Stochastic Analysis of Particle Movement over a Dune Bed By BAUM K. LEE and HARVEY E. JOBSON ' GEOLOGICAL SURVEY PROFESSIONAL PAPER 1040 UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON: 1977 UNITED STATES DEPARTMENT OF THE INTERIOR CECIL D. ANDRUS, Secretary GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress Cataloging in Publication Data Lee, Baum K. Stochastic analysis of particle movement over a dune bed. (Geological Survey Professional Paper 1040) Bibliography: p. 41 1. Sediment transport—Mathematical models. 2. Stochastic processes. 3. Sand-dunes. I. Jobson, Harvey E., joint author. II. Title. 111. Series: United States Geological Survey Professional Paper 1040 GBSGZ.L43 551.3’7'5 77—16278 For sale by the Superintendent of Documents, US. Government Printing Office Washington, DC. 20402 Stock Number 024-001-03024-1 CONTENTS Abstract ................................................ Introduction Acknowledgment ........................................ Background ............................................. Theoretical models .................................. Experimental studies ................................ Remarks ............................................ Development of theory ................................... Characteristics of particle movement over a dune bed . . Estimation of the probability distributions of the eleva- tions of deposition and erosion ...................... Estimation of the probability distributions of the rest periods ........................................... thimation of the probability distributions of the step lengths ............................... . ............ Bed-material transport equations ..................... 5-? fl UIUIUInhNMMp-‘H 10 14 Development of theory -— Continued General two-dimensional stochastic model for dispersion of bed-material sediment particles .................. Analysis and discussion of results ......................... Experiment and basic data ........................... Probability distributions of the elevations of deposition and erosion ....................................... Rest period distributions ............................. Step length distributions ...................... g ....... Bed-material transport ............................... Variation of various statistics with flow conditions and a relation between the step length and the rest period Two-dimensional stochastic model for dispersion of bed- material sediment particles ........................ Summary and conclusions ................................ References cited ......................................... ILLUSTRATIONS FIGURE 1—20. Graphs showing: WNH J8 . Typical yx(t) record illustrating the class marks for deposition and erosion ........ . ........................ . Typical yx(t) record illustrating the conditional rest periods of a particle ................................... . Statistic, {xi ”1,, } for the step length of a particle ........................................................ . Method for estimating the percentage of volume occupied by dunes between elevations 111- and 171-+1 ............ 5. Size distribution curve of bed material ................................................................ 6. Frequency histograms, triangular density function, and truncated Gaussian density function for the elevation of deposition and erosion ........................................................................... . Sample probability mass functions of the conditional rest periods with fitted two-parameter gamma functions for: 7. Run 4A ............................................................................................. 8. Run 16 .............................................................................................. 9. Run 17 .............................................................................................. 10. Variation of the conditional mean and variance of rest periods with bed elevation ............................... 11. Frequency histograms for the marginal rest period and exponential fits ......................................... 12—14. for: Sample probability mass functions of the conditional step lengths given the elevation of erosion is 0.0 with Gamma fits 12. Run 4A ............................................................................................. 13. Run 16 .............................................................................................. 14. Run 17 .............................................................................................. 15. 16. 17. 18. 19. 20. Variation of the conditional mean and variance of step lengths with bed elevation ............................... Frequency histograms for the marginal step length with Gamma fits ........................................... Effective volume ratio as a function of bed elevation .......................................................... Variation of bed-load discharge with bed elevation ............................................................ Mean transport speed of a bed-load particle as a function of bed elevation ...................................... Ratio of the conditional variance of step lengths to the conditional variance of rest periods as a function of bed eleva- tion III Page 18 22 22 23 25 30 36 39 39 40 4 1 Page 11 16 22 24 26 27 28 29 31 32 33 34 35 37 39 40 40 IV CONTENTS TABLES [Tables 1 and 73 —75 are in text; all others follow references cited] Page TABLE 1. Basic data and computed parameters ......................................................................... 23 2. Sample probability mass functions of elevations of deposition and erosion ....................................... 44 3 —5. Sample conditional probability mass function of rest periods, pT\ yD(tu\ y,) for: 3. Run 4A ............................................................................................. 45 4. Run 16 .............................................................................................. 46 5. Run 17 .............................................................................................. 47 6. Variation of conditional mean and variance of rest periods with elevation of deposition; E [T\ YD = y] and Va’fr [T\ YD = y] ............................................................................................... 48 7. Estimates of parameters and the results of goodness of fit test for the conditional rest periods (two-parameter gamma) ~ .......................................................................................................... 48 8 ‘10. Sample joint probability mass function of rest periods and elevation of deposition, pT, YDUa, yi) for: 8. Run 4A ............................................................................................. 49 9. Run 16 .............................................................................................. 5O 10. Run 17 .............................................................................................. 51 1 1 —24. Sample conditional probability mass function of step lengths, p X\ YE YD(xB\ yi, yj) (run 4A) ......................... 52-55 25 —42. Sample conditional probability mass function of step lengths, p X\ YE YD(x,3\ yi, yj) (run 16) .......................... 55-59 43 —56. Sample conditional probability mass function of steplengths, p X\ YE yD(xB\ yi,yj) (run 17) .......................... 60-63 57-59. Conditional means and variances of step lengths; E[X\ YE = yi, YD = yj] and var [X\ YE = yi, YD = yj] for: 57. Run 4A ............................................................................................. 64 58. Run 16 .............................................................................................. 65 59. Run 17 .............................................................................................. 66 60 —62. Estimates of parameters describing two-parameter gamma distribution for conditional step lengths for: 60. Run 4A ............................................................................................. 67 61. Run 16 .............................................................................................. 68 62. Run 17 ............................................... , ............................................... 6 9 63 —65. Results of goodness of fit test for conditional step lengths for: 63. Run 4A ............................................................................................. 70 64. Run 16 .............................................................................................. 70 65. Run 17 .............................................................................................. 70 66 —68. Sample conditional probability mass function of step lengths, p X\ YD(xB \ yj) for: 66. Run 4A ............................................................................................. 71 67. Run 16 .............................................................................................. 71 68. Run 17 .............................................................................................. 71 69. Variation of conditional mean and variance of step lengths with elevation of deposition; E[X\ YD = y] and szr [X\ YD = y] .............................................................................................. 71 70 —72. Sample joint probability mass function of step lengths and elevation of deposition, 17}: YD(xB ,3!) for: 70. Run 4A ............................................................................................. 72 71. Run 16 .............................................................................................. 72 72. Run 17 .............................................................................................. 72 73. Variation of various statistics with stream power ............................................................. 38 74. Comparison of the effective volume ratios at elevation yj; «Ej, from y,(x) record and C}, from yx(t) record ............... 38 75. Comparison of measured and computed total bed-material transport rates ...................................... 38 METRIC-ENGLISH EQUIVALENTS Metric unit English equivalent Meter (m) = 3.28 feet Centimeter (cm) 3.28 X 10‘2 foot Tonne per day per meter (t/day -m) .336 ton per day per foot Dyne per square centimeter (d/cm2) = 2.09 X 10'3 pound per square foot Dyne per centimeter per second (d/cm -S) = 6.85 X 10‘5 pound per foot per second "‘aru NdUi) Ne(yi) N(t) CONTENTS SYMBOLS Constants in estimating the conditional mean of the rest periods Constants in estimating the conditional variance of the rest periods Dimensionless Chezy discharge coefficient Concentration of total bed-material discharge Depth of flow Median sieve diameter of bed material Sieve diameter of bed material for which 84 percent is smaller Sieve diameter of bed material for which 16 percent is smaller Event that a particle eroded from elevation yi passes v dune crests before it is deposited at elevation yj Ehrent that a particle passes v dune crests before it is deposited Mathematical expectation and its estimate, respec- tively Froude number of the flow Probability density and distribution functions, respectively n-fold convolution of the probability density, fl .) Gravitation acceleration Average depth of the zone in which bed-material movement occurs determined from the y,(x) record Scale parameter of the conditional step length dis- tribution Scale parameter of the conditional rest period dis- tribution Length of the yx(t) record Length of the y, (I) record Total number of bed forms contained in the y,(x) record for which the upstream side intersects the elevation yi and the downstream side intersects the elevation yj Total number of possibilities of the event Ei tained in the y, (x) record Total number of bed forms contained in the y,(t) record and which also contain some erosion in the class interval associated with the elevation y,- Total number of bed forms contained in the yx(t) record and which also contain some deposition in the class interval associated with the elevation yj Total number of bed forms contained in the yx(t) record and which also contain both an up-crossing and a down-crossing at the elevation yj Number of steps taken by a particle or number of class intervals for the realization of YD and YE Total number of particles per unit area deposited in time t Total number of particles per unit area eroded in time t Total number of particles per unit area deposited within the class interval (In ,niH] in time t Total number of particles per unit area eroded with- in the class interval (ni Jim] in time t Counting process describing number of steps taken by a particle in time t ,j,v con- Tln) Var[-],V€r[-] w x x x5 xth xaiuk X(n) if> 4 a 9' a ! Ta+1 X 53:)" Vertical fall of the bed in the class interval associ- ated with y, for the kth erosion period of the y,(t) record Specific weight of bed material Gamma function Effective volume ratios Lower and upper class limits for yj, respectively Bulk porosity of the bed material in place lower and upper class limits for x5, respectively Estimate of correlation coefficient Geometric standard deviation of particle size Mean bed shear stress Lower and upper class limits for ta, respectively Critical value of chi-square statistic Number of particles per unit volume of the bed Number of particles per unit volume of the bed associated with yj STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED By BAUM K. LEE and HARVEY E. JOBSON ABSTRACT Stochastic models are available that can be used to predict the transport and dispersion of bed-material sediment particles in an alluvial channel. These models are based on the proposition that the movement of a single bed-material sediment particle consists of a series of steps of random length separated by rest periods of random duration and, therefore, application of the models requires a knowledge of the probability distributions of the step lengths, the rest periods, the elevation of particle deposition, and the elevation of particle erosion. In the past, it has proven impossible to estimate these distributions except by use of tedious and time-consuming single particle experiments. By considering a dune bed configuration which is composed of .uniformly sized particles, the probability distributions of the rest period, the elevation of particle deposition, and the elevation of parti- cle erosion are obtained from a record of the bed elevation at a fixed point as a continuous function of time. By restricting attention to a coarse sand, where the suspended load is negligible, the probability distribution of the step length is obtained from a series of “instan- taneous" longitudinal bed profiles in addition to the above informa— tion. Using these probability distributions, three bed-material transport equations and a two-dimensional stochastic model for dis- persion of bed-sediment particles are developed. The procedure was tested by determining these distributions from bed profiles formed in a large laboratory flume with a coarse sand as the bed material. The elevation of particle deposition and the eleva— tion of particle erosion can be considered to be identically dis- tributed, and their distribution can be described by either a “trun- cated Gaussian” or a “triangular” density function. The conditional probability distribution of the rest period given the elevation of parti- cle deposition closely followed the two-parameter gamma distribu- tion. The conditional probability distribution of the step length given the elevation of particle erosion and the elevation of particle deposi- tion