim Canyon; it dips about 60° cauldronward and could be interpreted as the sole of a landslide. In detail, the explosion tuff breccia differs from the breccias along the cauldron wall in that it is much less altered and is composed of about 50 percent blocks as much as 150 m (500 ft) long in a matrix of grayish—yellow tuff containing 0—20 percent angular clasts smaller than block size. More than 80 percent of the blocks are derived from Pah Canyon and Yucca Mountain Members. More— over, the blocks contain a light-gray devitrified facies of these members that is typical of these units inside the cauldron. This yellow tuff breccia is similar in appearance to that in rhyolite vent tuff breccia cones; moreover, the grayish-yellow frothy matrix of rhyolitic Tiva Canyon mineralogy locally grades into a dark-gray pumiceous scoria. We would interpret these yellow tuff breccias that contain intracauldron clasts as originally at or near vent sites, probably those that erupted the Tiva Canyon Member. Inward, about 2 km away from the cauldron wall (fig. 12), the tuff breccia thins and intertongues with the tuff of Chocolate Mountain and lies between the tuffs of Chocolate Mountain and Pinyon Pass. More than 3 km (2 mi) inward the tuff breccia tongues pinch out. These thin tongues of tuff breccia resemble the larger massive exposures to the south, except that the fragments are smaller and the tongues consist of a significant minor fraction of pre-Tiva Canyon rocks other than Pah Canyon and Yucca Mountain Members. TUFF OF PINYON PASS The tuff of Pinyon Pass is confined to the Claim Canyon cauldron segment and is here included as an informally named unit of the Paintbrush Tuff. The tuff is a multiple— flow simple cooling unit as much as 150 m (500 ft) thick, distinguished in the field by its generally moderate, rather than dense, welding, lack of vitrophyres, and strati- graphic position above the tuff of Chocolate Mountain. The tuff of Pinyon Pass is mapped with the tuff of Chocolate Mountain in figure 12, but is shown separately in section C-C’ (fig, 13), and on the geologic map (Byers and others, 1976). The lower nonwelded to partially welded rhyolitic zone, as much as 90 m (300 ft) thick, is commonly brilliant pink or light red where partially welded. Where the unit is thickest, however, the upper quartz latitic, most densely welded devitrified tuff is very much like the Chocolate Mountain in outcrop appearance; indeed, the flattened quartz latitic pumice lentils of the Pinyon Pass and Chocolate Mountain are identical in composition. The Pinyon Pass everywhere is concordant on the tuff of Chocolate Mountain or on tuff breccia that intertongues between the units. Very slight erosion of the upper part of the tuff of Chocolate Moun- tain before deposition of the tuff of Pinyon Pass is indicated locally by channeling and by absence of the non- welded top of the tuff of Chocolate Mountain. The tuff of Pinyon Pass is a compositionally zoned compound cooling unit and petrographically resembles the main part of the Tiva Canyon Member outside the Claim Canyon cauldron segment. The lower rhyolitic subunit, however, contains sparse dark, phenocryst-rich quartz latite lenticles, as much as 15 cm (6 in.) in length, typical of the underlying tuff of Chocolate Mountain. Hornblende is also very sparse or lacking (fig. 9) in the lower rhyolitic subunit of the tuff of Pinyon Pass, in contrast to the universal occurrence of hornblende and lack of biotite in the lower rhyolitic subunit of the Tiva Canyon Member. The middle unit is a high—silica rhyolite that contains a little hornblende and could be mistaken for either the lower or middle part of the Tiva Canyon. The upper crystal-rich compositional zone is more silicic than the analogous part of the Tiva Canyon (fig. 9) but con— siderable overlap in phenocryst percentages makes dis— tinction by means of thin—section modes highly uncertain. The stratigraphic position of the tuff of Pinyon Pass above the tuff of Chocolate Mountain is the best criterion for identification. Chemical analyses of the unit (fig. 9; W. D. Quinlivan and P. W. Lipman, written commun., 1974) corroborate the observed compositional zonation from rhyolite, upward to high-silica rhyolite, to low-silica rhyolite in the crystal-rich caprock. The close resemblance of the tuff of Pinyon Pass to that of the Tiva Canyon and the presence of Chocolate Mountain-type pumice lenticles in the lowermost part of the tuff of Pinyon Pass, despite slight local erosion of the underlying tuff of Chocolate Mountain, suggest that the tuff of Pinyon Pass was erupted within the Claim Canyon cauldron only a short time, perhaps within a few thousand years, after the Tiva Canyon Member was erupted. The lack of any bedded tuff breccia other than the single tongue that thins away from the cauldron wall suggests that no rhyolitic lava volcanism intervened between the tuff of Pinyon Pass and the underlying tuff of Chocolate Moun- tain. The tuff of Pinyon Pass, therefore, probably predates post-Tiva Canyon rhyolite lavas in the north wall of Timber Mountain caldera (fig. 8). POST-TWA CANYON RHYOLITE LAVAS Two rhyolite lavas related to but not part of the Paint— 36 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA brush Tuff are exposed in the north wall of Timber Mountain caldera (fig. 8) above the Tiva Canyon Member. These lavas are similar in physical appearance except that the lower flow has visibly more and larger biotite flakes. Petrographically, they are easily distinguished, for the upper rhyolite lava has sparse small quartz euhedra, whereas the more mafic lower rhyolite lava has none. These two flows were mapped separately in the Scrugham Peak quadrangle (Byers and Cummings, 1967). Outcrop thicknesses of the individual flows average about 150 m (500 ft), but within a few miles downdip under Pahute Mesa the quartz-bearing flow attains 347 m (1,140 ft) in a drill hole. The flows extend a few miles north and northeast under Pahute Mesa, where they have been penetrated in drill holes (fig. 8). Drill records indicate that approximately 64 km2 (25 mi?) of the mesa is underlain by either one or both of the rhyolites. The thickening of these lavas under Pahute Mesa, as inferred for the earlier lavas intercalated with the Paintbrush, is doubtless related to the existence of the Silent Canyon caldera, into which these later lavas also flowed. The lower biotite-bearing lava exposures in the north wall of Timber Mountain caldera (fig. 8) are instructive relative to the source of the lava and the location of the Claim Canyon cauldron wall. The lava has fused under- lying bedded tuffs and locally the basal vitrophyre zone cuts across the bedded tuff. Relict bedding visible below the flow within this vitrophyre indicates that the vitro- phyre is a fused tuff. Tongues of vitrophyric agglutinated ash-fall tuff are also intercalated in the bedded tuff near the fused zone, suggesting a nearby source for these pyro- clastics and the associated flow. N ortherly flow azimuths (Cummings, 1964; David Cummings, written commun., 1964) indicate the source to the south, probably just inside the present wall of Timber Mountain caldera. Petro- graphically, the lavas clearly are late effusives related to the Paintbrush Tuff (fig. 9). It is reasonable to assume that these flows were fed by radial fissures near the cauldron ring fault similar to feeders of the pre-Rainier Mesa lavas in the south wall of Timber Mountain caldera (fig. 12), thus providing additional corroborative evidence that the north wall of the Claim Canyon cauldron is probably buried within a few kilometres south of the north wall of Timber Mountain caldera. Petrographically and chemically, the two post-Tiva Canyon lavas are generally similar (fig. 9). Chemical an- alyses (W. D. Quinlivan and P. W. Lipman, written com- mun., 1974) reflect the minor petrographic differences between them. Probably of significance is the presence of euhedral quartz, ranging from 0.1 to 0.5 mm, amounting to several phenocrysts per thin section in the upper quartz- bearing flow. The small euhedral character of these quartz phenocrysts and their presence in all thin sections examined indicates that the magma from which the quartz-bearing flow was extruded had just attained the quartz-feldspar cotectic curve (Tuttle and Bowen, 1958), marking the change from the predominantly quartz-free volcanic rocks to overlying quartz-rich rocks of the Timber Mountain Tuff and related lavas. CAULDRON SUBSIDENCES RELATED TO ERUPTION 0F ASH-FLOW TUFF MEMBERS The fact that nearly all units of the Paintbrush Tuff and intercalated, petrologically similar lavas thicken greatly (table 3) within the Claim Canyon cauldron segment suggests accumulation within a caldera. The notable exception is the Topopah Spring Member, the basal unit of the Paintbrush Tuff; from study of other calderas in the area, we presume that it is present at depth beneath the thick infilling units of the cauldron (fig. 12). The great thickness of the post-Topopah Spring units within the Claim Canyon cauldron segment plus the areal distribution of the Topopah Spring Member (fig. 8) and also the presence of the middle xenolithic zone and the bedded upper vitrophyre within a few kilometres of the rim of the Claim Canyon segment (compare with Lipman, Christiansen, and O’Connor, 1966, fig. 18) indicate that the segment is an eroded portion of a cauldron whose earliest subsidence was probably associated with the eruption of the Topopah Spring Member. The northern boundary of this cauldron is probably not too far south of the present north wall of Timber Mountain caldera, for lavas related to the Paintbrush were poured northerly into Silent Canyon caldera (fig. 10) from vents at or just south of this wall. Additional suggestive evidence on the possible location of the north wall of Claim Canyon cauldron comes from a core taken at 1,402 m (4,600 ft) in drill hole UE18r. Debris flows associated with collapse of Timber Mountain caldera contain a few fragments of crystal-rich Topopah Spring caprock with granophyric pumice, typical of slowly cooled thick accumulations of ash-flow tuff within calderas. The occurrence of this granophyric Topopah Spring in the Timber Mountain caldera debris flows suggests that an intracauldron facies of the Topopah Spring Member was exposed within the present boundary of Timber Mountain caldera at the time of Timber Mountain caldera collapse. The northern limit of the collapse area associated with the eruption of the Topopah Spring Member, therefore, was probably north of drill hole UE18r, but south of the present north wall of Timber Mountain caldera (fig. 8). Cataclysmic subsidence of the Claim Canyon cauldron segment during eruption of the Tiva Canyon Member is indicated by the thick accumulation of the main part and the tuff of Chocolate Mountain of the Tiva Canyon within the cauldron. Within the Claim Canyon cauldron segment, the tuff of Chocolate Mountain intertongues with tuff breccias related to the Tiva Canyon (section C—C’, fig. 13), possibly of vent breccia origin, for in ——i PAINTBRUSH TUFF AND ROCKS RELATED TO CLAIM CANYON CAULDRON 37 many places they are modally like the Tiva Canyon and locally grade into welded Tiva Canyon quartz latite or tuff of Chocolate Mountain. These tuff breccias are coarse and thick just inside the cauldron wall but are thin and become finer away from the wall of the Claim Canyon cauldron segment. They were possibly largely derived from near- wall vents that were active during the emplacement of the Tiva Canyon. Moreover, these breccias contain pre-Tiva Canyon blocks of intracauldron facies as young as the Yucca Mountain Member, which immediately underlies the Tiva Canyon Member and may be considered an early eruptive phase of the Tiva Canyon. In a very few places within a kilometre of the cauldron wall, giant crackled blocks of dense welded Tiva Canyon occur in the massive tuff breccia and may have been autobrecciated in place during emplacement and subsequent collapse. Alter- natively, these blocks may have been an early solidified phase of the Tiva Canyon that calved off the caldera rim during late stages of eruption and collapse, similar to the origin of the caldera breccias of the western San Juan Mountains, Colorado (Lipman, 1976). Under either hypothesis, however, the cauldron tuff breccias that are seen today just within the southern border of the Claim Canyon segment are a product of volcanism related to the Tiva Canyon and not to the Topopah Spring Member. Indirect evidence also suggests that the Claim Canyon cauldron segment is a part of the collapse area related to eruption of the Tiva Canyon. The main part of the Tiva Canyon within the segment locally contains fluidal flow folds like a rhyolite lava, suggesting flowage while a caldera wall slope was being formed during emplacement and compaction related to welding. Only within the cauldron segment have flow folds been observed within the Tiva Canyon. The walls of the two caldera collapses associated with the eruptions of the Topopah Spring and Tiva Canyon Members are probably not far apart but are not readily identified because later structural movements in the vicinity of the cauldron wall have obscured evidence of earlier collapses. Evidence of the wall related to Topopah Spring collapse is probably preserved in the vicinity of Claim Canyon just outside the fault that repeats the thickened Pah Canyon and the rhyolite lava under it (section C—C’, fig. 12). Unfortunately, this part of the wall is covered by the alluvium flooring Claim Canyon. Unraveling the sequence of events along the cauldron wall probably can be accomplished only by detailed geologic mapping at a scale significantly larger than that (l:24,000) of the quadrangle maps (fig. 2). Specific collapses related to the eruptions of the Fab Canyon and Yucca Mountain Members are also difficult to document. They and the intercalated lavas have a much greater thickness, fivefold to sixfold, inside the cauldron than outside; the total thickness of the units between the Topopah Spring and Tiva Canyon Members is 180 m (600 ft) on the cauldron rim (table 3) compared to about 1,000 m (3,300 ft) within the cauldron segment. We would interpret from sections B—B’ and C—C’°(fig. 13) that the top of the Topopah Spring within the main part of the cauldron segment would be at least 820 m (1,000 m minus 180 m) below its top outside just after collapse related to the Yucca Mountain Member. The preceding line of argument becomes even more compelling when the difference in thicknesses of the Tiva Canyon, inside and outside the cauldron, namely about 800 m (2,600 ft), is added to the 820 m (2,700 ft) net difference in thickness between the Topopah Spring and Tiva Canyon Members (table 3). That is, assuming no resurgence of the cauldron, the top of Topopah Spring must have been more than 1,600 m (about 1 mi) below its altitude on the rim of the Claim Canyon cauldron after the eruption of the Tiva Canyon Member. It is difficult to con— ceive of this much cauldron subsidence relief on the Topopah Spring immediately after its eruption, because the unit has a relatively small volume (200 km3) outside the cauldron. The Claim Canyon cauldron more likely preserves a record of episodic subsidence (fig. 13) associated with eruption of each of the members of the Paintbrush Tuff. Analogous to the caldera relief of the Topopah Spring Member of the Paintbrush Tuff, the Grouse Canyon Member of the Belted Range Tuff within the Silent Canyon caldera is today about 2 km below its elevation outside the caldera (Orkild and others, 1968, 1969), owing to later subsidence related to eruptions of many post-Grouse Canyon units. In a related paper, Christiansen and others (1976) consider the regional structure and the overall pattern of distribution and thicknesses of members of the Paint— brush Tuff and suggest the Oasis Valley caldera segment, in contrast to the Claim Canyon cauldron segment, as the main area of collapse related to eruption of the Yucca, Mountain and Tiva Canyon Members. The differences between the conclusions expressed by these authors (four of whom are also authors of the present report) and those expressed here are minor and involve mainly emphasis on possible contrasting source areas between the two lower and the two upper ash-flow tuff members of the Paintbrush Tuff. POSSIBLE MAGMATIC RESURGENCE OF CLAIM CANYON CAULDRON The Claim Canyon cauldron segment, in contrast to nearly all other penecontemporaneous cauldron structures of the southwestern Nevada ash-flow field, today stands topographically high, like many central Nevada cauldron structures that have been elevated by both resurgence and basin-range faulting. The notable local exception is the Timber Mountain resurgent dome to be described later. The cauldron segment is not 38 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA structurally high in the sense that older rocks are exposed within it; in fact, generally younger rocks of the Paintbrush Tuff are exposed within the cauldron segment, analogous to the resurgent doming of the Ammonia Tanks Member of the Timber Mountain Tuff within Timber Mountain caldera. The present authors will attempt to point out some geometric relations that in our opinion suggest that magmatic resurgence of the Claim Canyon cauldron occurred prior to emplacement of the Rainier Mesa Member and its petrologically related underlying lavas, an idea already outlined in an abstract (Byers and others, 1969). Christiansen and others (1976) present structural evidence that the present position of the Claim Canyon cauldron segment is not due to magmatic resurgence related to Claim Canyon cauldron. The geologic map shown on figure 12, which is generalized from the geologic map of Timber Mountain caldera (Byers and others, 1976), is included in the present report merely to show the great intensity and trend of faults within the cauldron segment, especially in that part where the tuffs of Chocolate Mountain and Pinyon Pass (K-Ar age, 12.5 m.y.) of late Paintbrush age are exposed. The Rainier Mesa Member (K-Ar age, 11.3 my) and the underlying petrologically similar lava are affected mainly by basin-range faults that are approximately radial to the intensely faulted wall of the Claim Canyon cauldron segment; that is, the basin-range faults strike more north- easterly where the wall faults trend northwesterly at the west end of the segment. As pointed out by Christiansen, Lipman, Orkild, and Byers (1965, p. B46), movement on these radial basin-range type faults also occurred during broad doming of the Timber Mountain volcanic center, as well as during regional basin-range extension. Many of these radial faults can indeed be traced into the cauldron segment, but most terminate against the circumferential cauldron faults. Closely spaced circumferential faults and radial cross faults shown in figure 12 are localized within the cauldron segment and probably are representative of the much larger Claim Canyon cauldron, most of which is buried by the later collapses of Timber Mountain caldera and the Oasis Valley caldera segment. The strong intensity of this fault system confined within the cauldron segment is interpreted as a marginal fault pattern related to resurgent doming, and it bears close similarity to the fault pattern in the ring fracture zone at the periphery of Timber Mountain resurgent dome (Carr, 1964, fig. 2; Carr and Quinlivan, 1968, fig. 1; Byers and others, 1976). That this resurgence took place before the eruptions of pre-Rainier Mesa lavas and the Rainier Member is shown by the near absence of circumferential faults in these later units (fig. 12). Further strongly suggestive evidence that the Claim Canyon cauldron resurged following eruption of the Paintbrush Tuff is the structurally high position of the Pah Canyon Member and its reasonable former extension upward, prior to erosion (fig. 13). Also the main part and the tuff of Ch0colate Mountain of the Tiva Canyon are structurally higher within the cauldron than outside the cauldron or just inside the wall (section B—B’, fig. 13). Minor faults of very little displacement are not shown in the cross sections in order to emphasize the stratigraphy and structural position of the units. The present authors are also aware that these observed relations between the intracauldron and extracauldron rocks could also have come about, without magmatic resurgence, by extensional basin-range faulting related to regional right-lateral trans- current movement along the Walker Lane (Locke and others, 1940)——the cauldron segment remaining structurally high, while the radial extracauldron blocks subsided. We do not favor this alternative hypothesis as the local control, because the intense circumferential fault pattern within the cauldron segment indicates more move- ment than in the blocks outside and, moreover, is typical of an outer arcuate fault pattern of a magmatically resurgent dome. TIMBER MOUNTAIN TUF F AND ROCKS RELATED TO TIMBER MOUNTAIN CALDERA ORIGINAL DEFINITION AND REDEFINITION OF TIMBER MOUNTAIN TUF F The Timber Mountain Tuff (Orkild, 1965, p. A46—A47) was originally defined to include the composite sequence shown in the left column of figure 7. This sequence comprised the Rainier Mesa Member at the base successively overlain by the tuff of Cat Canyon, the tuff of Transvaal (tuff of Camp Transvaal of Lipman, Quinlivan, and others, 1966), and the Ammonia Tanks Member. Thin bedded tuff sequences at the base of the ash- flow members were included with the overlying member. The Timber Mountain Tuff is here redefined to include all quartz-bearing ash-flow tuffs and thin ash-fall tuffs erupted about 11 my ago (Marvin and others, 1970) from the Timber Mountain caldera center—the type area (Orkild, 1965). The tuff includes in ascending order, the Rainier Mesa and Ammonia Tanks Members of the original definition and, as here redefined, the tuff of Buttonhook Wash (Carr and Quinlivan, 1966) and the tuff of Crooked Canyon (Byers and others, 1976). The known areal distribution of the two newly added informal units is confined within the Timber Mountain—Oasis Valley caldera complex (fig. 7). Outside the complex, in all surface exposures, the Ammonia Tanks Member is the upper unit of the Timber Mountain Tuff, as in the original definition (Orkild, 1965). The Ammonia Tanks Member is redefined on the basis of field and petrographic studies to include the tuff of Cat Canyon and the tuff of Transvaal (fig. 7, first column; Orkild, 1965; Carr and Quinlivan, 1966; Lipman, Quinlivan, and others, 1966). TIMBER MOUNTAIN TUFF AND ROCKS RELATED TO TIMBER MOUNTAIN CALDERA 39 These units are equivalent to part or all of the Ammonia Tanks, as mapped in many of the 7%-minute quadrangles of the Nevada Test Site region (fig. 2). Thin sequences of bedded ash-fall tuffs that are intercalated with the ash-flow units are included in the Timber Mountain Tuff and were generally included with the overlying ash-flow unit on quadrangle maps of the Nevada Test Site (fig. 2). A bedded ash-fall sequence above the Ammonia Tanks Member was penetrated in a few drill holes in Yucca Flat (fig. 7, column 2). This bedded tuff, however, is less than 5 m (15 ft) thick and has not been recognized in surface exposures. Quartz phenocrysts as much as 3 mm long are common to abundant in both the ash-flow and associated thin ash- fall tuffs, in contrast to the absence or near-absence of quartz in the underlying and overlying stratigraphic units. An exception to this general statement is the upper- most several meters of the underlying bedded tuff, in which quartz crystals increase in abundance and size upward to the Paintbrush-Timber Mountain contact. Several lithologic units, closely associated with the Timber Mountain Tuff in genesis or in time, were not included in the original definition (Orkild, 1965) and are here likewise excluded—analogous to the exclusion of similar units from the Paintbrush Tuff. These units include debris flows that intertongue with the Rainier Mesa Member, alkali-calcic rhyolite lavas, and tuff dikes and other intrusives on Timber Mountain resurgent dome. The ash—flow tuff members and informal tuff units of the Timber Mountain Tuff and their relations to coeval debris flows, lavas, and intrusive rocks, are discussed on following pages. RAINIER MESA MEMBER The Rainier Mesa Member, the basal member of the Timber Mountain Tuff, was originally described by Hinrichs and Orkild (1961, p. D97) from excellent exposures at Rainier Mesa (fig. 14). These authors included the Rainier Mesa Member as the uppermost member of the Oak Spring Formation. Poole and McKeown (1962, p. C61 ), a year later, included the Rainier Mesa Member as the uppermost unit of their Piapi Canyon Formation of the Oak Spring Group. Still later, Orkild (1965, p. A49) defined Timber Mountain Tuff (fig. 7) and placed the Rainier Mesa Member at the base. 5 An isopach map of the Rainier Mesa Member generalized from Byers and others (1968) is shown in figure 14. The Rainier Mesa is more than 500 m (1,500 ft) thick, and the base is not exposed in the west wall of Timber Mountain caldera, where the unit is densely welded and is strongly devitrified with granophyric texture in the pumice. The coarse textural features suggest that at least another hundred metres may be present in the unexposed lower part in the west wall. A thickness of 396 m (1,300 ft) was penetrated in a drill hole on Pahute Mesa. The total volume of the Rainier Mesa Member, as calculated from the isopach map, is in excess of 1,200 km3 (300 mi?) The upper quartz latitic caprock of the Rainier Mesa Member was locally mapped along the wall of Timber Mountain caldera as the tuff of Falcon Canyon (Hinrichs and others, 1967; Orkild and others, 1969). This usage is not retained. This upper quartz latitic subunit is exposed within the caldera near the northeastern wall at test well 8 (fig. 14) and also in the bottom of Beatty Wash where it cuts the southwestern wall of the caldera. In both localities the quartz latite intertongues with debris flows (figs. 3 and 7). The debris flows pinch out northward in the south— west caldera wall and are absent where 450+ m (1,500+ ft) of Rainier Mesa Member is exposed in the Transvaal Hills (fig. 14). Here the quartz latitic subunit of the Rainier Mesa is slightly thicker than elsewhere outside the caldera, but only about 15 m (50 ft) of quartz latitic caprock was penetrated by drill hole UE18r. The upper quartz latitic subunit was emplaced during late stages of caldera collapse (Byers and others, 1969, p. 94—95), as exhibited by the intertonguing relations with the debris flows. The Rainier Mesa Member is a compositionally zoned compound cooling unit consisting of an extensive high- silica rhyolite tuff overlain with a partial cooling break by a considerably thinner quartz latitic tuff that is restricted to the vicinity of Timber Mountain caldera. The rhyolitic subunit, where exposed outside the caldera, is commonly pink nonwelded shard tuff at the base grading upward into densely welded black vitrophyre (Byers and others, 1968, p. 93). The vitrophyre grades upward into dense brown to pale—red devitrified tuff with common white flattened pumice lenticles, ranging from 2.5 to 10 cm (1—4 in. ). In the west wall of Timber Mountain caldera (fig. 14), the devitrified zone is coarser than elsewhere and the pumice is granophyric. In general, the devitrified zone grades upward into a light-gray crystalline zone with spherulitic sanidine and cristobalite in the pumice, typical of Smith’s (1960b) vapor-phase zone. The vapor-phase zone of the high-silica rhyolite subunit commonly grades within a few feet into a vitrophyric or dark-brown crypto— crystalline quartz latite, which in turn grades upward into a second light—colored vapor phase zone, visible in only a few places because of erosion or colluvial cover. Dark quartz latitic pumice lenticles in the quartz latite are some- what larger than the white rhyolitic pumice in the underlying rhyolite. Where the Rainier Mesa Member is 396 m (1,300 ft) thick in the drill hole in the west part of Silent Canyon caldera, the main cooling unit described above is underlain by a nonwelded phenocryst-poor pink shard tuff about 150 m (500 ft) thick. The phenocryst ratios are similar to those of the lower rhyolitic part of the over- lying main cooling unit of the Rainier Mesa. These two tuffs may be separate cooling units, and the Rainier Mesa, like the overlying Ammonia Tanks Member, may be a composite sheet (Smith, 1960a, p. 812). 40 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA a7? "7°I5' “7° 3, STONEWALL I MOUNTAIN EXPLANATION THICKNESS OF RAINIER MESA MEMBER 0—150 m 150—300 In 300—450 In \: >450 m 37? I5x PRE-RAINIER MESA RHYOLITE LAVAS High-silica rhyolite lava Rhyolite lavas of Windy Wash f\ M Feeder dike Limit of information other than UE18r caldera walls 0 Drill holes used for control ~ Nume bered holes discussed in text Line of geologic cross section 37° V ,. : BULLFROG HPLLSé ’Iumy..mm‘“’“""”v _ W‘Imunn” .3“ Mn, «WWW MOUNTAIN (U C l— 1"”: 2’ 112, § 0 5 10 15 20 MILES i; I I I I I I I I 0 5 10 15 20 KILOMETRES FIGURE l4.—-Generalized isopach map of the Rainier Mesa Member of Timber Mountain Tuff and The unique petrography of the Rainier Mesa Member distinguishes it from cooling units of the underlying Paintbrush Tuff, as well as from overlying units of the Timber Mountain Tuff (fig. 15). Analyses of 10 high-silica rhyolite samples show an average of about 77 percent silica, a percentage similar to that of the subunits of the TIMBER MOUNTAIN TUFF AND ROCKS RELATED TO TIMBER MOUNTAIN CALDERA 45' “6° 30' 15‘ u‘e° “.InIlItu-HIIH‘“”In N15 cquTv LINCOLN COUNTY SLEEPING / BUTTE CALDERA I SEGMENT // : / OASIS VALLEY *CALDERA 1;, mmmm J’Mm9_.___-___-_—_ TIMBER as é 5' 3 “”3 N? MOUNTAIN ‘0? MOUN AIN 5° I vs, v: Don/IE ~ 0 ' — —. -‘ , : CALDERA “mum WAHMONlE-SALYER VOLCANIC CENTER NEVADA TEST SIT E ~knu." ._x m mum“ Luthwp Wells s‘”"‘ H SPECTER ' , 5mm RANGE a» u, 30‘ E l \ I distribution of petrologically related, pros-Rainier Mesa lavas. Geologic cross section shown in figure 23. 5 Irv/— 100+ Lake / I / ARK COUNTY um», \ mm, ‘5 RANGER \ MOUNTAINS {J _OLN _COUN'LY_ \“mmun, — t «,H 41 37? I5 36° 45' Paintbrush Tuff (W. D. Quinlivan and P. W. Lipman, contains quartz, in contrast to crystal-poor quartz-free written commun., 1974, table 6; Lipman, 1966), The rhyolites of the Paintbrush Tuff. The quartz latitic Rainier Mesa rhyolite, however, is crystal rich and caprock contains a mixture of hornblende-rich and 42 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA ‘ ,, ,4 FELSIC PHENOCRYSTS AS PERCENTAGE A OF TOTAL PHENOCRYSTS ‘ STRATIGRAPHIC UNIT {SUBUNIT ‘ k f A) l‘ Id Number of thin sections (N) PHENOCRYSTS AS PERCENTAGE Quartz a I e Spar in parentheses OF TOTAL ROCK Plagioclase » l l l l l l l fil f Upper tuff (7) _ ‘ Tuffs of Crooked Canyon 4 i ’ i j) ’ L Lower tuff (8) _ m l : r a r l . l a ' . ) Quartz latite (14) ” Tuff of Buttonhook Wash P High-silica rhyolite (3) , A *‘l / t: Rhyolite (5) (mostly « S high-silica rhyolite) r “ s a Ammonia 30' Quartz latite (31) ‘ Tanks n‘ 1‘ Member 12’ Rhyolite (41) (includes 3 high-silica rhyolite) *4 E 3 ~« " _; Quartz latite (27) *i ‘ Pre-Ammonia Tanks 1 rhyolite lavas (9) ‘ Quartz latite (5) (intracaldera) Rainier Mesa Quartz latite (12) ‘ 1 Member (extracaldera) 4( ‘ High-silica rhyolite (37) v " ‘i { High-silica rhyolite ‘ Pre-Rainier lavas (5) (a A Mesa lavas _ fl Rhyollte lavas of Windy Wash (9) A l l l l I l l I 0 10 20 30 40 50 0 80 100 PERCENT 44 FIGURE 15,—Modal and silica ranges of units of Timber Mountain Tuff and petrologically related lavas in order of stratigraphic succession. clinopyro‘xene-rich mafic scoria pumices with sub- ordinate iquartz-rich rhyolitic pumice similar to that in the underlying rhyolite subunit (Byers and others, 1968, p. 95). The Rainier Mesa Member, like the Topopah Spring Member, differs from the overlying ash-flow units of the Timber Mountain Tuff in containing no sphene. The ratio of plagioclase to total feldspar phenocrysts is also 7 higher in the Rainier Mesa high-silica rhyolite than in the rhyolites of overlying ash-flow units of the Timber Mountain Tuff. This feldspar phenocryst variation is (1962), Lipman (1966), and w. D. Quinliva’n " similar to that in rhyolitic subunits of members of Paina brush Tuff (fig. 9). Sanidine phenocrysts in the high-silica“ rhyolitic subunit of the Rainier Mesa are relatively potassic and have cryptoperthite rims in contrast to micro: perthitic sodic sanidine and anorthoclase in the rhyolites,.. of younger cooling units of the Timber Mountain Tffff (O’Connor, 1963). 4 The Rainier Mesa Member of the Timber Mountain r Tuff is the same unit as cooling unit 4 of Cornwall arid Kleinhampl (1964, p. J10) in the Bullfrog Hills (fig. 14), TIMBER MOUNTAIN TUFF AND ROCKS RELATED TO TIMBER MOUNTAIN CALDERA 43 x r L MAFIC PHENOCRYSTS AS PERCENTAGE ‘ OF TOTAL PHENOCRYSTS b - Total mafics Hornblende NUMBER OF SILICA PERCENTAGE RANGE V . . . . B'°"‘e C""°°V’°Xe“e SPHENE PHENOCRYSTS (Recalculated H20 and Ca003 free) ‘_ (R) Resorbed PER TH|N SECTION Number of analyses (N) in parentheses 4——I——l——l I I I I I I I I I I 4 S 4% (0) .. IR) 4 El ‘ ® m (0) ‘ I fl I fl I I I I I I #I * I ‘ - I § (1) + -_ g I ‘ (1) I I I a I I I I I ‘ I I l . lg I0) L W 3 _ W L _ (R4) (18) V M m (24) ’ , m- :I h-—4 _\ (4) E (R) | I I gI I I I I I I H g H v (3) ———I——I—_I I I I I I I I I I I I - @ Egg I V (1) V w :3 # v m (6) V I § 3 I—'I ,. (10) I I I I I I I I I I d L I § (2) . § 3 T I I 4 I | I I I I I ( ) I I i‘ 40 5 10 15 0 2.5 5.0 0 10 20 64 66 68 70 72 74 76 78 PERCENT PERCENT b Modal analyses by F. M. Byers, W. D. Quinlivan, and J. T. O’Connor, and from Cornwall (1962, table 2). Silica percentages from Cornwall ' and P. W. Lipman (written commun., 1974). The K-Ar age is 11.1 m.y. (Marvin and others, 1970), and _. the thermal remanent magnetization is reverse (G. D. Bath, written commun., 1965). h AMMONIA TANKS MEMBER "1" he Ammonia Tanks Member was defined by Orkild (1965) as the uppermost unit of the Timber Mountain Tuff; its type locality is at Ammonia Tanks (fig. 16). ' Because of evidence to be presented in the following para- graphs, the Ammonia Tanks is here redefined to include .( V the tuff of Cat Canyon on Timber Mountain resurgent dome (Carr and Quinlivan, 1966) and the tuff of Transvaal (Orkild, 1965; tuff of Camp Transvaal of Lipman, Quinlivan, and others, 1966). (See fig. 7). An intracaldera section of the Ammonia Tanks is displayed at West Cat Canyon (fig. 16), where the tuff of Cat Canyon consists of a lower part of two mapped subunits, unit A and unit C (Lipman, Quinlivan, and others, 1966, section B-B’), which are more than 600 m (2,000 ft) thick and which are overlain by an upper part (unit E and unit F) about 300 m 44 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA 37° "7°I5' "7° 30' l t : | 22 STONEWALL um MOUNT JACKSON m MOUNTAIN EXPLANATION THICKNESSES OF AMMONIA TANKS MEMBER , ISOPACHS DASHED WHERE APPROXI- MATELY LOCATED 0—150 m 150—300 m 37? 300—450 In ‘5 450—600 In >600 m /////4 PIC-Ammonia Tanks rhyolite lavas —'”‘— Limit of information \“ ---------- Inferred buried wall of caldera that formed with eruption of Ammonia Tanks Member E 8r 0 Drill holes used for control — Num- bered holes discussed in text D> Geologic section 37 5 10 15 20 MILES l l l 5 10 15 20 KILOMETRES O——O FIGURE 16,—Generalized isopach map of Ammonia Tanks Member of Timber Mountain Tuff, showing great thickness within Timber Moun- shown in (1,000 ft) thick. An extracaldera section is in the Transvaal constitute the lower part of the Ammonia Tanks Member Hills (fig. 16; see also fig. 23), where two mapped subunits (Byers and others, 1976, section B-B’). The upper part of of the tuff of Camp Transvaal total about 150 m (500 ft) in the Ammonia Tanks at this section rests with slight thickness (Lipman, Quinlivan, and others, 1966) and angular discordance butwithoutacomplete cooling break TIMBER MOUNTAIN TUFF AND ROCKS RELATED TO TIMBER MOUNTAIN CALDERA “5° 30' I5‘ "6° OAK SPRING BUTTE s WAHMONIE-SA LYER VOLCANIC CENTER ‘_.A._NEflPA_.TET_$'T_E \. nmnl““‘\\:5PECTER ,, RANGE \_ Lothrop Wells : _"'7 45 Groom r— um- / I ,a./ / ‘l ’ LARK COUNTY 'V RANGER Wt ‘ MOUNTAINS 37° 15' 37° 36? "m 45 tain caldera, and associated lava flows that are petrologically similar to Ammonia Tanks high-silica rhyolite subunits. Geologic cross section figure 23. on the lower part of the Ammonia Tanks (Byers and includes the tuffs of Cat Canyon and Transvaal. The others, 1976, section B—B’). Ammonia Tanks is more than 900 m (3,000 ft) thick Arevised isopach map of the Ammonia Tanks Member beneath the Timber Mountain resurgent dome and (fig. 16; compare with Byers and others, 1968, fig. 5) slightly more than 450 m (1,500 ft) thick at Oasis 46 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA Mountain just inside the wall of the Oasis Valley caldera segment. Elsewhere, the member is generally less than 150 m (500 ft) thick. The total volume of the redefined Ammonia Tanks Member is about 900 km3 (230 mi3)—somewhat less than the volume of the Rainier Mesa Member. The following field and subsurface relations show that the Ammonia Tanks Member outside Timber Mountain caldera is the same cooling unit as the tuff of Cat Canyon on the Timber Mountain resurgent dome inside the caldera. The tuff of Buttonhook Wash overlies the Ammonia Tanks with complete cooling break both on the resurgent dome and on the west rim of Timber Mountain caldera. The stratigraphic succession penetrated in drill hole UE18r (fig. 16) on the buried north flank of the dome includes from top to bottom: the tuff of Cat Canyon (now part of the Ammonia Tanks Member), rhyolite lava, debris flows, and the upper part of the Rainier Mesa Member. This succession is the same as in surface exposures outside the collapse area of the caldera associated with the eruption of the Ammonia Tanks (fig. 16). The Ammonia Tanks in the drill hole is directly correlative with the lower part of the tuff of Cat Canyon in surface exposures several kilometres ( a few miles) south on Timber Mountain dome. The contact between the Ammonia Tanks Member and the overlying tuff of Buttonhook Wash is marked by about 10 cm (several inches) of fine ash-fall tuff at exposures within and outside the caldera. The contact zone with the underlying rhyolite lava consists of about a metre (a few feet) of bedded ash-fall tuff, both inside and outside the area of subsidence associated with the eruption of the Ammonia Tanks Member (fig. 16). Moreover, the rhyolite lava is a high—silica rhyolite, petrologically similar to high-silica rhyolite of the Ammonia Tanks (fig. 15). Where the rhyolite lava is missing, the Ammonia Tanks is separated from the Rainier Mesa Member by debris flows and bedded tuff inside the caldera and by not more than 20 m (70 ft) of thick—bedded tuff outside the caldera. The Ammonia Tanks Member, therefore, is tightly bracketed inside and outside the caldera by the underlying units and by the overlying tuff of Buttonhook Wash. The Ammonia Tanks Member is compositionally far more complex than the Rainier Mesa Member. The Ammonia Tanks is probably a composite sheet as defined by Smith (1960a, p. 812—813), for a local complete cooling break in the extracaldera Ammonia Tanks on Pahute Mesa (Noble and others, 1967) must be represented by one of several partial cooling breaks at the intracaldera section on Timber Mountain dome. The Ammonia Tanks Member is divided into upper and lower parts in order to show fault displacement on the geologic map of the Timber Mountain caldera area (Byers and others, 1976). These two parts can be further subdivided on the basis of composition into quartz latite and (high-silica) rhyolite, as shown in figure 15. The tuff of Cat Canyon (now part of the Ammonia Tanks), originally included six map units on Timber Mountain resurgent dome (Carr and Quinlivan, 1966). In general, our lower quartz latite includes Carr and Quinlivan’s units A and B; our lower rhyolite, their units C and D; our upper quartz latite, their unit E; and our upper rhyolite, their unit F; complex inter- tonguing of these units on the dome, however, makes direct correlation difficult in some localities. Chemical analyses of 25 intracaldera specimens and 23 specimens of extracaldera specimens were made available by W. D. Quinlivan and P. W. Lipman (written commun., 1974). The four compositional subunits of the Ammonia Tanks Member (fig. 15) are not everywhere present outside the area of Ammonia Tanks subsidence (fig. 16), and are not precisely correlative over the entire extent of the Ammonia Tanks. A few extra lenses of the upper quartz latite intertongue with the upper rhyolite on Timber Mountain dome (Carr and Quinlivan, 1966; 1968, p. 101). In the Dead Horse Flat quadrangle (fig. 2) on eastern Pahute Mesa, Noble, Krushensky, McKay, and Ege (1967) found a complete cooling break between their lower part, a quartz latite, and their main part, consisting of rhyolite overlain by quartz latite. In the area of the extracaldera section on the west rim of Timber Mountain caldera, Lipman, Quinlivan, Carr, and Anderson (1966) mappeda 10° angular unconformity between a lower rhyolitic part, which shows a compound cooling zonation (their tuff of Camp Transvaal), and an upper part, which consists of rhyolite and quartz latite (their Ammonia Tanks Member). These two parts are seemingly fused together in many places but locally a few centimetres of nonfused bedded tuff separates them (P. W. Lipman, oral commun., 1969). On the east flank of Timber Mountain dome a gentle angular discordance with a partial cooling break occurs between a faulted quartz latite and an overlying unfaulted local high-silica rhyolite, both within the upper part of the Ammonia Tanks (fig. 17). The lowest compositional subunit (fig. 15), where present, is a quartz latitic vitrophyre with sparse mafic lava xenoliths, especially in the western outcrop area of the Ammonia Tanks (fig. 16). This subunit extends eastward to Shoshone Mountain and eastern Pahute Mesa (fig. 16). In a few localities bordering the tuff dike zone on the eastern flank of Timber Mountain dome, the presumed equivalent subunit is a moderate brown devitrified tuff with conspicuous biotite; the base is not exposed. The overlying subunit, the rhyolite of the lower part(fig. 15), is typically a light-gray high-silica rhyolite with sparse mafic lava xenoliths and wedge-shaped phenocrysts of sphene. On Timber Mountain dome this subunit is compositionally gradational through hundreds of feet downward through low—silica rhyolite into the under- lying quartz latite. Peripheral to Timber Mountain in the northern moat TIMBER MOUNTAIN TUFF AND ROCKS RELATED TO TIMBER MOUNTAIN CALDERA 47 wag. FIGURE l7.—Structura1 unconformity between uppermost high-silica rhyolite and underlying quartz latite of upper part of Ammonia Tanks Member. North wall of tributary to Brushy Canyon on east flank of Timber Mountain resurgent dome. Timber Mounmin caldera wall in far right background. Timber Mountain dome had begun to rise (left or west side of photograph) before uppermost unit involved in doming was extruded, probably through tuff dike zone. The upper (light) unit here has nearly a complete cooling break with the faulted dark welded quartz latite tuff below, whereas a short distance to the west there is only a partial cooling break. area of Timber Mountain caldera (fig. 16), a thin extra- caldera rhyolite, less than 15 m (50 ft) thick, is composed of a pink nonwelded basal zone grading up through a few feet of partly welded gray vitrophyre into an upper light- gray devitrified zone. This rhyolite reappears as the lower subunit of the local main part on eastern Pahute Mesa (Noble and others, 1967). Thin-section modes of this subunit, a high-silica rhyolite, are pooled with the lower rhyolitic subunit of Timber Mountain dome (fig. 15) because the two subunits have similar stratigraphic positions within the Ammonia Tanks. This high-silica rhyolite is overlain by an upper quartz latite (as many as three quartz latites occur at this stratigraphic position on Timber Mountain dome) that can scarcely be dis- tinguished lithologically or petrographically from the lower quartz latite of the Ammonia Tanks (fig. 15). Locally between Ammonia Tanks and test well 8 (fig. 16) this upper quartz latite subunit becomes more mafic than elsewhere, having relatively large phenocrysts, including clinopyroxene, as much as 2.5 mm, and biotite books, as much as 4 mm, which are poikilitic with apatite and zircon. The uppermost subunit, the upper rhyolite (fig. 15), is found only on the east flank of Timber Mountain dome (fig. 17). The rhyolite is largely white, devitrified, and has spherulitic pumice, characteristic of the vapor phase zone. It is petrographically identical to most of the high-silica rhyolite dikes of the tuff dike zone and contains brown xenoliths of the lower quartz latite subunit, which the dikes intrude. The average of several K-Ar ages determined on the Ammonia Tanks Member is 11.1 m.y. (Kistler, 1968; Marvin and others, 1970). The Ammonia Tanks Member is equivalent to cooling unit 5 of Cornwall and Kleinhampl (1964, p. J10) in the Bullfrog Hills (fig. 16). Both the extracaldera and intracaldera facies of the Ammonia Tanks have a normal thermal remanent magnetization (G. D. Bath, written commun., 1965). TUFF OF BUTTONHOOK WASH The tuff of Buttonhook Wash of Carr and Quinlivan (1966) is a thin compositionally zoned cooling unit above the Ammonia Tanks Member and is here included with the redefined Timber Mountain Tuff. The best exposures 48 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA 116°30’ \ l llll,’ ‘ P A H U T E M E S A gull,” \ 3: \|"""""”"' Any/mu, \ \\\\\‘|Iv//: \ /, E_SLEEPING - 2: 2, BUTTE , \ mi: E 5/ (D S I E \u‘ H 9 E 5 5 ’5 0r ‘2 3, S 5 :1, A C; an? :: :: ,//’/I\\\“flm\ E Q , 37° \ 9 b ' ....... ‘5 oo' ‘ 5— BULLFB‘OG 3 HILLSg‘ , \V‘ 'ROSPECTQRSwo, 0“ S /’ ‘39 :\|'Iu\‘\\\ E 9 Q‘ \N‘ Q’ _\ - x S E 3/ \\"¢ ms 2, a o“ a», a " é E ’3 § 0°)“ 5 i , x"; I: a E E Z: 1; \\\\“\ ‘0 ’71, ‘3 I ”I i ::\C :_ 5: 7||I||\\\\‘H 4'? E /\\ 2/07 :3 ’1, 6 :0: I Hllé l5 0’4 p ’0, = = , ,, 2 ll///,, 4,0 :1 E I ”WAS“ o 5 CRATER , = ’1’) g FLAT 4/ 3 c, w, z 1 ’1’ o, "‘ : 3 (l) 1f) 2'0 MILES | l | O 10 20 KILOMETRES EXPLANATION _ 27A Intrusives cutting resurgent dome — Thickness of intracaldera tuffs — Dashed H1 Direction of dip in degrees where approximately located 70 A A’ Geologic cross section FIGURE 18.—Combined thicknesses of tuffs of Buttonhook Wash and Crooked Canyon and intrusive rocks on Timber Mountain resurgent dome. Tuff of Buttonhook Wash has not been recognized in eastern Timber Mountain caldera and overlying tuffs of Crooked Canyon have not been recognized in the Oasis Valley caldera segment. Geologic cross section shown in figure 22. are in Buttonhook Wash and in West Cat Canyon, on the southwest flank of Timber Mountain dome (fig. 18). The distribution of the tuff of Buttonhook Wash together with that of the overlying tuffs of Crooked Canyon is shown in figure 18. Both tuffs are largely confined to the western part of the Timber Mountain—Oasis Valley caldera complex. The tuff of Buttonhook Wash thins over Timber Mountain resurgent TIMBER MOUNTAIN TUFF AND ROCKS RELATED TO TIMBER MOUNTAIN CALDERA 49 dome, mainly by pinchout ofits quartz latite caprock (fig. 3). The thickness of the tuff of Buttonhook Wash on the western flank of the dome is about 60 m (200 ft); if the downdip thickening is projected under the caldera moat, the tuff may thicken to more than 150 m (500 ft). The tuff of Buttonhook Wash is found locally above the Ammonia Tanks Member in Oasis Valley and apparently just outside the Oasis Valley caldera segment in the eastern Bullfrog Hills, where the tuff may have flowed a short distance against an eroded wall. The tuff of Buttonhook Wash is not present in the intracaldera drill hole UE18r (fig. 18), nor in the eastern and northern moat areas of Timber Mountain caldera. The tuff of Buttonhook Wash is separated from the underlying Ammonia Tanks Member on Timber Mountain dome by about 10 cm of fine ash-fall tuff, and the lower part is glassy and nonwelded, indicating a brief but complete cooling break. On the west wall of Timber Mountain caldera, the tuff of Buttonhook Wash is also separated from the underlying Ammonia Tanks by about 10 cm of ash-fall tuff. On Timber Mountain dome the tuff of Buttonhook Wash is separated from the overlying tuffs of Crooked Canyon by a very slight angular uncon- formity that is probably related to resurgent doming of Timber Mountain. Less than 60 cm of bedded shard tuff separates the two units. The tuff of Buttonhook Wash consists of a lower light- gray moderately welded high-silica rhyolite and an upper reddish- to purplish-brown densely welded quartz latite that is locally vitrophyric. The high-silica rhyolite is indistinguishable from high-silica rhyolites of the Ammonia Tanks Member (fig. 15). The quartz latite caprock, however, is distinguished from Ammonia Tanks quartz latites by (1) plagioclase consistently in excess of alkali feldspar phenocrysts (fig. 15), (2) clinopyroxene in excess of biotite, and (3) all phenocrysts rarely exceeding 1.5 mm in contrast to phenocrysts as much as 4.0 mm long in the Ammonia Tanks. Sparse hornblende where the tuff is not oxidized also distinguishes the tuff of Buttonhook Wash from the upper quartz latite of the Ammonia Tanks. These petrographic features of the quartz latite caprock make the tuff of Buttonhook Wash an excellent strati- graphic marker. The high SiOz content of the quartz latite, nearly 73 percent (fig. 15), is probably due to slight silicification 0f the devitrified rock. One K-Ar age of 10.5 m.y., determined on sanidine by Kistler (1968, p. 254), is available from the tuff of Button- hook Wash. Kistler (1968) referred to the unit as “tuff of Cat Canyon, upper cooling unit,” a logical designation at that time. This K-Ar age is believed to be slightly young because of slight sericitization of the rock (Marvin and others, 1970, p. 2666). The slight thinning of the tuff onto the Ammonia Tanks Member (average K-Ar age of 11.1 m.y.) over the resurgent dome suggests that resurgent doming had already started before eruption of the tuff of Buttonhook Wash. However, this slight angular dis- cordance is not believed to represent a major time break, for slight angular discordances without a complete cooling break are present within the Ammonia Tanks Member on the east flank of the dome (fig. 17). TUFFS OF CROOKED CANYON The tuffs of Crooked Canyon are here included as an uppermost intracaldera unit of the redefined Timber Mountain Tuff (fig. 7) asshown on the map of the Timber Mountain caldera area (Byers and others, 1976). On the east flank of Timber Mountain dome the unit was mapped as tuff of Buttonhook Wash by Carr and Quinlivan (1966) and by Byers, Rogers, Carr, and Luft (1966). The exposed nonwelded distal edges of two ash—flow tuffs constituting the tuffs of Crooked Canyon are each less than 8 m (25 ft) thick where they onlap the northeast flank of the dome (fig. 18) in Crooked Canyon. Downdip in the subsurface off the northern flank of Timber Mountain, the tuffs thicken rapidly to more than 250 m (800 ft) toward the outer edge of the caldera and become partly welded; they were penetrated in the interval from 335 to 593 m (1,100 to 1,945 ft) in drill hole UEl8r in the northern moat of the caldera (figs. 3, 18). The total thickness of the upper tuff in the drill hole is about 180 m (600 ft); the lower tuff, 75 m (245 ft) thick, was not cored. The great increase in thickness, from the exposed flank of Timber Mountain dome outward to drill hole UE18r, indicates that the tuffs of Crooked Canyon were emplaced after the resurgent central dome was well advanced. The tuffs of Crooked Canyon overlie the tuff of Button- hook Wash or the Ammonia Tanks Member with a minor angular unconformity (fig. 3) and overlie the tuff of Buttonhook Wash in the central graben of the Timber Mountain resurgent dome. The contact zone between the tuffs of Crooked Canyon and the underlying Ammonia Tanks Member or tuff of Buttonhook Wash consists of l m or less of uniformly finegrained ash-fall shard tuff; more- over, the contact zone between the two ash-flow tuffs of Crooked Canyon contains 10—60 cm of fine ash-fall tuff, in contrast to as much as 18 m of variegated thin-bedded fine to coarse ash-fall tuff separating the tuffs of Crooked Canyon from the overlying post-Timber Mountain cooling units of the caldera fill (Byers and others, 1976). No compositional zoning in either of the tuffs of Crooked Canyon is apparent. The lower tuff is a simple nonwelded to moderately welded cooling unit of uniform rhyolitic composition, almost everywhere devitrified and pinkish to purplish gray, except at the distal edges flanking Timber Mountain dome. It contains fewer and smaller phenocrysts, with slightly more mafic pheno- crysts than the high-silica rhyolite in the upper part of the Ammonia Tanks (fig. 15). Near the tuff dike zone, it 50 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA contains distinctive brown xenoliths characteristic of the lower quartz latite of the Ammonia Tanks Member in the tuff dike zone. A few thin tuff dikes (fig. 18), identical in composition to the lower tuff, also contain these brown xenoliths, indicating the probable source of the tuff (see footnote description in Byers and others, 1966). The upper tuff is orange brown, glassy, and nonwelded near its pinchout on the northeast flank of Timber Mountain dome and likewise contains brown xenoliths from the lower quartz latite of the Ammonia Tanks in the tuffdike zone, suggesting a possible source similar to that of the lower tuff. The upper tuff contains, in addition, diagnostic phenocryst-poor glass xenoliths and phenocryst-rich pumice lumps. In drill hole UE18r, the upper tuff is partly welded and, except for the glass xenoliths and its stratigraphic position above the lower tuff and the Ammonia Tanks, is almost indistinguishable petrographically from similar parts of the Ammonia Tanks (fig. 15). The K-Ar age of the tuffs of Crooked Canyon is bracketed by the 10.5-m.y. age of the tuff of Buttonhook Wash and by the K-Ar age of 9.5 my on the overlying tuff of Cutoff Road (Lipman, Quinlivan, and others, 1966; tuff of the caldera fill of Kistler, 1968, p. 254). The 10.5-m.y. K- Ar age of the tuff of Buttonhook Wash, however, is probably slightly young because of alteration of the rock (Marvin and others, 1970, p. 2666). More likely the K-Ar age of the tuffs of Crooked Canyon is close to 11 my, as suggested by the stratigraphic relations: less than 60 cm of ash-fall occurs between the tuffs of Crooked Canyon and the underlying units of the Timber Mountain Tuff, whereas about 18 m (60 ft) of thin-bedded ash-fall tuff separates the overlying, 9.5-m.y., unit from the tuffs of Crooked Canyon near their pinchout on the east flank of Timber Mountain dome. On the west side of the Oasis Valley caldera segment, about 300 m (1,000 ft) of volcanic rocks, comprising bedded tuff, three ash-flow tuff cooling units, and a lava flow, occurs within this same stratigraphic interval in caldera fill (fig. 23). Moreover, the cooling units within the 300-m caldera-fill interval are petrologically dissimilar from the underlying quartz- bearing tuffs of Crooked Canyon and other units of the Timber Mountain Tuff as herein defined. From this strati— graphic and petrologic evidence, the tuffs of Crooked Canyon apparently are close in K-Ar age to the Ammonia Tanks Member, possibly about 11 my The petrologic and inferred close-time relations between the upper units of the Timber Mountain Tuff are consistent with the relatively small volume of the tuffs of Crooked Canyon, their lack of compositional zoning, their generally nonwelded to partly welded character, and their stratigraphic position as the youngest tuffs of the Timber mountain eruptive sequence. All these features indicate a marked decrease in the eruptive activity of the Timber Mountain magma chamber (fig. 3) following voluminous eruptions of the Rainier Mesa and Ammonia Tanks Members and the onset of resurgent doming of Timber Mountain. DEBRIS FLOWS The term “debris flows” as used here designates clastic deposits emplaced by rapid flowage of coarse and fine detritus. Such deposits are exposed within Timber Mountain caldera near the eastern and northern walls and within the Oasis Valley caldera segment in the Transvaal Hills and near the west wall just west of Oasis Mountain (figs. 14, 16). The unit overlies and locally intertongues with the uppermost part of the Rainier Mesa Member in the vicinity of test well 8 and in Beatty Wash where it crosses the southern end of the Transvaal Hills (fig. 14) but is not considered part of the redefined Timber Mountain Tuff. The debris flows underlie the Ammonia Tanks Member and pre-Ammonia Tanks rhyolite lava in drill hole UE18r (fig. 3). The unit has not been found outside Timber Mountain caldera or the Oasis Valley caldera segment. The debris flows are best exposed on the east side of the Timber Mountain caldera south of test well 8 (fig. 16) where the unit consists mostly of blocks of welded tuff as much as 6 m (20 ft) long in a yellowish-gray to grayish- orange tuffaceous matrix. The lower part of the deposit is nonsorted and nonstratified and apparently represents a single debris flow, but the highest part of the unit contains well-sorted and bedded lenses of reworked tuff which separate several thinner flows. These thinner flows contain more fine tuffaceous matrix with fewer angular cobbles than does the single flow, and the thinner flows may have been emplaced as successive mudflows. All rock types present in the caldera wall, including the lower rhyolitic subunit of the Rainier Mesa Member, are present in the coarse detritus. Densely welded tuff of the Paint- brush Tuff, Grouse Canyon Member of the Belted Range Tuff, and rhyolite of the lower part of the Rainier Mesa Member occur as blocks in the debris flows along the east wall of Timber Mountain caldera; fragments of rhyolite lavas predominate inside the north wall where pre- and post-Tiva Canyon rhyolite lavas are exposed. In the southern part of the Transvaal Hills, fragments of intra- cauldron Paintbrush Tuff are most abundant and indicate a source from the Claim Canyon cauldron segment. The debris flows are exposed in places inside the Oasis Valley caldera segment west and south of Oasis Mountain and contain an assemblage of blocks similar to those of the debris flows inside Timber Mountain caldera. Debris flows related to caldera collapse were penetrated in drill hole UE18r from a depth of 1,183 to 1,442 m (3,880 to 4,730 ft). They are overlain by pre-Ammonia Tanks rhyolite lava and underlain by the Rainier Mesa Member. Blocks many feet across are contained in a tuffaceous matrix. Most of the debris is welded tuff of the Paintbrush Tuff, but rhyolite lava intercalated with Paintbrush and TIMBER MOUNTAIN TU FF AND ROCKS RELATED TO TIMBER MOUNTAIN CALDERA 51 blocks of rhyolitic Rainier Mesa are also present. No rocks older than Paintbrush Tuff were identified. The stratigraphic position of the debris flows, their intertonguing relation with the uppermost Rainier Mesa quartz latitic tuff, the presence of rhyolitic blocks from the lower part of the Rainier Mesa, and finally the location of the exposed ends of the flows just within caldera walls all indicate that the bulk of the debris flows were generated catastrophically within a short period of time as a result of caldera collapse related to the eruption of the Rainier Mesa Member. The debris flows contain assorted discrete blocks of the lower rhyolitic subunit of the Rainier Mesa, indicating the presence of already solidified Rainier Mesa in the caldera wall. This relation further suggests that the Rainier Mesa Member is a composite ash-flow sheet in Smith’s (1960a) terminology. LAVAS PETROLOGICALLY RELATED TO TIMBER MOUNTAIN TUFF PRE-RAINIER MESA LAVAS Lava flows underlie the Rainier Mesa and Ammonia Tanks Members. The lava flow immediately under each member is petrographically similar to the member which overlies it. The pre—Rainier Mesa lavas (fig. 14) are divisible into two petrographic types: low—silica rhyolite lavas with sphene—the rhyolite of Windy Wash of Christiansen and Lipman (1965)—and high-silica rhyolite lavas without sphene and with few mafic pheno- crysts, petrographically and chemically like the high- silica rhyolitic tuff of the Rainier Mesa Member (fig. 15). The stratigraphic relations between these two rhyolites are not certain, but the fact that the high-silica rhyolite is petrographically more similar to the overlying Rainier Mesa Member than to the rhyolite of Windy Wash suggests that the high-silica lava postdates the rhyolite of Windy Wash. Moreover, the rhyolite of Windy Wash bears some petrologic affinity to the underlying Tiva Canyon Member of the Paintbrush Tuff and to the post-Tiva Canyon rhyolite lavas, mainly in sphene content and in the composition of opaque iron-titanium oxides (Lipman, 1971). The rhyolite lavas of Windy Wash commonly have a dark basal vitrophyre overlain by a pale- red devitrified interior. Feeder dikes of these flows are exposed as radial dikes in the south wall of Timber Moun- tain caldera (Byers and others, 1976), and one dike can actually be traced into the flow on the rim of the caldera. A rhyolite lava identical in lithology and chemistry (W. D. Quinlivan and P. W. Lipman, written commun., 1974) to the rhyolite of Windy Wash was penetrated in two drill holes in Silent Canyon caldera (fig. 14). This lava was called “quartz-rich lava of Scrugham Peak quadrangle” by Byers and Cummings (1967) and by Orkild, Sargent, and Snyder (1969; see composite diagram of their usage in fig. 7). The nearly identical thin-section modes of flows in these two areas on opposite sides of Timber Mountain caldera are shown in figure 19. The close similarity in composition of the lavas in these two areas suggests that the lavas were erupted at about the same time from the same compositional layer in the underlying magma chamber. The high-silica rhyolite lava crops out on the south- west side of the Timber Mountain—Oasis Valley caldera complex in the vicinity of Beatty (fig. 14). It has a typical glassy envelope with a light-gray devitrified interior. The lava is rather sparsely porphyritic and contains about equal numbers of quartz and sanidine phenocrysts; the sanidine has cryptoperthite(?) reaction rims similar to SOUTH 17 miles NORTH FLANK FLANK (5 specimens) (4 specimens) 3, 30 c —— — RANGE WITHIN L3 SAMPLE SUITE 5' 2 20 7 _ Sphene (U .. g g Apatite B ‘ 1O — 4 \ — V . § § le’ch E I § § Allanite (perrierite?) O .3 5— —— _ > s 4 — w E .3 3, _ Biotite '5 _ _ V E, 2 Hornblende a 1» — 8 0°: 0 m 60 — —— — 3 2 g E Alkali feldspar b g (Percentage of phenocrysts) g E 40 m _ _ z . Q‘s Plagloclase 3 % 'Core .Core (Percentage of phenocrystsl a: 93 U) z .2 £3 20— IRim IR' ‘ I Estimated mol percent g 3 'm anorthite in plagioclase g: ‘0 phenocrysts E 0 30 E E I Total phenocrysts I g (Percentage of total rock) N H C 20 o a, \ : '5. Quartz 0 ~23 (Percentage of phenocrysts) % a E .2 10 Total mafic phenocrysts g 8 (Percentage of phenocrysts) u) \— m at 0 FIGURE 19.—Phenocryst mineralogy of rhyolite lava of Windy Wash (pre-Rainier Mesa lavas) on opposite flanks of Timber Mountain caldera. Lava flow on north flank penetrated in drill holes (figs. 3, l4). 52 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA those of sanidine phenocrysts in the Rainier Mesa Member. The pre-Rainier Mesa rhyolite lavas correlate in part with “rhyolite flows and intrusives” of Cornwall and Kleinhampl (1964, p. J11; pl. 4). A K-Ar age determined on the high—silica rhyolite under the Rainier Mesa Member east of Beatty is 11.3 m.y. (Kistler, 1968, p. 255), the same age as the Rainier Mesa Member (Marvin and others, 1970, p. 2261). PRE-AMMONIA TANKS LAVAS IN TIMBER MOUNTAIN CALDERA MOAT The pre-Ammonia Tanks lavas are exposed near the north and southwest walls of Timber Mountain caldera (fig. 16) beneath thin glassy exposures of the Ammonia Tanks Member. These outcrops are in the outer annular moat which subsided during the eruption of the Rainier Mesa Member but which was not affected by collapse related to the eruption of the Ammonia Tanks Member (fig. 16). The lavas in the northern moat area of Timber Mountain caldera were called “rhyolite lavas of Timber Mountain caldera moat” on the map of the Scrugham Peak quadrangle (Byers and Cummings, 1967; fig. 2). In addition to the surface exposures, a lava flow was penetrated in the interval from 1,082 to 1,183 m (3,550 to 3,880 ft) in drill hole UE18r (figs. 3, 16) between the Ammonia Tanks Member and the underlying debris flows which partly filled the collapse area related to eruption of the Rainier Mesa Member. The pre-Ammonia Tanks lavas are glassy in the lower parts and have subjacent locally fused bedded tuff. The interiors of the flows are purplish gray laminated to light gray microcrystalline with conspicuous quartz pheno- crysts. The upper parts are light gray, porous, glassy, and commonly brecciated, typical of most rhyolite lavas. The lavas contain sphene and are petrochemically (W. D. Quinlivan and P. W. Lipman, written commun., 1974) generally similar to rhyolitic parts of the Ammonia Tanks but contain fewer phenocrysts. Thin-section modes (fig. 15) of the northern lava are the same as those of the southern lava (fig. 16), indicating a close magmatic and temporal relationship similar to that between the northern and southern bodies of the rhyolite of Windy Wash of pre-Rainier Mesa age (fig. 19) already described. The cycle consisting of eruption of lava followed by eruption of tuff petrologically similar to the lava appears to have been repeated. A core of the uppermost zeolitized brecciated part of pre- Ammonia Tanks rhyolite lava from drill hole UE18r contained about 1 percent phenocrysts, consisting of sparse quartz, sanidine, and 1 phenocryst of sphene. This nearly aphyric lava may have been extruded at a slightly different time than the lavas exposed in the northern and southern moat areas of Timber Mountain caldera, but the sample may not be truly representative. The age of the pre-Ammonia Tanks rhyolite lavas is bracketed by K-Ar ages of 11.3 m.y., determined on the Rainier Mesa Member (Marvin and others, 1970), and 11.1 m.y., determined on specimens of the Ammonia Tanks Member of the Timber Mountain Tuff (Kistler, 1968). The thermal remanent magnetization is reverse like that of the underlying Rainier Mesa but unlike that of the overlying Ammonia Tanks Member (G. D. Bath, written commun., 1965). INTRUSIVE ROCKS AND RELATION TO RESURGENT DOMING Intrusive rocks on Timber Mountain resurgent dome that are largely post-Ammonia Tanks comprise an outer tuff-dike zone, an inner microgranite-porphyry ring dike (Carr, 1964, p. B17), and intrusive rhyolite mainly in the central part of the dome (fig. 18). These intrusive rocks are petrologically closely similar to either high-silica rhyolitic or quartz latitic subunits of the Timber Mountain Tuff; these similarities, together with position and field relations of the intrusive rocks on Timber Mountain resurgent dome, imply a close genetic relation to the Timber Mountain Tuff. Presumably the tuffs, the related lavas, and the intrusive masses were derived from the same underlying magma that resurged and caused the doming of Timber Mountain. The tuff dike zone crops out on the outer eastern flank of the Timber Mountain dome and parallel to it (fig. 18). The dikes dip steeply inward toward the center of Timber Mountain (fig. 20). They are mostly high-silica rhyolite in composition and range in thickness from 0 to 8 m (26 ft). They intrude a distinctive brown devitrified facies of lower quartz latite of the Ammonia Tanks in an area about 150 m (500 ft) wide. Some dikes appear to be feeders to the uppermost high—silica rhyolite subunit of the Ammonia Tanks, for this subunit is not only petrographically identical with these dikes (fig. 21), but also contains the same distinctive brown xenoliths of the lower quartz latite subunit of the Ammonia Tanks. These dikes have a sharp contact with the Ammonia Tanks but show no evidence of chilling, indicating that the Ammonia Tanks was still hot at the time of intrusion. A few dikes are petrologically similar to the two tuffs of Crooked Canyon, which also contain xenoliths of Ammonia Tanks quartz latite. The slightly finer grain size of these dikes, at the contact indicates chilling and suggests that the dikes were intruded after the Ammonia Tanks had cooled somewhat. Although it is not possible to distinguish petrographically between dikes that may have been feeders for either the lower or upper tuff of the tuffs of Crooked Canyon, the presence of xenoliths in the tuffs indicates that their vent source was probably the tuff dike zone. No modal ranges of these dikes are shown in figure 21, because the feldspar phenocrysts are too altered to allow distinguishing sanidine and plagioclase. TIMBER MOUNTAIN TUFF AND ROCKS RELATED TO TIMBER MOUNTAIN CALDERA FIGURE 20.