also closely followed the two-parameter gamma density function. For a given flow, the scale and shape parameters describing the gam- ma probability distributions can be expressed as functions of bed elevation. The bed-material transport equations were tested for three flow conditions. The errors in the predicted mean total bed-material transport rates were —3.0, +3.5, and 80.1 percent for equation 55, and —1.7, +26.9, and +64.1 percent for equation 63. For the run with the large error, the mean total load concentration was small (8.9 milligrams per liter), and flow conditions were somewhat out of equilibrium. INTRODUCTION The movement of sediment in alluvial streams is so complex a process that it may never be subjected com- pletely to a deterministic solution. It represents, in fact, an extreme degree of unsteady, nonuniform flow, since the streambed as well as the water surface may be con- tinuously changing with time and position. Numerous formulas and equations have been developed to predict sediment transport rates. Most of these developments ignore the actual nature of sedi- ment movement and have assumed that the sediment transport rate can be described by a deterministic func- tion of certain flow parameters. Unfortunately, after decades of searching, no universally accepted sediment transport equation has been found. The theories of probability, statistics, and stochastic processes have been used to describe the kinematics of a single bed- sediment particle in an alluvial channel flow and to pre- dict the dispersion characteristics of a group of such particles. These theories have clearly demonstrated a great potential for development of stochastic models of sediment transport and dispersion. Most of the stochastic models (Shen and Todorovic, 1971; Grigg, 1969; Yang, 1968; Sayre and Conover, 1967; Hubbell and Sayre, 1964; Crickmore and Lean, 1962; Einstein, 1937) are based on the proposition that the movement of bed-sediment particles consists of a series of steps separated by rest periods, so that deter- mination of the probability distributions for the step lengths and the rest periods of a bed-sediment particle plays the major role in quantifying the bed-sediment transport. While this movement concept can easily be verified through laboratory observations, Einstein (1937) was the first to use it. He developed a one-dimen— sional probabilistic model for bedload transport. More recently, Sayre and Conover (1967) derived a two- dimensional stochastic model by introducing the prob- 2 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED ability distribution of the elevation at which a bed-sedi- ment particle is deposited. The probability distributions of the step lengths and the rest periods of a bed-sediment particle have been estimated from single particle experiments (Grigg, 1969) or by using a group of tracer particles (Yang, 1968; Hubbell and Sayre, 1964; Crickmore and Lean, 1962). Because of the considerable effort required to conduct such experiments, it seems clear that some way must be found to estimate the probability distributions from more readily accessible data if significant further progress is to be expected. To apply the Sayre-Conover (1967) two-dimensional stochastic model, the prob- ability distribution of the elevation at which a bed-sedi- ment particle is deposited must be known. A method for estimating this distribution is developed in this report. The objectives of this study are: 1. To present a method of estimating the following probability distributions for dune-bed conditions using only sounding records of the bed elevation — (a) prob- ability distributions (note that there are two separate distributions) of the elevation at which a bed-sediment particle is eroded and deposited, and (b) conditional probability distributions of the step lengths of a bed- sediment particle given the elevations at which the par- ticle is eroded and deposited. A method estimating the conditional probability distribution of the rest periods of a bed-sediment particle given the elevation at which the particle is deposited has been presented by Sayre and Conover (1967). 2. To develop bed-material transport equations based on the above probability distributions and to compare the results with the experimentally measured values. 3. To derive a two-dimensional stochastic model for dispersion of bed-sediment particles as a function of the above probability distributions. The probability distributions of the elevation at which a bed-sediment particle is eroded and deposited, and the probability distribution of the rest periods, con- ditioned on the elevation of deposition, will be obtained from a continuous record of the bed elevation at a par- ticular point as a function of time. The probability dis— tribution of the step lengths, conditioned on the eleva- tion of erosion and the elevation of deposition, will be obtained from a series of “instantaneous” longitudinal bed profiles. With these distributions obtained, various related probability distributions of vital interest will be estimated, and a relation between the rest periods and the step lengths of a bed-sediment particle will be in- vestigated. Three experimental runs are analyzed and the rela- tions between the statistics describing the postulated probability distributions and the hydraulic conditions are investigated. All data were obtained from a tilting recirculating flume of rectangular cross section 61 m long, 2.4 m wide, and 1.2 m deep. The bed material used in these experiments was screened river sand with a median sieve diameter equal to 1.13 mm and a geometric standard deviation equal to 1.51. ACKNOWLEDGMENT The data contained herein are essentially the same as those contained in a dissertation by Lee (1973). The data were collected under the general supervision of the second author. Special thanks are due E. V. Richardson, D. B. Simons, C. F. Nordin, D. C. Boes, and R. P.- Osborne. BACKGROUND THEORETICAL MODELS Einstein (1937) treated the movement of a single sediment particle over an alluvial bed as a stochastic process described by an alternating sequence of two in- dependent random variables, namely, step lengths and rest periods. Considering the particle movement in the distance-time plane on a Galton’s board (Parzen, 1960), Einstein derived exponential probability density func- tions for the step lengths and the rest periods, -kx ,x>0 (1) it II ;r- m and ,t>0 ' (2) E H W (b respectively, where X,T= random variables describing the step lengths and‘rest periods of a parti- cle, respectively; x,t = distance and time, respectively; fX(x), fT (t) = common probability density functions of the step lengths and rest periods, respectively; and k1, k2 = positive constants. For a sediment particle introduced into the stream at distance x = 0 in such a way that it takes its first step at time t = 0, Einstein obtained the probability density function of the total distance traveled by the particle at time tto be on —1 -kx-kt (kx)" . _ 1 2 l “““he E—W‘ n=l x>0, t>0,(3) n ~- 1 (kzt) I‘(n) ' in which I’ ( -) denotes the gamma function. Equation 3 also represents the concentration distribution of a group of identical sediment particles with respect to BACKGROUND longitudinal position, x, as a function of time, t. The probability density function for the case when the particle is initially (t = 0) at rest at x = 0 also was obtained by a similar procedure, —kx— kt 1 1e1 22 n=1 (k lnx) l"(n) n (k2 t) f(x;t)=k m. x>0, t>0. (4) It should be noted that equation 4 applies only to the particle that has taken at least one step. Einstein (1950) also developed his well-known bed- load equation by considering the dynamic lift force as a random variable. The idea is that the probability of a sediment particle being eroded from the bed surface is equal to the probability that the lift force exerted on the particle exceeds its submerged weight. He obtained BMW ‘11] o _2 A*d>* p=1-L eZdZ:W—+ 1(5) fi-B‘Y-l ** *~k — where 0 p = probability of a sediment particle being eroded; n0,A*,B* = constants; ‘I’* = intensity of shear for an individual particle size; and (13* = intensity of transport for an individual par- ticle size. Solving equation 5 for (1),“ which is a function of the bed- load transport rate, one obtains the bed-load discharge for individual particle sizes from hydraulic parameters and sediment properties. Hubbell and Sayre (1964) presented a one-dimen- sional stochastic model for the longitudinal dispersion of bed-material particles in an alluvial channel. The results are identical to Einstein’s (eqs. 1—4). The assumptions are: (1) the flow is in equilibrium (Simons and Richardson, 1966); (2) the particle always moves in a downstream direction with a series of alternate steps and rests; (3) the duration of movement is insig- nificant compared to the rest periods; and (4) the stochastic processes describing the number of steps taken by a particle in a distance interval and a time in- terval are independent of each other and both are homogeneous Poisson processes (ParZen, 1967). These assumptions are essentially the same as those of Ein- stein’s (1937) although stated in a different way. Based on the concept of continuity, Hubbell and Sayre (1964) proposed the transport equation for the bed material of a certain characteristic, (QT) :1 (y ) (1 - e)wh(°t£)c , (6) C C SC where QT = bed-material discharge in weight per unit time; ic = ratio of the volume of particles possessing the characteristic size to the volume of bed- material particles in the zone of particle move- ment; ys = specific weight of the bed material; 0 = bulk porosity of the bed in place; W = width of channel; h = average depth of the zone in which particle movement occurs; a? = average distance traveled by bed material in time t; t = measure of time; and c = subscript that denotes terms associated with particles possessing a certain characteristic size. Combining equation 6 with the result from the Hubbell- Sayre one-dimensional stochastic model gives the total bed-material discharge for all particle sizes, k _ . _ 2 QT— E [Cm/pea e)Wh(k—1) . (7) C C in which k1 and k2 are defined in equations 1 and 2, respectively. Sayre and Conover (1967) extended the one-dimen- sional stochastic model derived by Hubbell and Sayre (1964) to two dimensions by introducing the vertical level at which particles are deposited. Their analysis led to the joint probability density function for the event that a particle has, at time t, traveled a distance equal to x and is located at an elevation equal to y, f(x y; t) =f (V) f f (2') f (r\y)dtdt', (8) YD§1X(J)C)antf_t,T\1/D where , y) =pro a 11ty ens1ty unctlon or t e f)[)( b b’l' d ' f ' f h m (n elevation of particle deposition; fX (x) fT(t)— — n- fold convolutions of fx( (x) and f7(t respectively; fT\ YDU \y) = conditional probability density function for the rest periods given the eleva- tion at which the particle is deposited; and t’ = sum of the first n rest periods, If a group of identical sediment particles are released simultaneously at x = 0, y =y0, and t= 0, equation 8 gives the concentration of the particles, which were in- itially at rest and have moved from their respective in- itial positions, with respect to longitudinal position, x, and vertical position, _v, as a function of time, t. 4 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED In order to apply equation 8, the density functions, fYD( y) fT\ YD (t \ y) and fix (x) must be specified. The un- conditionalD density function, fT(t) is related to fT\ YD(t \ y) and fyD( y) by the relation ymax mm = f rm «\wa may, (9) y min D D in which ymax and ymin are the highest and lowest eleva- tions at which particles can be deposited, respectively. The marginal case of equation 8 is °° (n) ft {(n) (n + 1) Z (X (x) Tm - rTm a: n— - l °° (n) = E rxm P[N(t) = n]. (10) n = l in which P[N(t) = n] denotes the probability that a par- ticle takes n steps in a time interval t. Equation 10 is a general one-dimensional stochastic model where only longitudinal dispersion is considered. One may note that the substitution of equations 1 and 2 into equation 10 reduces to equation 4. Yang (1968) assumed the step lengths are gamma distributed with a shape parameter, r, and the common density function, kl -k x fx (x) = r‘ —(—,.) (k 1ac) e f(x;t) = (11) and the rest periods are exponentially distributed with the common density function given in equation 2. Substituting equations 2 and 11 into equation 10, he ob- tained -kx- k2: °° k sow—1 (kzt)n f(x;t)=k1e 2% m, n= x>0,t>0. (12) Since the gamma distribution reduces to the exponen- tial distribution when r = 1, equation 4 is actually a special case of equation 12. Shen and Todorovic (1971) generalized the Hubbell- Sayre one—dimensional model given in equations 1, 2, and 4. The essential difference between the two models is that the former was based on the nonhomogeneous Poisson processes (Parzen, 1967), while the latter was based on the homogeneous Poisson processes. In the Shen-Todorovic model, the probability density func- tions of the step lengths and the rest periods are, respectively, x -f (s)ds a“o fX(x) = kl(x) e ,x> O , (13) and t -f (s)ds t me = k2(t) e 0 ,t> 0 , (14) where k1(x), k2(t) = functions of x and t, respectively; and x0, t0 = initial position and time, respectively. The probability density function of the total travel dis- tance of a particle, which was initially at rest and has moved from its initial position, x0, was found to be x t -f k1(s)ds—ft k2(s)ds f(x;t) = k1(x)e x0 0 (15) x n -l t n a [£0 k1(s)ds] [ftokzmms] 2 PM I‘(n+ll) ,x>0,t>0. n = 1 It is seen from equations 13 and 14 that the mean num- ber of steps taken by a particle in (x0, x] and (to, t] are ka1(s)ds and 13kg“) ds, x0 0 respectively, whereas those of Hubbell-Sayre’s model are k1(x—x0) and k2(t—t0), respectively. The Hubbell- Sayre (1964) one-dimensional model is a special case of the Shen-Todorovic model. EXPERIMENTAL STUDIES Hubbell and Sayre (1964) conducted concentration distribution experiments both in the field and laborato- ry to evaluate the one-dimensional stochastic model given by equation 4. The bed configurations in these ex- periments were large dunes in the field and ripples in the laboratory flume. Using radioactive tracer parti- cles, a series of longitudinal concentration-distribution curves were obtained at different times for a given flow condition. The longitudinal concentrationdistribution function, @(x;t), is defined to be the weight of tracer particles per unit volume of bed material as a function of longitudinal distance and time and is related to fix; t) by W alf(x')tl Wh (16) (I) (x; t): in which WT is the total weight of the tracer particles placed in the channel, W is the channel width, h is DEVELOPMENT OF THEORY 5 average depth of the zone of bed material movement, and f(x;t) is given by equation 4. Based on equation 16, the parameters k1 and k2 were estimated. With these estimates, Hubbell and Sayre reported that the theoretical and observed concentration-distribution functions agree reasonably well. Yang (1968) carried out a set of experiments using radioactive tracer particles to verify the model given by equation 12. Experiments were performed with ripple and dune bed conditions in a laboratory flume 0.6 m wide by 18.3 m long. He reported that the shape of the experimental longitudinal dispersion curves are fairly well represented by equation 12. Yang also made preliminary runs with a single plastic particle in a small flume and found that the step lengths very closely follow a gamma distribution with the parameter r ap- proximately equal to 2 and that the rest periods follow an exponential distribution very closely. The first intensive experimental study on the move- ment of single particles was done by Grigg (1969). The experiments were conducted in a laboratory flume with two bed material sizes. The bed configurations were rip- ples and dunes. Using single radioactive tracer parti- cles, he measured the step lengths and the rest periods directly and found the step lengths to be approximately gamma distributed and the rest periods to be approx- imately exponentially distributed as proposed by Yang. Grigg found interesting correlations between: (1) Various properties of the step length distribution, the stream power (product of mean bed shear stress and mean flow velocity), and the distribution of bedform lengths; and (2) various properties of the rest period distribution and statistical properties derived from the variation of bed elevation with respect to time. Based on an idea suggested by Hubbell and Sayre (1965), Grigg also made some progress toward experi- mentally testing the Sayre-Conover two-dimensional stochastic model. By analyzing a record of the bed elevation as a function of time, he showed that the con- ditional probability density function of the rest periods can be approximated by the exponential function, _k t fT\yD(t\y) = k3 (y) e 30)) , (17) and (18) in which a and B are constants and y measures bed elevation in terms of the standard deviation about mean bed elevation. REMARKS Based on the review given in the previous sections, the following remarks are offered. 1. The Sayre-Conover model given by equation 10 is the most general one-dimensional model. The rest of the one-dimensional models, which were previously dis- cussed, can be obtained from this model by proper substitutions. Therefore, it may be rated as the best ex- isting one-dimensional model. 2. The Sayre-Conover model given by equation 8 is the only existing two-dimensional stochastic model. The derivation of the Sayre—Conover model has been discussed by Lee (1973). To verify equation 8, a method of estimating the probability distribution of the eleva- tion at which particles are deposited must be known. One of the purposes of this report is to present such a method. 3. In order for the stochastic model to serve a predic- tion purpose, the relation between flow conditions and the parameters describing the probability distributions must be known. Without such knowledge the stochastic models cannot contribute much to the prediction problem. 4. A great deal of effort is required to perform disper- sion and single particle experiments. If another method can be developed to estimate the necessary probability distributions from more readily accessible data, con- siderable savings would result. The methods developed in this report require only records of bed elevation. DEVELOPMENT OF THEORY CHARACTERISTICS OF PARTICLE MOVEMENT OVER A DUNE BED Dunes are one of the most common bed forms in alluvial channels. Field observations by Simons and Richardson (1966) indicated that dunes may form in any alluvial channel, irrespective of the size of bed material, if the stream power is sufficiently large to cause general transport of the bed material without ex- ceeding a Froude number of unity. The longitudinal profile of a dune is approximately triangular in shape with a gentle upstream slope and steep downstream slope. The upstream slope depends somewhat on flow conditions, whereas the downstream slope is more de- pendent on the angle of repose of the bed material. The length of a dune ranges from about 0.61 m to several hundred meters, depending on the scale of the flow system. The Chezy discharge coefficient, CA/E, ranges from 8 to 12, and the total bed-material discharge con- centration ranges from 100 to 1,200 milligrams per liter for dune flow conditions. For further information readers may refer to Simons and Richardson (1966). For dune flow conditions, a record of the bed eleva- tion as a function of time at a particular location reveals an alternating sequence of periods during which either erosion or deposition is occurring. This 6 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED type of record is commonly obtained from the output of a depth sounder which is at a fixed location and hereafter will be referred to as the yx(t) record, that is, the elevation of the bed, y, positive upward, as a» func- tion of time, t, at a fixed location, x. When deposition oc- curs, [dyx(t)/dt] >0, and when erosion occurs, [dyx (t)/dt] < 0, provided these derivatives exist. An in- stantaneous longitudinal bed profile may be charac- terized by an alternating series of erosion and deposi- tion reaches. An instantaneous longitudinal profile can be obtained by mounting a depth sounder in a boat, pro- vided that the speed of the boat is large relative to the speed of bed forms. These bed profiles will hereafter be referred to as the y,(x) records, that is, the elevation of the bed, y, positive upward, as a function of the longitudinal coordinates, x, at a given time, t. The longitudinal coordinate will be assumed to increase in the downstream direction; therefore, the reaches with positive slopes, [dyt(x)/dx] >0, represent the upstream or stoss sides of the dunes, and the reaches with nega- tive slopes, [dy,(x)/dx] <0, represent the downstream or slip faces of the dunes. The dune crest is defined by a local maximum in the y,(x) record, and the dune trough is defined by a local minimum in the record. Anyone who has an opportunity to observe closely the movement of sediment is aware that dunes move downstream owing to erosion from their. upstream face and deposition on their downstream face. That is, the bed forms migrate downstream because deposition oc- curs on the downstream face, where [dy,(x)/dx] <0, and erosion occurs on the upstream face, where [dyt(x)/dx] > 0. It will be assumed throughout this report that no deposition occurs on the upstream sides of dunes and no erosion occurs on the downstream faces of dunes. This assumption is not strictly true physically but is necessary for the determination of the condi- tional step length distributions. If the assumption is true, each sediment particle on the stoss side of a dune must make a step in the downstream direction before it is deposited on the slip face of any dune. Once deposited it rests there until the dune has migrated downstream and it becomes reexposed on the stoss side. In other words, sediment particles are transported downstream in an alternating sequence of steps and rests of random length and duration. The frequencies and magnitudes of these steps and rests are of basic interest in under- standing the nature of the movement of the sediments. Because particles must be eroded from and deposited on the surface of the bed, the step length of a particular particle depends only on the elevation from which it is eroded, the elevation at which it is deposited, the num- ber of dune crests which it passes before being deposited, and the scale and shape of the bed surface (yt(x) record) during the time of its movement. Likewise the rest period of a particular particle depends on the scale and shape of the yx(t) record and on the elevation at which the particle is deposited. If the bed material size is not uniform, the elevation of deposition or ero- sion may also depend on the size of particles because of vertical sorting. The intimate relationship between the bed-form shape, as measured by the yx(t) and y,(x) records, and the step lengths and rest periods of a bed-material par- ticle allow the probability distributions of the step lengths and the rest periods to be estimated from the bed-form data. In the following three sections, a method of estimating the probability distributions of the rest periods, step lengths, and elevations at which a particle is deposited or eroded using the yx(t) and y,(x) records will be presented. In the last two sections the bed— material transport equations and a general two-dimen- sional bed-material dispersion equation will be derived as functions of these probability distributions. In the next chapter the transport equations will be tested using data from three fiume runs and the results will be discussed. ESTIMATION OF THE PROBABILITY DISTRIBUTIONS OF THE ELEVATIONS OF DEPOSITION AND EROSION The probability that particles are deposited between the elevations ”01‘ and n j+1 may be written as PM]. < YD 3711,11 number of particles deposited = lim within the interval (71]., n]. + l] in time t (19) t-> °° . number of particles deposited ' over all intervals in time t where P H = probability; YD = random variable describing the elevations at which particles are deposited; 77,, 71141 = lower and upper class limits associated with the class mark of the elevation y], respec- tively; and t = time during which the observations were made. The elevation at which particles are deposited will hereafter simply be referred to as the elevation of deposition, YD. If the number of particles per unit volume of the bed, (I , is constant, the flow is stationary (statistical sense), and both erosion and deposition cannot occur at the same point at the same time, the numerator and the denominator of equation 19 can be obtained from the yx(t) record. The total number of particles deposited per unit area within the class interval (7),, ”0141] in time t, denoted by Nd(yJ-) is given by DEVELOPMENT OF THEORY 7 "‘1' + .=Q A. ;'=1,2,..., , 20) Nd(y]) 2:1 y1,k] n ( k: where yj = class mark for the realization of YD ; n = number of class intervals for the realization of Y0; Ayfk=vertical rise of the bed in the class interval associated with yj for the kth deposition period; and mj = maximum number of bed forms contained in the yx(t) record and which also contain some deposition in the class interval associated with yj. Figure 1 illustrates the class marks, yJ-, and the verti- cal rise of the bed, Axfk, within the class intervals, ij, for a typical yx(t) record. It is clear that 1339ka ij=nj+1 —nj.The total number of particles per unit area deposited over all intervals, the denominator of equation 19, is designated by Nd and is obtained by summing equation 20 over all class marks: n i=1 Equation 19 now becomes yxlt) . N (y-) _ 11m (1 l P[nj°° Nd lim 1 ‘m.—>oo_T— ‘92) J n 1 Z N Z yj,k j=l k=1 Similarly an analysis of the erosion periods can be used to estimate the probability that particles are eroded between the elevations 7) land 1) i+1, ]= lim Ne(yi) i + l m‘.-> o0 N 1 e P[ni< YEST] m‘l. E ”inc k=1 m'. n z E Z Ayi‘ k i=1 k=l = lim m'.—> co 7. (23) where YE = random variable describing the elevations at which particles are eroded; y = class mark for the realization of YE; 71,- , “qt-+1 = lower and upper class limits of yi , respec- tively; T ical erosion period Typical dc osition peri FIGURE 1. — Typical yx(t) record illustrating the class marks for deposition and erosion. 8 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED m; = maximum number of bed forms contained in the yx(t) record and which also contain some erosion in the class interval associ- ated with y,- ; Ne( y,) = total number of particles per unit area eroded from the interval (17,-, 71i+ll cen- tered at y,- ; N, = total number of particles per unit area eroded over all intervals; = number of class intervals for YE ; and Aygk = amount of erosion which occurred during the kth erosion period in the vertical class in- terval associated with y,- (fig. 1). In the limit as mjand m,’ approach infinity for a station— ary record, the distributions of P [r],- < YDS n j+1] and P[~q,~ < YES 77141] must be identical. The elevation at which particles are eroded will hereafter simply be referred to as the elevation of erosion, YE. To repeat, the following assumptions are necessary for equations 22 and 23 to be valid: (1) Flow is in equilibrium such that both deposition and erosion proc- esses are stationary with respect to time t; (2) both ero- sion and deposition do not occur at the same point dur- ing the same time period; and (3) the number of parti- cles per unit volume of the bed is constant. If the measuring equipment were sensitive enough to detect the movement of single particles, the second assumption would not be necessary because it would be physically impossible for one particle to be eroded and another to be deposited at the same point and at the same time. For yx(t) records obtained from less sensitive equipment, of course, the assumption may not be strictly true. When the bed material is not uniform in size, equations 20 through 23 are not strictly true because the number of particles per unit volume of the bed, (2, is a function of elevation due to a vertical sorting. However, equations 22 and 23 should serve as first approximations to the true prob- abilities, PM, < YDSnJ-H] and Pm, < YESTII'H], even for the nonuniform bed material. If the number of particles per unit volume were known as a function of elevation, y, assumption (3) could be dropped. In this case, the counterpart of equa- tion 22 is P[r]jx FIGURE 3. —Statistic, {xi,j,v.k; i,j = 1,2,. . ., n; U = 1, 2, . . .; k = 1, 2, .. mun} for the step length ofa particle. E, = event that a particle passes 12 dune crests in which before it is deposited (fig. 3). m- The corresponding probability density function, fX\ YE, YDva(x\y,’ 3”“), may be determined from PX\ YE,YD,EU(x3\yi , y,» 0 ), and its mean and variance are estimated to be m. . 1,],V \ A — : : 1 E[X\YE_yi' YD yj'Ev] m. .v E xi,j,v,k’ w, k=1 and Var [Xxx/15:321., YD=yj,EV] H36) mi,j,v mi,j,v 2 2; x - 1 x mijvk i],,)2vk mid-Wk: 1.],V.k J i,j,v 1 Z 2 A 2 (x.. ) =E[X \Y =y., y =y., m.. lIJIV k=1 l,],V,k E l D J E]. V If YE, YD, and EU are mutually independent (it seems to be reasonable that after a particle passes the crest of a dune it has probably lost track of where it came from), the density function [fX\ YD'EU(x\y, 0)] of the step lengths given that a particle is deposited at elevation y after passing v dune crests is estimated from the sam- ple conditional mass function which is given by px\ YD’EV (xBWJ-N) =P[AE (38) and Var [X\YD = yj’Ev] _* 2 _ _ ‘ _ 2 'Elx \YD_yleV] (E[X\YD-yleV]) I in which A 2 - E[X \YD — yj,EV] A 2 _ .- = E E[X \YE — yi, YD - yj.EV]pYE(yi) i = 1 Likewise, the following sample probability mass func- tions and corresponding means and variances are ob- tained; =p[:B(53) ,. E é[X'\EV]P[EV] g X U Ev =_v=2_°°__ v = 2 E P[Evl V = 4 Using equations 49, 51, 52, and 53, ,[ 1 2 mm 1pm} , E X\E _ VT_ . 