——Rhyolitic tuff dike cuts basal quartz latite of Ammonia Tanks Member and dips inward (westerly) along arcuate zone around the east flank of Timber Mountain dome. Dikes were prob- ably feeders for uppermost high-silica rhyolite of Ammonia Tanks and tuffs of Crooked Canyon. Flattened pumice (P; barely visible) and biotite foliation in quartz latite host rock are parallel to walls of dike. Note fragments of dark quartz latite tuff in dike and white (recrystallized) nonflattened pumice in quartz latite host rock. Ex- posure in Brushy Canyon (fig. 18). The microgranite porphyry ring dike consists of several alined discordant bodies (fig. 18) that intrude the lower part of the Ammonia Tanks Member. The dike is in an inner ring-fracture zone closer than the tuff dike zone to the apex of the Timber Mountain dome. Here the Ammonia Tanks is extremely thick (fig. 16) and may have extended as much as 600 m (2,000 ft) above the highest level penetrated by the ring dike. The contacts of the various bodies are abrupt, and the porphyry shows no change in texture or mineralogy at the borders. The lack of chilled borders suggests that the Ammonia Tanks host rock was still hot when the dikes were emplaced. Several of the dikes and irregular bodies broaden downward, presumably to connect at depth. The ring-dike system is generally parallel to an arcuate fault zone, a short distance to the southeast. Both the arcuate fault zone and also the ring dike dip steeply southeast (fig. 18), away from the central part of Timber Mountain resurgent dome. The intrusion of the porphyry dike probably accompanied the resurgent doming of Timber Mountain (Carr, 1964; Carr and Quinlivan, 1968, p. 103—104). The microgranite porphyry is nearly uniform in general appearance, texture, and color; small bodies differ only 53 slightly in texture from the largest. It is medium- to light- gray nonfoliated fairly homogeneous porphyritic rock containing about 50 percent phenocrysts of abundant feldspar as much as 1 cm across, minor biotite, and clino- pyroxene (fig. 21) in a very finely granular groundmass. Quartz and sphene phenocrysts are absent. The large feldspar phenocrysts consist of small plagioclase cores less than 3 mm in size, rimmed by a thick jacket of alkali feldspar. The groundmass is largely quartz and alkali feldspar, 0.1 to 0.2 mm in size and partly as granophyric intergrowths. The microgranite porphyry is almost chemically identical with the more mafic quartz latitic upper part of the Ammonia Tanks that contains 66-68 percent silica (W. D. Quinlivan and P. W. Lipman, written commun., 1974). The microgranite porphyry, however, had attained a more advanced stage of crystal- lization, as evidenced by (1) larger and slightly more abundant phenocrysts (fig. 21) and (2) more abundant alkali feldspar in the form of thick outer jackets enclosing plagioclase in the larger phenocrysts. Mafic quartz latites of the Ammonia Tanks and also the microgranite porphyry contain little if any quartz and sphene. The name “microgranite porphyry” was applied to these rocks prior to obtaining chemical analyses (W. D. Quinlivan and P. W. Lipman, written commun., 1974). In addition to microgranite porphyry dikes, pipelike and pluglike bodies of rhyoli te, and probably quartz latite, intrude the upper part of the Ammonia Tanks Member on Timber Mountain dome. Contact zones are commonly glassy and gradational through a horizontal zone of as much as 90 m (300 ft); considerable mixing and remobilization occurred between the glassy intrusives and the tuff subunits of the Ammonia Tanks. A few intrusives are coextensive with small extrusive domes that rest on, and show a complete cooling break from, the upper part of the Ammonia Tanks. Two petrologic types in different structural settings on the dome are included in the intrusive rhyolite. The rhyolite of Parachute Canyon (fig. 18), probably a quartz latite, is on the north flank of the dome near the inner ring-fracture zone, which is not well exposed here. In contrast, the rhyolite lavas of East Cat Canyon are composed of high-silica rhyolite, and were extruded from vents along faults of the central graben of the resurgent dome. The rhyolite of Parachute Canyon is a plug dome, as evidenced by vertical foliation near steep contacts. Vertical foliation is lacking in areas where the rhyolite reached the surface and flowed onto the upper part of the Ammonia Tanks Member. The interior of the body is light gray and porphyritic, and contains abundant small lithophysae, but it is encircled by a darker, finer grained border zone that is locally glassy, particularly on its northwest side where it apparently flowed onto the Ammonia Tanks. The border zone appears to consist of a few hundred metres of 54 TIMBER MOUNTAIN'OASIS VALLEY CALDERA COMPLEX, NEVADA ROCK UNIT } SUBUNIT Number of thin sections (N) in parentheses PHENOCRYSTS AS PERCENTAGE OF TOTAL ROCK FELSIC PHENOCRYSTS AS PERCENTAGE OF TOTAL PHENOCRYSTS Quartz lg Alkali feldspar Plagioclase East Cat Canyon; high silica (3) Intrusive rhyolite Parachute Canyon (3) Microgranite porphyry ring dike (4) Tuff-dike zone high — silica rhyolite (4) High — silica Ammonia Tanks rhyolite (5) Member - — Upper part Quartz Latite (31) PERCENT FIGURE 2l.—Modal and silica ranges of intrusive rocks on Timber Mountain compared with ranges of high-silica rhyolite and quartz latite Ammonia Tanks near the apex of the Timber Mountain resurgent dome (fig. 18). The pipes are localized along the faulted margin of a northwest-trending central graben, described by Carr and Quinlivan (1968, p. 105). The central parts of the pipes are light gray and devitrified, whereas the border zones are generally glassy or finely spherulitic. In a few places a crystallized transitional border zone as much as 15 m (50 ft) wide contains alternating bands of rhyolite having different phenocryst contents and probably represents a zone of intermixed intrusive high-silica rhyolite and remobilized quartz latite wallrock of the Ammonia Tanks Member. The modal range of three specimens and the silica content of 76.8 percent approach the composition of a typical high-silica rhyolite of the Ammonia Tanks Member (fig. 21). The inferred relations among the Ammonia Tanks Member in the Timber Mountain resurgent dome, the petrologically related intrusives, and an inferred under- lying compositionally zoned magma chamber are shown in figure 22. The two-layer compositional zoning of the upper part of a large magma chamber is based on models suggested by Quinlivan and Lipman (1965), by Lipman, Christiansen, and O’Connor (1966), and by Smith and Bailey (1966). The inward-dipping tuff dikes formed from high-silica rhyolite magma near the apical part of the magma chamber and are somewhat analogous to cone sheets (Anderson, 1936). The rhyolite intrusives of East Member of Timber Mountain Tuff. Silica analyses from mixed flow-banded rhyolite and probably fused mobilized quartz latite wallrock of the upper part of the Ammonia Tanks Member. Parts of the intrusive mass are quartz rich, and others are quartz poor, thus indicating a pronounced internal compositional variation that may reflect composite intrusions. The porphyritic central part of rhyolite of Parachute Canyon, whose mode is shown in figure 21, is more likely a quartz latite. Phenocrysts of alkali feldspar, as much as 8 mm across, enclose cores of subordinate plagioclase, with minor embayed quartz and biotite, all in a micro- crystalline groundmass. This central part petro- graphically shows a striking textural and modal similarity to the microgranite porphyry dike (fig. 21), except for smaller groundmass crystals and somewhat fewer pheno- crysts, and it probably represents a more chilled fine- grained upward extension of similar microgranite porphyry at depth. The quartz and sphene in the central part may represent mixture at one stage or another of chilled rhyolitic tuff of the Ammonia Tanks Member or of similar magma composition. The Parachute Canyon vent is also located near the inner ring—fracture zone on the flank of the Timber Mountain dome, a structural posi- tion similar to that of the porphyry ring dike on the south side (fig. 18). Pipelike intrusives of high-silica rhyolite, one grading upward into a plug dome, cut the upper part of the TIMBER MOUNTAIN TUFF AND ROCKS RELATED TO TIMBER MOUNTAIN CALDERA MAFIC PHENOCRYSTS AS PERCENTAGE OF TOTAL PHENOCRYSTS - Total mafics Hornblende Biotite Clinopyroxene NUMBER OF SPHENE PH ENOCRYSTS 55 SILICA PERCENTAGE RANGE (H20 and CO: free) (R) Resorbed PER THIN SECTION Number of analyses (N) in parentheses | l I | | I I | I | | | (I) M VIII/A I ‘I I I I I I I I I I I (R) I l w— _% (1) i I I I I I I I I I I I g (0) l I I I I I I I I I I I g (0) (a) (From lower part) 3 _ w m— ( A m (.8, I W I I I I I I I I I I I I 0 5 10 15 0 2.5 5.0 0 10 20 64 66 68 70 72 74 76 78 PERCENT PERCENT subunits of upper part of Ammonia Tanks Member. Age relations uncertain except that intrusive rocks generally postdate Ammonia Tanks W. D. Quinlivan and P. W. Lipman (written commun., 1974). A SOUTH MILES 2 Microgranite prophyry (68 percent SiOzI TIMBER MOUNTAIN RESURGENT DOME CENTRAL GRABEN CALDERA MOAT EXPLANATION Post-Timber Mountain rocks Post-Ammonia Tanks intracaldera tuffs of Buttonhook Wash and Crooked Canyon of Timber Mountain Tuff High-silica rhyolite intrusives and plugs of East Cat Canyon NORTH A’ KILOMETRES 3 CALDERA MOAT Drill hole UE18r 2 SEA LEVEL Ammonia Tanks Member of Timber Mountain Tuff Pre-Ammonia Tanks rocks FIGURE 22.—Generalized interpretive section across Timber Mountain resurgent dome, showing inferred relations between compositionally zoned magmatic source of Timber Mountain Tuff and related intrusives cutting Timber Mountain resurgent dome. Line of section shown in figure 18. 56 TIMBER MOUNTAIN -OASIS VALLEY CALDERA COMPLEX, NEVADA Cat Canyon, localized along marginal faults of the central graben, also formed from the high-silica rhyolite in the upper part of the magma chamber. The outward-dipping microgranite porphyry ring dike formed from quartz latitic magma, which averaged about 68 percent silica, and lay below the high-silica rhyolite interface. This idealized model of the magma chamber would probably be attained prior to and reestablished a short time after the voluminous extrusion of the Ammonia Tanks Member with its complex compositional grada- tions of high-silica rhyolite and quartz latite. Intermixing of these two compositions within the Ammonia Tanks Member inside the Timber Mountain caldera indicates disturbance of the interface (fig. 22) and probably the extrusion of the Ammonia Tanks Member from many vents, as originally suggested by Quinlivan and Lipman (1965). CALDERA COLLAPSES RELATED TO ERUPTIONS OF RAINIER MESA AND AMMONIA TANKS MEMBERS Discussion of areas of collapse related to eruptions of the Rainier Mesa and Ammonia Tanks Members of the Timber Mountain Tuff has been deferred until all the petrologically related quartzose igneous rocks that are defined as belonging to the Timber Mountain caldera center have been described. Timber Mountain caldera, as shown in figure 1 and other illustrations of this report, was originally defined not on the basis of its extrusive products but as a volcano—topographic feature in the Williams (1941) sense. The perspective diagram (see frontispiece) illustrates the extent of a subcircular caldera wall enclosing a moatlike annular area surrounding the central dome. Only in the northwestern part, where Thirsty Canyon has been incised, is the caldera wall lacking. These topographic features, indicative of a caldera and central resurgent dome, were recognized by R. L. Smith as early as 1960, and the name Timber Mountain was applied to the caldera and resurgent dome after the topo- graphically prominent central mountain. There are, however, not one but two voluminous coextensive ash- flow sheets, petrologically similar and erupted Within a few hundred thousand years of one another, that are dis- tributed peripherally to Timber Mountain caldera. Did Timber Mountain caldera as defined topographically result as a collapse feature related mainly to eruption of the Rainier Mesa Member, or is it a composite caldera resulting from two or more eruptions of Timber Mountain Tuff, or, still a third possibility, did other adjacent areas subside, producing a volcano-tectonic depression, during eruption of the members? These alter- natives and their possible combinations are considered next. The known areal extents of both the Rainier Mesa and Ammonia Tanks Members are closely similar, as a comparison of figures 14 and 16 will show. The extra— caldera thickness of the Rainier Mesa Member, however, is locally two to three times thicker than the Ammonia Tanks in former topographic depressions such as Silent Canyon caldera; elsewhere the Rainier Mesa Member is generally 50—100 percent thicker than the Ammonia Tanks (Byers and others, 1968, p. 91 and 93). Inasmuch as the intracaldera thickness of the Rainier Mesa Member is not known, only a minimum volume of 1,200 km3 (300 mi?) of the Rainier Mesa Member can be inferred, compared with a more accurately known total volume of 900 km3 (230 mis) of the Ammonia Tanks. The intra- caldera thickness of the Rainier Mesa might well be twice that of the Ammonia Tanks, and the total volume of the Rainier Mesa Member might well be nearly twice that of the Ammonia Tanks. Obviously, the approach of comparing the extracaldera volumes of the members with the areas and volumes of collapse is fraught with many unknowns, including the intracaldera thickness of the Rainier Mesa. A further complicating factor is the possibility that an inner Ammonia Tanks ring-fracture zone may have also been loci for some collapse along an inner zone during eruption of the Rainier Mesa Member. As a first approach, however, it seems likely that the area of collapse related to the Rainier Mesa eruption would be somewhat larger than that related to the Ammonia Tanks eruption. Fortunately, more direct evidence bearing on this problem is available. The area of collapse related to the eruption of the Ammonia Tanks Member is considered first, because thicknesses, facies, and other geologic relations are some- what better known than those parameters of the Rainier Mesa Member. The areas of collapse related to eruption of the Ammonia Tanks Member are inferred mainly from sequences more than 300 m (1,000 ft) thick that are strongly devitrified and contain granophyric pumice. Inside Timber Mountain caldera the main area of collapse is outlined in figure 16. Another probable area of Ammonia Tanks collapse is indicated in the Oasis Valley caldera segment by the 450-m-thick (1,500-ft) strongly devitrified Ammonia Tanks exposed on Oasis Mountain. The thick granophyric facies of the Ammonia Tanks Member on Timber Mountain resurgent dome was penetrated in drill hole UE18r (fig. 3), but core from the drill hole lacks the upper part of the Ammonia Tanks, suggesting that the area occupied by Timber Mountain dome further collapsed within an innermost ring-fracture zone, inside the area of outer collapse enclosed by the 300- m isopach (fig. 16). The area around the large mass of pre- Ammonia Tanks rhyolite lava (fig. 16) is obviously outside the caldera related to Ammonia Tanks collapse, although it is inside the prominent north wall of Timber Mountain caldera. Between this north wall and the inferred wall related to Ammonia Tanks collapse (fig. 16), the Ammonia Tanks Member is glassy, partly welded, and TIMBER MOUNTAIN TUFF AND ROCKS RELATED TO TIMBER MOUNTAIN CALDERA 57 less than 60 m (200 ft) thick—typical of the extracaldera facies. Similar relations occur just inside the south wall of Timber Mountain caldera along a narrow annular zone about 1.5 km wide in which a pre—Ammonia Tanks lava flow is overlain by thin glassy Ammonia Tanks (fig. 16). This narrow zone, obviously, was not within the collapse area related to the Ammonia Tanks Member, but this zone is within the caldera wall continuous with that related to the eruption of the Rainier Mesa Member. Oasis Mountain (fig. 16) occurs in a large westerly salient or “scallop” of the Oasis Valley caldera segment. The mountain is composed of at least 450 m (1,500 ft) of dense devitrified ash—flow tuff of the Ammonia Tanks having very low porosity. The general attitude of the Ammonia Tanks as determined by the basal contact (fig. 23) and by a prominent parting in the middle is 30°-35° eastward, but in the upper part stretched granophyric pumice dips 70°‘—80°. The upper part probably slid downdip, possibly as a result of caldera collapse farther east in the caldera segment. The Ammonia Tanks thins to about 150 m (500 ft) over moderately westward-dipping Rainier Mesa in the Transvaal Hills (fig. 23) on the east edge of the Oasis Valley caldera segment; a 10°—15° angular discordance occurs between the Ammonia Tanks and Rainier Mesa. Locally, an angular unconformity as much as 10° occurs between the westerly dipping upper and lower parts of the Ammonia Tanks. The thinner upper part chilled to a vitrophyre and is locally “frozen” to the thicker more devitrified lower part. The local small angular discordance and the implied partial cooling break indicate minor contemporaneous westward tilting during emplacement of the Ammonia Tanks Member. In summary, the area of collapse related to the eruption of the Ammonia Tanks was mainly peripheral to the Timber Mountain resurgent dome, because that is where the Ammonia Tanks is thickest and where the resurgence took place. High-silica rhyolite lavas that petrologically resemble parts of the Ammonia Tanks are present in the Timber Mountain caldera moat just outside the area of collapse. Thicknesses of more than 450 m ( 1,500 ft), steeply dipping stretched pumice at Oasis Mountain, and a local 10° angular unconformity within gently westward dipping Ammonia Tanks in the Transvaal Hills suggest that the floor of the Oasis Valley caldera segment partially subsided and tilted westward with the eruption of the Ammonia Tanks. This subsidence, however, was considerably less than that peripheral to Timber Mountain dome. The approach to the problem of the area of Rainier Mesa subsidence is more indirect and perhaps less convincing than in the case of the Ammonia Tanks. Several years ago D. C. Noble (written commun., 1966) suggested that the area of collapse related to the eruption of the Rainier Mesa Member included not only Timber Mountain caldera, but also Oasis Valley caldera segment and most of Sleeping Butte segment as well. There was little direct evidence to support his hypothesis at the time, but as geologic mapping and field checking became completed in the caldera segments west of Timber Mountain caldera, the hypothesis of a larger area of subsidence related to the Rainier Mesa eruption has become more attractive. There can be little doubt that the topographically defined Timber Mountain caldera was the main site of collapse caused by eruption of the Rainier Mesa Member. Under the description of debris flows, emphasis was placed on the unique intertonguing relation of the debris with the uppermost part of the Rainier Mesa Member, and on the fact that the debris flows lap onto the north and east walls of the topographically defined Timber Mountain caldera. The debris flows, moreover, contain blocks locally derived from the adjacent caldera wall. Clearly then, the debris flows, at least the basal parts of them that intertongue with the uppermost Rainier Mesa, record the late stages of collapse related to the eruption of the Rainier Mesa (Byers and others, 1968). The onlapping relation of the debris flows to the caldera wall outlines the limits of structural movement related to extrusion of the Rainier Mesa. Although the Rainier Mesa and the debris flows are buried by younger deposits elsewhere around the continuation of the caldera wall, the continuation of the wall and onlapping thin Ammonia Tanks at a few places just inside the wall indicate that Timber Mountain caldera, as defined herein, was also the site of collapse associated with eruption of the Rainier Mesa Member as well as the Ammonia Tanks. The Transvaal Hills (fig. 14) form the exposed west wall of Timber Mountain caldera as topographically defined—yet here the Rainier Mesa Member is more than 450 m (1,500 ft) thick, and the base is not exposed. The unit has a gray granophyric texture, typical of thick ash-flow tuffs. Debris flows intertongue with the uppermost Rainier Mesa and overlie it in the same relation as within Timber Mountain caldera. Although the Rainier Mesa dips about 25° W., suggesting a dip outward away from the caldera wall, the debris flows exposed between the Rainier Mesa and the Ammonia Tanks thin and pinch out north- ward in the Transvaal Hills alonga line about 5 km (3 mi) north of the south wall of the Oasis Valley caldera segment. Moreover, the debris flows are exposed under the Ammonia Tanks for a few miles westward just inside the south wall of Oasis Valley caldera segment, analogous to the relations along the north and east walls of Timber Mountain caldera. At Oasis Mountain (figs. 14, 16), in the westerly salient or “scallop” of the Oasis Valley caldera segment, the thick eastward-dipping Ammonia Tanks rests on at least 90 m (300 ft) of debris flows, whose base is not exposed (see fig 58 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX. NEVADA A WEST FEET Geologic relations 6000 — . proiected under alluvium from _ outcrop to south 035's Mountain Amargosa CALDERAf—Efi Valley RIM , 4000— Tmr QTa Pre-Rainier \ Mesa rocks 2000“ / LEE/EL_ >/// BASIN-RANGE FAULTS 2000 ~ // \\< // OASIS \ Tmr \ V \ /\ VALLEY CALDERA SEGMENT Tuff of Fleur-de-Iis Ranch projected from outcrop Trc Oasis Valley // \\ Pre-Rainier Mesa rocks \ \’S\7 Tmr C CALDERA FAULT // \\ \(A 4000 \ EXPLANATION QTa Alluvium and minor intercalated tuff Tcr Tuff of Cutoff Road Trc Rhyoli’te lava of West Cat Canyon 7 In- tertongues with tuffs of Fleur—de-lis Ranch Tuffs of Fleur-de-lis Ranch — Includes underlying tuffaceous sedimentary rocks Timber Mountain Tuff Tuff of Buttonhook Wash Tmb Ammonia Tanks Member Tmr Rainier Mesa Member Debris flow — Intertongues with upper- most part of Rainier Mesa Member FIGURE 23 (above and right).—Geologic section across Oasis Valley caldera segment, showing great thickness of Rainier Mesa and Ammonia Tanks Members and presence of debris flows of Timber Mountain caldera on east and west sides of segment. Line of section shown in figures 14 and 16. Geology of Oasis Valley area from P. P. Orkild, K. A. Sargent, and R. L. Christiansen (unpub. data, 1971). 23). The debris flows contain rhyolite blocks of the Rainier Mesa and Tiva Canyon Members similar in lithology to the same units exposed in the Bullfrog Hills to the west. Again, these relations are analogous to those inside Timber Mountain caldera, where blocks in the debris flows are derived from the adjacent wall. In fugure 23, we have interpreted the western limit of Rainier Mesa collapse to lie just west of Oasis Mountain. The deeper part of the cauldron subsidence was probably a few miles farther east under Oasis Valley, as indicated by a strong gravity gradient (D. L. Healey, written commun., 1969). Alternatively, the thick exposure of the Rainier Mesa sheet in the Transvaal Hills may not have accumulated in its own cauldron subsidence, but may have accumulated in an earlier Oasis Valley segment of the Claim Canyon cauldron, as suggested by Christiansen and others (1976). We recognize this probability, because the outline of the buried Claim Canyon cauldron shown in figures 8 and 10 includes the Transvaal Hills. However, the additional evidence, just cited, concerning the extent of the debris flows around the edges of Oasis Valley caldera segment suggests partial collapse during the Rainier Mesa eruptions. Also suggesting that the west wall of the subsidence associated with the Rainier Mesa eruption may be west of the Transvaal Hills are the two, Rainier Mesa-related, high-silica rhyolite lavas just east and north of Beatty (fig. 14). Other lavas previously discussed, especially the pre- Ammonia Tanks lavas (fig. 16), are generally within a few kilometres of the cauldron subsidence associated with the petrologically related tuffs. The high-silica rhyolite petro- logically related to the Rainier Mesa Member (fig. 15) in the Bullfrog Hills (fig. 14) would be more than 15 km (9 mi) away from a caldera wall located in the Transvaal YOUNGER INTRACALDERA ROCKS, TIMBER MOUNTAIN AND OASIS VALLEY CALDERAS 59 OASIS VALLEY CALDERA SEGMENT Tuff of Buttonhook Wash outcrop projected from north Bend in section Oasis Valley lntertonguing relations projected from south // Pre-Rainier Mesa rocks 8.4-km break in section — Alluvium and minor intercalated tuff at surface EAST A’ Kl LOMETRES -2.0 TIMBER Transvaal Hills MOUNTAIN CALDERA 0T3 Tcr —1.5 Trc Units exposed on west flank of Timber Mountain 1‘0 resurgent dome Pre-Rainier Mesa rocks " \/ —0.5 \ Tmr ‘ _ SEA \ LEVEL l l / l / v —0.5 BASIN-RANGE FAULTS CALDERA FAULT 1.0 Hills. It is far more likely, on the basis of comparison of the location of other rhyolite lavas to known related cauldron subsidences, that the Oasis Valley caldera segment collapsed during eruption of the Rainier Mesa. The availability nearby of high-silica rhyolite magma indicates that the volatile-rich magma near the top of magma chamber probably underlay Oasis Valley. D. C. Noble (written commun., 1966) interpreted the westward-dipping thick Rainier Mesa Member in the Transvaal Hills to be the tilted remnant of a central resurgent dome. According to the resurgent dome hypo- thesis, post-Rainier Mesa magmatic pressure related to an underlying mobile body of batholithic proportions may have caused the uplift and westward tilt, possibly forming a central resurgent dome that foundered with the erup- tion of the Ammonia Tanks Member. Christiansen and others (1976), however, emphasize several observations that are detrimental to the hypothesis. In conclusion, we believe the evidence strongly suggests that westward tilting and partial subsidence of the Oasis Valley caldera segment may have occurred along with sub- sidence of the Timber Mountain caldera during eruptions of both the Rainier Mesa and Ammonia Tanks Members. These repeated composite subsidences would thereby outline a volcano-tectonic depression, as originally defined by van Bemmelen (1930, 1939). The Timber Mountain caldera and its adjoining Oasis Valley caldera segment probably overlay the apical part of a magma chamber of batholithic magnitude. 2MlLES l I 3 KlLOMETRES YOUNGER INTRACALDERA ROCKS IN TIMBER MOUNTAIN AND OASIS VALLEY CALDERAS Following the climactic eruptions and resurgence at the Timber Mountain center there was a brief period of relative quiescence with local small lakes in the caldera moat. The earliest bedded tuffs on the flank of the dome are sandy, well sorted, and cross-laminated. These features suggest that the ash-fall tuff was reworked by water and that little, if any, volcanic activity took place. Moreover, the earliest mafic lavas are palagonitized, which suggests interaction with water. Locally as much as 30 m (100 ft) of sediments accumulated, including calcareous sandstone and siltstone in both Oasis Valley and Timber Mountain calderas. Upon renewal of volcanic activity, the composi- tion of the first intracaldera tuffs and lavas had drastically changed from highly silicic rhyolites of the Timber Mountain Tuff, to plagioclase-rich low-silica rhyolites, rhyodacite, and mafic lavas. The chemical character, however, of the post-Timber Mountain rocks remained alkali-calcic. RHYODACITIC AND MAFIC LAVAS The oldest post-Timber Mountain rocks to be extruded on the flank of the resurgent dome were intermediate to mafic lavas and relatively restricted in volume. Two of the largest areas of these lavas are on the east and south flanks of Timber Mountain dome (fig. 24). The smaller lava flow on the southern flank is an olivine-bearing trachyandesite (latite of Lipman, Quinlivan, and others, 1966) which has 60 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA 1 16°30’ \ | ‘PAHUTE MESA nu, 7min], \ l I I, ~ \\ lulu, \ .mm m H \\‘ "’“U'c \ J' . ESLEEPING HIIIH \|II||I/ \ , \ / 1m“ ”x, ’I\\ \l Ea/BUTTE 3‘ SLEEPING 5%, [6' : 5 ’: BUTTE 1...”; é Q § 5 ‘ 0 “MW/— \ — : 5 CALDERA MOUNTAIN E Q’ 2: 1 5 SEGMENT CALDERA :, k ‘3 "“5 < : / \un — : : 0”,”th :, 0. 37°; : 00' 5_ ‘ e r «v\ 1E'ROSPECTO‘R’Sv/o, s WPASS‘“ _ «9 5 \ "w \“sg/ V s“ / _\ 2 ¢ :1 \u, ms 0‘61" g s / a,“ .= a, g I" (Beatty /\;/%O nu; 2m“: 3: IM “ 5 5.) :\ 8‘ a a “‘0‘ ’ ; : » \H v ‘ an E, 4|I||\\\ 456‘ : ‘ I g If: 3” Quilt/é 5 Z 1/ [p 1/, ’1” 3‘ : : I”, A E 4’0 3 CRATER 5 '2 '5‘ ’0’, o E 'E I 4927 i FLAT ”a / ': 28’» I , 4’ ‘ 2 1 0 1o 20 MILES l | I l I o 10 20 KILOMETRES EXPLANATION \ . . Thicknesses of tuffs of Cutoff Road and Rhyolite lavas 0f Fortymlle Canyon Fleur-de-lis Ranch — Isopachs dashed % Rhyolite lavas of Beatty Wash . . > V V where inferred 0‘150 m Pattern dropped at 150_300 m overlap with lavas \V Rhyolite lavas of West Cat Canyon Rhyodacitic and mafic lavas — Relations with above lavas uncertain J-L-H-J-J— Boundary of Timber Mountain resurgent dome FIGURE 24.—Distribution of younger intracaldera rocks of Timber Mountain-Oasis Valley caldera complex. Rhyolite lavas of Beatty Wash in northern moat are inferred from drill-hole data and negative aeromagnetic anomaly. YOUNGER INTRACALDERA ROCKS, TIMBER MOUNTAIN AND OASIS VALLEY CALDERAS 61 10-percent phenocrysts consisting of 90-percent plagio- clase and lO-percent olivine, biotite, clinopyroxene, and orthopyroxene. The larger lava flow on the east flank is more exposed and has been more extensively sampled than the smaller flow. The larger lava flow is a phenocryst-rich rhyodacite whose mode and silica content are shown graphically in figure 25. A complete chemical analysis of this rock was furnished by W. D. Quinlivan and P. W. Lipman (written commun., 1974). TUFFS OF FLEUR-DE-LIS RANCH AND RELATED RHYOLITE LAVAS The tuffs of Fleur—de-lis Ranch are most completely exposed on the west side of Oasis Valley just west of the Fleur-de-lis Ranch (fig. 24) where three petrographically similar ash—flow tuff cooling units include an inter- calated lava flow between tuff units 2 and 3. Tuff units 1 and 2 crop out only at this 10cality where the aggregate thickness of all three tuffs and the lava is about 300 m (1,000 ft). Tuff unit 3 and the underlying lava extend southward; Oasis Valley is a narrow gorge where it cuts through these units. The only other place where a similar tuff is exposed is beneath a lava plug dome in an old vent south of Beatty Wash (fig. 24). Here a partly fused tuff, possibly a vent-related hot ashfall, dips inward toward the center of the plug dome and encloses it. The lavas at both localities are modally closely similar to the enclosing tuffs of Fleur-de-lis Ranch and also to the rhyolite lava at West Cat Canyon (figs. 24, 25), where the tuffs are absent. For convenience, all these lavas are designated the rhyolite lavas of West Cat Canyon. The tuffs of Fleur-de-lis Ranch and the related rhyolite lavas of West Cat Canyon rest on the oldest bedded tuff and sediments that fill Oasis Valley and Timber Mountain calderas and are overlain by the tuff of Cutoff Road, which is included with the tuffs of Fleur-de-lis Ranch in figure 24. Both the tuffs and the associated lavas are probably thicker and more extensive at depth under the Oasis Valley caldera segment. The tuffs have not been found flanking Timber Mountain dome, but the presence of the modally similar rhyolite lava at West Cat Canyon on the southwest flank of the dome (fig. 24) suggests the possibility of a con- cealed onlap (Byers and others, 1976, sec. B—B’). The similar modes of the tuffs and lavas are shown graphically in figure 25. Individual specimens vary some- what with respect to total phenocrysts, but the most common phenocrysts are plagioclase, biotite, and clinopyroxene—very similar to those of the dominant mafic pumice in the quartz latitic caprocks of the Paint- brush and Timber Mountain Tuffs. One specimen from the intercalated lava west of Fleur-de-lis Ranch north of Oasis Valley gorge contains hornblende and clino- pyroxene; specimens collected south of the gorge contain only clinopyroxene. Possibly two separate lava flows at the same stratigraphic position were sampled. Chemical analyses of the uppermost tuff at Fleur-de—lis Ranch and the rhyolite lava at West Cat Canyon are from W. D. Quinlivan and P. W. Lipman (written commun., 1974). The high alkalis and silica and high normative quartz and alkali feldspars with low normative anorthite indicate rhyolitic compositions, in contrast to more quartz latitic compositions of crystal-rich caprocks of Paintbrush and Timber Mountain Tuffs. The magnetic polarity of the upper tuff is normal (G. D. Bath, oral commun., 1964). TUFF 0F CUTOFF ROAD AND RELATED RHYOLITE LAVAS The tuff of Cutoff Road and the rhyolite lavas of Beatty Wash, which are compositionally equivalent, are the youngest ash-flow tuff and related lava confined to the Timber Mountain and Oasis Valley calderas. The tuff occurs in the Timber Mountain caldera moat area and around the exposed edges of the Oasis Valley caldera segment (fig. 24). The rhyolite lavas of Beatty Wash are exposed in Beatty Wash in the southern part of Timber Mountain caldera moat (fig. 24) and extend eastward under younger rhyolite lavas of Fortymile Canyon. A lava flow of the same modal petrography crops out in the bottom of Fortymile Canyon (fig. 24) beneath the rhyolite lavas of Fortymile Canyon but is included with these overlying rhyolite lavas on the geologic map of the Timber Mountain caldera area (Byers and others, 1976). Another lava flow, about 150 m (500 ft) thick, was penetrated in drill hole UE18r in the northern caldera moat; its inferred extent is based partly on a negative aero- magnetic anomaly (G. D. Bath, written commun., 1968) and partly on structural interpretation of the moat area. The tuff of Cutoff Road does not exceed 60 m (200 ft) in thickness where exposed; commonly it is less than 30 m (100 ft) and nonwelded, except in Oasis Valley on the west side of the Oasis Valley caldera segment. Near the distal nonwelded edge on the east flank of Timber Mountain dome, a maximum thickness of 26 m (85 ft) is exposed where the tuff dips gently eastward under cover. The rhyolite lavas of Beatty Wash are 150 m (500 ft) thick in the northern moat and greater than 60 m (200 ft) thick in the southern moat. The lavas are probably a significant fraction of the total, owing to the greatly reduced volume of tuff erupted from post-Timber Mountain Tuff vents of the Timber Mountain—Oasis Valley caldera complex. The tuff of Cutoff Road locally onlaps the rhyolite lava of Beatty Wash in Beatty Wash, but there is less than 3 m (10 ft) of coarse locally derived ash-fall tuff between the units. The tuff overlies the tuffs of Fleur-de-lis Ranch or its associated lavas, but the rhyolite lava of Beatty Wash has 62 TIMBER MOUNTAIN -OASIS VALLEY CALDERA COMPLEX, NEVADA STRATIGRAPHIC UNIT Number of thin sections (N) in parentheses \z SUBUNITS J PHENOCRYSTS AS PERCENTAGE OF TOTAL ROCK Rhyolite lava ”I" of Lipman E and others (1966) (3) g —. 