1P[El]+v—2 . E[T] BIT] = VBP[E1] + VS 2 P[EV] v = 2 = VBpusl] + VS(1— P[El]) (54) If all bed material particles have identical transport characteristics, which is reasonable for uniformly sized bed material, the mean total bed material discharge is obtained by use of the continuity concept, QT = 78 (1 — e)wnv (55) T . where QT = estimate of the mean total bed material dis- charge in weight per unit time; ys = specific weight of the bed material; 9 = porosity of the bed; W = width of the channel; h = average depth of the zone in which bed material movement occurs; and Tis given in equa- tion 49. Similarly, estimates of the mean bed-load discharge and suspended load discharge are, respectively, QB = ysa - 9)WhVBP[Ell (56) and Qs=ys(l -9)WhV (l -P[El]) , (57) S where 93 and [Vsare given in equations 51 and 52, respectively. From equations 54 through 57, QT = QB + €35 . (58) Although equations 55, 56, and 57 have the form of a continuity equation, the concept of continuity applies only in a statistical sense, because particles move only when they are exposed on the stoss side of a dune or when they are in suspension. Hubbell and Sayre (1964) proposed that the average depth of the zone of bed material movement, It, be estimated from the y,(x) record. For this method, the length of the reach for which h is to be determined is divided into sections. Starting from the upstream end, each section of length 4extends from the dune trough at which the section begins to the first trough downstream that is deeper relative to a line parallel to the plane of the mean bed surface. After sectioning, a mean depth of sand above the projected line for each section, hiis determined, and the h for the total reach, L3,, is computed as the weighted average of the h,’s for each section. Expressed mathematically, m _ l h—L—x E gin, (59) i=1 The reasoning behind the procedure is based upon the assumption that although the individual dunes may change shape as they progress downstream, a statisti- cal constancy of form exists over a long reach. Hence, quantitatively the particles subject to movement are those that would move if the entire profile were to progress downstream without changing form, and the depth of bed material movement is defined by lines that are parallel to the mean bed surface and extend downstream from the deepest trough. If all bed material particles are assumed to be transported as the bedload, éT = QB = ys(l - mwnfiB , (60) where [VB is determined from equation 51. For coarse sand P [E] is expected to be very close to unity because the suspended load is negligible compared to the bed load. For a fine sand for which P[El] #1, equation 60 would give only an approximation to the total load. 16 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED Equations 55, 56, and 57 can be used with measured yx(t) and y,(x) records to compute the various transport rates. However, in order to apply the equations to the prediction of the bed material transport rate where the 3/14,“) and y,(x) records are not available, the relations, E[X\ YE = y,, YD = yj, EU], Em YD = yji, P[E.,], py (yj), MM, and h, to perti- nent hydraulic and se ’ment parameters must be established. The mean transport speed of bed material particles deposited at elevation yj, which is denoted by VT(j), (more precisely, deposited between the elevations "'7; and le+1> centered at yJ-) may be estimated A E[X\Y =y.] VT”) =_A__2___J_ , (61) E[T\YD=yj] where 9T (D estimates VT(j ), E[X\YD=yj] m. . °° n 1,],V _ 1 ‘2 E m.. E xi,j,v,k pYE(yi)P[Ev] ' v=l i=1 "1"” k=1 and m. . 1:1 —1 Z ‘ m.. ',k ' 1,1 I k=1 E[T\YD = yj] = Based on equation 61, another transport equation can be developed as follows. Let Q denote the percentage of volume between elevations ’le and 1; 1+1 occupied by (x) yi ’ FLOW "m )‘i.3 )‘M M2 dunes over a given reach; then, 35,- can be estimated from the y,(x) record (fig. 4), :L A I 51' Lx Z j.k k where L, is the total length of y,(x) record, and Mk is defined in figure 4. Applying equations 61 and 62, the mean total bed material discharge can be expressed as (62) n Q'i‘ = Ysu — em 2 VT(j)Ay]-§j i=1 n A E[X\Y =y.] ='Y (1_9)w __D__J_ S j: ij§j , (63) 1 E[TWD =-yj] where . ; = estimate of mean total bed material dis- charge; and = nonstandardized class width associated with elevation yj(ij = "lb-+1 — nj). Ay- J Equation 63 takes into account the local variation of the depth of the zone of bed material movement with respect to the elevation of deposition, and demonstrates to what extent each elevation contributes to the total transport rate. Similarly, the mean transport speed of a bed-load particle deposited at elevation yj, which is denoted by VB( j), can be estimated from equation 61 by considering ME] = 1 and P[Ev] E 0 for v 2 2. It can be shown, of course, that equations 54, 56, and 57 also apply at each elevation j as well as to depth-averaged values. ‘ 1.5 L7 1.5 3.9 AYJ‘"j+r’7i Yj ---__-=E_.'.---_quf. -__ - --_-------_-------------- -------- -.-_-“- - - ":7,- ”1+1 Lx 0—'—'_’x . . . . 1 FIGURE 4. — Method for estimating the percentage of volume occupied by dunes between elevations "j and 7U+li g = L— 2 A _ k x J, k DEVELOPMENT OF THEORY 17 The third method to compute the mean total bed material discharge is based on the following reasoning: Number of particles Weight QT = deposited per x per unit time and area particle Mean distance Width x traveled by x of (64) a particle channel where QTis the mean total bed material discharge in weight per unit time. Restricting the attention to eleva- tion yj, the terms in equation 64 are estimated as follows: [Number of particles deposited per] 1 unit time and area at y]. m. J + Q. A . , 2: M- : k = l Lt ’ _ (65) Weight per particle _ Ys(1 e) * — ——-———— , and at y]. 9]. Mean distance traveled by a particle which is deposited at elevation y]. = é[X\YD = yj] ,) where L, is the total length of yx(t) record, and all other symbols have been defined previously. Summing the product of the terms in equation 65 over all elevations, m . . v (1 — e) w n J + II _ S QT L: E mxw 1 E “’1', k ,(66) j = l k = l where Q; = estimate of mean total bed-material discharge; and W = width of channel. Equation 66 also illustrates the‘contribution of each elevation to the total transport, but its primary distinc- tion is that the transport rate is computed from the sounding records with a minimum number of computa- tions. The relationship between the three transport equa- tions, 55, 63, and 66, will now be demonstrated. First, the comparison of equations 55 and 66 is demonstrated. Combining equations 22, 25, and 66, n . vs (Ll-em )dN QT=——————— [{Elxv =y.1pYDz} (75) where { .} = events; fl = intersection of events; 71. 2 Ti; and l : T,- = random variable describing the duration of ith rest period of a bed-material sediment particle. T(n) = By virtue of equation 75, it follows that P[N(t) = n] =P[T(n) it, T(n+l) > t] =P[T(n) gt, Tn+l>t- T(r1)l .(76) For further simplification of equation 74, the follow- ing assumptions are made: (1) X(n) and N(t) are mutually independent for every n. (2) Xifor i? 1 are independently and identically distributed according to the probability density function fX(x), where DEVELOPMENT OF THEORY 19 0 s x < oo. Outside this range, fX(x) = 0. (3) Xiis inde- pendent of YD (1) fori =/= j. (4) YD(L) for i 2 1 are indepen- dently and identically distributed according to the prob- ability density function fYD (y), where ymm< \ y< ymax. Outside this range, fYD ( y) 0. (5) T for i/ > 1 are inde- pendently and identically distributed according to the probability density function fT(t), where 0 S t < 00. Outside this range, fT(t) = O. (6) Tiis independent of YD (i — 1) for i =# j. In other words, assumption 1 states that the total distance X(n) traveled by a sediment par- ticle after n steps should not depend on which time in- terval within [0,t] that these n steps occurred. The step lengths are always positive so that the particle always moves in downstream direction (part of assumption 2). Each step length depends on the elevation at which the particle is deposited at the end of that step (assumption 3). The elevation at which the particle is deposited at the end of any step does not depend on the elevation at which it was deposited at the end of any previous step (assumption 4). Finally, the duration of each rest period depends on the elevation at which the particle was deposited at the end of the previous step (assumption 5). Utilizing assumption 1, equation 74 becomes y F(x,y;t) =f Z0 {P[X(n )= y'] ymin 'P[N(t) = n\YD(n) =y'1fYD (n) (y'de' y =f P[X(O) gwaw) =y'1 ymin ’ PlNlt) = 0\YD(O) ‘=y'lfYD(0)(3")(13’I y 00 +f Z {P[X(n)§_x\YD(n) =y'l ymin ":1 -P[N(t) = n\YD(n) = yllfYDm) (y')}dy' .(77) Under assumptions 2, 3, and 4, and using the concepts of joint and conditional probability, P[X(n) £x\YD (n) = y'l = P[X(n - 1) + X" §x\YD(n) = y'] =f [X(n — 1) + Xn\YD(n) (x'\y')dx' o xl x zf dx'f fX (n —1),Xn\YD(n)(€’xl'§\yI)d§ 0 0 and using assumptions 3 and 4, P[X(n) _<_x\YD(n) =y'] x x' =f dx'f fX(n— 1)(§)I"X\Y (x' —C\y')d§ D 0 0 (n-)l =fxdx'f fXC) fX\YD ('Wx—my d; (78) in which n-l X(n-l)= 2: X1. x0=0, i=0 (n-l) fY(n—l)(§) :fxm g (n-2) =f/"X(9) fX(§—e)de; n=3, 4, 5, 0 and (n) fxm =fX(x> n=2 (79) In equations 78 and 79, firm 1, XMYD ,(€,x— §\y) denotes the joint probability density function of X(n — 1) and X", conditioned on YD(n), fX({) is the (n— 1)- fold convolution of the probability density func- tion for the length of a single step, and it is equal to the probability density function for the distance traveled by the particle in (n— 1) steps, and fX\ yD(x \y) is the con- ditional probability density function for a single step length given that the particle is deposited at elevation y. Turning now to the other part of equation 77 and using equation 76 and assumptions 4, 5, and 6, P[N(t) = n\YD(n) = y'] =PlT(n) it, Tn+1>t— T(n)\YD(n) = y'l too too I (n) °° = f fT(t dt'f fT\Y (t\y )dr (80) O 20 in which Tl T(n)= z Tl.‘ i=1 n-l) (n) "( . fT(n)(t)=fT(t)=ffT(e) fT(t -s)de, o and (n) fT(t') =fT(t) ; n=l In the above, Tn+1 is the random variable describing the duration of the (n+1)th rest period of a particle, fnn),Tn+1\YD(n)(t"T\y') is the joint probability density function of T(n) and Tn“, conditioned on YD(n), fTw (t \ y) is the conditional probability density func- tion Ifor the duration of a rest period given that the par— (n) ticle was deposited at elevation y, and fT(t) is the n-fold convolution of the probability density function, fT(t), for the duration of a single rest period and is equal to the probability density function for the duration of n suc- cessive rest periods. Similarly, the terms for n = 0 in equation 77 become: (0) = y'l =1 (82) because X(0) = X0 = 0 and 0 S x < co, and P[X(O) £x\YD PM“) = 0\YD(0) =y'] =P[T1>t\YD(0) = y'l =ffT.Y (t'\y')dt' , (83) t \ D where T1 is the random variable describing the dura- tion of the first rest period in time t. It is important to note that the initial condition, X(O) = X) = 0, implies that the particle starts its first rest period at t = 0. Introducing equations 78, 80, 82, and 83 into equation 77, y as F(x.y; t) =f f (y')dy'f f (t'\y')dt' YD t T\YD ymin y M x x' (n-l) +f ryow'my' E fdx' f fxm y n = 1 0 0 min cc t (n) fwa‘x' — §\y')d§‘ffT(t')dt'f fT\YD(I\y')dt .(84) 0 t- t' STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED The first term of equation 84 represents the joint prob- ability that the particle has not moved from its initial position and that its initial elevation is equal to or less than y, and it is not a function of x. Hence, y 2 2 3-— [F(x -t)1 =——a— f (y')dy' axay 'y' n = o axay YD ymin f (t'\y')dt' E 0 ,[ T\YD The corresponding density function is therefore 2 f(x.y;t) =5§57F(x.y;t) °° x (n - l) = fYD(y) 2: f fxm rwaoc — §\y)d§ n = 1 0 ,f‘fm d (t') t' 0 T If a large number of identical particles are initially at rest at x = 0, y = yo, equation 85 expresses the longitudinal and vertical distribution at time tof the particles which have moved from their respective in- itial positions. It should be noted here that f(x, y; t) is not a true probability density function because I ' t' D °° ymax f dx f f(x,y;t)dy =1 — P[N(t) = 0] <1 . (86) 0 ymin L That is to say, equation 85 applies only after the parti- cle has moved from its initial position. The expression f(x, y; t) does not exist for x = 0. If we assume that Xiis independent of YD( j) for all i and j and drop assumption 3, equation 85 reduces to equation 8, Way; t) no t f ffllbdt'f 0 t-t' °° (n) =f (3’) fix) Y, '2: i n=l fT\ YD (t\y) dt , (8) which is the Sayre-Conover (1967) two-dimensional stochastic model. The difference between equations 85 and 8 is that equation 85 takes some of the dependence between X and YD into account whereas equation 8 is based on the independence of X and YD. Hence, the Sayre-Conover model is a special case of equation 85. DEVELOPMENT OF THEORY 21 The marginal case of equation 85 gives the longitudinal distribution at time tof the particles which have moved from their initial positions. Integrating equation 85 over y, ymaX ft1 ={T(n)gt}—{T(n+11>t}c = {T(n) 5t} - {T(n +1) St} where {T(n+1) > t}c denotes the complement of the event {T(n+1) > t}. Because {T(n+1) s vent of {T(n) $ t}, it follows that t} is a sube- P[N(t) = n]=P[T(n)St] —P[T(n+1)§t] (n+1) =fmei') )dt'- ffT( t) dt' .(90) From equations 89 and 90, we have the marginal prob— ability density function, °° (n) E fX(x)P[N(t) = n] n=l t 0° (n + 1) =2 fxriic )f [Tm —fT(t) dt .(91) n = l 0 Equation 91 is identical to the Sayre-Conover (1967) one-dimensional stochastic model which is given in equation 10. As with equation 85, here also flx;t) is not a true probability density function because f f(x;t)dx=l —P[N.(t) =0] <1 , where P[N(t) = 0] is the probability that the particle has not moved from its initial position. In order to apply equations 84 or 85, the probability density functions fYD (y), fT\yD(t \y), and fX\ yD(x \y) must be known. These density functions are estimated from equations 25, 28, and 39. The probability density functions fT(t) and fX(x) are determined by the rela- tions 3’ max fr“) = f fTwDUWVYDiyldy (92) ymin and ymax f .(x) = f. (x\y)f MW (93) ,\ X\YD YD ymin where ymin and ymax are estimated from the yx(t) record. . Equations 92 and 93 are the continuous forms corres- ponding to equations 29 and 40 (or 41), respectively. With fT(t) and fly (x) determined, the corresponding convolutions, fT(t) and fX (x) are determined from equa- tions 81 and 79, respectively. 22 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED ANALYSIS AND DISCUSSION OF RESULTS EXPERIMENT AND BASIC DATA Three dune runs were made in a tilting recirculating flume of rectangular cross section, 61 m long, 2.4 m wide, and 1.2 In deep. The flume has been described in detail by Williams (1971). The bed material used in these experiments was a screened river sand (Cherry Creek sand), with a med- ian sieve diameter, (150 = 1.13 mm, and a geometric standard deviation, 0g = 1.51. The size distribution, shown in figure 5, was obtained by a sieve analysis of 3,000 grams of bed material. After an equilibrium flow, as defined by Simons and Richardson (1966), was established, the yx(t) and yt(x) records, the total bed-material discharge, and the hy- draulic properties of interest were measured. The methods and procedures of the measurements have been described in detail by Lee (1969). The summary of measured and derived data is given in table 1. The values of the water discharge, depth, energy slope, bed shape, and total bed-material concentration presented in table 1 are the average of several individual measurements. The sampled load was measured by a DH-48 sampler. The number of measurements was the same as the number of y,(x) charts. __ l l l l l l l l I 99.0 .. dso - |.|3 mm _ d - |.72mm 95.0 - '3“ — d5 = 0.75mm 90.0 - _ a' = '(9”+9&)=15| 800 _ 5 dso d.. ' Sieve Percent Mean ' diameter finer fall velocity 5 _ (mm) — (cm/sec) z E 60.0 — ,_ _ 2.83 99.35 22.60 2 8 40.0 .. 2.00 9|.60 l9.00 tr 3" r |.4|0 69.|2 l8.|0 20.0 - LOO 39.86 l4.60 0.707 l2.87 | HQ l0.0 - 0.500 2.63 8.45 5.0 — 0.350 0.78 — 2-0 * 0250 020 -— |-0 5 0:77 0.05 —— 0.5 — E = 0.2 l l l l I l l l I 0.4 0.6 08 LC 2.0 3.0 SIEVE DIAMETER, IN MILLIMETERS FIGURE 5. — Size distribution curve of bed material. ANALYSIS AND DISCUSSION OF RESULTS TABLE 1. — Basic data and computed parameters 23 Water discharge Flow depth Energy slope Water temperature QW. m3/s d, cm Se T, “C Ru" 3: a a s: d d Meandwvxelomy s: d d St d d n ar ,cm s Mean degiation Mean deglIatidn Mean deSiIatIdn Mean deiilatidn 4A 0.464 .003 31.1 0.6 61.3 0.00167 0.00009 20.0 0.3 16 1.24 .006 90.8 0.9 55.8 0.00029 0.00021 22.8 0.2 17 1.53 .008 89.3 0.6 70.4 0.00047 0.00005 22.0 0.2 Total Bedmaterial discharge Sampled load Run Concentration Mean total concentration Mean Chezy Mean bed Mean Stregmpower CT, mg/L load mg/L resistance shear stress shear velocity , qT,t/day-m coefficient Tb, d/cm2 U.,cm/s d/cm-s Mean 3233333 Mm 33.31%: W? 4A 168.6 53.6 2.77 1.5 6.1 8.6 50.8 7.13 3110 16 8.9 3.1 0.39 0.0 0.0 12.1 25.9 4.91 1450 17 29.7 10.8 1.61 6.3 10.2 11.0 41.2 6.43 2900 3350) Record ytu) Record Mean Time interval of Range of Run Froudenumber Lengthof record Laginterval Lengthof record Numberof measurements laginterval F, L,, hours min 1, m charts hours cm 4A 0.35 312 2.4 1235 31 6 3.7 — 13.0 16 0.19 80 » 1.2 1006 33 1 8.6 — 13.6 17 0.24 109 0.6 983 30 1 7.2 — 11.8 The yt(x) charts were obtained by mounting a sonic fY (y) = fY (3’) } depth sounder on the instrument carriage and travers- D E ing it along the centerline of the flume in the upstream 1 2 direction. The sonic depth sounder has been described 1 e 2' y 1 Z by Karaki, Gray, and Collins (1961). Although the _ ‘/_2n ~ 1 017 1 e' 7 y duration of traverse was approximately 5 minutes, the _ 2 4 1 2 ' ' J23? yt(x) record was considered to be instantaneous. The 1 ' 73’ d H94) yx(t) record was obtained by locating a sonic depth m e y sounder at the centerline of the flume 42.1 m '2-4 downstream of the headbox. Both the yt(x) and yx(t) . .. for-Z.4' tr I— . (7) 0 5‘3 N? o o o N g o‘ a: 3g —. m 0 o O. 0 Run I7 Triangular . In ------ Truncated GaussIon O' I I l I I I I I -I I I I F I Vie - o '0. - _ o N o'-- -I at— .— Q I IV I I ‘ ‘ I O -3.0 -2.0 —|.0 0.0 LC 20 3.0 -3.0 -2.0 -I.O 0.0 |.O 2.0 3.0 STANDARDIZED ELEVATION, y FIGURE 6. — Frequency histograms, triangular density function, and truncated Gaussian density function for the elevation of deposition and erosion. ANALYSIS AND DISCUSSION OF RESULTS 25 The mean and variance for the truncated Gaussian density are E[YD] =E[YE] = O , and VarlYD] = Var[YE] = E[Y12)] (96) 2.4 = 13(11le 51.017 yzg (y) dy = 0.891 -2.4 l 2 ‘ 2y . . where g (y) = e . For the triangular denSIty 1t E[YD] =E[YE] = 0 and Var[YD] = Var[YE] = E[Y12)l (97) 2 _ 2 _ 2.4 _ -E[YEl ‘ T - 0.960 The variances of these distributions are quite sensitive to the assumed truncation limits. A goodness of fit test using the chi-square statistic in- dicated that neither model would be rejected for runs 16 and 17 at a significance level of 0.05. For run 4A, however, both models were rejected at the same level of significance. As seen in figure 6, the truncated Gaus- sian density appears to give a slightly better approx- imation to fYD( y) and fYE( y); but, the triangular density is much easier to handle in analytical treatments. The variance of the triangular distribution is even more sensitive to the assumed limits than is the variance of the truncated Gaussian distribution. Therefore, the triangular distribution probably should not be used in predicting the variance. For stationary processes, continuity requires that the probability of erosion equal the probability of deposition for all elevations. Therefore, the density functions for the elevations of deposition and erosion must be identi- cal. The mean and variance of sample histograms as well as the total number of points available for analysis, Emi, are shown in table 2. Little data were available for run 16, only 134 crossings compared to 2,167 for run 4A and 708 for run 17. Although run 16 was continued for 33 hours, the very low transport rate (table 1) and slow movement of the bed forms limited the number of cross- ings available for analysis. It should also be pointed out that equilibrium flow was never attained for this flow which was barely above the initiation of motion stage. REST PERIOD DISTRIBUTIONS The sample conditional probability mass function of the rest periods were computed by determining the difference betweeen the time of reexposure and move- ment and the time of burial of the center of each class mark for each crossing event, mj’j , that occurred in the yx(t) record (fig. 2). The results of the measurements are presented in tables 3 through 5, and examples of the mass functions are presented in figures 7, 8, and 9. The standardized yx(t) record was used and the class width of the elevation was taken to be the same as that used in determining the probability distribution for the elevation of deposition, 0.4. The mean and variance of the conditional rest periods were computed using equation 30, and the results are presented in table 6. These results are also plotted as a function of bed elevation in figure 10. As can be seen from figure 10, both the conditional mean and variance of the rest periods decrease with increas- ing elevation of deposition. Inspection of figure 2 indi- cates that the conditional mean should decrease with increasing elevation of deposition. However, the decrease of the variance is not so obvious. Because the mean value is decreasing with increasing elevation, the decrease in the variance is not too meaningful. The coefficient of variation (standard deviation/ mean) is probably a better measure of the variability of the rest periods. Restricting our attention to runs 4A and 17, for reasons to be discussed later, the coefficient of variation remained roughly constant in the range of 0.6—0.75 for elevations above the mean bed elevation, and it increased with decreasing elevation to a value of about 1.5 at 2.4 standard deviations below the mean bed elevation. Thus the variability of the rest period, as measured relative to its mean, also decreases with increasing elevation at least up to the mean bed eleva- tion. As seen from figure 10, both the mean and variance of the conditional rest periods may be approximated by an expression of the form, 1-f[T\,YD = y] = Ae‘By and (98) vérITwD = y] = Ce‘Dy The constants A, B, C, and Din equation 98 were deter- mined by a regression analysis of the data plotted in figure 10, and the resulting values are presented in the figure. The values A and C represent measures of the mean and variance of the rest period, respectively, for the mean bed elevation. The values of B and D are measures of the rate of change of the mean and variance of the rest period with bed elevation, respec- tively. The distributions of the conditional rest periods were approximated by the two-parameter gamma probability 26 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED 0.03 o 50 IOO T, IN MINU‘IES OBSERVED DENSITY 0.08 0.04 0 O 20 40 T, IN MINUTES FIGURE 7. —- Sample probability mass functions of the conditional rest periods with fitted two-parameter gamma functions (run 4A). ANALYSIS AND DISCUSSION OF RESULTS 0 400 800 T, IN MINUTES 0.02 0.006 . YD =-0.8 YD =O.8 . K2 = 0.006 A K2 =0.023 >_ 0004.. r2 = L353 r 2 = L407 : m. .:'5 0.0' m. '=|8 g |,| A 1,53 8 0.002- A A 8 q A A Ci 0% o ' A $ 0 800 0 ICC 200 8 Y =-|.6 0.008 ~ A 0 K2 = 0.003 0'02 ‘ r 2 = 2.494 0.004- mm: 6 0,0. 0 .A r 0 0 IOOO 2000 T, IN MINUTES FIGURE 8. — Sample probability mass functions of the conditional rest periods with fitted two-parameter gamma functions (run 16). 27 28 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED 0.03 30.02 :9 r: "" 0.0l 0 T, IN MINUTES 0.02 4 0.04 a YD ='0.8 '2, 0.03 >— t m 0.02 2 LU O D 0.0I LIJ > E 0 U) m 0 .0l 5 ‘ 0 Y0 = “LG 0 K2 = 0.003 0-0'0‘00 r 2 = 0.524 0 500 I000 0 20 40 T, IN MINUTES FIGURE 9. — Sample probability mass functions of the conditional rest periods with fitted two-parameter gamma functions (run 17). ANALYSIS AND DISCUSSION OF RESULTS 29 '04 I I I I I A B ORUN 4A 40.4 0.9l ARUN l6 I342 [.08 O A '03 _ DRUNI? 4|.6 |.lO . A D A 8 D- O A s g A A 2 2 _ - EIO B A A <2; 8 G A A E 5 8 8 g A no1 - 8 I 8 a D IOO l J 1 l 1 -2.0 -l.O 0 LG 2.0 b I I T l c 1 D ORUN 4A |2|O 2.!8 '06 - D ARUNIG l0l60 2.93 _ DRUNIT mo 2.79 A 9 a a A A _ A 2 4 - .. 3 I0 8 :5 8 A A m 8 A Z. 8 o A LIJ ‘5 I02.— 8 B - g o o o ‘>‘ c: A 100 ~ - Cl 1 1 1 l l -2.0 -IO 0 IO 2.0 STANDARDIZED ELEVATION FIGURE 10. — Variation of the conditional mean and variance of rest periods with bed elevation. density function which has the form, k2 2 "2 y f (My) = ' (k t) ’ T\YD I‘(r2,y) 2.y -1—(k e )t 2,3) (99) where l"(-) = gamma function; and k2! y, r2, y = scale and shape parameters, respectively. The scale and shape parameters were estimated by using the method of moments, k = E[T\YD=y} 20’ vér[T\YD=y] and (100) (filT‘xYD =yl)2 A r = A——— =k E[T\Y =y] 21y Var[T\YD=y] 2,y D and the data contained in table 6. The variation of k2, y and r2_ywith bed elevation are presented in table 7 along with the results of a chi-square goodness of fit test. The ability of the two-parameter gamma distribu- tion to fit the measured mass functions is illustrated in figures 7, 8, and 9. From table 7, as well as from figures 7, 8, and 9, both the scale and shape parameters increase with increas- ing bed elevation, with a few exceptions for the shape parameter. The shape of the conditional density of the rest periods (figs. 7, 8, 9) approaches a J-shape and becomes more peaked as bed elevation decreases. Therefore, the exponential density might fit better than the two-parameter gamma density below the mean bed elevation (y < 0). The exponential density function is a special case of the gamma density with r2”, = 1. The better fit of the exponential density seems to be consistent with the fact that all rejections of the chi-square test (6 rejections out of 22 at a significant level of 0.05) occurred below the mean bed elevation (table 7). It would appear that the exponential form for the conditional rest period as proposed by Grigg (1969) is only valid for elevations below the mean bed eleva- tion. A major factor in determining the degree of fit be- tween the measured density functions and the fitted curves in figures 7, 8, and 9 appears to be the number of points available from which the distribution was con- structed. In general, if more than 100 points were available, m,_,~, the fit is pretty good. The weakness of the data for run 16 is very apparent. Even at the mean bed elevation, only 18 crossing events were observed. 30 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED Combining equations 98 and 100, the scale and shape parameters can be estimated using only the constants A, B, C, and D. and .(101) AZeDy r213): C—2_e By = k2,yAe _By The sample joint probability mass functions of the rest period and the elevation of deposition were com— puted from equation 33 using the results presented in tables 2 through 5. The results of these computations are presented in tables 8, 9, and 10. The correlation coefficients were computed by using equation 34, along with the data contained in tables 2, 6, 8, 9, and 10. The values of the correlation coefficients were —0.27, -0.53, and —0.26 for runs 4A, 16, and 17, respectively. The rest period and the elevation of deposition are negatively correlated, but the degree of their linear association is not strong. The sample marginal probability 'mass functions, pT(ta), were computed by use of equation 29 and the data contained in tables 2, 3, 4, and 5. The results of these computations are also presented in tables 8, 9, and 10. The sample frequency histograms for the marginal rest periods are plotted in figure 11. The mean and variance of the marginal rest periods were computed by use of equation 31. These results are also presented in figure 11. The variance values appear to be extremely large. For example, the standard deviation for run 4A is almost four times the mean value. The computed variance values are extremely dependent on the long rest periods, the extreme events generally oc- cur at low bed elevations. For example, by ignoring rest periods of greater than 2,000 minutes, which have a probability of occurrence of only 0.0015, the variance is reduced from 42,000 to 12,000. Also shown in figure 11 are exponential density func- tions with a mean equal to the computed marginal mean. The exponential density function fits the data reasonably well; however, there would appear to be room for improvement. A gamma density fitted by the method of moments would be an extremely poor fit of the data. A gamma distribution, estimated by the max- imum likelihood method may fit the data reasonably well. The marginal distribution of the rest periods could also be estimated by 2.4 f(t)= f f - x = fll—X—L (xk .) l'y'ye 13’ y (103) V1 .) l,y.y ,y/y where y and y' = arguments of YE and YD, respec- tively; and k,yyyy.andrlyy‘y.=scale and shape parameters, respectively. The parameters k1, y‘ y. and r1, y, ylwere estimated by the OBSERVED DENSITY ANALYSIS AND DISCUSSION OF RESULTS 31 0.03 " RUN 4A A E T =53.e ' 0.02 ~ [1 m < V’a‘r [T]: 42,IIO min2 \ 0.0: - 0 v ‘- t 0.03 RUN 16 /E\[T] - I60 5 ' 002 . - . mm vé‘r[T]= 93,077 min2 0.0l ‘ o —1_ M 1 0.03 RUN I? 0.02 ‘ /E\[T] =60.l min vé‘rfi]: 60,469 min2 0.0: - o d 0 I00 200 REST PERIOD, IN MINUTES FIGURE 11. — Frequency histograms for the marginal rest period and exponential fits. 300 32 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED 2 2 >_ 4 YE = 0.0 >_ t YD =-0.8 t: m ‘2 Z T O . .= g Tn", 440 g o H o LLJ LIJ & ' a 34 % m m 0 d O O I I l I I I O 1 2 O 1 2 STEP LENGTH, IN METERS STEP LENGTH, IN METERS 2 OBSERVED DENSITY STEP LENGTH, IN METERS FIGURE 12. — Sample probability mass functions of the conditional step lengths given the elevation of erosion is 0.0 with Gamma fits (run 4A). OBSERVED DENSITY ANALYSIS AND DISCUSSION OF RESULTS 2 YE = 0.0 - YD ='O.8 > d mm: 376 : $3 DJ 0 0 LL] > I [U U) (D O YD =0.0 STEP LENGTH, IN METERS OBSERVED DENSITY 2 0 YE =0.0 YD =0.8 0 I I I o 1 2 STEP LENGTH,IN METERS 2 STEP LENGTH, IN METERS 33 FIGURE 13. — Sample probability mass functions of the conditional step lengths given the elevation of erosion is 0.0 with Gamma fits (run 16). 34 OBSERVED DENSITY STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED 2 OBSERVED DENSITY STEP LENGTH, IN METERS OBSERVED DENSITY O 2 1 - O I l I r: 0 1 2 O l T I l 1 2 STEP LENGTH, IN METERS FIGURE 14. — Sample probability mass functions of the conditional step lengths given the elevation of erosion is 0.0 with Gamma fits (run 17). ANALYSIS AND DISCUSSION OF RESULTS 35 method of moments, using data contained in tables 57, 58, and 59 and the expressions é[X\yE=y, YD=y'] l'y'y' vérpm/ =y, y =y‘] E D and ‘ ,, _ _. 2 (104) (Elx‘ya'y'yb'yll vér [xxx/E = y, YD = y'l Plouy' = ElX\YE = y, YD = y']kl'yly, The variation of k1, y' y. and r1, y, y. with the elevations of erosion and deposition are shown in tables 60, 61, and 62. These approximations are also shown in figures 12, 13, and 14. The chi-square test for goodness of fit was used to test these gamma approximations. The results of these tests are summarized in tables 63, 64, and 65. None of the 81 distributions tested could be rejected at the 0.05 level of significance. In other words, there is no good statistical reason to reject the hypothesis that the prob- ability density functions for the step lengths, given the elevation of deposition and erosion, are distributed ac- cording to the two-parameter gamma distribution. The LO 4 A L A A 08 A ' - O A A D A A A D 8 o A 6 I D O U) D o a: D o Eoe~ g o o o D 2 D D D n E z" 25 0.4 - 2 0 RUN 4A A RUN IS 02‘ o RUN I? c 0 I -2.0 0 2.0 fitted gamma distributions are also plotted and the example mass functions presented in figures 12, 13, and 14. These figures also help to illustrate the abili- ty of the two-parameter gamma distributions to fit the measured conditional step length distributions. The sample conditional mass functions, given the elevation of deposition, were computed based on equa- tion 37 and the data contained in tables 2 and 11—56. These mass functions are presented in tables 66, 67, and 68. The corresponding conditional means and variances were computed using equation 38 and are presented in table 69 as well as being plotted in figure 15. Again, the general decrease in the expected value of the step length with an increase in the elevation of deposition is apparent. The sample joint probability mass function of the step length and the elevation of deposition was com- puted by equation 43, and the results are shown in tables 70, 71, and 72. The correlation coefficients were computed by equation 44, and their values were —0.15, -0.15,and—O.20 for runs 4A,16, and 17, respectively, in- dicating that the step length and the elevation of deposi- tion are negatively correlated, but the degree of their linear associations is not strong. The sample marginal probability mass functions, pX(xB), computed using equation 40, are also shown in tables 70, 71, and 7 2. The 0.3 ORUN4A - ARUNIG g . uJ URUNI? ,— Lu 2 0.2- ll? 1 300 0°C 0 oo 3 DDDDDDDD 1 o I O 2.0 0-1 - 2.0 STANDARDIZED ELEVATION FIGURE 15. — Variation of the conditional mean and variance of step lengths with bed elevation; E[X\ YD = y]. 