7 ?%l~*+?—4 E Rhyolite lava overlying : mafic lavas of Dome I o c g. 5 Mountain (2) 8 g ‘Rhyolite lava of I 1, 2° Pinnacles Ridge (4) E ‘— 5- 3 Rhyolite lava of - ‘5 .E Comb Peak (4) IL l— »— u— o o Rhyolite lava of — E E Waterpipe Butte (5) N 1. ‘93 8 Rhyolite lava of - f, 3 Buried Canyon (2) > E .E n: E Rhyolite lava of I Delirium Canyon (2) (Lavas Rhyolite lava of Vent Pass (5) Tuff of Cutoff Road (7) (Tuff probably intertongues with unit below) Rhyolite lavas of Beatty Wash (6) Tuffs of Fleur-de-lis Ranch (7) (Tuffs intertongue With unit below) Rhyolite lavas of West Cat Canyon (7) ?%?\? Rhyodacite lava on Timber Mountain dome (4) - 0 10 20 30 FELSlC PH ENOCRYSTS OF TOTAL PHENOCRYSTS Quartz Plagioclase Alkali Feldspar VIII/IA MAFIC PH ENOCRYSTS AS PERCENTAGE OF TOTAL PHENOCRYSTS - Total mafics Hornblende Biotite Clinopyroxene AS PERCENTAGE L\\\\\V PERCENT FIGURE 25.—Moda1 and silica ranges of younger, intracaldera volcanic rocks of Timber Mountain—Oasis Valley caldera complex, uncertain field relation. Rhyolite lava “V” (Lipman, Quinlivan, and others, 1966) may correlate with rhyolite of Pinnacles cryst is absent. Silica analyses from W. D. Quinlivan and P. W. Lipman (written commun.,1974). not been observed in direct stratigraphic contact with these units. The tuff of Cutoff Road and the rhyolite of Beatty Wash are inferred to be slightly younger than the tuffs of Fleu‘r-de-lis Ranch and the rhyolite lavas of West Cat Canyon, as indicated partly by superposition of the tuff units and partly by inferred near-equivalence in time of petrographically similar tuff and rhyolite lava. The thin-section modes of the tuff of Cutoff Road and of the rhyolite lavas of Beatty Wash are nearly identical, even in the relative abundance of sphene (fig. 25). The only slight difference is the presence of sparse quartz in a few of the tuff thin sections. Both rocks are rhyolites having identical silica content at 74.5 percent (fig. 25) and there- fore can be correlated as one stratigraphic unit. A mean K-Ar age on the tuff of Cutoff Road is 9.5 m.y. as reported by Kistler (1968, table 1, ”tuff of the caldera fill”). The magnetic polarity for both the tuff and the rhyolite of Beatty Wash is reversed from the Earth’s present field (G. D. Bath, written commun., 1968). RHYOLITE LAVAS OF FORTYMILE CANYON The last activity of the Timber Mountain center resulted in a sequence of rhyolitic lava flows, domes, and closely associated air-fall pyroclastic rocks erupted from a zone along and just outside the south and west rims of the Timber Mountain caldera. The lavas flowed mainly southward across the outer flank of the caldera structure but also flowed into the caldera moat, and at one place lapped onto the Timber Mountain resurgent dome (fig. 24). These rocks have been designated collectively the rhyolite lavas of Fortymile Canyon (Christiansen and Lipman, 1965; Orkild and O’Connor, 1970). The rhyolite lavas of Fortymile Canyon include eight known individual lava flows and domes (fig. 25), each associated with a sequence of pyroclastic rocks, mainly bedded tuffs. These flow units and their tuffs have been mapped individually on the ”6-minute quadrangle maps (fig. 2; Christiansen and Lipman, 1965; Orkild and O’Connor, 1970). No stratigraphic thickness is meaningful for the unit as a whole because of the limited areal extent and the large variations in thickness of indivi- dual flows. Flows are as thick as 300 m (1,000 ft) where they fill old valleys. The pyroclastic rocks in most instances thin outward from the vent areas, where piles of tuff and agglomerate associated with a single flow are as much as SUMMARY OF GEOLOGIC HISTORY 63 SILICA PERCENTAGE RANGE (Recalculated H20 and CO2 free) Number of analyses (N) in parentheses v, vitrophyre; c, microcrystalline and groundmass NUMBER OF SPHENE PHENOCRYSTS PER THIN SECTION ti I I I I I I ?) H? (1) (0) (2)(2) VCVC (Lava)H T (Dike) VVCC HI“ (4) V C H (2) V C H (2) C (H: Phenocrysts-rich (3)(18 percent) lava *Ii 1 C #)——1'1 | (1) VC PERCENT in order of stratigraphic succession, except as noted. Query indicates Ridge (Christiansen and Lipman, 1965, 1966). Blank indicates pheno- 250 m (800 ft) thick. The rhyolite lavas of Fortymile Canyon contain about 20 km3 (5 mi3) of erupted material. The content of silica nad quartz phenocrysts in the rhyolite lavas of Fortymile Canyon increases upward from quartz-poor flows at the base to quartz-rich flows at the top (fig. 25). The quartz trend seems to repeat a similar trend in the Paintbrush-Timber Mountain Tuff sequence. The silica content of the lavas increases gradually upward to high—silica rhyolite (fig. 25), also similar to the trend in the tuffs and rhyolites of Area 20 that fill the Silent Canyon caldera. The petrochemical trend of these lavas toward high quartz and silica continues the trend of the older postresurgence lavas and tuffs in the caldera moat. Rhyolite “V” of Lipman, Quinlivan, Carr, and Anderson (1966) is a high-silica rhyolite (figs. 24, 25) with thin- section modes almost identical to those of the rhyolite lava of Pinnacles Ridge (fig. 25), suggesting either approxi- mate contemporaneity or the tapping of high-silica rhyolite magma near the top of the magma chamber at different times. The rhyolite lavas 0f Fortymile Canyon are petro- chemically and structurally related to volcanism of the Timber Mountain center and are probably the last erup- tive products of that episode. Their age is bracketed by a K- Ar age date of 9.5 m.y. on the underlying tuff of Cutoff Road and by a K-Ar date of 7.5 m.y. on the overlying Thirsty Canyon Tuff (Kistler, 1968) from the Black Mountain caldera center. The petrochemical trend of the rhyolite lavas both repeats the main eruptive sequence of the Timber Mountain-Oasis Valley caldera complex and continues the trend of the younger intracaldera rocks, thereby concluding the activity related to the Timber Mountain center about 9 m.y. ago. SUMMARY OF GEOLOGIC HISTORY Prior to about 16 m.y. ago the area of the Timber Mountain—Oasis Valley caldera complex was probably a terrane of basin-range-type block-faulted mountains of relatively low relief, separated by valleys partially filled with alluvium and tuff. The vents from which the tuff was erupted were outside the area. The youngest of these old tuffs, the calc-alkalic Fraction Tuff, is slightly more than 16 m.y. old (all K-Ar dates from Kistler, 1968, or Marvin and others, 1970); it flowed into the ancestral valleys of the Timber Mountain-Oasis Valley area from vents to the north (Ekren and others, 1971). The earliest known volcanic activity in the Timber Mountain—Oasis Valley caldera complex was the eruption of predominantly hornblende-bearing calc-alkalic ash- flow and bedded tuffs and subordinate rhyolite lavas, probably related to the Sleeping Butte caldera whose northwest wall is exposed in northern Oasis Valley. Little is known of the extent of this early caldera, and for that reason only the extracaldera extrusive products are shown in diagram 1 of figure 26. The principal extrusives from the Sleeping Butte caldera include the Redrock Valley and Crater Flat Tuffs. These ash flows possibly constituted as much as 1400 km3 (350 mi3) of tuff deposited around and within the Timber Mountain-Oasis Valley caldera complex. The Bullfrog Member of the Crater Flat Tuff is the most extensive, possibly extending into Death Valley to the west. The extracaldera extent of the overlying Prow Pass Member is on the south side of the complex. The Prow Pass Member is the only known tuff with ortho— pyroxene as the chief mafic phenocryst. These eruptive events took place 16 to 14 million years ago. A major collapse within the Silent Canyon caldera (diagram 2, fig. 26) occurred 13.8 m.y. ago with the eruption of the peralkaline Grouse Canyon Member of the Belted Range Tuff. The Silent Canyon caldera is not part of the Timber Mountain—Oasis Valley caldera complex, from which only calc-alkalic and alkali—calcic rocks are known. Following the early peralkaline eruptions, the Silent Canyon caldera was the source of calc—alkalic roeks, mainly the tuffs and rhyolite lavas of Area 20 inside the caldera and the Stockade Wash Tuff outside the caldera. The Stockade Wash was formerly the lowermost member of the alkali-calcic Paintbrush Tuff but differs litho- 64 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA SOUTH NORTH Pahute Mesa Tp Tma TIMBER MOUNTAIN CALDERA Tma Trnr Tma Tmr ,.__—_—__ Tp Tp ’ ' A 7 Tra Extrusion of pre-Ammonia Tanks lavas followed by eruption p ,, of Ammonia Tanks Member and tuff of Buttonhook Wash. , , Further cauldron subsidence G pTl’ T _ 4V :1 11 1 ranitic ‘ l _ , _ . m.y. rock Magmajl Tp LTmrJr Tmr ‘1'?er rMagma Gr‘agétkic CLAIM CANYON Subsidence CAULDRON SEGMENT Pahute Mesa 7% 6 Eruption of Rainier Mesa Member accompanied by multiple collapse and emplacement of debris from caldera wall onto floor of caldera and into ring fissures '/\,;_\wm 11.3 m.y. 5 Broad doming accompanied by extrusion of pre-Rainier Mesa lavas 11.3»12 m.y. SILENT Tp CANYON CALDERA 4 Possible resurgence of Claim Canyon cauldron at early stage of magmatic doming 12—125 m.y. Resurgencel?) SILENT CANYON CLAIM CANYON CALDERA CALDERA CLAIM CANYON 3 CAULDRON Eruption of Paintbrush Tuff and related lavas with recurrent subsidence within Claim Canyon cauldron 12.5—13 m.y. S b ' u srdence SILENT Ts CANYON CALDERA 2 Eruption of Grouse Canyon Member and main collapse within Silent Canyon caldera followed by eruption of tuffs and rhyolites of Area 20 inside caldera and by eruption of Stockade Wash Tuff outside caldera. Episodic cauldron subsidence 13—1 4 m.y. 1 Eruption of pre-Grouse Canyon tuffs and lavas on eroded pre- Tertiary rocks. Probable subsidences at Sleeping Butte caldera west of line of section 14—1 6 m.y. PX Approximately 40 km (25 mil RA SUMMARY OF GEOLOGIC HISTORY 65 SOUTH TIMBER Pinnacles Ridge Yucca Yucca Mtn J: MOUNTA|N Drill hole 10 Present Timber Mountain caldera after deposition of caldera fill and erosion 0—9 m.y. - '.sv/:3~IA\\7A ;. 9 Extrusion of intracaldera tutfs and lavas, concluding activity of caldera complex. Crystallization of magma chamber into granitic batholith 9—11 m.y. Further crystallization of magma TIMBER MOUNTA|N RESURGENT DOME Pahute Mesa 8 Magmatic resurgence, forming Timber Mountain dome accom- panied by ring dikes, intrusive rhyolite, and tuff dikes (cone sheets?) and followed by eruption of tuffs of Crooked Canyon 11 my. Resurgence EXPLANATION Post caldera-complex rock—Largely caldera fill INTRACALDERA TUFFS AND LAVAS Rhyolite lavas of Fortymile Canyon and feeder dike Tuff of Cutoff Road Rhyolite lavas of Beatty Wash and West Cat Canyon, and feeder dike TIMBER MOUNTAIN TUFF AND RELATED IGNEOUS ROCKS Tmc Tuffs of Crooked Canyon Microgranite porphyry ring dike, intrusive rhyolite, and tuff dike Tuff of Buttonhook Wash and Ammonia Tanks Member Pre-Ammonia Tanks rhyolite lavas and feeder dike Rainier Mesa Member and associated debris flows and breccia Pre-Rainier Mesa rhyolite lavas and feeder dike ROCKS OF CLAIM CANYON CAULDRON , Paintbrush Tuff and related rhyolite lavas and feeder dike ROCKS 0F SILENT CANYON CALDERA Stockade Wash Tuff and tuffs and rhyolites of T5 T’a Area 20 Ts, Stockade Wash Tuff (extracaldera) Tra, tuffs and rhyolites of Area 20 (intracaldera) Grouse Canyon Member of Belted Range Tuff PRE-GROUSE CANYON ROCKS VIII/[IIIII/A Crater Flat Tuff Pre-Belted Range tuffs, lavas, and volcaniclastic sedimentary rocks—Intercalated with Crater Flat and Redrock Valley Tuffs Redrock Valley Tuff pTr Pre-Tertiary rocks FIGURE 26 (left and above).—Sequence of interpretive diagrams through Timber Mountain caldera, illustrating volcanic history of the Timber Mountain-Oasis Valley caldera complex. 66 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA logically and petrographically from members of the Paint- brush Tuff, as now defined, and is more closely related in its petrography to the intracaldera tuffs and rhyolites of Area 20. The xenolithic inclusions of intracaldera Grouse Canyon and the areal distribution of the Stockade Wash indicate a source within the Silent Canyon caldera. The total erupted volume of the Stockade Wash was about 20—40 km3 (5—10 mi3). Recurrent minor subsidences probably occurred within Silent Canyon caldera as the thick sequence of tuffs and rhyolites of Area 20 accumulated inside the caldera, with the result that the underlying Crater Flat Tuff was eventually downdropped more than 2 km (diagram 2, fig. 26). The members of the Paintbrush Tuff and petro- logically related alkali-calcic lavas were erupted from the Claim Canyon cauldron (diagram 3, fig. 26) resulting in a caldera of unknown extent. Only a small segment of the Claim Canyon cauldron is now exposed, but the general size and location of the collapse structure can be inferred, from distribution of tuffs and lavas north and south of Timber Mountain caldera, to be about 25 km (15 mi) in diameter and to be centered several kilometres west of the present summit of Timber Mountain. The oldest member of the Paintbrush Tuff, the Topopah Spring, was erupted about 13 m.y. ago with a total volume of about 250 km3 (60 mi’) Recurrent subsidence within the cauldron followed eruption of the Topopah Spring Member of the Paint- brush Tuff, and probably a minor collapse followed the eruption of the Pah Canyon and Yucca Mountain Members. Within the caldera they accumulated to several times their thickness outside the caldera. The Tiva Canyon Member, representing the climactic eruption in the evolution of the Claim Canyon cauldron, flowed out about 12.5 m.y. ago mainly to the west of the Claim Canyon cauldron. The tuff of Chocolate Mountain, which is about 1,000 m (3,300 ft) thick, was erupted as a late quartz latitic phase of the Tiva Canyon and was confined within the Claim Canyon cauldron, suggesting that collapse was occurring during its eruption, although Christiansen and others (1976) infer a collapse to the west of the site of Oasis Valley. Tuff breccia, similar petro- graphically to the Tiva Canyon, was generated at vents along or near the wall of the Claim Canyon cauldron segment. The volume of the Tiva Canyon, including the intracauldron tuff of Chocolate Mountain, may have totaled as much as 1,000 km3 (250 mi3). The youngest separate cooling unit of the Paintbrush, the tuff of Pinyon Pass, followed very soon after the Tiva Canyon was confined to the newly formed caldera. After eruption of the ash-flow sheets of the Paintbrush Tuff and after the formation of the caldera, intricate faulting occurred within the Paintbrush cauldron, similar to that on Timber Mountain. The fault pattern is inter- preted by the present authors (see also Christiansen and others, 1976) as owing to magmatic resurgence with uplift of the cauldron block from its former subsided position (diagram 4, fig. 26). The entire Paintbrush cauldron later participated in a broad magmatic doming preliminary to the culminating stages of volcanism—the eruption of ash- flow sheets of the Timber Mountain Tuff. Quartz-bearing pre-Rainier Mesa rhyolite lavas were extruded from fractures probably related to this broad doming (diagram 5, fig. 26). Gas-charged silicic rhyolitic magma accumulated at the top of the domical chamber, and tuffs were assimilated from the roof as crystallization proceeded lower in the chamber. The Timber Mountain Tuff chapter of the caldera complex began 11.3 m.y. ago with eruption of the voluminous Rainier Mesa Member from the central part of the broad dome that had formed over the magma chamber. The domed roof over the magma chamber ruptured sufficiently to cause a considerable reduction of pressure and to trigger vesiculation which greatly increased the volume of the gas-charged magma escaping through fissures. As the eruption proceeded, collapse of the roof occurred, and the Rainier Mesa accumulated to a greater thickness inside the subsidence area than outside (diagram 6, fig. 26). As the caldera deepened with continuing eruption a more crystal-rich, but less gas-charged, quartz latitic magma was tapped, whose eruptions were largely confined within the caldera. Poorly sorted debris flows containing large blocks of welded tuff and lava slid off the newly formed oversteepened caldera walls and inter- tongued with the upper quartz latitic part of the Rainier Mesa Member. A total volume of about 1,200 km3 (300 mi3) of tuff had been extruded by this time, and a large volcano- tectonic depression formed, including the Timber Mountain caldera and probably the adjacent Oasis Valley caldera segment. There is little evidence to indicate whether or not there was central resurgent doming of the Rainier Mesa Member within this depression, except for the westward-tilted block in the Transvaal Hills west of Timber Mountain caldera. During the brief interval between the eruptions of the Rainier Mesa Member and Ammonia Tanks Members of the Timber Mountain Tuff (11.3—1 1.1 m.y.), pre- Ammonia Tanks rhyolite lavas were extruded within the caldera that had resulted from the Rainier Mesa eruptions (diagram 7, fig. 26). The Ammonia Tanks Member of the Timber Mountain Tuff was erupted 11.1 m.y. ago. Nearly half its total volume of 900 km3 (230 mi?) probably accumulated within a concomitantly subsiding caldera, as evidenced by the fact that no debris flows or breccias are known to intertongue with the tuff. This episode of cauldron subsidence of Timber Mountain caldera occurred within an area some- what smaller than the Rainier Mesa collapse area (diagram 7, fig. 26) and was centered farther south, with REFERENCES CITED 67 the result that a crescent—shaped northern moat block did not participate in the Ammonia Tanks subsidence. The Ammonia Tanks eruptions reflect compositional trends with time that were more complex than earlier ash-flow tuff eruptions. In the ash flows of the Ammonia Tanks that were spread outside the caldera the trends progress upward from quartz latite to rhyolite and back to quartz latite. This general sequence outside the Ammonia Tanks collapse area was represented within the collapse area by a much greater thickness of complex intertonguing rhyolite and quartz latite, suggesting multiple source vents and concomitant eruption and collapse. Central resurgent doming of the Ammonia Tanks immediately occurred during the emplacement of the late intracaldera units of the Timber Mountain Tuff. The doming may even have begun before the final ash-flows of the Ammonia Tanks had been extruded. The tuff of Buttonhook Wash was erupted very shortly after the Ammonia Tanks and was confined within the Timber Mountain—Oasis Valley caldera complex. During or just before this eruption about 11 my ago, magmatic resurgence continued and resulted in slight moderate doming of the Ammonia Tanks and intrusions of ring dikes, rhyolite plugs, tuff dikes, and possibly cone sheets into the dome (diagram 8, fig. 26). Doming continued and the weakly welded tuffs of Crooked Canyon of the Timber Mountain Tuff were erupted into the newly formed Timber Mountain caldera moat at a late stage of the doming. A northwest-trending apical graben formed at the crest of the resurgent dome owing to stretching of the domed mass. The culminating chapter in the volcanic history of the Timber Mountain—Oasis Valley caldera complex was now over; subsequent eruptions were small and confined to the caldera moat (diagram 9, fig. 26). The tuffs of Fleur-de-lis Ranch and Cutoff Road and related lavas were erupted 11 to 9.5 m.y. ago and were confined within the Timber Mountain—Oasis Valley caldera complex. They probably had source vents to the west of Timber Mountain caldera in the Oasis Valley caldera segment. Local caldera collapses probably occurred within the segment accompanying these eruptions, accounting for the low relief of the segment and the caldera fill that postdates the tuff of Cutoff Road. Finally, the rhyolite lavas of Forty- mile Canyon accumulated on the rim and in the moat of Timber Mountain caldera and concluded the volcanic history of the complex. The geologic history subsequent to the last volcanic activity of the Timber Mountain—Oasis Valley caldera complex includes the partial filling of the caldera moat with mafic lavas, ash-flow tuffs and lavas from the neighboring Black Mountain center, and the deposition of fan gravels (diagram 10, fig. 26). The Timber Mountain caldera walls and the resurgent dome were somewhat modified by erosion. REFERENCES CITED Anderson, C. A., 1941, Volcanoes of the Medicine Lake highland, Cali- fornia: California Univ. Dept. Geol. Sci. Bull., v. 25, p. 347-422. Anderson, E. M., 1936, The dynamics of the formation of cone sheets, ring dikes, and cauldron subsidences: Royal Soc. Edinburgh Proc., v. 56, pt. 2, p. 128—157. Barnes, Harley, Christiansen, R. L., and Byers, F. M., Jr., 1965, Geo- logic map of the Jangle Ridge quadrangle, Nye and Lincoln Counties, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—363. Barnes, Harley, HOuser, F. N., and Poole, F. G., 1963, Geology of the Oak Spring quadrangle, Nye County, Nevada: U.S. Geol. Sur- vey Geol. Quad. Map GQ-214. Bemmelen, R. W., van, 1930, The origin of Lake Toba (North Sumatra): Pacific Sci. Cong., 4th, Batavia-Bandoeng (Java) 1929, Proc., v. 2a, p. 115—124. 1939, The volcano-tectonic origin of Lake Toba (North Sumatra): De Ingenieur Ned.-Indi'e', v. 6, no. 9, p. 126—140. Billings, M. P., 1943, Ring-dikes and their origin: New York Acad. Sci. Trans., ser. 2, v. 5, no. 6, p. 131—144, Branch, C. D., 1966, Volcanic cauldrons, ring complexes, and associated granites of the Georgetown Inlier, Queensland: Australia Bur. Mineral Resources, Geology and Geophysics Bull. 76, 158 p. Buddington, A. F., 1959, Granite emplacement with special reference to North America: Geol. Soc. America Bull., v. 70, p. 671—747. Burchfiel, B. C., 1966, Reconnaissance geologic map of the Lathrop Wells 15-minute quadrangle, Nye County, Nevada: U.S. Geol. Survey Misc. Geol. Inv. Map 1-474. Byers, F. M., Jr., and Barnes, Harley, 1967, Geologic map of the Paiute Ridge quadrangle, Nye and Lincoln Counties, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-577. Byers, F. M., Jr., Carr, W. J., Christiansen, R. L., Lipman, P. W., Orkild, P. P., and Quinlivan, W. D., 1976, Geologic map of the Timber Mountain caldera area, Nye County, Nevada: U.S. Geol. Survey Misc. Geol. Inv. Map I-891. Byers, F. M., Jr., Christiansen, R. L., Lipman, P. W., Marvin, R. F., and Orkild, P. P., 1969, Cauldron collapse and resurgence of Paintbrush volcanic center, southern Nye County, Nevada, in Abstracts for 1968: Geol. Soc. America Spec. Paper 121, p. 493. Byers, F. M., Jr., and Cummings, David, 1967, Geologic map of the Scrugham Peak quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—695. Byers, F. M., Jr., Orkild, P. P., Carr, W. J., and Quinlivan, W. D., 1968, Timber Mountain Tuff, southern Nevada, and its relation to cauldron subsidence, in Nevada Test Site: Geol. Soc. America Mem. 110, p. 87—97. Byers, F. M., Jr., Rogers, C. L., Carr, W. J., and Luft, S. J., 1966, Geologic map of the Buckboard Mesa quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—552. Carr, W. J., 1964, Structure of part of the Timber Mountain dome and caldera, Nye County, Nevada; in Geological Survey research 1964: U.S. Geol. Survey Prof. Paper 501-B, p. B16—B19. Carr, W. J., and Quinlivan, W. D., 1966, Geologic map of the Timber Mountain quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—503. 1968, Structure of Timber Mountain resurgent dome, Nevada Test Site, in Nevada Test Site: Geol. Soc. America Mem. 110, p. 99—108. Christiansen, R. L., and Lipman, P. W., 1965, Geologic map of the Topopah Spring NW quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—444. 1966, Emplacement and thermal history of a rhyolite lava flow near Fortymile Canyon, southern Nevada: Geol. Soc. America Bull., v. 77, p. 671—684. 68 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA Christiansen, R. L., Lipman, P. W., Carr, W. J., Byers, F. M., Jr., Orkild P. P., and Sargent, K. A., 1976, The Timber Mountain— Oasis Valley caldera complex of southern Nevada: Geol. Soc. America Bull. v. 87. ' Christiansen, R. L., Lipman, P. W., Orkild, P. P., and Byers, F. M., Jr., 1965, Structure of the Timber Mountain caldera, southern Nevada, and its relation to Basin-Range structure, in Geological Survey research 1965: U.S. Geol. Survey Prof. Paper 525—B, p. B43—B48. Christiansen, R. L., and Noble, D. C., 1965, Black Mountain volcanism of southern Nevada [abs.]: Geol. Soc. America Spec. Paper 82, p. 246. 1968, Geologic map of the Trail Ridge quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-774. Christiansen, R. L., Noble, D. C., Orkild, P. P., and Sargent, K. A., 1968, Cogenetic sequences and source areas in silicic volcanic terranes [abs.]: Geol. Soc. America Spec. Paper 101, p. 39. Clough, C. T., Maufe, H. B., and Bailey, E. B., 1909, The cauldron- subsidence of Glen Coe and the associated igneous phenomena: Geol. Soc. London Quart. Jour., v. 65, p. 611—678. Colton, R. B., and McKay, E. J., 1966, Geologic map of the Yucca Flat quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-582. Colton, R. B., and Noble, D. C., 1967, Geologic map of the Groom Mine SW quadrangle, Nye and Lincoln Counties, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-719. Cornwall, H. R., 1962, Calderas and associated volcanic rocks near Beatty, Nye County, Nevada, in Engel, A. E. J., James, H. L., and Leonard, B. F., eds., Petrologic studies (Buddington volume): Geol. Soc. America, p. 357—371. Cornwall, H. R., and Kleinhampl, F. J., 1961, Geology of the Bare Mountain quadrangle [Nye County], Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—157. 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: U.S. Geol. Survey Prof. Paper 454-], p. J1—J25. Cummings, David, 1964, Eddies as indicators of local flow direction in rhyolite, in Short papers in geology and hydrology: U.S. Geol. Survey Prof. Paper 475-D, p. D70—D72. Eckel, E. B., 1968, Development of geologic knowledge at Nevada Test Site, in Nevada Test Site: Geol. Soc. America Mem. 110, p. 5—10. Eggler, D. H., 1968, Virginia Dale Precambrian ring-dike complex, Colorado-Wyoming: Geol. Soc. America Bull., v. 79, p. 154941563. Ekren, E. B., Anderson, R. E., Orkild, P. P., and Hinrichs, E. N., 1966, Geologic map of the Silent Butte quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-493. Ekren, E. B., Anderson, R. E., Rogers, C. L., and Noble, D. C., 1971, Geology of northern N ellis Air Force Base Bombing and Gunnery Range, Nye County, Nevada: U.S. Geol. Survey Prof. Paper 651, 91 p. Ekren, E. B., Rogers, C. L., Anderson, R. E., and Botinelly, Theodore, 1967, Geologic map of the Belted Peak quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-606. Ekren, E. B., and Sargent, K. A., 1965, Geologic map of the Skull Mountain quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad Map GQ-387. Fernald, A. T., Corchary, G. S., and Williams, W. P., 1968, Surficial geologic map of Yucca Flat, Nye and Lincoln Counties, Nevada: U.S. Geol. Survey Misc. Geol. Inv. Map 1-550. Gibbons, A. B., Hinrichs, E. N., Hansen, W. R., and Lemke, R. W., 1963 Geology of the Rainier Mesa quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-215. Hamilton, Warren, and Myers, W. B., 1967, The nature of batholiths: U.S. Geol. Survey Prof. Paper 554-C, 30 p. Hinrichs, E. N ., 1968, Geologic map of the Camp Desert Rock quad- rangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—726. Hinrichs, E. N., Krushensky, R. D., and Luft, S. J., 1967, Geologic map of the Ammonia Tanks quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-638. Hinrichs, E. N ., and McKay, E. J., 1965, Geologic map of the Plutonium Valley quadrangle, Nye and Lincoln Counties, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—384. Hinrichs, E. N., and Orkild, P. P., 1961, Eight members of the Oak Spring Formation, Nevada Test Site and vicinity, Nye and Lincoln Counties, Nevada, in Short papers in the geologic and hydrologic sciences: U.S. Geol. Survey Prof. Paper 424-D, p. D96—D103. Holmes, Arthur, 1965, Principles of physical geology [revised ed.]: New York, Ronald Press 00., 1288 p. Houser, F. N ., and Poole, F. G., 1960, Preliminary geologic map of the Climax stock and vicinity, Nye County, Nevada: U.S. Geol. Survey Misc. Geol. Inv. Map 1-328. Kistler, R. W., 1968, Potassium—argon ages of volcanic rocks in Nye and Esmeralda Counties, Nevada, in Nevada Test Site: Geol. Soc. America Mem. 110, p. 251—262. Lipman, P. W., 1966, Water pressures during differentiation and crystallization of some ash-flow magmas from southern Nevada: Am. Jour. Sci., v. 264, p. 810—826. 1971, Iron-titanium oxide phenocrysts in compositionally zoned ash—flow sheets from southern Nevada: Jour. Geology, v. 79, p. 438—456. 1976, Caldera collapse breccias in the western San Juan Moun- tains, Colorado: Geol. Soc. America Bull., v. 87. Lipman, P. W., and Christiansen, R. L., 1964, Zonal features of an ash-flow sheet in the Piapi Canyon Formation, southern Nevada, in Geological Survey research 1964: U.S. Geol. Survey Prof. Paper 501-B, p. B74—B78. Lipman, P. W., Christiansen, R. L., and O’Connor, J. T., 1966, A compositionally zoned ash-flow sheet in southern Nevada: U.S. Geol. Survey Prof. Paper 524-F, p. F1—F47. Lipman, P. W., and McKay, E. J., 1965, Geologic map of the Topopah Spring SW quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-439. Lipman, P. W., Quinlivan, W. D., Carr, W. J., and Anderson, R. E., 1966, Geologic map of the Thirsty Canyon SE quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-489. Locke, Augustus, Billingsley, P. R., and Mayo, E. B., 1940, Sierra Nevada tectonic patterns: Geol. Soc. America Bull., v. 51, p. 513—540. Luft, S. J., 1964, Mafic lavas of Dome Mountain, Timber Mountain caldera, southern Nevada, in Geological Survey research 1964: U.S. Geol. Survey Prof. Paper 501-D, p. Dl4—D2l. Marvin, R. F., Byers, F. M., Jr., Mehnert, H. H., Orkild, P. P., and Stern, T. W., 1970, Radiometric ages and stratigraphic sequence of volcanic and plutonic rocks, southern Nye and western Lincoln Counties, Nevada: Geol. Soc. America Bull., v. 81, p. 2657—2676. McKay, E. J., and Sargent, K. A., 1970, Geologic map of the Lathrop Wells quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-883. McKay, E. J., and Williams, W. P., 1964, Geology of the Jackass Flats quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—368. McKeown, F. A., 1976, Geologic map of the Yucca Lake quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad Map GQ—1327. REFERENCES CITED 69 Noble, D. C., 1970, Loss of sodium from crystallized comendite welded tuffs of the Miocene Grouse Canyon Member of the Belted Range Tuff, Nevada: Geol. Soc. America Bull., v. 81, p. 2677—2688. Noble, D. C., Anderson, R. E., Ekren, E. B., and O’Connor, J. T., 1963, Thirsty Canyon Tuff of Nye and Esmeralda Counties, Nevada, in Short papers in geology and hydrology: U.S. Geol. Survey Prof. Paper 475-D, p. D24—D27. Noble, D. C., Bath, G. D., Christiansen, R. L., and Orkild, P. P., 1968, Zonal relations and paleomagnetism of the Spearhead and Rocket Wash Members of the Thirsty Canyon Tuff, southern Nevada, in Geological Survey research 1968: U.S. Geol. Survey Prof. Paper 600-C, p. C61—C65. Noble, D. C., and Christiansen, R. L., 1968, Geologic map of the southwest quarter of the Black Mountain quadrangle, Nye County, Nevada: U.S. Geol. Survey Misc. Geol. Inv. Map I-562. Noble, D. C., Haffty, Joseph, and Hedge, C. E., 1969, Strontium and magnesium contents of some natural peralkaline silicic glasses and their petrogenetic significance: Am. Jour. Sci., v. 267, p. 598—608. Noble, D. C., Kistler, R. W., Christiansen, R. L., Lipman, P. W., and Poole, F. G., 1965, Close association in space and time of alkalic, calc-alkalic, and calcic volcanism in southern Nevada [abs.]: Geol. Soc. America Spec. Paper 82, p. 143—144. Noble, D. C., Krushensky, R. D., McKay, E. J., and Ege, J. R., 1967, Geologic map of the Dead Horse Flat quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-614. Noble, D. C., Sargent, K. A., Mehnert, H. H., Ekren, E. B., and Byers, F. M., Jr., 1968, Silent Canyon volcanic center, Nye County, Nevada, in Nevada Test Site: Geol. Soc. America Mem. 110, p. 65-75. Nockolds, S. R., 1954, Average chemical composition of some igneous rocks: Geol. Soc. America Bull., v. 65, p. 1007—1032. O’Connor, J. T., 1963, Petrographic characteristics of some welded tuffs of the Piapi Canyon Formation, Nevada Test Site, Nevada, in Short papers in geology and hydrology: U.S. Geol. Survey Prof. Paper 475-B, p. B52—B55. 