36 sample frequency histograms for the marginal rest periods are plotted in figure 16. The mean and variance of the marginal rest periods were computed by use of equation 41. These results are also presented in figure 16. The range of the means is fairly small, only 0.610 to 0.799 m. The mean dune lengths, as measured by the distance between trough points, for the three runs were 1.19, 1.66, and 1.23 m respectively for runs 4A, 16, and 17. The mean step lengths were, therefore, 54, 48, and 49 percent of the mean dune lengths. Grigg (1969) found the mean step lengths of single tagged particles to be about 60 percent of the mean dune length. Of course, Grigg was working with a much finer sand, .33 to .45 mm, as compared to 1.15 mm for this study. Also shown in figure 16 are gamma density functions for which the parameters k and r were determined from the mean and variance shown in the figure. The gamma functions appear to fit the data very well for all three runs. The value of the parameter r ranged from 4.05 for run 4A to 4.59 for run 17. This is slightly more than twice the value estimated by Yang (1968) from the step length distribution of a single plastic particle. BED-MATERIAL TRANSPORT The following assumptions and conditions were used to estimate the mean total bed-material transport rate by equations 55, 63, and 66: (1) Because the bed material was coarse sand (fig. 5), all sediment par zicles are assumed to be transported as bed load. Expressed mathematically, P[E1]= 1. (2) 'yS (1 — G)— - 1602 kg/m3. (3) Ay= 0. 43y everywhere. By virtue of item 1, it follows that = 153, VT(j)= —V},(j), and QT: QB. In item3, s, is the standard deviation of the bed elevation computed from the yx(t) record. All parameters and statistics which are required by equations 55, 63, and 66 are summarized in tables 73 and 74. The average depth cf the zone of bed material movement, h, was determined by equation 59. It was found that one chart of the y,(x) record (about 34 m) is sufficient to obtain a reliable value of h, although over 30 charts of the y,(x) record were used in this study. Each chart contained about ten dunes. Using equation 62, E 1, the percentage of volume between eleva- tions 7)} and 711+) occupied by dunes (hereafter will be referred to as the effective volume ra~ tio) was obtained from the y,(x) record. The results are presented in table 74 and plotted in figure 17. As shown in figure 17 {j is nearly independent of flow con- dition. As long as the bed forms are dunes, EJ- does not change appreciably. It is also shown in figure 17 that EJ- is nearly unity and zero at yJ- = - 2.4 and yj = + 2.4, respectively. This is partial justification for the upper and lower limits of the elevations of erosion and deposi- tion used in equations 94 and 95. STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED Another effective volume ratio can be obtained from the yx(t) record. Denoting this ratio by Q, J l g]. —Tt E tj'k (105) k = 1 where L, = total length of yx(t) record; mj = maximum number of bed forms contained in the yx(t) record which also contains some deposition at elevation yj; and t“ = measurement of the conditional rest period. There is no significant difference between 15 and g (table 74) except for depths greater or less than 2.0 standard deviations from the mean. The longitudinal profiles ( y,(x) records) appear to contain a larger num- ber of extreme events than the time record at a given point (the yx(t)). The explanation for this is probably that the flow was fairly stationary but that it was not longitudinally uniform. A comparison of measured and computed total bed- material transport rates is shown in table 75. It is seen that: 1. For run 4A, all three equations provide excellent estimates to the observed mean total bed-material dis- charges. 2. Equation 55 provided an excellent estimate to the mean total bed-material discharge for run 17 . However, the other two equations overestimated the discharge by more than 25 percent. The reason for the differences in the equations is not understood. 3. None of the equations gave good estimates of the mean total bed-material discharge for run 16. The con- sistently overestimated discharge ranged from 64 per- cent for equation 63 to 80 percent for equation 55. It should be remembered, however, that the mean total bed-material discharge was less than 9 mg/L during this run, that the flow was not in equilibrium as illus- trated by the large variation of energy slope (table 1), and that very few rest period statistics were available for analysis (fig. 8). Taken as a whole, the results are very encouraging. Although equation 55 gave the most accurate results for run 17, it should be noted that equations 63 and 66 gave very consistent results for all runs when they are compared one with the other. The discharge predicted by equation 66 was 8.3, 8.4, and 8.5 percent larger than that predicted by equation 63 for runs 4A, 16, and 17, respectively. Although equation 66 is probably simpler to evaluate than equation 63, it appears that some acv curacy has been sacrificed. The main difference be- tween equations 63 and 66 is the way in which the effective depth or effective volume ratio (eq. 62, 71) is ANALYSIS AND DISCUSSION OF RESULTS 37 I5 RUN 4: I.O . E [x] = 0.649 m \ Vé‘r[X]=0.lO4m2 0.5 - \ \i— 0 I I I I I o l 2 3 LS RUN I6 .>_- o /E\[X] = 0.799m .. I. . 3 \ var [x]: 0.I4O m2 3 / \ O U > $05 - \ 3 \ o \ _ o / I I x— 1 o | 2 3 Is - RUN I7 I /E\[X] = O.6|O m Lo 4 var [X]=0.08| m2 0.5 - \ o I T T 1' I o I 2 3 STEP LENGTH, IN METERS FIGURE 16. - Frequency histograms for the marginal step length with Gamma fits. 38 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED TABLE 73. — Variation of various statistics with stream power Run b 13%] Var2[Xl £1511) vérlTZI VT = fl V+ h s (d/Cm‘s) (m i ' (min ) EU] (cm/s) (cm) (cm) (cm/s) 4A 3,110 0.649 0.104 53.8 42,110 00201 0.0378 9.66 4.26 16 1,450 0.799 0.140 180.5 93.077 0.0073 0.0166 6.89 3.01 17 2,900 0.610 0.081 60.1 60,469 00169 0.0372 7.13 3.61 TABLE 74.-—Comparison of the effective volume ratios at elevation yj; of the similarity of equations 55 and 63, it would be 5 7170’”th record andgrfi'omyxm record diffith to say one was more accurate than the other. Run 4A Run 16 Run 17 Their relative accuracy probably depends on chance oc- ID=yj E C {I z z z currence of extreme events in one or the other records .7 .7 .7 .7‘ .7 .7 of bed elevation. -2.8 ”68 _____ .ooo _____ 1.000 _____ In table 75, 6T is the mean total bed-material dis- -“ .963 _____ _995 _____ .998 _____ charge in weight per width and time, and it was ob- _2_() .949 0.903 _975 ..... .989 _____ tained by div1d1'ng equations 55, 63, and 66 by the Wldth -1.6 .925 .883 .935 0.890 .964 0.943 of the channel, W. '1-2 -377 “5 ~365 -840 -9°° -867 If we define q’B (j) as the mean bed-load discharge ‘0' 8 '791 '787 ‘77s ‘719 ‘8‘” '75" associated with elevation yj, then based on equation 63, —0.4 .668 .675 .670 .656 .669 645 0.0 512 .529 .535 .567 .522 .522 “ A 0.4 .344 .379 .385 .388 .364 .380 (11'3”) = 73 (l _ (”VB (1') €1.ij (106) 0.8 .189 .222 .230 .230 .202 .234 i: '3: 3:: '3: 3:: '22: '3: where 4‘B( j ) estimates q’B( j ) and 17154 j ) is an estimate of 2.0 .001 .009 .012 .008 .002 .013 the mean transport speed of a bed-load particle at 2,4 .000 _____ .004 .001 .000 .00, elevation yj. The mean transport speed, VB( j), is given 2.8 .......... .001 ————— .000 ..... by equation 61 provided that the suspended load is negligible. With equation 106, the variation of bed-load discharge with bed elevation may be investigated. This computed, and these functions were similar (table 74); variation is shown in figure 18 for all three runs. It is therefore, the consistency of their final result was ex- seen that the maximum bed-load discharge is associ- pected. Equation 55 had the lowest average absolute er- ated with the mean bed elevation and that an insignifi- ror for all three runs; however, equation 63 gave the cant portion of the bed-load movement appears to occur most accurate result on two out of three runs. Because for yj S —2.4 and y, 2 +2.4. TABLE 75.—Comparison of measured and computed total bed- material transport rates Measured total bed-material discharge (t/day-m) Run Number of 5.1ax1mum Minimum Standard Mean measurements deviation 4A 54 5,30 0.91 0.88 2.77 16 32 0.72 0.18 0.14 0.40 17 32 3.04 0.59 0.59 1.61 Computed mean. 0T (t/day-m) Computed mean/measured mean Run Eq. 55 Eq. 63 Eq. 66 Eq. 55 Eq, 63 Eq. 66 4A 2.68 2.72 2.94 0.970 0.983 1,066 16 0.70 0.64 0.67 1.801 1.641 1.725 17 1.67 2,05 2.18 1.035 1.269 1.354 ANALYSIS AND DISCUSSION OF RESULTS 39 I I 1’ I I I I TI I . 0. ' ,3— Explanahon “ d 0 Run 4A E ®_ 4 d A Run I6 .1 a: D Run I? L; o 3 O‘- I‘ .1 S , (85) The one-dimensional model as a marginal case of equa- tion 85 was °° (n) t (n) (n+1) E rxmf [Tn-1 -fT(t') av .(91) =1 0 Note that y is the standardized elevation. In order to ap- ply equations 85 and 91, the probability density func- tions, fYDU’), fT\ YDUV’) fr“) fr“) fT(t) fX\YD(x\y)1fX(x) fX(x), and fX(x) must be specified. Although probability density functions for all these distributions have not been determined in this report, the measured probability mass functions have been presented in tables 2, 3—5, 8-10, and 66—68, respec- f(x; t) = SUMMARY AND CONCLUSIONS Stochastic models were developed which can be used to predict the transport and dispersion of bed-material sediment particles in an alluvial channel. These models are based on the proposition that the movement of bed- 40 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED In 6 r v I l 1* I 1 I I I I Explanation A ’ 9, - 0 Run 4A ,I ~ 0 A Run I6 El Run l7 Va(y), IN METERS PER MINUTE 0.05 A 0.0 l l 1 1 l l 1 l l —3.2 —2.4 -|.6 -0.8 0.0 0.8 L6 2.4 3.2 STANDARDIZED ELEVATION, y FIGURE 19. — Mean transport speed of a bed-load particle as a func- tion of bed elevation, y. l l T l l I I T 7 _ Explanation i 0 Run 4A A Run I6 0 Run l7 IN SQUARE METERS PER SQUARE MINUTE 200 300 v |00 y : Var [T1Y0:y] v’ar 0.0 —0.8 0.0 08 LG 2.4 3.2 STANDARDIZED ELEVATION, y —3.2 -2.4 —I.6 FIGURE 20. — Ratio of the conditional variance of step lengths to the conditional variance of rest periods as a function of bed elevation. tively. Equations for determining the n-fold convolu- tions of pT(t) and p X(x) can be obtained from equations 82 and 79 with proper substitutions (Parzen, 1967). Further progress in the solution of either equation 85 or 91 could proceed along either of two lines. First, all probability density functions could be replaced with the corresponding sample probability mass functions, the integrals approximated by summations, and the solu- tions obtained numerically. Alternately, the mass func- tions could be fitted by density functions of some assumed form and an analytical solution attempted. Lee (1973) used various fitting procedures to obtain all the probability density functions required to solve equa- tion 85, but the integration of the equation appears quite formidable. material sediment particles consists of a series of steps separated by rest periods and, therefore, their applica- tion requires a knowledge of the probability distribu- tions of the step lengths, the rest periods, and the eleva- tion of particle deposition and erosion. The probability distribution of the rest periods, condi- tioned on the elevation of particle deposition and the probability distributions of the elevation of particle ero- sion and deposition, were obtained from a record of the bed elevation at a fixed point as a continuous function of time [yx(t) record]. The necessary assumptions were: (1) Equilibrium flow; (2) both erosion and deposition do not occur at the same point during the same time period; and (3) the number of particles per unit volume of the bed is constant. The probability distribution of the step lengths, con- ditioned on the elevation of particle erosion and the elevation of particle deposition, was obtained from a series of instantaneous longitudinal bed profiles [y,(x) record]. The required assumptions were: (1) All bed- material sediment particles which are eroded from the upstream face of a dune will be deposited on the downstream side of the same dune; and (2) no deposi- tion occurs on the upstream sides of dunes, and no ero- sion occurs on the downstream faces of dunes. These assumptions appeared to be reasonable at least for a dune-covered bed composed of a coarse sand. Introducing an additional assumption that the eleva- tion of particle erosion and the elevation of particle deposition are mutually independent, various related probability distributions were obtained. These distribu- tions included: (1) The marginal distributions of the rest periods and the step lengths; (2) the joint distribu- tion of the rest periods and the elevation of particle deposition; and (3) the joint distribution of the step lengths and the elevation of particle deposition. A two-dimensional stochastic model for dispersion of bed-sediment particles was then derived (eq. 85). In order to apply the model, the probability distributions of (1) the step lengths given the elevation of particle deposition; (2) the rest periods given the elevation of particle deposition; and (3) the elevation of particle deposition, must be known. The mass functions of these distributions were estimated; however, the integrations required by the model remained unsolved. Applying the concept of continuity, three bed- material transport models were presented. Application of these models requires the estimation of: (1) The con- REFERENCES CITED 41 ditional means of the rest periods and the step lengths; (2) the probability distribution of the elevation of deposition; (3) the average depth of the zone of bed- material movement; and (4) the effective volume ratio. These were all obtained from the yx(t) and y,(x) records. In the derivation of the models, the bed load was defined as that part of bed material which is deposited on the downstream face of the dune from which it is eroded, and the suspended load was defined as that part of bed material which passes two or more dune crests before being deposited. These definitions are very pre- cise compared to the definitions prepared by the Task Committee on Preparation of Sedimentation Manual (1962). Based on flume experiments with a coarse sand, the following conclusions were drawn: 1. The elevation of particle erosion and the elevation of particle deposition can be considered to be identically distributed, and their distribution can be approximated by either a truncated Gaussian density function or a symmetric triangular density function. In general, the truncated Gaussian density provides slightly better results; although the triangular density is much easier to handle analytically. 