1965, A classification for quartz-rich igneous rocks based on feldspar ratios, in Geological Survey research 1965: U.S. Geol. Survey Prof. Paper 525-B, p. B79—BS4. O’Connor, J. T., Anderson, R. E., and Lipman, P. W., 1966, Geologic map of the Thirsty Canyon quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—524. Orkild, P. P., 1963, Geology of the Tippipah Spring quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—213. 1965, Paintbrush Tuff and Timber Mountain Tuff of Nye County, Nevada, in Cohee, G. V., and West, W. 8., Changes in stratigraphic nomenclature by the U.S. Geological Survey, 1964: U.S. Geol. Survey Bull. l224-A, p. A44—‘A5l. 1968, Geologic map of the Mine Mountain quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—746. 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 Nevada Test Site: Geol. Soc. America Mem. 110, p. 77—86. Orkild, P. P., and O’Connor, J. T., 1970, Geologic map of the Topopah Spring quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—849. Orkild, P. P., Sargent, K. A., and Snyder, R. P., 1969, Geologic map of Pahute Mesa, Nevada Test Site and vicinity, Nye County, Nevada: U.S. Geol. Survey Misc. Geol. Inv. Map I-567. Peacock, M. A., 1931, Classification of igneous rock series: Jour. Geology, v. 39, p. 54—67. Poole, F. G., 1965, Geologic map of the Frenchman Flat quadrangle, Nye, Lincoln, and Clark Counties, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—456. Poole, F. G., Carr, W. J., and Elston, D. P., 1965, Salyer and Wahmonie Formations of southeastern Nye County, Nevada, in Cohee, G. V., and West, W. 8., Changes in stratigraphic nomenclature by the U.S. Geological Survey, 1964: U.S. Geol. Survey Bull. 1224-A, p. A36—A44. Poole, F. G., Elston, D. P., and Carr, W. J., 1965, Geologic map of the Cane Spring quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad Map GQ-455. Poole, F. G., and McKeown, F. A., 1962, Oak Spring Group of the Nevada Test Site and vicinity, Nevada, in Short papers in geology, hydrology, and topography: U.S. Geol. Survey Prof. Paper 450—C. p. C60—62. Quinlivan, W. D., and Lipman, P. W., 1965, Compositional variations in some Cenozoic ash-flow tuffs, southern Nevada, in Abstracts for 1964: Geol. Soc. America Spec. Paper 82, p. 342. Ratté, J. C., and Steven, T. A., 1967, Ash flows and related volcanic rocks associated with the Creede caldera, San Juan Mountains, Colorado: U.S. Geol. Survey Prof. Paper 524-11, 58 p. Richey, J. E., 1932, Tertiary ring structures in Britain: Geol. Soc. Glasgow Trans., v. 19, p. 42—140. 1961, British regional geology; Scotland—The Tertiary volcanic districts [3d ed.], with revision by MacGregor, A. G., and Anderson, F. W.: Edinburgh, Scotland, Dept. Sci. and Indus. Research Geol. Survey and Museum, 120 p. Rittman, Alfred, 1952, Nomenclature of volcanic rocks: Volcanol. I, ser. 2, v. 12, p. 75—102. Rogers, C. L., Anderson, R. E., Ekren, E. B., and O’Connor, J. T., 1967, Geologic map of the Quartzite Mountain quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-672. Rogers, C. L., Ekren, E. B., Noble, D. C., and Weir, J. E., 1968, Geo- logic map of the northern half of the Black Mountain quad- rangle, Nye County, Nevada: U.S. Geol. Survey Misc. Geol. Inv. Map I-545. Rogers, C. L., and Noble, D. C., 1969, Geologic map of the Oak Spring Butte quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—822. Ross, C. S., and Smith, R. L., 1961, Ash-flow tuffs—their origin, geo- logic relations and identification: U.S. Geol. Survey Prof. Paper 366, 81 p. Sargent, K. A., 1969, Petrography and heavy minerals of three groups of rhyolitic lavas, Pahute Mesa, Nevada Test Site, in Geological Survey research 1969: U.S. Geol. Survey Prof. Paper 650-C, P. C18’—C24. Sargent, K. A., Luft, S. J., Gibbons, A. B., and Hoover, D. L., 1966, Geologic map of the Quartet Dome quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-496. Sargent, K. A., McKay, E. J., and Burchfiel, B. C., 1970, Geologic map of the Striped Hills quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ—882. Sargent, K. A., Noble, D. C., and Ekren, E. B., 1965, Belted Range Tuff of Nye and Lincoln Counties, Nevada, in Cohee, G. V., and West, W. S., Changes in stratigraphic nomenclature by the U.S. Geological Survey, 1964: U.S. Geol. Survey Bull. 1224-A, p. A32—A36. Sargent, K. A., and Orkild, P. P., 1973, Geologic map of the Wheel- barrow Peak—Rainier Mesa area, Nye County, Nevada: U.S. Geol. Survey Misc. Geol. Inv. Map I-754. Sargent, K. A., and Stewart, J. H., 1971, Geologic map of the Specter Range NW quadrangle, Nye County, Nevada: U.S. Geol. Survey Geol. Quad. Map GQ-884. Shand, S. J., 1947, Eruptive rocks, [3d ed.]: New York, John Wiley and Sons, 488 p. Smith, R. L., 1960a, Ash flows: Geol. Soc. American Bull., v. 71, p. 795—842. Bull. 70 TIMBER MOUNTAIN-OASIS VALLEY CALDERA COMPLEX, NEVADA 1960b, Zones and zonal variations in welded ash flows: U.S. Stock, Chester, and Bode, F. D., 1935, Occurrence of lower Oligocene Geol. Survey Prof. Paper 354-F, p. 149—159. mammal—bearing beds near Death Valley, California: Nat. Acad. Smith, R. 1..., and Bailey, R. A., 1966, The Bandelier Tuff—A study Sci. Proc., v. 21, no. 10, p. 571—579. of ash-flow eruption cycles from zonal magma chambers: Bull. Volcanol., v. 29, p. 83—103. Tuttle, O. F., and Bowen, N. L., 1958, Origin of granite in the light Smith, R. L., Bailey, R. A., and Ross, C. S., 1961, Structural evolu- of experimental studies in the system NaAlSiaosKAISiSOG-Siof tion of the Valles caldera, New Mexico. and its bearing on the 1-120: Geol. Soc. America Mem. 74, 153 p. emplacement of ring dikes, in Short papers in the geologic and hydrologic sciences: U.S. Geol. Survey Prof. Paper 424-D, Williams, Howel, 1941, Calderas and their origin: California Univ. 1). D145—D149. Dept. Geol. Sci. Bull., v. 25, p. 239—346. crus GOVERNMENT PRINTING OFFICE: l976—677-340/95 @E76’ 5M1";— $5 (2.92.0 Metamorphism and Plutonisrn Around the Middle and South Forks of the Feather River, California GEOLOGICAL SURVEY PROFESSIONAL PAPER 920 ‘ 4 DECZ81976 / ‘1‘). ’i I n (ax {99wa £83) " 7 1975 f5.D.‘ Metamorphism and Plutonisrn Around the Middle and South Forks of the Feather River, California By ANNA HIETANEN GEOLOGICAL SURVEY PROFESSIONAL PAPER 920 Petrologic and structural studies of metamorphic and igneous rocks of part of the N evadan orogenie belt in the northwestern Sierra Nevada UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1976 UNITED STATES DEPARTMENT OF THE INTERIOR THOMAS S. KLEPPE, Secretary GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress Cataloging in Publication Data Hietanen, Anna Martta, 1909— Metamorphism and plutonism around the Middle and South Forks of the Feather River, California. (Geological Survey Professional Paper 920) Bibliography: p. 29. Supt. of Docs. no.2 I 19.16920 1. Metamorphism (Geology)—Ca1ifornia—Feather River watershed. 2. Intrusions (Geology)~Ca1ifornia—Feather River Watershed. 3, Petrology—Ca1ifornia—-Feather River watershed. I. Title. II. Series: United States Geological Survey Professional Paper 920. QE475.A2H53 552’.03 76—608227 For sale by the Superintendent of Documents, US. Government Printing Office Washington, DC. 20402 Stock Number 024—001—02889—1 CONTENTS Page Page Abstract ___________________________________________________ 1 Plutonic rocks , __________________________________________________ 14 Introduction _______________________________________________ 1 Lumpkin pluton and related rocks ______________________ 15 Geologic setting ____________________________________________ 3 Cascade pluton __________________________________________ 16 Metavolcanic and metasedimentary rocks _____________________ 5 Hartman Bar pluton ___________________________________ 20 Franklin Canyon Formation _____________________________ 5 Merrimac pluton _______________________________________ 21 Horseshoe Bend Formation _______________________________ 7 Mass at Hampshire Creek _________________________________ 21 Structures ____________________________________________ 10 Mass in Indian Valley ____________________________________ 22 Age and correlation _____________________________________ 11 Bald Rock pluton ________________________________________ 22 Metamorphosed intrusive rocks _____________________________ 12 Composition of the plutonic rocks __________________________ 22 Serpentine and peridotite _______________________________ 13 Structures and evolution of the plutons ______________________ 24 Metagabbro __________________________________________ 13 Origin of magmas ________________________________________________ 25 Metadiorite ____________________________________________ 13 Island-arc-type volcanic rocks ___________________________ 25 Metatrondhjemite ______________________________________ 13 Monzotonalitic magma ___________________________________ 27 Metamorphism ____________________________________________ 14 Tertiary volcanic rocks _______________________________ 28 Magmas in space and time _____________________________ 29 References _____________________________________________________ 29 ILLUSTRATIONS page PLATE 1. Geologic map of the area around the Middle and South Forks of the Feather River, California ______________________ In pocket FIGURE 1. Index map showing location of the study area in northern California __________________________________________________ 1 2. Sketch map and section of the study area and vicinity, northern California ___________________________________________ 2 3—6. Photographs of: 3. Hexagonal plates of biotite in metarhyolite north of Mountain House, southeastern contact zone of the Merri- mac pluton (loc. 1470) ____________________________________________________________________________________ 9 4. Flat-lying bedding and poorly developed steep second cleavage in black phyllite on Slate Creek __________________ 10 5. Recumbent fold on southeast-plunging axis, indicating overturning to the southwest, in black phyllite at 4,200-foot elevation east of Slate Creek ___________________________________________________________________________ 10 6. Fine-pebble conglomerate 1 mile south of Clipper Mills ________________________________________________________ 12 7. Sketch map showing lineation in the Cascade pluton and vicinity and the location of stained specimens _____________ 19 8. Photomicrograph of tonalite, containing large biotite flakes, from a logging road along Rock Creek on the southwestern part of the Cascade pluton (loc. 1272) ___________________________________________________________________________ 20 9. Photomicrograph of foliated trondhjemite from Watson Ridge ______________________________________________________ 20 10. Diagram showing ratios of total feldspar, quartz, and femic minerals in stained specimens of plutonic rocks ____________ 23 11. Normative Ab-Or-An diagram showing points for new analyses in relation to the differentiation curve for the plutonic rocks in the Pulga and Bucks Lake quadrangles ______________________________________________________________ 24 124 Sketch showing evolution of magmas in the northern Sierra Nevada ________________________________________________ 26 TABLES Page TABLE 1. Percentage of major constituents measured in stained samples of plutonic rocks ____________________________________ 16 2. Chemical composition, molecular norm, and trace~element contents of plutonic rocks _________________________________ 18 III METAMORPHISM AND PLUTONISM AROUND THE MIDDLE AND SOUTH FORKS OF THE FEATHER RIVER, CALIFORNIA By ANNA HIETANEN ABSTRACT The area around the Middle and South Forks of the Feather River provides information on metamorphic and igneous processes that bear on the origin of andesitic and granitic magmas in general and on the variation of their potassium content in particular. In the north, the area joins the Pulga and Bucks Lake quadrangles studied previously. Tectonically, this area is situated in the southern part of an arcuate segment of the Nevadan orogenic belt in the northwest- ern Sierra Nevada. The oldest rocks are metamorphosed calc- alkaline island-arc-type andesite, dacite, and sodarhyolite with in- terbedded tufl" layers (the Franklin Canyon Formation), all probably correlative with Devonian rocks in the Klamath Mountains. Younger rocks form a sequence of volcanic, volcaniclastic, and sedimentary rocks including some limestone (The Horseshoe Bend Formation), probably Permian in age. All the volcanic and sedimen- tary rocks were folded and recrystallized to the greenschist facies during the Nevadan (Jurassic) orogeny and were invaded by mon- zotonalitic magmas shortly thereafter. A second lineation and metamorphism to the epidote-amphibolite facies developed in a nar- row zone around the plutons. In light of the concept of plate tectonics, it is suggested that the early (Devonian?) island-arc-type andesite, dacite, and sodarhyolite (the Franklin Canyon Formation) were derived from the mantle above a Benioff zone by partial melting of peridotite in hydrous con- ditions. The water was probably derived from an oceanic plate de- scending to the mantle. Later (Permian?) magmas were mainly basaltic; some discontinuous layers of potassium-rich rhyolite indi- cate a change into anhydrous conditions and a deeper level of magma generation. The plutonic magmas that invaded the metamorphic rocks at the end of the Jurassic may contain material from the man- tle, the subducted oceanic lithosphere, and the downfolded metamor- phic rocks. The ratio of partial melts from these three sources may have changed with time, giving rise to the diversity in composition of magmas. INTRODUCTION Petrologic and structural studies around Middle and South Forks of the Feather River extend the work begun in the Pulga and Bucks Lake quadrangles to the southern part of the northern arcuate segment of the western metamorphic belt of the Sierra Nevada (fig. 1). The study area covers some 750 square kilometres in the southern half of the old Bidwell Bar 30-minute quadrangle (Turner, 1898), now covered by five 71/2- minute quadrangles, the Brush Creek, Cascade, American House, Clipper Mills, and Strawberry Val- ley quadrangles in northern California (figs. 1, 2). The mapped area is between long 121°O’ and 121°221/2’, lat 39°30’ and 39°45’, adjoining the Bucks Lake and Pulga quadrangles to the north (Hietanen, 1973a). In the western part, it includes part of the Merrimac area (Hietanen, 1951) and the eastern contact zone of the Bald Rock pluton (Compton, 1955). Most of the area is underlain by plutonic rocks that include the Cascade pluton, parts of two other large plutons, the Merrimac and Bald Rock, and several small masses. The metamorphic rocks exposed be- tween the plutons are continuous with the metavol- canic and metasedimentary rocks in the Pulga and Bucks Lake quadrangles to the north and the Mer- rimac area to the west. Subdivision of the metamorphic rocks into three formations, the Calaveras, Franklin 0 124° 123° 122° 121° 120° 42 ' — ' — - —— - - M‘ ‘ — "f EXPLANATION 1 ' l Study area 0 SHASTA I 41 LAKE . D LAKE l Pulga and Bucks up; AlLMANOR I Lake quadrangles :1 I a j 40 ‘2 l 8 “I . 1 7.3. "i6 @0211er 1 <1 N 1‘: : fife? s 0 wer LAKE 39 8 V‘ ' TAHOE L“ é‘ to“ \ SACRAMENTO N“ \ \ ‘\ 38" l \) San Francisco (A, 37°30' l 1 o 50 100 M1 LES 0 50 100 150 KILOMETRES FIGURE 1.—Location of the study area in northern California. METAMORPHISM AND PLUTONISM, FEATHER RIVER AREA, CALIFORNIA 121°30' JONESVILLE QUADRANGLE 15' ALMANOR OUADRANGLE 121°oo' 40°00' fl; LU J m 0 4 L, E O w / Z (r 'I ‘V\ E <( ‘/ i I 7 < I D Meadow C 3 D * ( O < Lu 3 ¥ 0 < < .3 (3 —1 3 3 o ‘3- 3 m V ,\ \ < \\/K 50' -- /\\ / 45' 45' g u] LU $4 8.: mo IO (:2 z 0< 2< >0: lV3d 000M900 o E g Brush Creek 8 -; quadrangle m .E E I: 6 2 u. nnva )IVEId 1awv3 Cascade quadrangle Contact Dashed where approximately located Fault Dashed where qfigH ANS Jeeu llnEd approximately located 2 O l- D _l n. I.“ D <32 0 VJ < Horsehoe Bend Formation {I} .11an ONES 918 8000 4000' 1000 FIGURE 2.—Continued. part of the area are covered by Tertiary volcanic rocks. These rocks have been described in an earlier report (Hietanen, 1972) and will not be discussed further here. The major geologic events in the area were as fol- lows: (1) deposition of the Franklin Canyon Formation (Devonian?); (2) deposition of the Horseshoe Bend Formation (Permian?); (3) intrusion of synkinematic quartz diorite and trondhjemite and 0f ultramafic rocks, subsequent deformation and recrystallization culminating in Jurassic time; (4) continued faulting and emplacement of ultramafic rocks; (5) emplacement of postkinematic plutons (Late Jurassic and Early Cre- taceous); (6) uplift and erosion; (7) extrusion of Ter- tiary volcanic rocks. GEOLOGIC SETTING The study area lies in the innermost part of an ar- cuate segment of the northern Sierra Nevada; struc- tural trends in the metamorphic rocks south of the area are generally northerly, trends north and west of the area, westerly. The trends curve sharply from northerly to westerly around the Hartman Bar pluton just north of the map area and around the Bald Rock pluton in the southwestern part of the area (fig. 2; pl. 1). Local disruption in the arc is evident southeast of the Hartman Bar pluton where the westerly trends butt against the north-trending segment of the Camel Peak fault concealed by the elongate body of serpentine (fig. 2; pl. 1). This fault, well defined by elongate serpentine bodies (pl. 1), is a contact between two formations, the metavolcanic Franklin Canyon Formation to the northeast and the metasedimentary and the metavol— canic Horseshoe Bend Formation to the southwest. Another major fault zone, en echelon in part, passes through the metamorphic rocks north and northeast of the Bald Rock pluton where several faults, all accom- panied by serpentine bodies, were mapped. The faults and serpentine bodies are older than the youngest of the Cretaceous plutons, as shown by a tongue of trondhjemite that, in the canyon of the Middle Fork of the Feather River, cuts sharply the metamorphic rocks and all their structures (pl. 1). On the northwest side of the river, the fault zone is concealed by serpentine bodies; farther west, two parellel faults pass through a narrow strip of metamorphic rocks between the Bald Rock and Merrimac plutons. This fault zone is the eastern extension of the Big Bend fault (fig. 2). Two parallel faults occupied by thin bodies of serpentine are exposed on Watson Ridge (pl. 1) on the southeast side of the Middle Fork. The southeastern extension of this fault zone continues between the Lumpkin pluton and 4 METAMORPHISM AND PLUTONISM, FEATHER RIVER AREA, CALIFORNIA the Cascade pluton, where it is represented by three faults that are several kilometres long and about 1—2 km apart. The middle branch of this en echelon fault passes on the west side of Sugar Pine Point at the south end of Lumpkin Ridge, where it is marked by a brown-weathering fault gorge, several metres wide, and by local angular discordance of structures. This middle branch terminates at the border of the Lump- kin pluton, and another south—southeast-trending fault accompanied by a long thin body of serpentine is ex- posed about 1 km east of Sugar Pine Point passing southward to the Lost Creek Reservoir and to the east side of Barton Hill. Several north-northwest-trending faults and numerous long thin conformable serpentine bodies in the Horseshoe Bend Formation south and southwest of Lumpkin pluton suggest intensive slicing during the deformation in a wide zone east of the Bald Rock plu- ton. Metavolcanic rocks exposed to the west of the westernmost fault of this zone are less deformed than the rocks of the Horseshoe Bend Formation and are probably Mesozoic in age. T wo major fault zones divide the metamorphic rocks into three belts, each containing rocks of only one for- mation, from east to west, the Franklin Canyon For- mation, the Horseshoe Bend Formation, and the un- named Mesozoic rocks. In the easternmost belt (pl. 1), east of the Camel Peak fault, the Franklin Canyon Formation is a direct continuation of similar rocks in the Bucks Lake quadrangle (Hietanen, 1973a). The Franklin Canyon consists of potassium-poor metavol- canic rocks—specifically metasodarhyolite, metada- cite, meta—andesite, and associated metatuffs. It is probably correlative with the Devonian island-arc-type volcanic rocks in the Klamath Mountains. In the Bucks Lake quadrangle, this formation rests on the metasedimentary Calaveras Formation, which con- sists of interbedded metachert and phyllite and very little limestone originally similar to sediments depo- sited on ocean floors. The formation exposed around the Cascade pluton belongs to the central belt and consists of interbedded volcanic, volcaniclastic, and sedimentary strata in— cluding small lens-shaped bodies of marble. These rocks are a southeast extension of the Horseshoe Bend Formation of the Bucks Lake and Pulga quadrangles; structural relations there suggest that they are younger than the metavolcanic rocks of the Franklin Canyon Formation. They may be correlative with the Permian volcanic, volcaniclastic, and shallow-water sedimentary rocks of the Klamath Mountains. This correlation is supported by an occurrence of Permian(?) fossils in similar limestones just west of the study area (Creely, 1965). It is noteworthy that whereas the metarhyolites of the older formation, the Devonian(?) Franklin Canyon, are poor in potassium, some of those of the younger Horseshoe Bend Formation contain a considerable amount of potassium feldspar. A sequence of metabasalt and metatuff with inter— bedded metarhyolite, quartzite, and phyllite exposed north and east of the Bald Rock pluton in the south— western part of the central belt is similar to parts of the Horseshoe Bend Formation. It also resembles a similar sequence in the Pulga quadrangle, where it was dis- tinguished as the Duffey Dome Formation (Paleozoic?) underlying the metasedimentary rocks of the Horse- shoe Bend Formation. On the geologic map (pl. 1) this sequence is included in the Horseshoe Bend Forma- tion. The metavolcanic rocks bordering the Bald Rock pluton to the southeast are similar to the Mesozoic rocks southwest of the map area. Detailed description of these rocks is not included in this report. The bound- ary between the Paleozoic rocks of the Horseshoe Bend Formation and the Mesozoic rocks to the southwest probably lies in part along the northern and eastern contact of the Bald Rock pluton. The northern contact zone of this pluton is strongly sheared and in aline- ment with a long thin body of serpentine and talc schist in the Big Bend area (Hietanen, 1951). This lineament is 1 km south of the Big Bend fault and may represent a southern branch of this fault. In each of the fault blocks, geographically defined by the three belts, the metasedimentary and metavol- canic rocks are isoclinally folded; the folds in general are overturned to the southwest. In the easternmost belt, the axes of major folds plunge 25°—45° to the southeast, and there is a strong lineation parallel to the axes. In the central belt, the early southeasterly structural trends have been modified by the Cascade, Merrimac, p and Hartman Bar plutons. Lineation around the northwestern end of the Cascade pluton plunges to the northeast, at right angle to that in the eastern belt. A second folding around this northeast lineation is strongly developed between the Cascade and Merrimac plutons, where the rocks of the Horse— shoe Bend Formation were squeezed at right angles to the preplutonic trends and refolded around the north- east axes. Distribution of the rock types on the north— east side of the Cascade pluton indicates that the major folds, strongly overturned to the southwest, are on the southeast-plunging axes. The attitude of bedding planes and the easterly plunge of lineation in the nar- row belt of metamorphic rocks between the Cascade and Bald Rock plutons also indicates overturning to the southwest. These structural relations are shown in a north- eastward-trending cross section from Bald Rock pluton METAVOLCANIC AND METASEDIMENTARY ROCKS 5 across the Melones fault into the Shoo Fly Formation at the northeast corner of the Bucks Lake quadrangle (fig. 2). The movement along each of the five faults—— Melones, Rich Bar, Dogwood Peak, Camel Peak, and Big Bend—was down on the southwest side, supporting the hypothesis of subduction along the late Paleozoic and early Mesozoic continental margin, as suggested by Burchfiel and Davis (1972). The five faults can be interpreted as Mesozoic surface features of the subduc- tion that earlier (during the Paleozoic) gave rise to an extensive island-arc-type volcanism in this area. With few exceptions, the faults are preplutonic, as shown by sharp crosscutting relations in the canyon of the Mid- dle Fork of the Feather River south of American Bar. In age, the plutons average 130 million years (Grommé and others, 1967); the older plutons are in the eastern part. The largest is the composite Cascade pluton, which, partly covered by Tertiary volcanic rocks, underlies an area of about 195 square kilome- tres in the central part of the area (pl. 1). Parts of the Merrimac and Bald Rock plutons are exposed in the westernmost part of the area, and the southern part of the Hartman Bar pluton is along the northern border of the area. The conformable contacts of the Cascade pluton and its elongate curved shape that conforms with the arc in the trends of the metamorphic rocks suggests that this pluton was emplaced somewhat ear- lier than the Merrimac and Bald Rock plutons. This is confirmed by a tongue of younger trondhjemitic rocks that extends from the Bald Rock pluton to the Cascade pluton, cutting its structures discordantly. Two small masses of altered diorite, the Lumpkin pluton and a mass at Frey Creek exposed between the Cascade and Bald Rock plutons, are probably somewhat older than the Cascade pluton. They consist of diorite that is more thoroughly recrystallized than the quartz diorite in the large plutons. A small mass of trondhjemite at Shute Mountain is similar to the trondhjemite in the Bald Rock pluton and is probably an offshoot of the same magma. The Cascade pluton was emplaced in an anticli— norium (sections A—A’ and B—B’), the broken apex of which is near Sky High in the northwestern part of the area of plate 1 and the western limb of which was squeezed between the Cascade and Merrimac plutons and refolded around northeast-trending axes. Gener- ally the trends curve around the plutons, giving an impression of the shouldering effect of the invading magma. The details of the structural features are given after the petrographic description of the rocks. METAVOLCANIC AND METASEDIMENTARY ROCKS The lithology and petrography of the metavolcanic Franklin Canyon Formation and the interbedded metasedimentary and metavolcanic rocks of the Horseshoe Bend Formation are similar to lithology and petrography of these formations in the Pulga and Bucks Lake quadrangles (Hietanen, 1973a). Therefore only a brief description is given in this paper, the em- phasis being on differences rather than similarities. FRANKLIN CANYON FORMATION The Franklin Canyon Formation—considered Devo— nian(?) in age-—consists of meta-andesite, metadacite, metasodarhyolite, metatuff, and a small amount of phyllite. In its western part, it is mainly interlayered metadacite and metasodarhyolite that contain some interbedded metatuff. In its eastern part, a thick layer of andesitic metatuff is overlain by meta-andesite that includes thick flows, many showing pillow structures. The structural relations suggest that metadacite and interbedded metasodarhyolite are older and in part faulted against the meta-andesitic rocks. In the east- central part of the study area (pl. 1)—near American House and along Slate Creek—and to the south, thick discontinuous layers of fine-grained black phyllite are interbedded with andesitic metatuff. Thinner discon- tinuous layers of similar phyllite are interbedded with metatuff farther north along Lost Creek and its tributaries. These layers have well-preserved bedding that helped in mapping the structure. A few thin layers of metachert not shown on the map are inter- bedded with metatuff. South of American House and along Slate Creek, the andesitic metatuff grades through tuffaceous metasediment into black phyllite. Fragments of this phyllite and metatuff~well exposed in the canyon of Slate Creek—in overlying mafic meta- andesite prove that the meta-andesite is younger. A few discontinuous thin layers of granular gray quartz- ite are interbedded with phyllite and metavolcanic rocks in the southeastern corner of the area. The mafic meta—andesite along Simon Ravine and Gold Run is coarser grained than meta-andesite elsewhere, suggesting a thicker flow or closeness of a former vent. The existence of a vent is supported by the occurrence of swarms of inclusions of black phyllite in meta-andesite in the gorge of Gold Run. The individual fragments in these swarms, in length a fraction of a centimetre to several metres, are from a layer of black phyllite that extends southward to Simon Ravine and beyond. This phyllite is underlain by metatuff on the west and bounded by a fault on the east. Some small rounded fragments of metatrondhjemite—presumably a deep-seated equivalent of metamorphosed soda- rhyolite—are included in the coarse-grained meta- andesite along Simon Ravine, supporting the relative 6 METAMORPHISM AND PLUTONISM, FEATHER RIVER AREA, CALIFORNIA ages inferred from the structures of the older silicic "western and somewhat younger andesitic eastern part 0‘? this belt. Metadacite is a fine-grained greenish-gray rock in which some quartz can be identified with a hand lens. In many layers volcanic bombs and other pyroclastic structures similar to those described from the Bucks Lake quadrangle (Hietanen, 1973a) are common, indi- cating the explosive character of the eruptions. Thin sections show that most metadacite contains pheno- crysts of albite and occasionally some quartz embedded in the groundmass that consists of albite, actinolite, green hornblende, chlorite, epidote, clinozoisite, quartz, sparse biotite, leucoxene, magnetite, and py— rite. Albite includes numerous tiny crystals of actino— lite, epidote minerals, and some muscovite. Vesicles are filled by quartz. Amphibole prisms, many consist- ing of green hornblende with pale actinolitic ends, are either randomly oriented or show a weak parallel orientation. The interbedded tuffaceous layers have the same mineralogy but are clearly schistose. Many layers have fine-grained lapilli consisting of a mixture of sericite plus epidote. Metasodarhyolite is lighter in color than the metadacite and contains more quartz and albite. Thin sections show that quartz and albite occur as pheno- crysts; the albite includes numerous tiny grains of epi- dote and muscovite. Groundmass consists of quartz, al- bite, chlorite, and muscovite with or without biotite or actinolite. Leucoxene and magnetite are common ac— cessory minerals. lnterbedded tuffaceous layers are rich in quartz and muscovite and contain some chlo- rite, biotite, epidote, leucoxene, magnetite, and hema- tite. Large granulated phenocrysts of quartz are com- mon. Meta-andesite is a medium greenish-gray fine- grained tough—to-break rock in which no quartz can be seen under a hand lens. Clean outcrops along the riv- ers reveal both pyroclastic and pillow structures. Thin sections show that meta-andesite consists mainly of epidote minerals, green hornblende, actinolite, albite, and chlorite in varying proportions. Sphene, leuco- xene, magnetite, and pyrite are common accessory minerals. In places, quartz occurs in small round grains and secondary veinlets. Phenocrysts in the original andesite were pyroxene, hornblende, or plagioclase or any two of these miner- als. Most of the pyroxene phenocrysts have been al- tered to chlorite with some sphene or to chlorite and actinolite. In a few places, as at the Little Grass Valley Dam, phenocrysts of pyroxene are well preserved in massive parts of the flows and in centers of pillows. These phenocrysts, rimmed by chlorite, are embedded in a thoroughly recrystallized groundmass consisting mainly of epidote and amphiboles. Euhedral phenocrysts of hornblende, 1—6 mm long, occur in meta-andesite on the South Fork of the Feather River about 1 km below the Little Grass Val- ley Dam and to the south. The pleochroism of the hornblende in the phenocrysts is y=brownish green, B=green, a=pale yellowish green; green to pale-green hornblende occurs as an alteration product. Pyroxene has been pseudomorphosed to aggregates of actinolite + chlorite. Actinolite forms slender prisms that show parallel orientation and are separated by a mesh of fine-grained chlorite. Former plagioclase phenocrysts consist of epidote + albite. Groundmass is a fine- grained mixture of epidote, actinolite, hornblende, and albite. Phenocrysts of albite, 0.5—1.5 mm long, are common in meta-andesite at Lexington Hill. Ferromagnesian minerals in this rock are chlorite and hornblende. Some aggregates of chlorite include small grains of epidote and show outlines of former pyroxene crystals. Most epidote forms large aggregates irregular in out- line, but some grains are scattered. Many small frac- tures are filled by epidote crystals and quartz. Small amounts of quartz in tiny euhedral grains, in small oval aggregates, and as a fracture filling are common. The meta-andesite, some with pillow structure and some with augite phenocrysts, is rich in calcium (most of it contained in epidote), iron, and magnesium, and poor in silicon (Hietanen, 1973a) and can be best clas- sified as basaltic meta-andesite. It is intimately as- sociated with metadacite and metamorphosed sodarhyolite and with more silicic meta-andesites con- taining hornblende and plagioclase phenocrysts, all these rocks being differentiates of a potassium-poor andesitic magma. Most of the andesitic metatuff is distinctly bedded, the fine-grained dark-gray layers, 0.2—5 cm thick, al- ternating with brownish-gray coarse-grained layers of the same thickness. Contacts