2. The conditional probability distribution of the rest periods, given the elevation of deposition, can be well described by the two-parameter gamma density func- tion. The shape of the conditional density approaches a J-shape and becomes more peaked as bed elevation decreases. A. Both the conditional mean and variance of the rest periods increase with decreasing bed eleva- tion. These relations can be expressed by exponen- tial functions. B. Both the scale and shape parameters for the conditional distribution of the rest periods increase with increasing bed elevation, and they can be described by exponential functions of bed eleva- tion. C. The correlation coefficient between the rest periods and the elevation of deposition indicated that the rest periods and the elevation of deposition are negatively correlated, but the degree of their linear association is not strong. 3. The conditional probability distribution of the step lengths, given the elevation of deposition and the eleva- tion of erosion, can be approximated by the two— parameter gamma distribution. The shape of the condi- tional density is strongly dependent on the elevation of deposition and erosion. A. For a fixed elevation of deposition, both the double conditional mean and variance of the step lengths increase with decreasing elevation of ero- sion. In other words, longer step lengths are associ- ated with lower elevation at which a sediment par- ticle is eroded or deposited and vice versa. B. The correlation coefficient between the step lengths and the elevation of deposition indicates that they are negatively correlated, but the degree of their linear association is not strong. 4. All three bed-material transport models are found to be quite satisfactory except for run 16. A. The effective volume ratio can be obtained from either the y,(x) record or the yx(t) record, and it appears to be nearly independent of flow condi- tion. B. The maximum bed-load movement is associ- ated with mean bed elevation, and little movement occurs for y$ —2.4 and y? +2.4. 5. The mean transport speed of a bed-material parti- cle, the average depth of the zone of bed material move- ment, and the standard deviation of bed elevation in- creased with increasing stream power, whereas the marginal means and variances of the rest periods and the step lengths decreased with increasing stream power. Figures 10 and 15 suggest that the step lengths and the rest periods are positively correlated in an average sense, but the degree of linear association was not strong. REFERENCES CITED Crickmore, M. J ., and Lean, G. H., 1962, The measurement of sand transport by means of radioactive tracers: Proc. Royal Soc. of London, ser. A, v. 266, p. 402-421. Einstein, H. A., 1937, Der Geschiebetrieb als Wahrscheinlichkeits- problem [The bed-load movement as a probability problem]: Mit- teilung der Verschsanstalt fiir Wasserbau, an der Eidgeno'ssische Technische Hochschule in Ziirich, Verlag Rascher and 00., 110 p. 1950, The bed load function for sediment transportation in open channel flows: US. Dept. Agriculture Tech. 31111., no. 1026, 70 p. Grigg, N. S., 1969, Motion of single particles in sand channels: Ph. D. dissert., Colorado State Univ., Fort Collins, 162 p. Hubbell, D. W., and Sayre, W. W., 1964, Sand transport studies with radioactive tracers: Am. Soc. Civil Engineers Proc., v. 90, no. HY3, p. 39—68. 1965, Closure to: Sand transport studies with radioactive tra- cers: Am. Soc. Civil Engineers Proc., v. 91, no. HY5, p. 139—149. Karaki, S. 8., Gray, E. E., and Collins, J ., 1961, Dual channel stream monitor: Am. Soc. Civil Engineers Proc., v. 87, no. HY6, p. 1—16. Lee, B. K., 1969, Laboratory study of an alluvial stream at one-foot depth: M.S. thesis, Colorado State Univ., Fort Collins, Civil Eng. Dept., 57 p. __1973, Stochastic analysis of particle movement over a dune bed; Ph. D. dissert., Colorado State Univ., Fort Collins, 220 p. Parzen, E., 1960, Modern probability theory and its applications: New York, John Wiley and Sons, Inc., 464 p. 1967, Stochastic processes: San Francisco, Holden-Day, Inc., 324 p. 42 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED Sayre, W. W., and Conover, W. J., 1967, General two-dimensional alluvial channels: U.S. Geol. Survey Prof. Paper 422—J, 61 p. stochastic model for the transport and dispersion 0f bed- Task Committee on Preparation of Sedimentation Manual, Commit- material sediment particles: Internat. Assoc. Hydraulic tee on Sedimentation, 1962, Sediment transportation Research, 12th Cong., Fort Collins, 0010-, Procu V~ 2, P- 38—95. mechanics: Introduction and properties of sediment: Am. Soc. Shen, H. W., and Todorovic, P. N., 1971, A general stochastic model Civil Engineers Proc., v. 88, no. HY6, pt. 1, p. 78. for the transport of sediment bed material: 1st Internat. Sym- Williams, G. P., 1971, Aids in designing laboratory flumes: U.S. Geol. posium on Stochastic Hydraulics, ed. Chao-Lin Chin, Proc., Pitts- Survey open-file report, 294 p. burgh, Penn, 1. 426-448. Yang, T., 1968, Sand dispersion in a laboratory flume: Ph.D. dissert., Simons, D. B., and Richardson, E. V., 1966, Resistance to flow in Colorado State Univ., Fort Collins, 162 p. SUPPLEMENTAL DATA TABLES 2-72 44 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED Table 2. Sample probability mass Motion: of elevations of dnpoeitm‘ and erosion Run 4A Run 16 Run 17 Elevation Triangular Truncated y. p (14-) P (11-) p (11-) P (y‘) p 01-) p (y.) Density Gaussian 1. Int 1’51. I’D-1. IE7. YD; YE‘L -3.6 0.000 0.000 0.000 0.000 0.000 0.000 -3.2 .000 .000 .006 .006 .000 .000 ----- -2.8 .000 .000 .007 .007 .001 .002 ---------- -2.4 .006 .006 .011 .013 .005 .006 0.004 0.006 -2.0 .025 .032 .019 .025 .017 .017 .028 .022 - 1 .6 . 060 . 060 . 038 . 044 . 044 . 047 . 056 .046 -1.2 .091 .088 .062 .065 .089 .089 .083 .079 -0.8 .125 .129 .104 .109 .129 .129 .111 .118 -0.4 .152 .140 .122 .116 .153 .148 .139 .148 0.0 .160 .166 .133 .134 .154 .157 .158 .162 0.4 .156 .154 .197 .189 .154 .153 .139 .148 0.8 .120 .116 .137 .130 .117 .118 .111 .118 1.2 .070 .068 .097 .097 .073 .072 .083 .079 1.6 .025 .032 .046 .044 .041 .041 .056 .046 2.0 .009 .009 .016 .016 .019 .017 .028 .022 2.4 .001 .001 .005 .005 .004 .004 .004 .006 2.8 .000 .000 .000 .000 .000 .000 ---------- 2 m1. or Z m; 2.167 2.167 134 134 708 708 ---------- i 1i END] or ENE] - .130 - .133 .055 .006 - .039 - .043 ---------- VErUD] or v§r[yE] .513 .850 .999 1.046 .863 .870 ---------- Note: mi is the total number of bed forms Contained in the 1110?) record and which also contain some deposition in the class interval associated with the elevation 1111 . 14.3. elevation 141:. is the total number of bed forms contained in the yr(t) record and which also contain some erosion in the class interval nssoclated with the SUPPLEMENTAL DATA TABLES Table 3. Sample conditional probability mass flotation of rest periods, p1,” (tulyi) (Run M) D 1a 0 10 20 50 40 50 60 70 30 90 100 110 120 150 140 10"] 10 20 5o 40 50 60 70 30 90 100 110 120 150 140 150 ta 5 15 25 55 45 55 65 75 35 95 105 115 125 155 145 —2.3 0 o o 0 0 0 0 0 0 0 0 0 0 0 0 -2.4 0 0 0 0 0 o 0 0 0 0 0 0 0 .0909 o —2.0 .0443 .0746 .1045 .0396 .0443 .0293 .0149 .0293 .0293 .0149 0 .0293 .0149 .0149 .0149 .,. -1.6 .0411 .1241 .1022 .0657 .0657 .1095 .0453 .0146 .0565 .0219 .0146 .0146 .0075 .0219 .0219 =3 —1.2 .1051 .1445 .1136 .0725 .0979 .0670 .0412 .0253 .0561 .0105 .0154 .0052 .0509 .0206 .0561 5 -0.3 .0909 .2016 .1265 .0933 .0350 .0933 .0257 .0516 .0516 .0257 .0277 .0257 .0193 .0153 .0153 E —0.4 .1019 .2229 .1557 .1752 .1173 .0657 .0473 .0550 .0225 .0159 .0127 .0096 .0159 .0064 .0052 E 0.0 .1579 .2742 .2271 .1535 .0805 .0499 .0560 .0111 .0035 .0054 0 .0023 .0035 0 0 3 0.4 .2139 .5659 .2041 .1056 .0710 .0143 .0177 .0059 0 0 o 0 0 0 0 5 0.3 .5145 .5952 .1694 .0766 .0202 .0242 0 0 o 0 0 o 0 0 o 5 1.2 .4067 .5555 .1367 .0467 .0067 0 0 0 0 0 0 0 0 0 0 E 1.6 .5159 .5750 .0355 .0273 0 0 0 0 0 0 0 0 o 0 0 m 2.0 .7563 .2105 .0526 0 0 0 0 0 0 0 0 0 0 0 0 2.4 .6667 .5555 0 o 0 0 0 0 0 0 0 0 0 0 0 2.3 o 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Ta 150 160 170 130 190 200 250 500 550 400 500 600 1,000 2,000 13” 160 170 130 190 200 250 500 550 400 500 600 1,000 2,000 3,000 mi”; ta 155 165 175 135 195 225 275 525 575 450 550 300 1,500 5,000 —2.3 0 0 0 0 0 0 0 0 0 0 0 0 o o o -2.4 , 0 o 0 0 0 .2727 .0909 0 0 .0909 0 .0909 .0909 .2727 11 -2.0 0 0 .0149 .0443 o .0746 .0597 .0149 .0293 .0293 .0443 .0396 .0443 0 67 g: -1.6 .0453 .0219 .0146 .0075 .0219 .0219 .0219 .0453 .0219 .0292 .0 19 .0146 0 0 157 a -1.2 .0253 .0052 .0052 .0105 .0155 .0509 .0464 .0155 .0052 .0154 o 0 0 0 194 g -0.3 .0153 .0193 .0079 .0079 .0059 .0193 .0079 0 0 .0059 0 0 0 0 255 E -0.4 .0052 .0052 .0052 0 o .0052 .0052 0 0 o 0 0 0 0 514 E 0.0 0 o 0 0 o 0 0 0 0 0 0 0 0 0 561 3 0.4 0 0 0 o 0 0 0 0 0 0 0 o 0 0 553 g 0.3 0 0 0 0 0 0 0 0 0 0 0 o 0 0 243 E 1.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 150 a 1.6 0 0 0 0 0 o 0 0 0 0 0 0 0 0 72 2.0 0 o o 0 0 0 0 0 o 0 0 0 0 0 19 2.4 0 o 0 0 0 0 0 0 0 o 0 0 o o 5 2.3 0 0 0 0 0 o o 0 o 0 0 0 0 0 0 Note: Tu’ Towl' and ta are in minutes. "'12 1: is the total number of bed forms contained in the yr“) record and which also contain both an up-crossing ’ and a down-crossing at the elevation yi. 45 46 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED Table 4. Sarple conditional probability mass function of rest periods, PTIY (talyi) (Run 16) D a 0 20 40 60 80 100 120 140 160 180 200 to“ 20 40 60 80 100 120 140 160 180 200 220 a 10 50 50 70 90 110 130 150 170 190 210 —2 . 8 0 0 0 0 0 0 0 0 0 0 0 —2.4 0 0 0 0 0 0 0 0 0 0 o —2.0 0 0 0 o 0 0 0 0 0 0 0 gr - 1 . 6 0 0 o 0 0 0 o .1667 0 0 0 g -1.2 .2500 0 0 0 0 0 0 .1250 0 0 0 E -0.8 .1553 .0667 .0667 0 0 .0667 .0667 .0667 0 0 .0666 § -0.4 .1765 .0588 0 .0588 .0588 .0588 .0588 .0588 .0588 0 0588 E 0.0 .1111 .0556 .1111 .0556 .0556 .0556 .1111 .1111 .0555 0 0 2 0.4 .4138 .1034 .0690 .0345 .0545 .2069 .0545 0 .0545 0345 0 g 0.8 .2778 .1667 .1111 .1111 .0556 .0556 .1666 0 .0556 0 0 g 1.2 .5585 0 .2308 0 .2508 0 0 0 0 0 0 :3 1.6 .2000 .4000 .4000 0 o 0 0 0 0 0 o 2.0 .5000 .5000 0 o 0 0 0 0 0 0 0 2 . 4 1.0000 0 0 0 0 0 0 0 0 0 0 2.8 0 0 0 0 0 0 0 o 0 0 0 a 220 240 260 500 400 500 600 700 800 1,500 1“” 240 260 300 400 500 600 700 800 1,500 1,800 mi ,1. a 230 250 280 350 450 550 650 750 1,050 1,550 -2.8 0 0 0 0 0 0 0 0 0 o 0 -2.4 0 0 0 0 0 0 0 0 0 0 0 —2 . 0 0 0 0 0 0 0 0 0 0 1.0000 2 $4 .1.6 0 0 0 .1667 0 0 .1667 .1667 .1666 .1666 6 =- —1.2 0 0 .1250 0 .1250 0 0 .1250 .1250 .1250 8 .5 08 .0666 0 .0666 .1353 .0667 .0667 .0667 0 o o 15 g -0.4 0 .0588 .1176 .0588 .0588 0 .0588 0 0 0 17 75 0.0 0 .1111 .1111 0 0 .0555 0 0 0 0 18 E 0.4 .0545 0 0 0 0 0 0 o 0 0 29 E 0.8 0 0 0 0 0 0 0 0 0 0 18 g 1 . 2 0 0 0 0 0 0 o 0 0 0 13 g 1.6 0 0 0 0 0 0 o 0 o 0 5 2.0 0 0 0 0 0 0 0 0 0 0 2 2.4 0 0 0 0 0 0 0 0 0 0 1 2.8 0 0 0 0 0 0 0 0 0 0 0 Note: 0 m. n ' ‘ s. 1 , Ia*1, and ta are 1n mlnute . is the total number of bed forms contained in the yx(t) record and which also contain both an up—crossing and a down-crossing at the elevation yi. SUPPLEMENTAL DATA TABLES 161116 5. Sample conditional probability mass fmction of neat periods, pTIYDUuIyi) (Run 17) 1a 0 10 20 30 40 50 60 70 80 90 100 110 120 Ta” 10 20 30 40 so 60 70 80 90 100 110 120 130 ta 5 15 25 35 45 55 65 7s 85 95 105 115 125 -2.8 0 o 0 0 0 0 0 0 0 0 0 0 0 -2.4 0 0 0 0 0 0 0 0 0 0 0 0 0 -2.0 0 0 0 0 .1000 .1000 o 0 0 0 0 .1000 .1000 g» -1.6 .0322 .0322 .0322 .1290 .0968 .0322 0 .0322 0 .0323 .0323 .0968 o E -1.2 .0328 .0164 .1311 .0984 .1639 .0492 .1639 .0656 .0328 .0328 .0328 .0164 .0164 E -0.8 .0652 .0652 .1630 .1522 .1739 .1304 .0978 .0326 .0217 0 .0109 .0217 0 g -0.4 .0385 .1442 .2404 .2115 .1250 .1058 .0288 .0385 .0096 .0192 .0096 0 0 .1 0.0 .0789 .2281 .2544 .2105 .1228 .0351 .0526 0 0 .0088 .0088 0 0 E 0.4 .1892 .3063 .2703 .1261 .0631 .0270 .0180 0 0 o 0 0 0 E 0.8 .2625 .3250 .2500 .0875 .0500 .0250 0 0 0 0 0 0 0 $5 1.2 .4286 .2857 .1786 .0893 .0179 0 0 0 o 0 0 0 0 g 1.6 .6250 .2500 .1250 0 0 0 0 0 0 0 0 0 0 2.0 .7273 .2727 0 0 0 0 0 0 0 0 o 0 0 2.4 1.0000 0 0 0 0 0 0 o 0 0 0 0 0 2.8 0 0 0 0 0 0 o 0 0 0 0 0 0 1a 130 140 150 200 250 300 400 500 1,000 2,000 3,000 m1 140 150 200 250 300 400 500 1,000 2,000 3,000 4,000 mi”: ta 135 145 175 225 275 350 450 750 1,500 2,500 3,500 -Z.8 0 0 0 0 0 0 0 0 0 0 0 0 —2.4 0 .3333 0 .3334 0 0 0 0 0 0 .3333 3 -2.0 .1000 0 0 0 .1000 0 .1000 .1000 .1000 .1000 0 10 5',“ -1.6 .0322 .0322 .0322 .1613 .0322 .0322 .0645 .0645 0 0 0 31 =- -1.2 .0164 0 .0328 .0328 0 .0328 .0164 .0164 0 0 0 61 E -0.8 .0109 .0109 .0217 0 .0109 .0109 0 0 0 0 0 92 g 04 0 0 0 0 .0096 0 0 0 0 0 0 104 '5 0.0 0 0 0 0 0 0 0 0 0 0 0 114 E 0.4 0 0 0 0 0 0 0 0 0 0 0 111 E 0.8 0 0 0 0 0 0 0 0 0 0 0 80 E 1.2 0 0 0 0 0 0 0 0 0 0 0 56 g 1.6 0 0 0 0 0 0 0 0 0 0 0 32 2.0 0 0 0 0 0 0 o 0 0 0 0 11 2.4 0 0 0 0 0 0 0 0 0 0 0 3 2.8 0 0 0 0 0 0 0 0 0 0 0 0 Note: Ta’ 1091' and ta are in minutes. m. is the total number of bed forms contained in the 513(8) record and which also contain ‘1- i ’ both an up-crossing and a down—crossing at the elevatxon yi. 47 48 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED Table 6. Variation of conditional mean and variance of rest periods 1012f]: elevation of deposition; Emlyn-y] and Var[TlJ’D-y] A . A 4 Stagrigxgiiiid Emlyn-511.], 111171 Var[TIYD=yi], 11an yo 1' Run 411 Run 16 Run 17 Run 4A Run 16 Run 17 -2.8 ------------------------------------------------ -2.4 1,400.8 ------- 1,443.7 4,179,378 ------- 4,671,616 -2.0 252.2 1,550.3 610.6 109,148 84,679 762,079 -l.6 120.5 714.8 198.0 19,552 204,840 74,793 -1.2 83.4 505.6 92.5 3,696 246,658 14,193 -0.8 58.2 230.9 53.5 3,529 39,416 2,548 -0.4 40.2 185.8 40.4 1,288 28,686 1,129 0.0 27.4 151.7 29.8 409 16,715 316 0.4 21.0 64.5 22.3 193 4,030 188 0.8 16.7 61.5 19.0 127 2,691 145 1.2 13.9 36.0 14.0 82 1,039 107 1.6 11.4 35.6 9.4 49 351 50 2.0 3.5 19.5 7.5 30 2 18 2.4 7 1 6.9 2.6 29 2.3 ------------------------------ Em 53.3 180.5 60.1 --------------------------- V317] --------------------- 42,110 93,077 60,469 m. . 2,167 134 703 2,167 134 703 . 1.,1. Table 7. Estimates of parameters and the results of goodness of fit test for the conditional rest periods (two-parameter gm) Run 4A Run 16 Run 17 Elevation k l/ Goodness k Goodness k Goodness y- 2.17 .2_/ 2.0 2,y 1. r m. 11 of _3/ r2 i i of r2 mi 15 of min‘1 2"" 1” Fit Test min“ .y ’ Fit Test min'1 .14 ' Fit Test —2.4 0.0003 0.470 11 ----------------- 0 ------------------ 3 ....... -2.0 .0023 .583 67 ————————————————— 2 ------- 0.0003 0.439 10 ....... -1.6 .006 .742 137 :2 > an: 0.003 2.494 6 ------- .003 .524 31 x2 < I: -1.2 .010 .300 194 x1 > z: .002 1.036 s .007 .604 61 x2 > as: -o.s .016 .959 253 22 < I: .006 1.353 15 ------- .021 1.125 92 x2 > as: -0.4 .031 1.255 314 0:2 > x: .006 1.204 17 1:2 < z: .036 1.444 104 22 > x: 0.0 .067 1.835 361 x2 < 2:: .009 1.376 18 :2 < I: .094 2.311 114 22 < as: 0.4 .109 2.276 338 x2 < x; .016 1.031 29 :2 < z: .113 2.638 111 32 < an; 0.8 .131 2.193 248 3:2 < 2:: .023 1.407 18 :2 < z: .131 2.488 80 :2 < as: 1.2 .157 2.361 150 x2 < x: .035 1.247 13 ------- .131 1.327 56 :2 < an: 1.6 .233 2.680 72 3:2 < x: .101 3.611 s ------- .186 1.747 32 22 < an: 2.0 .283 2.408 19 ----------------- 2 .417 3.125 11 ------ 2.4 ----------- 3 ----------------- 1 ------------------ 3 ....... A 2 A 2 3/ l/ E[T| YD=y] _/ ([[lep=y]) ._ 2 . . ' k - 7—— r I —- z = critical chx-squlre value at 2.51 VariTIYD=yl Z-y firITlny] c a significant level of 0.05 SUPPLEMENTAL DATA TABLES Table 8. 