between the beds are usually gradational in both directions. Graded bedding is rare and can seldom serve as a marker for the tops of the beds. Thin sections show that in the coarse-grained layers angular to lens-shaped lithic fragments consisting of albite, epidote, actinolite, quartz, and muscovite in various proportions are embedded in matrix that con— sists mainly of epidote and actinolite. Some euhedral to subhedral prisms of brownish—green hornblende, sub- hedral grains of altered plagioclase and quartz, are scattered in the matrix. Leucoxene and magnetite are common accessory minerals. The fine-grained layers consist mainly of albite, epidote, and actinolite with 5—10 percent quartz and some chlorite and muscovite. Layers of tuffaceous metasediments that are inter— bedded with metatuffs contain more quartz and micaceous minerals and less epidote and actinolite METAVOLCANIC AND METASEDIMENTARY ROCKS 7 than the normal metatuffs. Scattered fragments of quartz and altered phenocrysts of plagioclase and hornblende are common. All black phyllite is fine grained, distinctly bedded, and has a strong b lineation. The major constituents are round to angular grains of quartz and some albite in a fine-grained matrix of quartz, albite, muscovite, chlorite, biotite, epidote minerals, leucoxene, magne- tite, and hematite. The dark color is imparted by dis- seminated magnetite in the matrix. The lighter colored layers are generally coarser grained and contain less magnetite. These phyllite layers are most likely sedimentary in origin. Quartzite in the southeastern corner of the area is thin bedded. Layers 1—5 cm thick of this light-gray granular quartzite are separated by micaceous laminae or interbedded phyllite. HORSESHOE BEND FORMATION The Horseshoe Bend Formation—considered Per- mian(?) in age—is exposed all around the Cascade plu- ton. In the north, a zone of the Horseshoe Bend about 1% km wide separates the Hartman Bar pluton from the Cascade pluton. Toward the east, the Horseshoe Bend Formation butts against the Camel Peak fault for a distance of 2 km (fig. 2; pl. 1), and farther south only a thin discontinuous strip of the Horseshoe Bend is exposed between the pluton and serpentine that ac- companies this fault to the southern border of the map area (pl. 1). On the west side of the pluton, the Horse- shoe Bend Formation is exposed between the Cascade and Merrimac plutons and, farther south, between the Cascade and Bald Rock plutons. The Lumpkin pluton was emplaced into this formation. The Horseshoe Bend Formation consists of inter- bedded metasedimentary and metavolcanic rocks. The metasedimentary rocks are quartzite and phyllite with discontinuous layers of marble. The metavolcanic rocks are mainly metabasalt and metatuff with inter- bedded discontinuous layers of meta-andesite, metada- cite, and metarhyolite, all more extensive and more numerous in the northern and southern parts of the area than in the central part, where only a few thin discontinuous layers of metarhyolites are present. Stratigraphy of the formation is difficult to establish because the rocks are tightly folded, overturned, and faulted. Distribution of the rock types and the attitude of bedding near Marble Cone along the Middle Fork of the Feather River north of the Cascade pluton suggest that the metasedimentary units on either side of the meta-andesite—metadacite unit belong to the same layer. The marble dips gently (30°—60°) under the vol- canic unit, whereas dips in the metasedimentary unit above are steep (70° to vertical). These structural fea- tures indicate a syncline overturned to the southwest (cross section B—B’, pl. 1). The quartzite-phyllite unit on the northeast side contains several thin discontinu- ous marble layers that may be lateral extensions of the thick marble layer at Marble Cone. Thin layers of metatuff and metabasalt underlie the interbedded quartzite-phyllite unit that includes marble lenses. Another unit of metabasalt, stratigraphically in the upper part of the formation, overlies the phyllite southeast of the Hartman Bar pluton. A similar sequence is exposed between the Cascade and Merrimac plutons and continues therefrom to the southeast between the Cascade and Bald Rock plutons. The metabasalt unit exposed at Shute Mountain and Vicinity is overlain by a thick sequence of interbedded quartzite and phyllite that includes discontinuous layers of white to gray marble. The tops of the beds as well as the plunge of the fold axes along the Little North Fork are to the northeast. Some thin layers of metatuff and lens-shaped bodies of metadacite and metarhyolite also are interbedded. The depositional sequence is interrupted by faults near Sky High, and it seems likely that the northern part of the section is overturned. If it is, the metabasalt at the northern bor- der of the area is equivalent to the metabasalt unit at Shute Mountain, and the thick layer of white to gray marble exposed in the canyon of the Little North Fork west of Sky High might be in the same stratigraphic horizon as the marble at Milsap Bar and its counter- part near the mouth of the Little North Fork. The extension of the Horsehoe Bend Formation be- tween the Bald Rock and Cascade plutons south of the dikelike body of trondhjemite that cuts through this narrow zone of metamorphic rocks consists of rock types similar to those exposed north of the trondhje- mite but contains only a few thin calcite—bearing layers. The metabasalt just east of the Bald Rock plu— ton north of Feather Falls is equivalent to the metabasalt at Shute Mountain and is overlain by in- terbedded quartzite and phyllite that include tuffa- ceous beds and discontinuous layers of metabasalt, metarhyolite, and meta-andesite. This sequence con— tinues southward to the west side of the southern tip of the Cascade pluton, where a quartzite and phyllite unit is overlain by basaltic metatuff and metabasalt. The western part of the section south of the Lumpkin pluton is probably thickened by several faults and may include rocks younger than the Horseshoe Bend For- mation. This section is well exposed along the deep gorge of the South Fork of the Feather River and on the roadcuts south of it. Several long thin bodies of serpen- tine in this section were probably emplaced along the faults, a concept supported by the repetition of se- quences of quartzite, phyllite, and metavolcanic rocks in narrow zones between these serpentine bodies. Metavolcanic rocks in the southernmost part of the 8 METAMORPHISM AND PLUTONISM, FEATHER RIVER AREA, CALIFORNIA area are similar to those along the northern boundary. Small lenseshaped bodies of meta—andesite, metada- cite, and metarhyolite occur with metatuffs and metabasalt. Potassium feldspar content of metarhyol- ites ranges from negligible to about 10 percent. The deposition of the Horseshoe Bend Formation seems to have begun with basaltic volcanism, followed by a quiet period during which sediments that pro- duced quartzite, phyllite, and marble were laid down. A second episode of volcanism followed and was period- ically interrupted long enough to allow intervening sedimentation. It is probable that the lower metabasalt unit north and east of the Bald Rock pluton is equivalent to the metabasalt exposed in the central part of the Pulga quadrangle called the Duffey Dome Formation (Hiet— anen, 1973a). This formation, like the metabasalt ex— posed between the Merrimac and Bald Rock plutons, includes some quartzite, metatuff, and metarhyolite. The Duffey Dome Formation underlies the metasedi— mentary unit of the Horseshoe Bend Formation at the headwaters of Marble Creek, Pulga quadrangle (Hietanen, 1973a), and thus seems to have the same stratigraphic position as the metabasalt in the Vicinity of Brush Creek and Shute Mountain. The quartzite of the Horseshoe Bend Formation is white to light gray, granular, and contains muscovite and biotite in varying amounts. Most of it is thin bedded with micaceous laminae and thin (112—1 cm thick) layers of phyllite. The beds in the gray quartzite are 1—10 cm thick and in the white quartzite 10-50 cm thick. There is every gradation from quartzite to phyl- lite and to micaceous layers with albite. Some pebbly layers and conglomerate are interbedded south of Lit- tle Marble Cone, west and northeast of Sky High, at Milsap Bar and on Fields Ridge. The pebbles are quartzite, more rarely metarhyolite or aplite. Some of the rocks mapped as quartzite are thin-bedded metachert, some metamorphosed weathered rhyolitic tuff. Phyllite is fine grained and dark gray, rarely black. It consists of quartz, biotite, muscovite, some epidote, albite, magnetite, and hematite. Small crystals of tourmaline occur in some layers. Numerous small crystals of red garnet occur in quartz-rich phyllite south of Little Marble Cone; elsewhere garnet is rare. Calcareous layers are common in the phyllite near marble layers. Beds with lenses of white marble and others with abundant actinolite and epidote are inter- bedded in several localities south of the Hartman Bar pluton. A part of the epidote in these calcareous layers is segregaged into small lenses, (3 by 5 cm in size) and pyrite is a common accessory mineral. Disseminated carbon makes some layers black. Marble lenses in the black phyllite exposed in the streambed of the South Branch of the Middle Fork of the Feather River 11/2 km northeast of Morgan Bar are 1—20 cm long and consist of either medium-grained calcite or calcite and quartz. In some of the small lenses, calcite and quartz grains have a spiral arrangement, giving an impression that they were fossils, but complete recrystallization makes further identification impossible. Thin sections show a granoblastic mixture of calcite and quartz devoid of all organic textures. Pebbly layers are interbedded with phyllite that ex- tends southeast from the vicinity of Maynards Ranch. East of the mouth of Lost Creek, this unit consists of dark-gray metagraywacke in which subangular to elongate clasts, 1—20 mm long, are metachert, quartzite, phyllite, and some metavolcanic rocks. Small fragments of quartz and occasionally albite are common in the matrix which consists mainly of micaceous minerals, quartz, and albite and is heavily dusted by carbonaceous material and iron oxides. In- terbedded are tuffaceous layers with clasts of metavol- canic rocks in a matrix that contains chlorite, am- phibole, epidote, and disseminated iron oxides. Fine-pebble conglomerate and associated lithic metagraywacke that are somewhat less deformed than the other rocks of the Horseshoe Bend Formation are exposed 1—3 km south of Clipper Mills between Oroleve and Empire Ridge. Outcrops along a logging road 1 km south of Clipper Mills show a poorly developed folia- tion parallel to the regional trends. Boulders of very hard fine-pebble conglomerate farther south show a well-developed parallel orientation of subangular elongate clasts, in which relict structures and textures of Paleozoic source rocks are preserved. Two-thirds of the clasts consist of metachert and quartzite; the re- maining one-third consists of phyllite, metavolcanic rocks, and marble. These clasts are similar to the rocks that make up the Calaveras Formation in the Bucks Lake quadrangle (Hietanen, 1973), about 30 km to the north and northeast. Only a few clasts consist of indi- vidual grains of quartz, hornblende, and plagioclase. The metavolcanic clasts consists of hornblende, biotite, epidote, albite, quartz, magnetite, and sphene altered to leucoxene. Hornblende, biotite, quartz, epidote can be identified in the matrix, which is heavily clouded with iron oxide. The similarity in the minenralogy of the volcanic clasts and the matrix suggests that the matrix contains tuffaceous material. A similar lithic metagraywacke with an interbedded pebbly layer is exposed on the east side of the North Yuba River, east of the plutonic mass along the river. A fault separates the graywacke from the fine-grained phyllite to the east. Because of the closeness of the pluton, the graywacke here is strongly deformed and recrystallized. Marble in the northern part of the area is distinctly v METAVOLCANIC AND METASEDIMENTARY ROCKS 9 bedded; thick (10—50 cm) coarse-grained layers of white calcite alternate with thinner light-gray layers that contain some muscovite, biotite, magnetite, and pyrite, and others that contain actionolite and epidote. Chemical analyses of marble near Milsap Bar (Hieta— nen, 1951, table 2) show that carbonate is calcite con- taining 0.8 percent MgO. Much of the marble in the canyon of the North Yuba River in the southern part of the map area has a lenticular structure. Lenses of coarse-grained white marble ranging from a few cen- timetres to several metres in length are embedded in dark amphibolite-rich matrix. Metabasalt is dark greenish gray to black and gen- erally well foliated. It was earlier called amphibolite (Hietanen, 1951), but relict textures, such as volcanic breccia and phenocrysts, prove its volcanic origin. Metabasalt consists of bluish-green hornblende, plagi- oclase, epidote minerals, and some quartz and biotite or chlorite. Accessory minerals are magnetite, sphene, leucoxene, and hematite. Plagioclase phenocrysts are common and usually include small crystals of hornblende. Phenocrysts of hornblende and aggregates of feltlike hornblende (presumably pseudomorphs after augite) occur in places. Vesicles are filled with calcite or epidote. Layers of metatuff associated with metabasalt are thin bedded, dark to medium gray, and consist of hornblende, plagioclase, quartz, epidote, and magne- tite. Near the plutons these layers are recrystallized to medium-grained hornblende gneiss. Thin layers of quartzite and phyllite are interbedded. Meta-andesite and metadacite are similar to the cor— responding rock types just north of the area in the Bucks Lake and Pulga quadrangles (Hietanen, 1973a, 1974); all consist mainly of hornblende, actinolite, epi- dote minerals, albite, and quartz, in varying propor— tions. _ A few thin discontinuous layers of metarhyolite are interbedded with metabasalt and metasedimentary rocks west of the Cascade pluton. These are light gray to white, rich in quartz and feldspar, and contain less than 15 percent biotite plus muscovite; some hornblende and epidote are common. Difference in the potassium feldspar content indicates that there are two types of metarhyolite: The occurrences around the Horseshoe Bend at the northern boundary of the area of plate 1 contain very little or no potassium feldspar, whereas in the lenses west of Brush Creek, potassium feldspar content is 10—15 percent by volume. Metarhyolite that contains numerous thin hexag- onal plates of biotite oriented parallel to a gneissoid texture (fig. 3) was described and illustrated earlier (Hietanen, 1951) from a locality about 1 km northeast of Mountain House, in the southeastern border zone of the Merrimac pluton. Since the earlier fieldwork, other FIGURE 3.—Hexagonal plates of biotite (black) in metarhyolite north of .vIountain House, southeastern contact zone of the Merrimac pluton (loc. 1470). lens—shaped bodies of this unusual rock have been ex- posed in new roadcuts. Most of these are elongate in— clusions in the quartz dioritic border zone of the Mer— rimac pluton; some metarhyolite is interlayered with phyllite and biotite quartzite that borders the pluton in this vicinity. The hexagonal biotite plates, 1—2 mm long and 0.1 mm thick, are embedded in a fine-grained groundmass that consists of albite, quartz, orthoclase, and muscovite. Biotite phenocrysts make up about 20 percent of this rock; quartz and albite phenocrysts are sparse. All phenocrysts are undeformed and are bounded by smooth euhedral crystal faces; in this, they contrast with the common deformed shape and in part sutured boundaries of the phenocrysts in the other metarhyolites of this formation. The biotite-rich metarhyolite is exceptionally well preserved or is perhaps younger, as similar undeformed textures are typical of Mesozoic metavolcanic rocks southwest of the area. (See section "Age and Correlation”) A mesonorm in molecular percentages was calcu- lated from the chemical analyses published earlier (Hietanen, 1951, table 1, analysis 406). It shows 25.9 quartz, 36.3 albite, 1.9 anorthite, 8.7 orthoclase, 11.5 muscovite, 14.4 biotite, 1.2 sphene, 0.2 apatite, and 0.1 magnetite, and it reflects the true mineral content of this rock, except that the actual percentage of biotite is higher (about 21) and that of muscovite lower (5), and that all iron and titanium are in biotite. The percent- age of biotite is much higher than in normal rhyolites; that of orthoclase, lower. The euhedral biotite pheno- crysts must have crystallized early in the magma, which was rich in iron (3.9 weight percent FeO) and contained more magnesium (1.6 percent MgO) and water than present in normal rhyolitic magma. The Mg/Fe ratio of biotite on the basis of rock analysis is 10 METAMORPHISM AND PLUTONISM, FEATHER RIVER AREA, CALIFORNIA 0.74, and its A1203 content about 17 percent. Small occurrences of metarhyolite rich in biotite elsewhere in the Horseshoe Bend Formation (as a small outcrop in loc. 1570) have textures typical of common metarhyolite. Biotite is in small flakes or in thin laminae parallel to the foliation. Phenocrysts of albite and quartz have sutured borders, and many are granulated or deformed. STRUCTURES In the Franklin Canyon Formation, bedding is dis- tinct and well preserved only in the tuffaceous and sedimentary layers (fig. 4). Pillow structures are well exposed on high vertical walls of a rock quarry at the south end of the Little Grass Valley Dam and along Slate Creek in the southeastern part of the area. The shape of the pillows indicates that the formation is right side up at these localities. Elsewhere, distribu- tion of the rock types brings out a discontinuous layer- ing and shows that all rocks are folded. In the north, folds are tight and isoclinal; toward the southeast, they become increasingly gentle, and in the eastern part of the mapped area (pl. 1), an S-shaped curvature is evi- dent in the metatuff and interbedded black phyllite exposed along Slate Creek and vicinity. A roadcut be- tween Slate Creek and Poverty Hill (fig. 5) exposes a recumbent fold on a southeast-plunging axis over— turned to the southwest. Lineation is well developed in the tuffaceous and sedimentary layers but absent in the massive parts of the flows. In the tuffaceous layers, lapilli are stretched parallel to it; in the sedimentary layers, it is an inter— section of bedding and foliation. It plunges 25°—60° to the south-southeast. Where folds are exposed, their FIGURE 4._Flat-lying bedding and poorly developed steep_second cleavage in black Phyllite on Slate Creek, eastern part of the map— ped area (pl. 1). FIGURE 5.~Recumbent fold on southeast-plunging axis, indicating overturning to the southwest, in black phyllite at 4,200-foot eleva- tion east of Slate Creek. axes coincide with this lineation, which is thus a b line- ation. Foliation is marked in the metatuff and phyllite where micaceous minerals and amphiboles are parallel or subparallel to it. The massive parts of the flows are in general surrounded by foliated rock in which de- formed blocks of massive lava are included. Foliation can be measured in most outcrops. In the metatuff and phyllite, it is parallel to the axial planes of folds and generally makes an angle with the bedding planes. In summary, the structures indicate that the Franklin Canyon Formation is isoclinally folded and has a well-developed axial plane foliation and a b lineation. Overturning to the west is common. A second lineation, presumably an a lineation, was observed in only a few outcrops, as on Slate Creek in the eastern part of the area. The outcrops of metatuff and phyllite along Slate Creek show a second cleavage that strikes about N. 20° W. and is mostly vertical (fig. 4). This cleavage is a set of closely spaced fractures re- sembling slaty cleavage. There is no mineral orienta- tion parallel to it, and it must therefore be later than the foliation. It is most likely related to the faulting. Structures of the Horseshoe Bend Formation differ in many respects from those of the Franklin Canyon Formation, owing in part to the material folded and in part to deformation by emplacement of nearby plutons. Bedding is prominent in most rock types of this forma— tion. In the quartzite and phyllite, thin beds, 1—5 cm thick, rich in either quartz or micas alternate or are separated by mica laminae. Individual beds in the marble (2—100 cm thick) are separated by thin dark- colored layers that contain some mica, magnetite, and pyrite. Thin (1—20 cm thick) phyllitic layers are inter- bedded in many places. Rocks mapped as metabasalt METAVOLCANIC AND METASEDIMENTARY ROCKS 1 1 may include tuffaceous layers and some thin layers of quartzite and phyllite. In metatuff, beds are a few mil- limetres to about 20 cm thick, rarely thicker. The meta-andesite and metadacite that occur in rather thin layers and lenses in the northern and southwestern parts of the area include some tuffaceous layers that yielded during the deformation but left the flows unde- formed. Two sets of folds and two lineations are apparent in the metamorphic rocks near the plutons. The trend of the major axis deviates from its regular northwesterly direction around the plutons, having been modified by them. The plunge of the major axis is commonly to the east or to the southeast. The folds on this axis are tight and isoclinal, and the rock masses are elongate paral- lel to it. A second set of folds have axes parallel to the linea- tion that plunges to the northeast or nearly so, about at right angles to the major fold axes. This lineation, an a lineation in relation to the major folds, is strongest around the northwestern end of the Cascade pluton, where thin-bedded phyllite has small folds around it. Large folds around the a lineation are evident in the rocks between the Cascade and Merrimac plutons. It seems reasonable to assume that this folding and line— ation are later than the major folding and were formed during the emplacement of the plutons. Compton (1955, pl. 1) mapped a similar lineation in the border zone of the Bald Rock pluton and in its metamorphic wall rocks south of Brush Creek. The parallelism of the lineation in the pluton and its wallrocks certainly proves a common origin. This second lineation, promi— nent in the Horseshoe Bend Formation but negligible in the Franklin Canyon Formation, is a later feature than the major northwest trends. AGE AND CORRELATION The metavolcanic and metasedimentary rocks can be tentatively correlated with units in nearby areas on the basis of lithology and continuation of major struc- tures. The southward continuations of the Franklin Canyon and Horseshoe Bend Formations are shown as Paleozoic metavolcanic and metasedimentary rocks on the geologic map of California, Chico sheet (Burnett and Jennings, 1962). The metabasalt around the southern panhandle of the Lumpkin pluton is shown as Mesozoic on that map. Detailed mapping for this study, however, suggests that this metabasalt is a part of the Horseshoe Bend Formation. Interbedded layers of quartzite, metatuff, and metarhyolite, similar to those in the Horseshoe Bend and its northward extension of andesitic metabasalt, underlie the quartzite of the Horseshoe Bend Formation east of Sugar Pine Point at the south end of Lumpkin Ridge. It was earlier pointed out (Hietanen, 1973a) that the succession of pyroclastic and sedimentary units of the Franklin Canyon and Horseshoe Bend Formations is broadly similar to the Devonian and Permian in the Taylorsville area as given by McMath (1966). Correla— tion of the tectonic elements of the northern Sierra Nevada and the Klamath Mountains (Davis, 1969; Irwin, 1966) suggests that deposition of the rock units west of the Melones fault may have been roughly con- temporaneous with units west of the Trinity fault in the Klamath Mountains. Specifically, the Franklin Canyon Formation could be correlative with Devonian pyroclastic rocks in the Klamath Mountains, and lime- stones of the Horseshoe Bend Formation could be cor- relative with the lenses of coarsely crystalline Permian limestone in the Paleozoic sequence in the Klamath Mountains. A Permian(?) age was assigned by Creely (1965) to similar limestone west of the study area. The Franklin Canyon Formation that underlies the Horseshoe Bend Formation in the Pulga quadrangle (Hietanen, 1973a) consists of a metamorphosed potassium-poor pyroclastic andesite-sodarhyolite se- quence typical of island arcs. These rocks overlie the metachert and interbedded phyllite mapped as the Calaveras Formation in the adjoining Bucks Lake quadrangle (Hietanen, 1973a). All these rocks are ex- posed just west of a serpentine belt along the Melones fault and together form a sequence that is common on the ocean floors. This stratigraphy supports the con- tention of Burchfiel and Davis (1972) that in Devonian time an island—arc system extended from the Klamath Mountains to the northern Sierra Nevada. In contrast to the potassium-poor andesite and sodarhyolite in the Devonian(?) Franklin Canyon Formation, the vol— canogenic rocks in the Permian(?) Horseshoe Bend Formation are basalt and in part potassium-rich rhyo— lite, and the interbedded metasediments include shallow-water limestones. These changes in composi- tion resulted from changes in conditions of formation of magmas and of deposition in a tectonically active belt. (Discussed under “Origin of Magmas.”) Comparison of the structures and textures of the Franklin Canyon Formation with petrologically simi- lar but much less deformed Mesozoic metavolcanic rocks southwest of the Big Bend fault (fig. 2) supports the suggestion of a Paleozoic age for the Franklin Can- yon Formation. The metavolcanic rocks southwest of the Big Bend fault form the northern end of the west- ernmost metamorphic belt of the Sierra Nevada. The metavolcanic rocks of this belt are generally consid- ered Jurassic in age (Taliaferro, 1943; Clark, 1964, Bateman and Wahrhaftig, 1966). The Mesozoic metavolcanic rocks southwest of the Big Ben fault are well exposed along the North Fork of 12 METAMORPHISM AND PLUTONISM, FEATHER RIVER AREA, CALIFORNIA the Feather River east of Las Plumas. A brief descrip- tion and chemical analyses of the major rock types in this section were included in the Merrimac report (Hietanen, 1951). These rocks belong to a calc-alkaline andesitic suite poor in potassium. The major rock types are meta-andesite and metadacite with interbedded layers of metatuff. Metamorphosed sodarhyolite and basaltic andesite are intercalated with metadacite in the eastern part of the section. Petrologically these rocks are similar to the Franklin Canyon Formation, but there are certain distinct structual, textural, and chemical differences. The most notable differences between these two an— desitic suites are in structures and textures of meta- andesite, metadacite, and metamorphosed sodarhyo- lite. In the Franklin Canyon Formation, all rocks are generally strongly deformed, and recrystallization has modified or obliterated the original textures. The vol- canic bombs and pillows are flattened parallel to the foliation and stretched parallel to the lineation. The phenocrysts have partly sutured boundaries and con- tain mineral inclusions, such as chlorite and am- phiboles in plagioclase, which indicate migration of ions during the metamorphism. In contrast the Mesozoic rocks are either undeformed or show only a weak foliation. The primary structures and textures such as volcanic bombs and phenocrysts have pre- served their original shape or are only slightly elon- gate parallel to the regional trends (see Hietanen, 1951, fig. 3; fig. 1, pl. 2 and fig. 1, pl. 3). Phenocrysts are bounded by well-developed smooth crystal faces, and their inclusions are products of isochemical recrystalli- zation (save addition of H20) such as epidote after anorthite component of plagioclase and muscovite after orthoclase originally contained in the plagioclase phenocrysts. Comparison of the chemical composition of the cor- responding members of these two andesitic suites shows that the rocks of the Franklin Canyon Forma- tion (Hietanen, 1973a, table 1, analyses 463, 464, 461) contain less Si02 and alkalies and more CaO and MgO than their less altered Mesozoic counterparts (Hieta- nen, 1951, table 1, analyses 2, 3, 4, 6). This is in agreement with the trend of migration of material in the Paleozoic formations during their regional meta- morphism. It was suggested in an earlier paper (Hietanen, 1973b) that the metamorphic rocks above a plutonic magma chamber were bled of elements (mostly Si, K, Na) needed for the formation of mon- zotonalitic magmas by differential melting and were enriched in residual elements (Ca, Fe, Mg). These chemical changes are reflected in the quantitative mineralogy of the rock units of the Franklin Canyon Formation. The metavolcanic rocks bordering the Bald Rock pluton around its southeast corner are probably cor- relative with rocks shown as Mesozoic metavolcanic rocks on thegeologic map of California, Chico sheet (Burnett and Jennings, 1962). Field observations on Sucker Run, on a small dirt road north of the South Fork of the Feather River, and on the Old Forbestown Road south of this river suggest that these Mesozoic(?) metavolcanic rocks are separated by a north-trending fault from the metasedimentary rocks of the Horseshoe Bend Formation in the east. ‘Metabasalt, metadacite, and metarhyolite were identified on the west side of the fault. Metadacite has long slender laths of plagio- clase and clusters of epidote embedded in a fine— grained groundmass consisting of quartz, feldspar, chlorite, muscovite, and some epidote and biotite. Il- menite, partly altered to leucoxene, is a common acces- sory mineral in this rock. The fine-pebble conglomerate and associated meta- graywacke south and southeast of Clipper Mills is younger than the Paleozoic formations (Calaveras?) from which the clasts were derived. These early Paleozoic rocks had undergone deformation and re- crystallization before they were broken up and the fragments transported into a basin that also received some tuffaceous material from erupting volcanoes nearby. The distance of transportation was short enough to preserve the subangular shapes of many clasts but long enough to allow considerable sorting according to the pebble size (fig. 6). The deposition of this graywacke was after a Paleozoic deformation and recrystallization but before the Jurassic (Nevadan) orogeny and before the Mesozoic volcanism. METAMORPHOSED INTRUSIVE ROCKS The most mafic among the regionally metamor- phosed synkinematic intrusive rocks are the perido- tites and pyroxenites that occur mainly as long t in W . _ , .. FIGURE 6.—-—Fine-pebble conglomerate 1 km south of Clipper Mills. METAMORPHOSED INTRUSIVE ROCKS bodies along the fault zones and are partly altered to serpentine, soapstone, and talc schist. Metagabbro, metadiorite, and metatrondhjemite in small bodies elongate parallel to the structural trends represent in— trusive equivalents of the metavolcanic rocks they in- trude and with which they were deformed and recrys- tallized. It has been shown (Hietanen, 1973a) that in chemical composition and trace-element content, the synkinematic intrusive rocks are similar to the metavolcanic series but different from the post- kinematic Late Jurassic and Early Cretaceous plutons. The medium-grained equigranular texture of the syn- kinematic intrusive rocks suggests emplacement at shallow depths. SERPENTINE AND PERIDOTITE Numerous elongate bodies of peridotite at various stages of alteration to serpentine, soapstone, and talc schist are enclosed in the metamorphic rocks, mainly along the fault zones. The Camel Peak fault is accom— panied by these ultramafic rocks for all of its exposed length (fig. 2; pl. 1). Most of the thick bodies consist of peridotite that is altered to serpentine minerals only along tiny cracks and fractures. The border zones of these bodies consist of serpentine, soapstone, and talc schist, the alterations being similar to those of ul- tramafic rocks described in the Bucks Lake quadrangle (Hietanen, 1973a). A layer of talc-tremolite rock is common along the contacts. The peridotite is dark greenish gray and fine grained. It consists of anhedral grains of olivine 0.02— 0.03 mm in diameter, prisms of enstatite and augite 0.1—0.2 mm long, grains of magnetite, and some ser- pentine minerals along tiny fractures. Colorless hornblende rich in magnesium (Hietanen, 1973a) is a primary mineral in some of the peridotite. Parts of the bodies are sheared and completely altered to serpen- tine and soapstone. METAGABBRO A large inhomogeneous body of metagabbro is ex- posed along Slate Creek just east of the serpentine that accompanies the Camel Peak fault. Three somewhat smaller (3—4 km long) bodies lie between the Cascade pluton and Lumpkin pluton. A fairly large mass is in the southeastern corner of the area, but elsewhere only a few small bodies occur next to serpentine and metadiorite. The largest body is well exposed along Slate Creek below the dam, which is 200 metres north of the southern border of Plumas County. The metagabbro is a dark coarse— to medium—grained slightly foliated rock in which hornblende and plagio— clase can be identified in the hand specimen. In thin sections the hornblende is blue green and strongly 13 pleochroic (y = blue green, [3 = green, 01 = very pale green) and in places forms clusters of small prisms. Plagioclase (An35) contains abundant small inclusions of quartz, epidote, and hornblende or consists of a mix- ture, mainly epidote with less albite. Quartz, actino- lite, and clinozoisite occur in varying amounts. The common accessory minerals are magnetite, ilmenite, and rutile. The variable ratios of the major-con— stituents give rise to a considerable inhomogeneity within the masses. Brecciated metagabbro at the west end of the bridge over Slate Creek, 150 metres north of the southern boundary of Plumas County, consists of angular fragments of fine-grained dark metagabbro and interstitial light—colored material consisting of plagioclase, epidote, and quartz with chlorite and some hornblende. METADIORITE The largest mass of metadiorite lies just east of thin masses of metagabbro and serpentine that accompany the Camel Peak fault. Most of the other masses intrude the Franklin Canyon Formation. Metadiorite is a medium-grained hornblende-plagioclase-epidote- quartz rock that shows a crude foliation. Hornblende is the chief dark mineral, constituting about 30 percent of the rock. It is blue-green and strongly pleochroic near the postkinematic plutons but pale green else- where. Actinolite occurs in tiny prisms in plagioclase and forms the ends of some green hornblende prisms. Plagioclase is albitic in masses far from the plutons, rich in anorthite near the plutons. Epidote minerals, clinozoisite and epidote, more rarely zoisite, constitute about 30 percent of the metadiorite outside 1—2 km wide contact zone of the plutons. Most metadiorite con- tains about 10 percent quartz, some leucoxene, magne- tite, and pyrite. METATRONDHJEMITE Several elongate conformable bodies of metatrond- hjemite occur in the eastern part of the area. In outcrop these rocks resemble metadiorite except that they con- tain less hornblende and more quartz. They are medium grained, light bluish gray, and slightly foliated. They contain about 60 percent albitic plagio- clase and 25—30 percent quartz. Potassium feldspar is absent. Dark constituents are either chlorite and epi- dote, or green hornblende, actinolite, biotite, and epi- dote. Muscovite occurs in places. Rutile, leucoxene, and magnetite are the common accessory minerals. Albitic plagioclase includes numerous small grains of epidote and some sericite. As hornblende increases and quartz decreases, these rocks grade through metatonalite to metadiorite. The texture is granoblastic, with large anhedral to subhedral grains of plagioclase and inter- stitial aggregates of small strained quartz grains that 14 METAMORPHISM AND PLUTONISM, FEATHER RIVER AREA, CALIFORNIA have interlocking boundaries. These aggregates were originally individual large grains of quartz that were granulated during deformation. The mineral content, and presumably the chemical composition, of metatrondhjemite is similar to that of the metamorphosed sodarhyolite. The coarse grain size and equigranular texture indicate deep-seated cooling of the trondhjemitic magma. The metatrondhjemite is thus a deep-seated equivalent of the metasodarhyolite, a relation similar to that described earlier between meta-andesite and metagabbro and between metada- cite and metadiorite in the Bucks Lake quadrangle- (Hietanen, 1973a). METAMORPHISM Metamorphism has equally affected the metavol- canic, metasedimentary, and the oldest intrusive rocks. All these rocks were recrystallized to the border- line of the greenschist and epidote-amphibolite facies during a regional metamorphism that accompanied the folding on the major northwest axes. A second (con- tact) metamorphism affected rocks in about a mile wide zone around the Jurrassic and Cretaceous plu- tons. In this zone, higher grade minerals such as staurolite, cordierite, and andalusite crystallized in the pelitic layers, and blue-green aluminum-rich hornblende, clinozoisite, and epidote crystallized in the metavolcanic rocks, indicating pressure-temperature conditions near the borderline of the epidote- amphibolite and amphibolite facies. The relation be- tween the deformation, recrystallization, and plutonism is similar to that described from the area farther north (Hietanen, 197 3a). The Franklin Canyon Formation is not exposed near the plutons, and its recrystallized mineral as- semblages are those typical of the low grade metamorphism. Primary minerals such as augite and hornblende are rarely preserved. The outline of pseudomorphs allow the identification of the original phenocrysts and together with the bulk composition serve to identify the original rock types. The typical mineral assemblage is epidote-c1inozoisite-albite— actinolite-green hornblende-chlorite with or without quartz. Biotite and muscovite with some garnet occur in the interbedded pelitic layers. These mineral as- semblages indicate recrystallization at temperatures and pressures near the borderline of the greenschist and epidote-amphibolite facies. The Horseshoe Bend Formation envelopes the Jurassic and Cretaceous plutons and sustained a higher grade metamorphism and stronger deformation than the Franklin Canyon Formation. The most com- mon amphibole in the metavolcanic rocks is a blue- green variety that in the Bucks Lake quadrangle oc— curs in the highest grade zone near the plutons and was found to be rich in aluminum and sodium (Hieta- nen, 1974). Most phyllite contains biotite but no chlor- ite; garnet occurs in only a few localities. Metatuff next to the Bald Rock pluton recrystallized as hornblende gneiss with a grain size coarser than that present in metatuff farther from the contact. Metabasalt and associated metatuff between the Mer- rimac and Bald Rock plutons were recrystallized as amphibolite and hornblende gneiss. A small lens of interbedded calcareous sediment (Ice. 1519) recrystal— lized as coarse-grained grossularite-diopside-epidote- plagioclase (An33)-quartz rock with some zoisite, clinozoisite, and sphene. Similar contact rocks are common in calcareous layers in the Pulga quadrangle, where the mineral assemblages in the interbedded pelitic layers (staurolite-andalusite-biotite-quartz in the outer and andalusite-cordierite-biotite-quartz in the inner contact aureole) indicate metamorphism at the lower border of the amphibolite facies (Hietanen, 1967, 1973a). PLUTONIC ROCKS A sequence of postkinematic plutonic rocks, deter- mined on the basis of field relations, structures, tex- tures, mineralogy, and composition, comprises three groups: earliest postkinematic intrusions, the main mass of the Cascade pluton and the Hartman Bar plu- ton, and trondhjemite. (1) The earliest postkinematic intrusions include an altered gabbro on Swain Hill and vicinity and related dioritic masses, the Lumpkin plu- ton, and a small mass of quartz diorite at Frey Creek. The westernmost marginal mass of the Cascade pluton resembles the quartz diorite at Frey Creek in its mineralogy and texture and may have been intruded during the same episode. (2) The main mass of the composite Cascade pluton was emplaced later than the westernmost marginal mass. The main mass grades from foliated quartz diorite along the borders to mas- sive monzotonalite in the southeastern interior. The Hartman Bar pluton is texturally and mineralogically similar to the main part of the Cascade pluton and presumably is comagmatic and was emplaced during the same episode. (3) A trondhjemitic tongue extends eastward from the northern part of the Bald Rock plu- ton and cuts discordantly the metamorphic wall-rocks between the two plutons and the western part of the Cascade pluton. The trondhjemite was therefore emplaced after the Cascade pluton had solidified. Each of the three groups has characteristic textural and mineralogic features that help in correlating these rocks with the plutonic rocks farther north, where the Bucks Lake, Grizzly, and Merrimac plutons were dated by Grommé, Merrill, and Verhoogen (1967) at 142 (Late Jurassic) to 128 (Early Cretaceous) million years. PLUTONIC ROCKS 15 LUMPKIN PLUTON AND RELATED ROCKS A large body of altered gabbro exposed around Swain Hill northwest of Feather Falls was briefly described by Compton (1955) in connection with his study of similar rocks—considered Late Jurassic by him— south of the Bald Rock pluton. Two small dioritic plu- tons, the Lumpkin pluton and the quartz diorite at Frey Creek, are partly surrounded by the altered gab- bro on Swain Hill. Theltotal area covered by these al- tered plutonic rocks is about 28 square kilometres. Several small masses of altered quartz diorite occur south of Lumpkin pluton. The texture of all these rocks is hypidiomorphic, with subhedral plagioclase and hornblende. As is common in the postkinematic plu- tons, only the border zones are foliated; in the centers, minerals are randomly oriented. These rocks, however, are partially recrystallized, as shown by abundant epi- dote, granulated quartz, and small prisms of bluish- green hornblende included in plagioclase. This partial recrystallization, due to reheating during the emplacement of younger plutons, was accompanied by addition of water. An outcrop of pyroxene gabbro that resembles the pyroxene diorite in the center of the Bucks Lake pluton, except for more pyroxenes in the gabbro, is exposed in a roadcut north of Swain Hill (loc. 1337, pl. 1). Because of their partial recrystallization, all these rocks are considered to be older than the large plutons but considerably younger than the deformed and regionally metamorphosed synkinematic intrusive rocks. The pyroxene gabbro north of Swain Hill consists of ortho- and clinopyroxene, hornblende, plagioclase (An7o), and some olivine, quartz, magnetite, and sphene. The texture is granoblastic, with rounded pyroxene grains. A few poikilitic hornblende crystals include round grains of pyroxene and plagioclase. Large pyroxene grains include rounded olivine grains, some partly altered to a mixture of chlorite, serpen- tine, and magnetite. Olivine that is not included in pyroxene and some large poikilitic magnetite grains have a narrow corona of green hornblende. Symplectic intergrowth of orthopyroxene and magnetite in the centers of some pyroxene crystals may indicate former olivine. Plagioclase (An7o) is in stubby subhedral crys- tals that are 1/2—11/2 mm long and show polysynthetic twinning. Hornblende gabbro is very dark gray to black, fine to medium grained, and consists mainly of hornblende and plagioclase. On Swain Hill green hornblende is rimmed by blue-green hornblende; similar hornblende occurs also as individual homogeneous crystals of vari— ous sizes. The centers of large grains and aggregates of green hornblende have inclusions of small round grains of quartz. In the Bucks Lake quadrangle, this texture was traced to pyroxene that had been altered to hornblende during a later phase of intrusion (Hieta- nen, 1973a). In hornblende gabbro near the Bald Rock pluton, all hornblende is intensely blue green. Some of the grains are large holoblasts that include quartz, plagioclase, epidote, magnetite, and numberous tiny grains of magnetite and scales of ilmenite. In most hornblende gabbro, plagioclase is An5z65; near the Bald Rock pluton, it is An40 and is accom- panied by many large grains of epidote and many tiny inclusions of quartz. Plagioclase grains are either sub— hedral or anhedral and irregular in outline. Complex twinning is common. Small flakes of biotite, some quartz and alteration products, epidote and chlorite are more abundant along the border zones than in the central part. Apatite in small euhedral crystals, magnetite, ilmenite, and sphene are the accessory minerals. The quartz diorite on Frey Creek is coarse- to medium-grained hornblende-plagioclase rock with a hypidiomorphic slightly foliated texture. Plagioclase (An36—37) constitutes about 55 percent of this rock, hornblende 20—30 percent, quartz 10—20 percent, and biotite about 1 percent. The mineral content measured in two stained specimens is shown in table 1, and the chemical analysis of one of these (sample 1267) is shown in table 2. Magnetite and sphene are common accessory minerals; chlorite, epidote, and muscovite occur as alteration products. Plagioclase includes grains of epidote; the centers of some large zoned crys- tals are altered to a mixture of epidote and muscovite. Hornblende includes small grains of quartz and mag- netite and is pleochroic: y=blue green, B=green, a=light green. The blue-green color is typical of the recrystallized hornblende of this rock and contrasts with the olive-green color of the igneous amphiboles in the younger plutonic rocks. The Lumpkin pluton, at its northernmost border zone, consists of coarse-grained hornblende gabbro that grades into a dioritic rock toward the south. This gabbro is mineralogically and texturally similar to the dioritic main mass, except for a higher percentage of hornblende. It represents a marginal accumulation of early crystallized dark constituents. In places the con- tact between this gabbro and the older enveloping one is obscure. The main part of this pluton consists of brownish-gray medium-grained diorite in which the major constituents—plagioclase (An27—35), epidote, hornblende, chlorite, and quartz—occur in varying proportions (table 1, locs. 1531, 1532). Most plagioclase grains have been altered to epidote with only a small amount of twinned oligoclase. Chlorite is a common alteration product after biotite and hornblende. Horn- blende is bluish green to green and occurs in large subhedral crystals. Magnetite and apatite are the ac- cessory minerals. 16 METAMORPHISM AND PLUTONISM, FEATHER RIVER AREA, CALIFORNIA TABLE 1.—Percentage of major constituents measured in stained samples of platonic rocks [Analyst M. B. Norman; M204 is from Hietanen (1951)] Specific Biotite and Rock type No. Locality gravity Plagioclase Quartz K-feldspar hornblende Cascade Pluton Hornblende-biotite quartz diorite ____________ 1 1552 2.71 58.9 10.6 7.3 23.2 0 ____________________ 2 1200 2.71 59.7 18.3 0 22.0 D0 ____________________ 3 1209 2.74 61.0 16.4 1.2 21.4 D0 ____________________ 4 1330 2.75 AD37 58.5 20.1 0 21.4 D0 ____________________ 5 1308 2.74 A1136 62.7 16.0 0 21.3 Do ____________________ 6 1224 2.69 63.6 16.8 0 19.6 D0 ____________________ 7 1207 2.75 Amsiso 60.4 17.2 3.6 18.8 D0 ____________________ 8 1206 2.76 Amo 61.3 16.8 3.3 18.6 Do ____________________ 9 1208 2.73 A1135 66.4 13.6 2.0 18.0 D0 ____________________ 10 1328 2.58 63.3 18.1 0 18.6 D0 ____________________ 11 1309 2.76 Augean 61.3 20.3 0 18.4 Tonalite __________________ 12 1211 2.71 61.1 20.3 1.5 17.1 D0 ____________________ 13 1248 2.75 Ame—43 60.2 23.8 0 16.0 Do ____________________ 14 1223 2.78 68.4 15.8 0 15.8 Do ____________________ 15 1370 2.77 65.6 15.6 3.1 15.7 Do .................... 16 1362 2.69 65.2 19.3 0 15.5 Monzotonalite ______________ 17 1555 2.68 61.6 15.2 7.8 15.4 Tonalite __________________ 18 1556 2.74 Ana546 61.7 23.2 .2 14.9 o .................... 19 1450 2.72 56.8 23.5 5.5 14.2 D0 ____________________ 20 1364 2.71 60.4 23.5 2.0 14.1 Do ____________________ 21 1270 2.68 An32 59.3 27.0 .1 13.6 Monzotonalite ______________ 22 1271 2.70 Anzoflgo 57.3 25.2 7.2 10.3 Tonalite __________________ 23 1272 2.69 Ams 59.4 25. 7 5.2 9.7 Do ____________________ 24 1367 2.71 181126732 60.0 25 5 5.7 8.8 Do ____________________ 25 1368a 2.68 61.1 26. 2 5.2 7.5 Trondhjemite _______________ 26 1368b ___ 66.8 24.8 2.8 5.6 Trondhjemitic tongue in Cascade Pluton Trondhjemite ______________ 27 1205 2.70 An17724 64.0 25.0 2.0 9.0 ____________________ 28 1327 2.59 57.0 28.7 9.3 5.0 Do ____________________ 29 1210 2.63 A1125 56.7 30.8 8.2 4.3 Do ____________________ 30 1313 2.65 A1114732 61.5 28.7 5.7 4.1 Do ____________________ 31 1311 2.61 67.6 24.1 4.3 4.0 Do ____________________ 32 1312 2.57 66.9 24.6 5.2 3.3 Do ____________________ 33 649 2.62 62.1 25.8 8.7 3.4 Do ____________________ 34 648 2.62 61.0 28. 7 7.2 3.1 Bald Rock pluton Hornblende quartz diorite __________________ 35 1451 2.64 Ame—35 49.3 22.1 6.9 21.7 Do ____________________ 36 1326 2.64 53.1 21.2 5.6 20.1 Monzotonalite ______________ 37 1492 2.67 Ams—ei 50.0 27.5 9.4 13.1 Trondhjemite ______________ 38 1452 2.68 A1125 54.6 36.9 2.3 6.2 Granite-trondhjemite ______ 39 M204 __1, Ann 53.2 30.7 9.9 4.5 Trondhjemite ______________ 40 1461 2.64 61.6 9.1 5.6 3.7 Shute Mountain mass Trondhjemite .............. 41 1462 2.53 61.6 29.5 0 8.9 Quartz diorite in small masses at Mount Hope and near Orleve in the southern part of the area resembles quartz diorite at Frey Creek in the outcrop but is somewhat finer grained and grades in places to a light—colored tonalite. Thin sections show that plagio- clase is altered to a mixture of epidote and 'sericite that includes some small grains of hornblende and chlorite. The percentage of quartz is higher than in the Frey Creek mass, generally ranging from 15 to 30 percent. CASCADE PLUTON Two distinctly different phases of intrusion can be recognized in the composite Cascade pluton: a quartz diorite main mass that grades through tonalite to monzotonalite in the eastern part and a later trondh- jemitic tongue that cuts the western part of the pluton. The western border zone near the mouth of the South Branch of the Middle Fork of the Feather River (secs. 1 and 2, T. 21 N., R. 6 E.) is probably somewhat older PLUTONIC ROCKS 17 TABLE 1.—Percentage of major constituents measured in i 0d sun '1. 1 of] ' s—Continued Specific Biotite and Rock type No. Locality gravity Plagioclase Q1. .11 K»feldspar hornblende Merrimac Pluton Diorite ____________________ 42 1557 2.8.: . Any—.55 68.2 6.0 0 25.9 Hornblende-biotite quartz diorite ____________ 43 1468 f? 77 AIBOAO 54.9 23 3 0 21.9 0 _____________________ 44 1521 359 55.2 22 6 1.6 20.6 D0 ____________________ 45 1604 2.71 Aims—45 55.2 21.0 3.2 20.6 Do _____________________ 46 161;: 2.71 54.4 21.0 4.6 200 Do _____________________ 47 1616 2.73 Ame—37 59.2 19.1 2.7 19.0 Do ____________________ 48 1522 2.70 A1132—36 51.0 23.9 5.7 19.5 Do ____________________ 49 1610 2.73 Anzo—ss 53.2 24.3 3.9 186 Do ____________________ 50 1611 2.72 53.7 25.0 2.9 18.4 Do ____________________ 51 1515 2.64 52.9 24.7 4.1 18.4 Monzotonalite ______________ 52 1606 2.71 52.3 22.8 7.7 17.2 Tonalite __________________ 53 1615 2.72 Arm—35 57.2 24.0 3.4 15.4 Do _____________________ 54 1559 2.74 61. 7 23. 2 .2 14.9 Do ____________________ 55 1608 2.67 53.1 29 6 3.7 13.6 Do ____________________ 56 1613 2.70 Arm—45 55. 9 25.3 5.3 13.5 Monzotonalite ______________ 57 1607 2.68 57.0 23.9 5.8 13.3 Tonalite __________________ 58 1614 2.68 Amaas 58.1 26. 7 5.6 9.6 Monzotonalite _________ , ,,,,, 59 1605 2.64 A1322747 45.0 28. 7 18.5 7.8 Hartman Bar Pluton Hornblende quartz diorite ___________________ 60 1214 2.77 Anss 61.9 17.7 0 20.4 Tonalite __________________ 61 1281 2.75 57.9 31.5 0 10.6 Trondhjemite ______________ 62 1282 2.67 53.2 39.5 0 7.3 Do ____________________ 63 1292 2.70 49.7 43.8 0 6.5 Lumpkin pluton Diorite ____________________ 64 1531 2.98 48.2 11.1 “A, 40.7 Quartz diorite ______________ 65 1532 2.90 55.9 13.8 --__ 30.3 Frey Creek mass Hornblende quartz diorite __________________ 66 1267 0 AIB6<37 56.0 14.0 0 30.0 Do ____________________ 67 1333 2.77 Ans2 60.3 20.3 0 19.4 Hampshire Creek mass Monzotonalite ______________ 68 1621 2.66 A1123 36.8 41.2 16.2 5.8 Indian Valley mass Tonalite __________________ 69 1910 2.72 Amo 57.0 23.3 4.7 15.0 than the border zone of the main mass. This western border zone, about 1—2 km wide, consists of altered quartz diorite that in many respects resembles the quartz diorite at Frey Creek. Moreover, it contains dis— continuous thin layers of metamorphic rocks and dike- like bodies of intrusive rocks—hornblendite, gabbro, and trondhjemite. On the south slope of the Middle Fork (10c. 1206, pl. 1), a fault separates this altered quartz diorite from the normal border-zone type in the east. A similar altered quartz diorite is exposed for 100 metres in a roadcut about 700 metres to the southeast. This large inclusion is enclosed in the quartz diorite of the normal border zone, indicating that the older al- tered quartz diorite formerly occupied a wider area in this Vicinity. The main part of the Cascade pluton is zoned in an irregular manner. The border zones and the west- central part consist of hornblende-biotite quartz diorite that is devoid of potassium feldspar (see table 1; fig. 7); in the southeast-central part, this rock grades through a lighter colored tonalite to monzotonalite. A tongue of light-colored trondhjemite that extends from the Bald Rock pluton through the metamorphic wall-rocks and western border phase into the main mass of the Cas- cade pluton continues northward and forms the north- western tip of the Cascade pluton. A sharp discordant contact between this youngest rock and the main mass is exposed in roadcuts on Watson Ridge. The western border zone, believed to be somewhat older than the main mass, consists of well-foliated 18 TABLE 2.-——Chernical composition in weight and ionic percentages, molecular norm, and trace-element contents of platonic rocks [Analystsz chemical analyses, Elaine Brandt; trace elements, Nancy Conklin} Sample ,,,,,,,,,,,,, 126 1313 7 1367 Locality , Frey Creek South Fork, Feather River Rock type ,,,,,,,, Hornblende Biotite tonalite Trondhjemite quartz diorite Cascade pluton Cascade pluton Weight Cation Weight Cation Weight Cation percent percent percent Si02 ,,,,,,,,,,,,,,, 58.28 56.46 66.76 62.26 72.62 67.42 Ti02 ,,,,,.,._ .W. .86 .62 .31 .22 .12 .08 A1203(Al03/2) ,,,,,, 17.86 20.40 17.06 18.75 15.72 17.20 F8203(F903/2) ,,,,,, 2.43 1.77 1.05 .74 .43 .30 FeO ... H. ..,,..,, 4.80 3.89 1.93 1.51 .66 .51 MnO .16 .13 .09 .07 .05 .04 MgO W, H 2.35 3.40 1.29 1.79 .33 .46 CaO ,.. 6.60 6.85 4.57 4.57 2.70 2.69 Na20(NaOi/2),,i.. ._. 2.92 5.49 4.40 7.96 5.25 9.45 K20 Om. , H ‘ r 25:0 :09»qu Eli! 3:268»: 3:0: - >o=m> baanbm \ \ 0:205? cocustgo \ Icl DISSVHHI 01 (DNVINOAEIG L uE—oczm 252:2 r 0 TT 3 V A S @883 buns—Exoamu E 32%: «3:25 0 Ssmh II D f. r 3832 Angufimxennu D “a H 22:: 3235 M H "NV W. 82.50 a L a S a II s v m .w _ 3 m m _ ZO~H 10 20 30 4O 50 0 FIGURE 10.——Ratios of total feldspar (F), quartz (Q), and femic minerals (M) in stained specimens of plutonic rocks. Numbers refer to those in the first column in table 1. Specimens containing 7—18 percent potassium feldspar are encircled. 24 METAMORPHISM AND PLUTONISM, FEATHER RIVER AREA, CALIFORNIA As shown by chemical analyses, the composition of biotite tonalite from the eastern part of the Cascade pluton (sample 1367, table 2) is similar to that of tonal- ite in the Granite Basin pluton in the Bucks Lake quadrangle. No analysis of the quartz dioritic western part of the Cascade pluton is available, but on the basis of its mineralogy, it is considered to be similar to the quartz dioritic border zones of the Bucks Lake, Grizzly, and Granite Basin plutons farther north (Hietanen, 1973a). The monzotonalitic rocks in the east-central part of the Cascade pluton (table 1, locs. 1271, 1450, 1555) are mineralogically and presumably also chemi— cally similar to the central monzotonalites in the Grizzly, Oliver Lake, and Granite Basin plutons. These interior parts contain 2.3.5 percent K20 and 4—4.5 per- cent N a20. The chemical analysis of trondhjemite (table 2, sam- ple 1313) shows more than 5 percent Na20 and only 1.5 percent K20, a part of which is contained in biotite and muscovite. The potassium feldspar content is about 5 percent. The calcic cores of many zoned plagioclase crystals raise the CaO content to 2.7 percent. Compari- son with light-colored biotite tonalite from the Cascade pluton (sample 1367, table 2) shows a lower percentage of MgO, FeO, and CaO and a higher percentage of SiO2 and NazO in the trondhjemite, which has a much lower content of hornblende, biotite, and anorthite compo- nent and a higher content of quartz. The potassium content of trondhjemite is only slightly lower than that in the biotite tonalite, about 1 percent lower than that in the monzotonalitic parts of the interior, but much lower than that found earlier (Hietanen, 1973a) in the granitic parts of the Grizzly pluton (fig. 2). The differentiation of the plutonic magma was from a quartz dioritic border to either a potassium-rich or a potassium-poor end product in the center. This split- ting of the magma into two branches of differentiation is best illustrated in an Ab-Or-An diagram (fig. 11), Where the new analyses are shown in relation to the trend line for the differentiation of the plutonic rocks in the Pulga and Bucks Lake quadrangles to the north . (Hietanen, 1973a). The trondhjemites plot toward the Ab corner far from this trend line. Dikelike bodies of bluish-gray medium-grained trondhjemite (loc. 1368b) in biotite tonalite represent the local end product dur- ing the crystallization of tonalitic magma. The scarcity of potassium feldspar in this end product is partly due to impoverishment of magma in K20 by crystallization of a large amount of biotite in the tonalite. Trace elements (table 2) show trends similar to those in the plutonic rocks in the adjoining Pulga and Bucks Lake quadrangles (Hietanen, 1973a), supporting the suggestion that all plutonic magmas have the same parent magma. Ba and Sr increase with the increasing Si02, whereas the V, Cu, Co, Y, and Sc decrease. In comparison with the quartz diorites farther north, the hornblende quartz diorite from the Frey Creek mass is exceptionally low in Cr and Ni. STRUCTURES AND EVOLUTION OF THE PLUTONS Foliation is well developed parallel to the contact along most border zones of large plutons but becomes obscure toward the cores. It is considered to have been produced by the movement and pressure of the magma that was still rising in the center while the borders were crystallizing. In most plutons, the last- crystallized part of core is not in the center but con— spicuously to one side, generally the east side, indicat- ing, together with flow structures, that the plutons are tilted to the west. The intrusion most likely followed preexisting eastward-dipping structures. From the description of the rock types in the Cascade pluton, it is clear that the history of this pluton is more complicated than that of most other plutons in the Feather River area (fig. 2). The zoned main mass has a foliation roughly concentric around the south-central part, the last to crystallize. The lineation plunges 45°—55° to the southeast, parallel to the major fold axes of the wallrocks, suggesting that their structures de- termined the direction of the invading magma. This structure, together with the strongly elongate shape, suggests that the pluton was emplaced earlier than most other plutons in the northwestern Sierra Nevada. The mineralogy, in particular the relative abundance An 35 // To // \\// / 1313; / Mto / \ \ \ \/\ \\ O \\ M204 \~ \ '\ \ Tr /Gr-tr / \\l\ \\\\\ / Gr / \\\\\ \\\\\\\ Ab 15 30 50 Or Tr=trondhjemite, Grrtr: granitertrondhjemite, Grzgranite, To=tonalite, Mto:monzotonalite, Qm=quartz monzonite, Qd=quartz diorite Gd=granodiorite, Di=diorite. FIGURE ll.—Normative Ab-Or-An diagram showing points for new analyses in relation to the differentiation curve for the plutonic rocks in the Pulga and Bucks Lake quadrangles. ORIGIN OF MAGMAS 25 0f epidote and other products of alteration, supports the concept of a somewhat older age and longer history. The trondhjemitic tongue that extends from the Bald Rock pluton eastward through its foliated border zone and through the metamorphic wallrocks, and farther east cuts the western border zone of the Cascade pluton discordantly, must have the age of similar trondhjem— itic rocks in the interior of the Bald Rock pluton. The trondhjemitic dikes that crystallized from the residual magma of the biotite tonalite in the eastern part of the Cascade pluton show crystallization differ- entiation in the Cascade pluton was from biotite quartz diorite to trondhjemite, a trend similar to that in the Bald Rock pluton except for a larger amount of biotite and less potassium feldspar. These relations suggest that the Cascade pluton is an earlier differentiate of the tonalitic magma from which the Bald Rock pluton was emplaced at a somewhat later phase. Together all the plutonic rocks probably span plutonic activity from the Late Jurassic to Early Cretaceous. In contrast to the trondhjemitic trend in the Cascade and Bald Rock plutons, the differentiation in the plu- tons to the north of the study area, as in the Grizzly pluton (fig. 11), was toward a granitic composition, with an increase in the potassium feldspar content to- ward the silicic end members. None of the plutons in the study area (fig. 2), however, have granitic cores; rather, the dominant rock type in the centers is mon- zotonalite with a higher content of anorthite in plagio- clase, a larger amount of biotite and hornblende, and less potassium feldspar than is common in granite. The percentage of potassium feldspar in the monzotonalite is similar to that in parts of the trondhjemitic core of the Bald Rock pluton; the percentages of dark con- stituents and the anorthite content of plagioclase are higher. ORIGIN OF MAGMAS The following three major groups of igneous rocks are recognized in the study area: Paleozoic metavol- canic rocks (early orogenic), Mesozoic plutonic rocks (late orogenic), and Tertiary volcanic rocks (post- orogenic, Hietanen, 1972). A similar sequence of magma generation from early volcanism to plutonism during the orogeny and a later postorogenic volcanism is common in most orogenic belts. These three groups have distinctive chemical and mineralogic characteris- tics that suggest different origins. A short summary of the geologic history of the Feather River area provides some information that could lead to a better under- standing of the relations between the magma types and the tectonic environment in which they were gen- erated. ISLAND-ARC-TYPE VOLCANIC ROCKS The earliest igneous rocks extruded during the dep- osition are volcanic and volcaniclastic rocks of the andesite-sodarhyolite suite, mapped as the Franklin Canyon Formation. Some of these rocks in the Bucks Lake quadrangle rest on a sequence of interbedded metachert and phyllite (the Calaveras Formation), in- dicating a younger age for the metavolcanic sequence. The main metavolcanic unit is southwest of the Calaveras belt and is bordered by faults, the Dogwood Peak fault on the northeast and the Camel Peak fault on the southwest. Displacement along the Dogwood Peak fault was probably minor, for in the Bucks Lake quadrangle (Hietanen, 1973a) meta—andesite similar to that of the Franklin Canyon Formation overlies the rocks of the Calaveras Formation on the northeast side of this fault. Moreover, there are no ultramafic rocks along it, as there are along all the other major faults. An inhomogeneous sequence of interbedded metavol— canic and metasedimentary rocks (the Horseshoe Bend Formation) that includes minor discontinuous lime- stone beds is exposed southwest of the Franklin Can- yon Formation. Field relations in the Pulga quad- rangle suggest that the Horseshoe Bend is the younger. The andesite-sodarhyolite suite (the Franklin Can— yon Formation), here considered probably correlative with the Devonian island-arc-type andesites of the Klamath Mountains, was deposited on a sequence of interbedded chert and shale, which are typical sedi- ments on the ocean floors. This stratigraphy is compat- ible with that in the tectonic environment of island arcs along the continental margin as postulated for the Devonian by Burchfiel and Davis (1972, p. 102). The andesitic volcanism in the Feather River area was fol- lowed by deposition of volcanic, volcaniclastic, and sedimentary rocks (the Horseshoe Bend Formation) that include minor carbonate layers indicating deposi- tion in shallow water. Most rhyolites of this period, probably Permian, are rich in potassium feldspar rather than in albite, as in the Franklin Canyon For— mation. The formations exposed in the successive fault blocks to the west (fig. 2) are progressively younger. This progression, together with overturning of the folds to the southwest, can be considered as surface expression of a large-scale underthrusting to the east, consistent with the concept of subduction of the Pacific Ocean floor under the North American continent dur- ing Paleozoic and Mesozoic time (fig. 12). The trace of the subduction zone would be west of the area of figure 2, most likely in the Coast Ranges, where the common occurrence of glaucophane schists indicates high pres- sures and low temperatures during metamorphism. The andesitic magmas that gave rise to island-arc-type volcanism in the Devonian originated above this sub— duction zone (Hietanen, 1973b). Before the introduction of the concept of plate tec— tonics, calc-alkalic andesite was generally considered 26 METAMORPHISM AND PLUTON ISM, FEATHER RIVER AREA, CALIFORNIA to be a derivative of basaltic magma either through fractionation (Kuno, 1968) or by contamination from sialic material. The hypothesis of contamination in the island arcs is refuted by the thinness of the crust there. Uniformly low Sr87/Sr86 ratios typical of all island-arc- type andesites also tend to refute this hypothesis, as pointed out by Taylor (1969) and McBirney (1969). In the light of plate tectonics, it has been suggested that calc-alkalic andesite magma is derived by melting of the oceanic lithosphere (Dickinson, 1970) or by partial melting of mantle peridotite under hydrous conditions (Yoder, 1969). Seismic studies on the modern island-arc systems show that the descending sea floor can be traced as a high-velocity lithospheric slab to depths of 350 and 700 km (Isacks and others, 1969; Barazangi and others, 1970; Yoder, 1971). The slab stays rigid and cooler than the surrounding mantle, making it improbable that the andesitic magmas of the island arcs would be generated by melting of the downgoing oceanic litho- sphere. It is clear that long before this slab reaches temperatures at which andesite begins to melt, it undergoes certain chemical changes such as loss of water and other volatiles from the sediments attached to the sea floor and from hydrous minerals in the oceanic crust. The water ascends to the peridotitic mantle of the upper plate above (1a in fig. 12 and Hietanen, 1973b), causing there two critical changes as shown by the experimental work of Yoder (1969): The water lowers the temperature at which the perido- Franciscan rocks Great Valley sequence . Early volcanism; a: Island are andesite is derived from the mantle peridotite of the conti- nental plate in hydrous conditions (Devonian?). b: Continued melting of the peridotite in anhydrous conditions and at greater depth yields basalt and potassium-rich rhyolite (Permian?). \ \\ 2. Plutonic magmas; melting of the downfolded metamorphic rocks, \ the subducted oceanic lithosphere, and the mantle produced large \ quantities of monzotonalitic magma (Late Jurassic and Cretaceous \\ plutonism). . . . . . \ 3. Early Tertlary volcanlsm; postorogemc basalt and andesrte are derived \ 3 \\ from the mantle below and above the subducted oceanic slab, being a product of differentiation or contamination by crustal material (Tertiary). \\\ 4. Cenozoic volcanism east of Sierra Nevada. tite begins to melt by about 250°C. The composition of the first melt is andesitic and not basaltic, as it would be in anhydrous conditions and at higher temperatures from the same parent peridotite (Yoder and Tilley, 1962). Both volcanic suites in the study area include silicic lavas, conventionally considered to be products of frac- tional crystallization. It has been pointed out that metasodarhyolite and metadacite appear to underlie the meta-andesite of the Franklin Canyon Formation, indicating that volcanism started with extrusion of sodarhyolite. This sequence of events agrees with the hypothesis of magma formation by fractional melting. In the Horseshoe Bend Formation, discontinuous lay- ers of metarhyolite containing a considerable amount of potassium are interlayered with metabasalt, suggesting more or less contemporaneous extrusion. Yoder (1973) has suggested a possible mechanism for the generation of silicic and basic magmas from a common parent rock, such as the rocks in the mantle. The first liquid to form at elevated temperatures is silicic. This silicic liquid continues to form until all available quartz is exhausted. With removal of the silicic melt, the composition of the remaining parent rock becomes increasingly more mafic, thereby requir- ing an elevation of temperature for the resumption of its melting. The liquid formed at this second stage of fractional melting is andesitic in hydrous conditions but basaltic in anhydrous conditions, the basaltic liquid requiring a higher temperature. FAULT 1 MELONES Basin and Range Province Pl, Paleozoic belt Mz, Mesozoic belt \\ \ Extensional deformation and volcanism in the Basin and Range province resulted from release of compressional stresses at the termination of subduction by the intersection of the subduction zone with the East Pacific Rise. FIGURE 12,—Evolution of magmas in the northern Sierra Nevada. ORIGIN OF MAGMAS Nothing was said by Yoder (1973) about the possible variation in the potassium content of the silicic melts formed in the first stage of fractional melting. This appears to be a crucial point in the generation of Devo— nian(?) and Permian(?) magmas in the study area, as it was during the Cenozoic volcanism in western North America in general (Dickinson and Hatherton, 1967; Dickinson, 1970; Lipman and others, 1972). Dickinson and Hatherton (1967) and Dickinson (1970) have plotted the potassium content in volcanic rocks with various silica contents relative to the depth of the inclined seismic zones beneath the volcanoes in active Cenozoic island arcs and marginal continental ranges of the circum—Pacific orogenic belt. Comparison of the metavolcanic rocks of the Feather River area with their diagrams would suggest, after allowing for the changes due to metamorphism, that the potassium—poor andesite-sodarhyolite suite of the Franklin Canyon Formation (Devonian?) originated from a shallower depth than the younger potassium— rich lavas of the Horseshoe Bend Formation (Per— mian?) The dacitic and sodarhyolitic rocks analyzed contain 1—0.2 percent K20 (Hietanen, 1951, 1973a), which unmodified, and applying Dickinson’s (1970) plots of potassium content (K) relative to depth of magma chambers (h) for active Quaternary volcanic island arcs, yields depths of 50—100 km for the loci of generation of silicic magmas of the Franklin Canyon Formation. In contrast, the rhyolites of the Horseshoe Bend Formation contain 2—4 percent K20, yielding depths of 150—300 km according to Dickinson’s scheme for the inclined seismic zone. The deeper level could have resulted from thickening of the upper plate above the magma chamber, either by accumulation of vol- canic and sedimentary material and deformation of the upper plate (Hietanen, 1973b) or by thicker parts of the wedge—shaped upper plate overriding the heat source, or both. Why does the potassium content of the magmas in- crease with depth of the magma chamber, thus with the increase in the pressure and temperature of the magma generation? The experimental work by Tuttle and Bowen (1958) on the fractional crystallization of granitic magma is illuminating. Tuttle and Bowen (1958, p. 78) showed that in the ternary system albite- orthoclase-quartz, the eutectic moves toward the quartz-albite sideline with increasing water-vapor pressure. The first liquid formed by fractional melting from the mantle in hydrous conditions would therefore be a sodium-rich rhyolite, followed at higher tempera- tures by an andesite liquid. The later (Permian?) magmas were generated from the mantle peridotite that had lost most of its water during the earlier period. The first Permian(?) liquids were therefore normal rhyolites with 2—4 percent K20, and the basic 27 magmas were basaltic. The lower water-vapor pres- sure at this second stage was coupled with higher total pressure that resulted from an increased load above. Higher temperature is indicated by the basaltic com- position of the melt. An alternative explanation is based on the stability relations of amphiboles and micas. Experimental work by Modreski and Boettcher (1972) and Allen, Mod- reski, Haygood, and Boettcher (1972) show that whereas amphibole is decomposed below depths of 75 km, phlogopite is stable to depths of 100—175 km under the conditions of the low geothermal gradients antici- pated in a relatively cool subducting oceanic litho— sphere. Accordingly, mainly Na and H20 would be re- leased into a melt at depths of 60—90 km, and K would be tied up in phlogopite and become available only at deeper levels. MONZOTONALITIC MAGMA The deformation caused by postulated subduction of the oceanic plate under the continental plate during the Paleozoic and early Mesozoic came to an end late in the Jurassic Period, about 148 my. ago, which was the approximate beginning of local major plutonism and a major Sierran plutonic event that culminated in the Cretaceous (Kistler and others, 1971). The problem of the origin of the granitic magmas from which the plutonic rocks crystallized is universal and has been much disputed. In the Feather River area, the mean composition of the Cretaceous plutonic rocks is not truly granitic, but monzotonalitic (Hietanen, 1961), containing less Si02 and K20 and more Ca0, FeO, and MgO than the normal eutectic granites (Tuttle and Bowen, 1958). The trace-element content of the plutonic rocks is different from that of the regionally metamorphosed igneous suites (Hietanen, 1973a), in- dicating a different origin for these two magma series. Partial melting of an oceanic plate as it descends to the mantle may take place eventually, giving rise to the generation of calc-alkalic magmas. The processes involved could be akin to those suggested by Green and Ringwood (1969) for the partial melting of quartz eclo- gite, basalt, and amphibolite on the basis of their high—pressure experimental studies. According to these investigators, calc—alkalic magma could be pro- duced in the descending oceanic slab by partial melting of quartz eclogite at depths of 80—150 km and by par- tial melting of amphibolite (derived from basalt) and gabbro at depths of 30—40 km. Chemical changes in the folded and recrystallized metavolcanic rocks in the study area, however, suggest that the metamorphic complex above the magma chamber was involved in the processes of magma generation. Comparison of the present composition of the metavolcanic rocks (Hietanen, 1951, 1973a) with 28 METAMORPHISM AND PLUTONISM, FEATHER RIVER AREA, CALIFORNIA Daly’s (1933), McBirney’s (1969) and Chayes’ averages (1969) for andesites shows that meta-andesites may have lost much of their potassium and silicon and have been greatly enriched in calcium, iron, and mag- nesium. The mineralogy and texture of the metavol- canic rocks suggest that these changes in composition took place during the metamorphism. For example, metadacites still have crystals of quartz that were slightly granulated during the deformation but are still recognizable as former euhedral phenocrysts. The rest of the rock is completely recrystallized, consisting now of epidote, amphiboles, albite, and quartz without any potassium-bearing metamorphic minerals. Large amounts of epidote and amphiboles that formed stable mineral assemblages with albite at low to medium metamorphic temperatures raise the percentages of CaO, FeO, and MgO to levels 3—5 percent higher than those common in average dacite. The sodium (and sili- con) content of the metadacite is considerably lower than that of an average dacite, and there is only a fraction of a percent K20 in the metadacite. There must have been an extensive migration of silicon and alkalies, particularly of potassium out of the metada- cite during its metamorphism. Similar chemical changes occurred during the metamorphism of ande- sites, which, relative to average andesites, were en- riched in CaO (4—9 percent), FeO (3.4 percent), and MgO (4 percent) and impoverished in Na20 (1—2 per- cent), K20 (2 percent), and Si02 (10 percent. A1203 con- tent remained remarkably unchanged. The mean composition of the plutonic rocks is close to monzotonalite that contains about 4 percent CaO, 2 percent FeO, 1.5 percent MgO, 4.5 percent Na20, 1.8 percent K20, 66 percent Si02, and 16 percent A1203. Comparison of this composition with that of an average andesite shows 1—2 percent less CaO, FeO, and MgO and more Si02 and Na20 in the monzotonalite. Had the monzotonalitic magma formed by melting of primary basaltic rocks, the differences would be even greater, and there would not be sufficient K20. It cannot be a mere coincidence that the elements lost (K, Na, Si) and gained (Ca, Fe, Mg) by the volcanic rocks during their metamorphism are those that should be added (K, Na, Si) and subtracted (Ca, Fe, Mg) from andesitic and basaltic rocks to yield a monzotonalitic composition. Rather, these changes tended to reestablish the geochemical equilibrium in the metavolcanic- metasedimentary complex when it was warped down to high temperatures during the Jurassic deformation and a eutectic melt started to form. The melting proc- ess, which produced large quantities of magma, must have occurred during the regional metamorphism (contemporaneous with the deformation) to account for the chemical exchange of elements between the magma and the surrounding metavolcanic-meta— sedimentary complex. An Sr87 /SI'86 ratio higher than that in the mantle-derived basalt and andesite, but considerably lower than that in the sedimentary rocks, as found by Hurley, Bateman, Fairbairn, and Pinson (1965) for the Sierra Nevada plutonic rocks, is in agreement with the postulated mixed origin. The metamorphic assemblages (andalusite—stauro- lite and andalusite-cordierite) in the pelitic layers in- dicate that the metamorphism was at pressures lower than the triple point of the aluminum silicates and at temperatures higher than the upper stability limit of staurolite. The temperatures about 600°C and pres- sures of 4 kb seem reasonable and could have been reached 12—15 km below the surface. The region of melting was at the lower level. Monzotonalitic magma would form by differential melting at about 700°C (Piwinskii, 1968) at depths of 25—30 km, assuming geothermal gradients of 30°—25°C/km. The mineralogic features of the plutonic rocks indicate that the mon- zotonalitic magma had a high H20 content and was most likely formed close to the solidus temperatures. Textures of the plutonic rocks suggest that this magma was never completely liquid but contained crystals of plagioclase (An40—45), epidote, hornblende, and biotite. Oscillatory zoning and resorption of early plagioclase phenocrysts indicates fluctuation of the physical condi- tions and possibly of the composition of the melt. A notable mineralogic feature is inclusions of euhedral epidote in biotite. This epidote must have crystallized earlier than the biotite that includes it. Moreover, the half-enclosed epidote crystals are bounded by crystal faces only on the biotite side; the side surrounded by plagioclase is anhedral, probably because of resorption. It is possible that the large amount of early epidote was inherited from metavolcanic rocks that were consumed by the magma. The excess of incorporated epidote could have recrystallized with euhedral shapes in a crystal mush in which most of the epidote was used up by plagioclase. The large round biotite flakes in the biotite tonalite are early and may have lost their euhedral shapes by resportion. The metarhyolite with euhedral (pseudohexagonal) biotite phenocrysts (fig. 3) north of Mountain House is a close effusive equivalent of the biotite tonalite. The biotite-rich rhyolite has only a few small phenocrysts of quartz and albite. Biotite crystal- lized early, and the rest liquid was impoverished in potassium, yielding a trondhjemitic composition to the last differentiate, as evident in the late trondhjemitic dikes that cut the biotite tonalite in the Cascade pluton. TERTIARY VOLCANIC ROCKS Volcanism in the northern Sierra Nevada recom— REFERENCES CITED menced in Miocene time by extrusion of andesitic pyroxene basalt flows, the Lovejoy Basalt, and was fol— lowed by the eruption of pyroclastic andesite in which phenocrysts are plagioclase, hornblende, and augite. The youngest flows are olivine basalt and two— pyroxene andesite. There is every gradation from olivine basalt through two-pyroxene basaltic andesite to silicic hypersthene andesite, as described earlier (Hietanen, 1972), proving that Kuno’s (1968) sugges- tion for the origin of andesite by differentiation from basaltic magma is valid here. The olivine basalt and its andesitic derivatives were extruded late in the Ter- tiary after the crust had thickened by deformation and plutonism. The postorogenic andesites have a different origin and more complicated history than the island- arc-type andesites. Using the potassium content of the parent basalt (0.8 percent, Hietanen, 1972) and Dick- inson’s (1970) K-h plot suggest depths of 120—230 km for the subduction zone near which the Tertiary basalt magma was generated. A 45° angle for the descent of the oceanic plate would bring the associated trench to the belt of glaucophane schists in the Coast Ranges (fig. 12). MAGMAS IN SPACE AND TIME This discussion has shown that in the mobile belts the composition of magmas depends on the tectonic en- vironment in which they are generated and that this composition changes with geologic time. Good exam- ples are the two distinctive types of andesites, the Paleozoic island-arc type and the postorogenic (Ter- tiary) andesites of this area. Moreover, the experimen- tal work by Yoder (1969) and Green and Ringwood (1969) suggests that the calc-alkaline andesite magma may be generated in different ways at different geologic times, the mode of origin changing with the changing physicochemical conditions in the orogenic belts. In general, four processes, supported either experi— mentally or by natural occurrences, may operate in orogenic belts through geologic time: 1. At an early stage, island—arc—type low-potassium andesite is generated from the mantle above the Benioff zone. Water from the descending sea floor ascends to the mantle above, lowering the melt- ing point of peridotite. The first melt is andesitic, as demonstrated by the experimental work of Yoder (1969) and Kushiro (1972). After exhaus- tion of water, the composition of the melt is basal- tic (Yoder and Tilley, 1962), and silicic lavas may be derived from this basalt by fractionation, or, with the presence of water, a silicic melt may be produced first, and after quartz is exhausted, the melt, which requires a higher temperature, will be basaltic (Yoder, 1973). 29 2. At a later stage, calc-alkaline andesite and related magmas may be generated by partial melting of quartz eclogite, amphibolite (derived from basalt), or gabbro in the downgoing oceanic lithosphere and in the base of the thickened crust (Green and Ringwood, 1969). Much of these magmas may not have vented to the surface but, rather, joined the plutonic magmas formed by partial melting of the crust and downfolded met- amorphic rocks to produce large quantities of monzotonalitic magmas. 3. Late-stage magmas are produced by fractionation of basaltic magmas in either hydrous or anhydrous conditions (Green and Ringwood, 1969; Kuno, 1968) or by contamination of basaltic magma by crustal material. Late Tertiary two-pyroxene andesites in the northern Sierra Nevada were de- rived from tholeiitic basalt magma by fractional crystallization of olivine (Hietanen, 1972). 4. The latest stage magmas, late Cenozoic basaltic lavas east of the Sierra Nevada, are related to an extensional deformation typical of interarc basins behind the island arcs (Karig, 1971). An interarc-type spreading in the Basin and Range province was a direct result of the termination of subduction of the Pacific floor under the North American continent. Scholz, Barazangi, and Sbar (1971) have suggested the following mechanism: Partially melted, material from the upper part of the subducting slab rises diapirically through the mantle, is trapped beneath the lithosphere, flat- tens there, and spreads outward. Extensional de- formation and volcanism occur when the stress is released. REFERENCES CITED Allen, J. C., Modreski, P. J., Haygood, C. and Boettcher, A. L., 1972, The role of water in the mantle of the Earth; the stability of amphiboles and micas: Internat. Geol. Cong, 24th, Montreal 1972, sec. 2, Proc., p. 231—240. Barazangi, M., Isacks, B., and Oliver, J., 1970, Propagation of seis- mic waves through and beneath the lithosphere that descends under the Tonga island arc: Geol. Soc. America Abs. with Pro- grams, v. 2, no. 7, p. 488—489. Bateman, P. C., and Wahrhaftig, Clyde, 1966, Geology of the Sierra Nevada, in Bailey, E. H., ed., Geology of northern California: California Div. Mines and Geology Bull. 190, p. 107—172. Burchfiel, B. C., and Davis, G. A., 1972, Structural framework and evolution of the southern part of the Cordilleran orogeny, west- ern United States: Am. Jour. Sci., V. 272, p. 97—118. Burnett, J. L., and Jennings, C. W., 1962, Geologic map of California (Chico sheet): California Div. Mines and Geology, scale 1:250,000. Chayes, F., 1969, The chemical composition of Cenozoic andesite, in McBirney, A. R., ed., Proceedings of the andesite conference: Oregon Dept. Geology and Mineral Industries Bull. 65, p. 1—12. 30 Clark, L. D., 1964, Stratigraphy and structure of part of the western Sierra Nevada metamorphic belt, California: US. Geol. Survey Prof. Paper 410, 70 p. Compton, R. R., 1955, Trondhjemite batholith near Bidwell Bar, California: Geol. Soc. America Bull., v. 66, no. 1, p. 9—44. Creely, R. S., 1965, Geology of the Oroville quadrangle, California: California Divi Mines and Geology Bull. 184, 86 p. Daly, R. A., 1933, Igneous rocks and the depths of the earth: New York, McGraw-Hill, 598 p. Davis, G. A., 1969, Tectonic correlations, Klamath Mountains and western Sierra Nevada, California: Geol. Soc. America Bull. 80, p. 1095—1108. Dickinson, W. R., 1970, Relations of andesites, granites, and deriva- tive sandstones to arc-trench tectonics: Rev. Geophysics and Space Physics, v. 8, no. 4, p. 813—860. Dickinson, W. R., and Hatherton, T., 1967, Andesitic volcanism and seismicity around the Pacific: Science, v. 157, p. 801«803. Evernden, J. F., and Kistler, R. W., 1970, Chronology of emplace- ment of Mesozoic batholithic complexes in California and west- ern Nevada: US. Geol. Survey Prof. Paper 623, 42 p. Green, T. H., and Ringwood, A. E., 1969, The high pressure experi— mental studies on the origin of andesites, in McBirney, A. R, ed., Proceedings of the andesite conference: Oregon Dept. Geol» ogy and Mineral Industries Bull. 65, p. 21—32. Grommé, C. S., Merrill, R. T., and Verhoogen, J., 1967, Paleomag- netism of Jurassic and Cretaceous plutonic rocks in the Sierra Nevada, California, and its significance for polar wandering and continental drift: Jour. Geophys. Research, v. 72, no. 22, p. 5661—5684. Hietanen, Anna, 1951, Metamorphic and igneous rocks of the Mer— rimac area, Plumas National Forest, California: Geol. Soc. America Bull., v. 62, no. 6, p. 565—608. 1961, A proposal for clarifying the use of plutonic calc-alkalic rock names, in Short papers in the geologic and hydrologic sci- ences: U.S. Geol. Survey Prof. Paper 424—D, p. 340—343. 1967, On the facies series in various types of metamorphism: Jour. Geology, v. 75, no. 2, p. 187—214. 1972, Tertiary basalts in the Feather River area, California, in Geological Survey research 1972: US. Geol. Survey Prof. Paper 800-8, p. B85—B94. 1973a, Geology of the Pulga and Bucks Lake quadrangles, Butte and Plumas Counties, California: US. Geol. Survey Prof. Paper 731, 66 p. 1973b, Origin of andesitic and granitic magmas in the north- ern Sierra Nevada, California: Geol. Soc. America Bull., v. 84, no. 6, p. 2111—2118. 1974, Composition of coexisting amphiboles, epidote minerals, chlorite and plagioclase in metamorphic rocks, northern Sierra Nevada, California: Am. Mineralogist, v. 59, p. 22~40. Hurley, P. M., Bateman, P. C., Fairbairn, H. W., and Pinson, W. H., Jr., 1965, Investigation of initial Sr87 /Sr86 ratios in the Sierra Nevada plutonic province: Geol. Soc. America Bull., v. 76, p. 165—174. Isacks, B., Sykes, L. R., and Oliver, J., 1969, Focal mechanisms of deep and shallow earthquakes in the Tonga-Kermadec Region and the tectonics of island arcs: Geol. Soc. America Bull., v. 80, p. 1443—1469. X}U.S. GOVERNMENT PRINTING OFFICE: 1976 - 690-036/97 METAMORPHISM AND PLUTONISM, FEATHER RIVER AREA, CALIFORNIA Irwin, W. P., 1966, Geology of the Klamath Mountains province, in Bailey, E. H., ed., Geology of northern California: California Div. Mines and Geology Bull. 190, p. 19—38. Karig, D. E., 1971, Origin and development of marginal basins in the western Pacific: Jour. Geophys. Research, v. 76, p. 2542—2561. Kistler, R. W., Evernden, J. F., and Shaw, H. R., 1971, Sierra Nevada plutonic cycle; Part 1, Origin of composite granitic batholiths: Geol. Soc. America Bull., v. 82, p. 853—868. Kuno, H., 1968, Origin of andesite and its bearing on the island arc structure: Bull. volcanol., v. 32, p. 141—176. Kushiro, Ikuo, 1972, Effect of water on the composition of magmas formed at high pressures: Jour. Petrology, v. 13, pt. 2, p. 311— 334. Lipman, P. W., Prostka, H. J., and Christiansen, R. L., 1972, Cenozoic volcanism and plate-tectonic evolution of the Western United States: Royal Soc. London Philos. Trans. ser. A., v. 27]., p. 217—248. McBirney, A. R., 1969, Andesitic and rhyolitic volcanism of orogenic belts, in The earth’s crust and upper mantle: Am. Geophys. Union Geophys. Mon. 13, p. SOL—507. McMath, V. E., 1966, Geology of the Taylorsville area, northern Sierra Nevada, in Bailey, E. H., ed., Geology of northern California: California Div. Mines and Geology Bull. 190, p. 173—183. Modreski, P. J., and Boettcher, A. L., 1972, The stability of phlogo- pite + enstatite at high pressures—a model for micas in the interior of the Earth: Am. Jour. Sci., v. 272, p. 852869. Piwinskii, A. J., 1968, Experimental studies of igneous rock series, central Sierra Nevada batholith, California: Jour. Geology, V. 76, no. 5, p. 548—570. Scholz, Ch. H., Barazangi, M., and Sbar, M. L., 1971, Late Cenozoic evolution of the Great Basin, Western United States, as an en- sialic interarc basin: Geol. Soc. America Bull., V. 82, p. 2979— 2990, Taliaferro, N. L., 1943, Manganese deposits of the Sierra Nevada, their genesis and metamorphism: California Div. Mines and Geology Bull. 125, p. 277—332. Taylor, S. R., 1969, Trace element chemistry of andesites and as- sociated calc-alkaline rocks, in McBirney, A. R., ed., Proceedings of the andesite conference: Oregon Dept. Geology and Mineral Industries Bull. 65, p. 43—63. Turner, H. W., 1898, Bidwell Bar, California: US. Geol. Survey Geol. Atlas, Folio 43. Tuttle, O. F., and Bowen, N. L., 1958, Origin of granite in the light of experimental studies in the system NaAlSiOsOs-KAlSi30s- SiOZ-H20: Geol. Soc. America Mem. 74, 153 p. Yoder, H. 8., Jr., 1969, Calc-alkalic andesites, experimental data bearing on the origin of their assumed characteristics, in McBirney, A. R., ed., Proceedings of the andesite conference: Oregon Dept. Geology and Mineral Industries Bull. 65, p. 77—99. 1971, Petrologic implications of plate tectonics: Science, v. 173, p. 464—466. 1973, Contemporaneous basaltic and rhyolitic magmas: Am. Mineralogist, v. 58, p. 153—171. Yoder, H. 8., Jr., and Tilley, C. E., 1962, Origin of basalt magmas—— an experimental study of natural and synthetic rocks systems: Jour. Petrology, v. 3, no. 3, p. 342~532. UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY 121,132,230” R r E, I _ Horseshooe Bend ’ 50 39340' lulu , - , . ,. 1 , - ‘v. :‘r . \ fi'fifi , . .. .1 , ' , ”it , J»: «111% _. .. -, 7‘ . ~ ~, 1,. W933 «a? w- . “" “w , 0y Scout My “3; ME‘ %{ // R ‘ . “(S/A / 4: /, I V , « , ‘ ‘ , ‘ ' " ‘ I a, d! " If» ‘ V ,7 ‘ 77 I w I I _ - I ' , H . ‘ ‘ I ' , , V ,I I , I .1 ‘ ._ {fieryoir 1W4; 1?" “v I Tyk‘fi‘ "' / A, , “13%;; \ J__ a... ,1 ”"77 II [\\ \)> Ir x 1: tv‘l § \11 ’\ ”~14. \ ' \ \ a T21 N "\ \’\\I \ "fl‘wl’x \ ”en hi I p < - I .\ § I . ‘1 .315 La I» y‘— Kay .1. o122°oo' 45' 30’ 15' 121°00' 45’ 120°30’ 4o 00' e V «£0,521 $37? A SC>é Q I (Ree me ea “3 $33? @5613” ‘35” 3: $4“ a e — R'CHARDSON - PARADISE PULGA BUCKS LAKE QUINCY —BLA|RSDF—N_ I SPRINGSe 1953 1950 193656 ,-‘, ,, 39?? 3% as F to \ .. §§ §§$§ a“? 51 cg? Q5 93,9} & 45' ‘ 5” \L 9% A e ‘v ‘0 ‘ 6‘3. (941‘? §A§q® 6x“0 58/ 0Q:9 9: \ 01 V‘ 01 Q~ Q % .. .1 see M We as“ ea 4 $3513} 3??) (3'; Q) ‘53 N CHICO BIG B E— SIERRA CITY T220 N. Q 1949 (v Q ,Q 3 1955 e , e we » v V (2 es, g ‘1 <1 \V & V15», 3&3 , <1 a? *6 o ‘3 N We? was Lie" $6” N1 e” _ \ \ \ N. 1' \I 30' 6 k a N S > M 0 «0% § Q? e V» 1: am e o c? <38 e 6” $9 $5) Se? $24“ $32 {9" N — BUTTE CITY—— GRIDLEY— BAN NY—-EMIGRANT GAP- 1954 g 1952 1955 (a e «9 1 x :3? «Q 0 Q. Q‘ o e G ‘v 9 <9 0 ‘ve (a Q Q, \Q \/ QQW , Q Q Q 9: Q Q Q a? Q w 0 9: Y N 1,? @VA 0 ‘0 o (3235,» ,5? 3’ (3)30) ‘2‘ ‘0’ 9&3? 993? 0%) e 39 15' 0 10 20 30 MILES .1..- I I I I I ‘ I ' I I I o 10 20 30 40 KILOMETRES INDEX MAP SHOWING AREA OF THIS REPORT I I. II 0 2 Lu 3 E APPROXIMATE MEAN DECLINATION, 1976 19 N SCALE 1:48 000 1 1/2 O , 1 2 3 MILES '- I—I I—I I-——-I I-——I I—I I—I I—I 3903a ‘ . '_ ‘ I 25 39030, 1 .5 o 1 2 ' 3 4 5 KILOMETRES 121 “I ‘ 5, R 8 E R 9 E} 1210 00’ I—I I—I I—I I—I I—I I—I . . I—I Base from US. Geological Survey Geology by Anna Hietanen 1969—73 CONTOUR INTERVAL 100 FEET Mooreville Ridge,and Big Bend Mtn., DATUM IS MEAN SEA LEVEL 1262 500 1948 Reservoirs as of 1973 GEOLOGIC MAP OF THE AREA AROUND THE MIDDLE AND SOUTH FORKS OF THE FEATHER RIVER, CALIFORNIA CORRELATION OF, MAP UNITS I Pliocene Miocene Eocene DESCRIPTION OF MAP UNITS EXTRUSIVE ROCKS OLIVINE BASALT — Light- to medium-gray porphyritic rock with flow structure and columnarjointing. Phenocrysts are olivine and augite. Groundmass consists of plagioclase, augite, magnetite, and some glass PYROCLASTIC ANDESITE i Mudflow breccia and pyroclastic material with round fragnents of andesite embedded in fine-grained light-gray debris. Phenocrysts of augite and oxyhornblende are in a groundmass of plagio— clase and augite LOVEJOY BASALT (Miocene) — Fine-grained black flows with some small pheno- crysts of olivine. Groundmass consists of small laths of plagioclase, tiny grains of augite, disseminated to dendritic magnetite, euhedral grains of magnetite, and interstitial glass PREVOLCANIC SEDIMENTARY ROCKS Ts AURIFEROUS STREAM DEPOSITS — Sand and gravel PLUTONIC ROCKS TRONDHJEMITE i Coarse-grained very light gray plagioclase (An 1 5-2 5)-quartz- biotite rock with some hornblende, muscovite, and potassium feldspar. Grades with increasing amount of hornblende and decreasing amount of quartz to tonalite and quartz diorite toward the borders of the pluton QUARTZ DIORITE AND TONALITE, UNDIFFERENTIATED — Coarse- grained plagioclase-quartz—biotite-hornblende rock that grades from quartz diorite at the borders to tonalite toward the interior of the plutons with decrease in hornblende and increase in quartz content. Tonalite contains large biotite crystals in places and grades to potassium feldspar-bearing monzo- tonalite toward the centers of the plutons DIORITE AND QUARTZ DIORITE, UNDIFFERENTIATED — Coarse- to medium-grained gray rock that consists of plagioclase, epidote, and hornblende with some quartz, biotite, and magnetite GABBRO — Coarse-grained dark-gray to black rock that consists chiefly of horn- blende, plagioclase (An35_65), epidote, and magnetite METAMORPHOSED INTRUSIVE ROCKS SERPENTINE AND PERIDOTITE — Peridotite consists of augite, enstatite, olivine, and magnesio-hornblende; partly altered to serpentine, talc schist, and soapstone METATRONDHJEMITE — Medium-grained, light—bluish—gray, equigranular, massive to foliated albite-quartz-biotite—hornblende rock with some actinolite, epidote, muscovite, and chlorite METADIORITE — Medium-grained, gray, equigranular, massive or slightly foliated rock consisting of hornblende albite, epidote, and quartz METAGABBRO AND HORNBLENDITE — Coarse- to medium—grained dark-gray to black equigranular hornblende-plagioclase-epidote rock, in places foliated. Includes masses of coarse-grained hornblende rock Contact HORSESHOE BEND FORMATION (Permian?) EHII Lower Cretaceous and Upper Jurassic Lower Jurassic(?) to Devonian(?) - TERTIARY _ CRETACEOUS ' AND JURASSIC e JURASSICC’) JURASSICC?) — JURASSIC(?) A TO DEVONIAN(?) - PERMIANC?) DEVONIAN(?) METAMORPHIC ROCKS PROFESSIONAL PAPER 920 PLATE 1 Metabasalt — Dark—gray to black foliated hornblende-albite rock with some epidote and magnetite Meta-andesite — Greenish-gray fine-grained massive to foliated epidote-actinolite- hurublendcarlbite rock with or without chlorite Ph Metadacite — Light—greenish-gray massive to foliated albite—quartz-actinolite- hornblende-epidote rock with or without chlorite Metarhyolite — White- to light-gray massive to foliated quartz-albite-biotite- muscovite-epidote rock with some actinolite. Some occurrences are rich in potassium feldspar, some others in biotite Metatuff — Foliated rocks consisting of quartz, albite, epidote, and actinolite with varying amounts of hornblende, chlorite, biotite, and muscovite. Color changes with increasing quartz content from dark gray through greenish or brownish gray to white. Includes layers of tuffaceous metasediment Marble — White to light-gray partly micaceous calcium carbonate rock with distinct bedding Quartzite and metachert, undifferentiated — Quartzite is thin bedded, light bluish gray to white , with some tremolite and micas. Metachert is thin bedded, white to gray, with micaceous laminae or dark gray and massive with white quartz veinlets. Grades to black phyllite Phyllite — Brownish—gray fine- to medium-grained foliated rock that consists of quartz, muscovite, biotite, and chlorite with or without epidote and calcite. Includes layers of lithic metagraywacke in the southern part Hornblende gneiss a Medium- grained gray foliated rock that consists chiefly of plagioclase, hornblende, and some quartz, biotite, and epidote FRANKLIN CANYON FORMATION (Devonian?) Meta-andesite — Includes pyroclastic material, pillow lavas, tuffs, and flows, all metamorphosed to greenish-gray epidote-actinolite-albite rocks with some chlorite, hornblende, and occasional remanents of augite and plagioclase pheno- V f crysts g‘wffg Phyllite — Black fine-grained quartz-albite-biotite-chlorite-muscovite-epidote- F without chlorite gm Metamorphosed sodarhyolite — White- to light—gray rocks with quartz and albite phenocrysts and a fine-grained groundmass consisting of quartz, albite, muscovite, and biotite with some actinolite or chlorite and epidote magnetite rock with bedding and slaty cleavage. Includes layers of lithic meta- graywacke in the southern part Quartzite — Thin bedded gray granular quartzite with micaceous laminae Metatuff and tuffaceous metasediment — Well-foliated dark-gray to white rocks consisting of albite, epidote, actinolite, hornblende, biotite, chlorite, muscovite, and quartz in varying proportions. Includes layers rich in carbonates Metadacite — Flows and pyroclastic material with tuff layers, metamorphosed to light-greenish-gray albite—quartz-epidote-actinolite-hornblende rocks with or 2000' 1000' SEA LEVEL VERTICAL EXAG ”fl GERATIONX2 Interior—Geological Survey, Menlo Park, CA. —197 6—G75 102 HARTMAN BAR PLUTON KJqd — 4000' 3000’ —>25 Bearing and plunge of lineation Fault — Dashed where approximately Strike and dip of joints located; dotted where concealed 3-0— Inclined —*40 Minor fold axis, showing plunge + Vertical 70 Strike and dip of beds 1312- Sample locality 4— Inclined . + Vertical — Dike, undifferentiated Strike and dip of fdiation _5.5_ Inclined + Vertical LOCALITIES OF SPECIMENS No. Section Township north Range east FNO. Section Township north Range east No. Section Township north Range east M204 20 21 6 1308 21 21 7 1470 33 22 6 648 19 22 7 I 1309 20 21 7 1492 26 21 6 649 19 22 7 I 1311 13 21 6 1515 33 22 6 1200 9 21 8 1312 13 21 6 1519 7 21 6 1205 36 22 6 1313 14 21 6 1521 6 21 6 1206 31 22 7 1326 22 21 6 1522 6 21 6 1207 31 22 7 1327 7 21 7 1531 16 20 7 1208 31 22 7 1328 8 21 7 1532 9 20 7 1209 6 21 7 1330 9 21 7 1552 19 20 8 1210 8 21 7 1333 l 20 6 1555 20 20 8 1211 34 22 7 1337 6 20 7 1556 12 20 7 1214 25 22 7 1362 33 21 8 1557 27 22 6 1223 8 21 8 1364 27 21 8 1559 27 22 6 1224 8 21 8 1367 32 21 8 1570 30 20 8 1248 ll 21 7 1368 29 21 8 1604 21 22 6 1267 36 21 6 1370 11 20 7 1605 21 22 6 1270 25 21 7 1450 19 20 8 1606 29 22 6 ~ 1271 24 21 7 1451 10 21 6 1607 29 22 6 1272 1 20 7 1452 10 21 6 1608 30 22 6 1281 30 22 8 1461 10 21 6 1610 32 22 6 1282 30 22 8 1462 9 21 6 1611 31 22 6 1292 30 22 8 1468 27 22 6 BALD ROCK E E CASCADE PLUTON 3‘. PLUTON E g; {52‘ § E: g e A S s ‘35 E I \J 4000 g f; E *3 I; LL o e: o L o in D? L‘ LE 2 2 ~= 3 N 3000' g E3, ‘S g E 9 i :9: E E m 2000' 1000' L SEA LEVEL