51mph joint probability ma.“ flotation of rest periods and elevation of dapoaitian, pl. 1' (tr-V12) (Run 4A) ' D 'ru 0 10 20 30 40 50 60 70 BO 90 100 110 120 130 140 ‘1’“.1 10 20 30 40 50 60 70 30 90 100 110 1 20 130 140 150 ta 5 15 25 35 45 55 65 75 85 95 105 115 125 135 145 -2 . 8 0 0 0 0 0 0 0 0 >0 0 0 0 0 0 0 -2 . 4 0 0 0 0 0 0 0 0 0 0 0 0 0 .0005 0 -2.0 .0011 .0019 .0027 .0023 .0011 .0008 .0004 .0008 .0008 .0004 0 .0008 .0004 .0004 .0004 - 1 . 6 . 0025 . 00 75 . 0062 . 0040 . 0040 . 0066 . 0026 . 0009 . 0022 . 0013 . 0009 . 0009 . 0004 . 001 3 . 0013 —1.2 .0093 .0131 .0107 .0065 .0089 .0061 .0037 .0023 .0033 .0009 .0014 .0005 .0028 .0019 .0033 -0.8 .0114 .0253 .0158 .0124 .0104 .0124 .0030 .0040 .0040 .0030 .0035 .0030 .0025 .0020 .0020 —0.4 .0155 .0338 0203 .0266 .0179 .0097 .0072 .0053 .0034 .0024 .0019 .0014 .0024 .0010 .0005 ‘L Standardized elevation, y. 0.0 .0253 .0440 .0365 .0222 .0129 .0080 .0058 .0018 .0013 .0009 0 .0004 .0013 0 0 0.4 .0341 .0567 .0318 .0161 .0111 .0023 .0028 .0009 0 0 0 0 0 0 0 0.8 .0377 .0473 .0203 .0092 .0024 .0029 0 0 0 0 0 0 0 0 0 1.2 .0286 .0248 .0131 .0033 .0005 0 0 0 0 0 0 0 0 0 0 1.6 .0127 .0093 .0021 .0007 0 0 0 O 0 0 0 0 0 0 0 2 .0 .0064 .0018 .0004 0 0 0 O 0 0 0 0 0 0 0 0 2.4 .0005 .0003 0 0 0 0 0 0 0 0 0 0 0 0 0 2.8 0 O 0 0 0 O 0 0 0 0 0 0 0 0 0 PT(tu) .1857 .2658 .1600 .1033 .0691 .0487 .0255 .0160 .0150 .0089 .0077 .0070 .0098 .0070 .0074 In 150 150 170 130 190 200 250 300 550 400 500 600 1,000 2,000 1” 1 150 1 70 130 190 200 250 500 350 400 500 600 1,000 2 . 000 s ,000 ta 155 165 175 185 195 225 275 325 575 450 550 300 1.500 5.000 .2. s 0 0 o 0 0 o o 0 0 o 0 0 0 0 -2 . 4 0 o 0 0 o .0015 .0005 o 0 .0005 0 .0005 .0005 .0015 -2.o o o .0004 .0011 0 .0019 .0015 .0004 .0008 .0005 .0011 .0025 .0011 o 33.. —1.6 .0026 .0015 .0009 .0004 .0013 .0015 .0015 .0026 .0013 .0010 .0015 .0009 0 o =- —1.2 .0025 .0005 .0005 .0009 .0014 .0020 .0045 .0014 .0005 .0014 o o 0 o g -0.s .0020 .0025 .0010 .0010 .0005 .0025 .0010 0 0 0 0 0 0 o g -o . 4 .0005 . 0005 .0005 o o . 0005 .0005 o 0 . 0005 o 0 0 0 '0‘ 0.0 o 0 o o 0 0 0 o o 0 0 0 o 0 E 0.4 0 0 0 o o 0 0 0 o 0 0 0 0 0 E 0. a 0 o 0 0 o 0 o o 0 0 o 0 o o g 1.2 0 0 o 0 0 0 o 0 o o 0 0 o o g 1.5 o o 0 0 o 0 o o o o o o o 0 2.0 o 0 0 0 o o 0 o o o o 0 o o 2.4 o 0 o o 0 o 0 0 o o 0 0 0 o 2.8 0 0 0 0 o 0 o 0 o o 0 0 0 0 pT(tu) .0074 .0048 .0032 .0035 .0032 .0105 .0090 .0044 .0026 .0049 .0024 .0037 .0017 .0015 Note: 1 Tml' and to: are m mlnutes. (1’ 50 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED u’ 1691’ and to. are in minutes. Table 9. Sarrpla joint probability mass function of rest periods and elevation of deposition, PT'YD(ta.yi) (Run 16) Ta 0 20 40 60 80 100 120 140 160 180 200 a” 20 40 60 80 100 120 140 160 180 200 220 ta 110 30 50 70 90 * 110 130 150 170 190 210 -Z.8 0 0 0 0 0 0 0 0 0 0 0 -Z.4 0 0 0 0 0 0 0 0 0 0 0 —2.0 0 0 0 0 0 0 0 0 0 0 0 3',“ -1.6 0 0 0 0 0 0 0 .0063 0 0 0 E —1.2 .0151 0 0 0 0 0 0 .0075 0 0 0 ‘3 -0.8 .0138 .0069 .0069 0 0 .0069 .0069 .0069 0 0 0069 g -0.4 .0216 .0072 0 .0072 .0072 .0072 .0072 .0072 .0072 0 0072 I; 0.0 .0148 .0074 .0148 .0074 .0074 .0074 .0148 .0148 .0074 0 0 E 0.4 .0817 .0204 .0136 .0068 .0068 .0409 .0068 0 .0068 .0068 0 E 0.8 .0381 .0229 .0153 .0153 L .0076 .0076 .0229 0 .0076 0 0 E 1.2 .0524 0 .0225 0 .0225 0 0 0 0 0 0 g 1.6 .0093 .0185 .0185 0 0 0 0 0 0 0 0 2.0 .0082 .0082 0 0 0 0 0 0 0 0 0 2.4 .0048 0 0 0 0 0 0 0 0 0 0 2.8 0 0 0 0 0 0 0 0 0 0 0 pTUa) .2598 .0915 .0916 .0367 .0515 .0700 .0586 .0427 .0290 .0068 0141 Ta 220 240 260 300 400 500 600 700 800 1,300 Ta+1 240 260 300 400 500 600 700 800 1, 300 1, 800 ta 230 250 280 350 450 550 650 750 1,050 1,550 -2.8 0 0 0 0 0 0 0 0 0 0 -2.4 0 0 0 0 0 0 0 0 0 0 «2.0 0 -0 0 0 0 0 0 0 0 .0191 $4 —l.6 0 0 0 .0063 0 0 .0063 .0063 .0063 .0063 E -1.2 0 0 .0075 0 .0075 0 0 .0075 .0075 .0075 E -0.8 .0069 0 .0069 .0138 .0069 .0069 .0069 O 0 0 E -0.4 0 .0072 .0144 .0072 .0072 0 .0072 0 0 0 E 0.0 0 .0148 .0148 0 0 .0074 0 0 0 0 E 0.4 .0068 0 0 0 0 0 0 0 0 0 . f: 0.8 0 0 0 0 0 0 0 0 0 0 E 1.2 o o o o o o o o o o g 1.6 0 0 0 0 0 0 0 0 0 0 2.0 0 0 0 0 0 0 0 0 0 0 2.4 0 0 0 0 0 0 0 0 0 0 2.8 0 0 0 0 0 0 0 0 0 0 p7,(tu) .0137 .0220 .0436 .0273 .0216 .0143 .0204 .0138 .0138 .0329 Note: 1' (1’ T a d t are in minut s. 0141’ n 01 e SUPPLEMENTAL DATA TABLES Table 10. Sample joint probability mass function of rest periods and elevation of daposition, PTJDUQ'yi) (Run 17) Ta 0 10 20 50 4o 50 6o 70 30 90 100 110 1m} 10 20 30 4o 50 60 70 30 90 100 110 120 ta 5 15 25 35 45 55 65 75 35 95 105 115 .2.3 0 o 0 0 0 0 o 0 0 0 0 o -2.4 0 o 0 0 o 0 o 0 0 0 o 0 -2.0 0 0 0 0 .0017 .0017 0 0 0 0 0 .0017 $4 -1.6 .0014 .0014 .0014 .0056 .0042 .0014 0 .0014 o .0014 .0014 .0042 a —1.2 .0029 .0015 .0017 .0033 .0146 .0044 .0146 .0053 .0029 .0029 .0029 .0015 E -0.3 .0035 .0035 .0211 .0197 .0226 .0169 .0127 .0042 .0023 0 .0014 .0023 g .0.4 .0059 .0220 .0367 .0323 .0191 .0162 .0044 .0059 .0015 .0029 0 .0029 u 0.0 .0121 .0351 .0592 .0524 .0189 .0054 .0031 0 o .0015 .0015 0 E 0.4 .0294 .0476 .0420 .0196 .0093 .0042 .0025 o o, 0 0 0 E 0.3 .0309 .0352 .0294 .0103 .0059 .0029 o 0 o 0 0 0 ‘1; 1.2 .0512 .0203 .0150 .0065 .0013 0 o o o 0 0 0 g 1.6 .0256 .0102 .0051 0 0 0 0 0 0 0 o 0 2.0 .0157 .0051 0 0 0 0 0 o 0 0 0 0 2.4 .0055 0 0 0 0 0 0 0 0 0 0 0 2.3 0 0 0 0 0 0 0 0 0 0 0 0 pT(tu] .1651 .1904 .1996 .1352 .0931 .0551 .0426 .0173 .0072 .0035 .0070 .0151 Ta 120 150 140 150 200 250 300 400 500 1,000 2.000 3,000 on] 150 140 150 200 250 500 400 500 1,000 2,000 5,000 4,000 ta 125 155 145 175 225 275 550 450 750 1,500 2,500 5,500 .2.3 0 0 0 0 o 0 0 0 0 0 0 0 .2.4 o 0 .0013 0 .0013 0 0 0 0 0 0 .0013 —2.0 .0017 .0017 0 0 0 .0017 0 .0017 .0017 .0017 .0017 0 g -1.6 o .0014 .0014 .0014 .0070 .0014 .0014 .0023 .0023 0 0 0 =~ -1.2 .0015 .0015 0 .0029 .0029 o .0029 .0015 .0015 0 0 . 0 E -o.3 0 .0014 .0014 .0023 o .0014 .0014 0 0 0 0 0 E -0.4 .0015 0 o 0 0 .0015 0 0 0 o o 0 7’ 0.0 0 '0 0 0 o o 0 0 0 0 0 0 E 0.4 0 0 0 0 o 0 o 0 0 o o 0 E 0.8 0 0 o o o 0 o o 0 0 0 o E3 1.2 0 0 0 0 o 0 o 0 o o 0 0 :2 1.6 0 0 0 0 0 0 o o 0 0 0 o 2.0 0 0 0 0 0 0 0 0 0 0 0 0 2.4 0 0 0 0 0 0 0 0 0 0 0 o 2.8 0 0 0 0 0 0 o 0 0 0 o 0 pT(ta) .0047 .0060 .0046 .0071 .0117 .0060 .0057 .0060 .0060 .0017 .0017 .0015 Note: T 51 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED 52 «1 3a now 38 u o» :33 a: mfimnoa no: 3w $53: mmunu 2:. "302 .m :a how uoou _ 3 3.50 0.5 mzuucua noun now “53: mmfiu 2:. "302 .--- 8°. vac. mno. Hue. Nug. Nun. wan. wuo. a Kinny ham hue. mma. wwn. cmv. <00. m Rina lam o o --.-.. N m .3 ...--- 2 -..-.. a 21: mmm~.a mfivo. moao. ouao. vmao. name. mNVA. mcnc. comm. oooo.~ Oman. «no». name. nova. NNOA. Nmmu. omm~. mnom.o omhn. nmvv. mmmn. ongv. mnnn. Hovv. ommm. hmwN. mnmfi. cemn.o oooo.~ 000m. noun. mm¢m. mvwn. «sum. «mam. mmh~. mom~.o QGOc. mmao. ~ono.o v.~ o.N o.H ~.H m.o v.0 c.o v.6. ”.0: ~.~u o.—u o.Nu v.Nn w.~| ~.nu oon::8332:N—eo mono. ammo. mnno. «H—H. wfimn. cooc.o coma. unvn. Anew. wocN. nnnn. onom. cocv. oocm.o vm—o. 00cm. 005V. NucN. eoov. unnu. wuhu. ocow. ooom.o ooco.~ oooo.~ upon. comm. mumn. «new. nnnu. mhnn. mama.o vono.o v.N o.N o.~ N." «.5 v.9 o.o v.0: n.9- N.uu 0.”: a.~. v.u. w.u' N.nu m.m m.n um .3 Jim: mmiu m.o c an Jun: was nub» u o.Nv n . Mafia a . m c: :3: Amemlnv a :5“ £5953 mm: kn ruins...“ anus Eusnfidaoxm Nuxomfimgu 335m. :m you 33 a 3 Have a: 295.3 auum you 233.: mmmfiu 2:. uuuoz .vH ofinmk a . 2v :3: A991“: » :3 guarfl m3» «8 Efieé 32. 3.3832“ N§o£%§ «35m 3. . bra» .m 3m you 38 m 3 1.53 a: “games noun you ESE: mwflu 2:. ”3oz .nn «unak ova. Hm”. awn. fimv. ooo.o a 1.333 15‘ Nmo. oco.a a Kind :xm ooNNMoothmcooc up: ooom.o oocv. Vunm. mnnn. coon. nnnn. 500w. oocm.o oeoo.H 000v. hmwN. nmmm. coon. nmnn.c v.~ o.~ o.~ N.H w.c v.o o.o v.0- v.0: N.«- o.~a o.~u v.Nu w.~u ~.n« oooooauchcoooo coon. ooce.u caoo.H v.~ o.~ w.~ N.a w.o v.0 c.o v.9. m.c. ~.~u n.~u o.- v.uu m.~. N.n. "E .. Ah .\ s' um .nu .xkuz mmaau mm» m.m um .wn st»: mnuS mum» c: :3: E; N. .3. u $2» Q .m m_uuv M »_km .mxum:w~ Amen kc rom¢u=xk wads muwflvaunozm “6:3vumwrou mgnsam .- awash n-. 1km £0”.me m . . . . . . N 8.. .8 §£u§x m3... 33.3583 “96.0%..8 .35 2 3 and. STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED 54 m :m wow uoou H 0» 12:3 0.3 usuacofi noun you £33: main 2:. ”302 .m :m .30 uoom a 3 ”5.50 a: mfiucfl 03m .80 933: 320 2:. "ouoz .... «8. 2°. 8°. SN. 8... «8. $1381? .8. N8. 29. 8a. SN. 2:. EN. Animuqzxn o .-.- .-..-- .... -.-.. ----. .--..- --.-.- ...N o ..-.. --.... ---.-. . ..... --..-. .-..-- . ..... -..... v.“ n. ...... ...... -..... --.... .--... ...... 83. SR. 9” q ....-. ...... ...... -..... .-...- -..I. 83. 82. ad 8 .....- -...-- .-.--- -..-.. ...-.. $8. 83. 5%. o; 3. ...... ...... ...... ..-... .....- 3.3. $8. 88. c; an .-.... -..-.. $8. 38. $8. 3.3. N; N...” ..... . -..... -. ..- 2.-.- $8. 83. $2. 83. N; a: ...... ...... -..--. -..-.. :8. 28. 8%.. 29.. wé a: ...... .................. 38. 83. $3. 3.8. 80 ohm . uuuuuuuuuuu nuuu: £00. «£0. QVNN. $00. vona. v.0 2m uuuuuuuuuuuuuuuuuu mmoo. mawo. oooH. anm. ooNN. «.0 nmm ..... n IIInIo 0.30. 020. ammo. 00mm. hmnm. mmmc. 0.0 woe :1-.- nnnnnn HN00. :ao. «000. N03. ommv. One. 0.0 NAN u-|.|u 13..-: v02... HNmm. Nmmv. mumn. wn—a. v.0: Ham uuuuuuuuuuuu nu:n. m0N0. mmNH. ova. New»: mmvo. v.0. .2: nnnnnnnnnnnnnnnnnn name. GuNN. owvv. wNmN. mm00. w.0| new uuuuuuuuuuuu mooo. wac. 0NON. wmwv. NmmN. mwoo. 0.0. mnH nnnnnnnnnnnn Qumc. Whmo. OQVN. who? owom. N500. N41 mNN uuuuuuuuuuuu ”5:0. NNOO. vvvn. 3.3... thN. 3.00.0 NA. mm uuuuuu n 11111 3H0. mNNO. mmmm. NNNm. «own. «:00 0.7 B2 ............ mnoo. :mo. hcmu. mmNm. «and. o.”- Hm uuuuuuuuuuuu 020. ammo. 5.4». 000mm. 0wm0.0 ...... 0.~- om ............ 0mg. 002. 00mm. mun... mnmo. ...... 0.Nu vN I uuuuuuuuuuu NHV0.0 0mNH.0 omhn. mmmv. nnnnnnnnnnnn v.~: on 111111 whN0.0 wnN0.0 mme. mmwm. 000m. ammo. nnnnnn in: 0 IIIIII 1.3-1: ::::::::::: a 000m. 000m. llllll I‘Ilnn w.~: .: uuuuuuuuuuuuuuuuuu ”Hm—.0 Som.0 onemé. 0000.0 uncut: m.~. N .................. 000m.0 000m.0 ...... N4»... N ........................ 00004 ~47 .0.qu mK m.o m.m m.v min m.N m4 m.0 WJE mK vie rim m.v m.m m.~ m4 m.0 u» LN» um .mH Jim: was aw .3“ Jim: mums n N; u bra» as . nmufi 2w 5.5 nhmfipm—muunx. 10$ aufimxg mu: k0 08305..“ 3!: 3336095 ngufiufirg 395m. .NN 023. $6 :3: Athm—méax. :5» {£953 mmuw Kc Eon—3.8% was $335093 “Sufism—Eu 335% .3 3an m :m how 908 0 3 12.30 0.3 mgumcoa nuum now £33: 320 of. "302 . . . . . . . . .N. c nix .... 8o :8 N8 8° 93 SN 3. N: h 2 3 m a ---.-. .-.--. --.--. --.--. .--.-- ---.-. ...--. .-..-- ...N v -..-.. ----.. ...-.- -..-.- --.... ..---- 88. 88. o." 9 .-.-: 8:. :3. 83. 04 EN .-...- ..--.- --.-.. 38. 28. 3.8. 82. .5... .NJ 3. ...... -----. ..-.-. 88. ES. 82. 33. 8:. ma 8“ ............ :8. :8. :8. 8:. $3. 8:. ...o .80 ...... S8. ...... 88. $8. :2. 8:. 32. ed a: --.-.. 28. $8. $8. 3:. 38. 3:: 88. «a. 8.. ...... :8. 38. :3. :8. 22.. KS. 28. ms. 38 £8... 88. 88. RR. 33.. 8:. 28. 2. m2 ..... . --.-.- $8. 28. 8:. SS. 3:. 38. e4. 8 ., ..... --.--- 28. 3.2. :2. SS. 38. :88 97 3 ..... - 38.0 83.9 3:. :3: $3.. :3. ti: TN. 2 .--.-- 826 2.8. 38.0 885 --.... «A. N ....................... . c884 ...... ----.- ...... N5. .54: mg mg m.m m; 2 mg m; We ”Sufi uw .9». Jim: mums Q N 2.6 :3: an». 30.080 a 10$ 6.00930 mmum kc tomuurxk was EmNU—fiaoam “6:333:00 mNAEm. .8 ~32. «.0 n wmuq .m :u .30 uoow H 3 :33 0.3 Eta—.3 Auum how £332 mun? 2:. Nunez .... . . . . . . . . .5 m mix 28 8 8o :0 N8 ”3 m2. 8 n 3 5 n o ...... ...... .1... .2... ...... .-...- ...... ..-... ...N v --..-. ........... . ---.-. .-...- 88. 88. c.~ an ................. . ...... 3.8. $8. 8:. $8. a; hmfi :- uu. . nmoc. Huno. 05:. 0va. mmmn. N4 nov uuuuuuuuuuu n mN00. 3.00. mmmo. manm. mnnm. nmmm. m.0 one uuuuuuuuuuuu mmoo. wmoo. ammo. mmON. vwmw. omMN. «.0 mm» uuuuuu wmeo. waoo. ~20. wvoc. mmmu. 23.. wvmm. 0.0 moo u: .- 580. 300. ammo. Om~H. 05:... “H0... 0500. v.0. nmv ~N00. uuuuuu 0000. 035. SAN. omen. A03. 030. 0.0: oaN 3.00.0 ...... 00.8. 0500. Raw. oqu. 0mg. ”000. N4. 50.. uuuuuuuuuuuu om~0. awnu. «NAM. HmNQ. mafia. 0000. 04: mm |||||||||||| manna. mnma. ”mom. nman. :01 N206 0.N- 0m nnnnnn 0m~0.0 mono. ~m0N. nonw. mean. memo. in-.. v.~u 3 ............ :36 32.0 $3. 83... :38 ...... w.~. N ...... ...--. .--..- ...... 884 ............ fin. .thE mK m6 m.m m4 m.m m.~ m4 m0 3TH» aw .mH Jim: was 0.0 n wan 3v :3: flnmLm—mfivqx. :xm ..mfimrmN mmum k0 gsuux‘fi. wads 3336020 Nurofiduxoo 305% .2 as: 55 SUPPLEMENTAL DATA TABLES .8 Zn you you... H 0» 3:3 PE gum—.2 noum How mzuvmz mmfiu 9E. nuuoz -a you Noam A 0» “gave ohm “gauze“ moum you msuuwz mma~u oak "ouaz a n N. N. -... ...... ...... m8. £8. 8,“. N8. mNH. ...... H.232 :xm .... ........... . ...... -...-. NmN. 23. m2. --..-- éfiué 1: o ............ . ................................... o.m o ........... . .................. ---... ---... ...... c.m e ...... ...... ...... .-.... ..-..- ........... . ...... NA o ...... ...... ...... . ........... ...-.. ...... -..... Nd o ............ .----. - ................. . ........... m.~ c ..... - ...... ..-.-. ---... ...... ....-. ------ a.~ o -.-... ........................ ...--- - ---... v.~ o .--.-. .-.--- -.-... -.-.-. ...... ----.. - ..... v.~ o .................. ...... .....- ................. . o.N o ...... -..... ...... ........... . --.... --..-. ....-- 9N a ...... ........................ --.... ........... - o; o .................. --.... ............ -..... . ..... o.— H ............ -.-..- .................. 88; ...... N; A .................. ...... ...... 884 ...... N; N ...... . .... . .... ...... 83. 88. ............ .8 H .......................................... a... N ...... .... . ...... 88. -...-. 88. ...... -...-. ...o H ............ .... ........................ ...o N ............ ...... 88.0 ...... 88.8 ............ a... N .............................. 88; ............ .8 N ............ 88... ...... 88. .................. To. N ........................ 884 .................. .....- A ........................ 884 . .... ............ ma. N ...... . .... ............ 88; .................. m6. 0 ................................................ N4. 8 .................. . ... ........... . N4. 0 ................................................ w.~. o ........................ .- -. ............ .. --- 0."- o ................................................ o.N. o ......................................... - ...... o.~- o .......................................... «.m. o ................................................ v.~- o .......................................... w.~- o .................. - .................. w.~. o .............................. -....- ...... Na. o .......................................... Na- 8 .......................................... ....-. ....m. o .......................................... 9n. “:8 mK m6 m.m mé m.m m.~ m4 m.o W 5.9 mé. m6 m.m m.v m.m m.~ m4 m6 . .E u » . .5 am .mH Jim: mans um .mH .45: mmaG N. N. N.m- n .mnfih 9m. u .MHQ» N H 8 8N8»: N. s m afimzx 3” :35 ASE».— Hv a 333:3 A»: kc ton—unfit. anus 53332“ Nuromumuroo 393m .3 392. ”3 =55 TAT».— 5 n .mxumrfi mmum Kn 53.33% was 3938593 8.63%:8 386% .mm 392. .m Ada new uoom N cu finsvc mun mcumcoa noun yew mauvwx unmau ugh Nunez Nan you uoow N 0» .asco mum mzuuco~ moum how mzuvwz mmu_u och ”ouoz N. m nix N. a mix -... ...-.. ...-.. 8°. 3;. 81 N2. Em. NZ. :2 .3 m ..-. ...... ...... 88. N8. m8. 3:. man. 2N. ".3 a m o -..... . ..... -...-. ..--.. . ..... ...... .-..-. ..... - ...N o ..... . ...... .....- .--..- ...-.. --..-. ...... ...... TN .. -..-.. - ..... --.... ...-.. ..-..- ...... 83. 8?. o.N v ...... -.--.. ..... . ............ -.-... 8:. 8r. 9N v ...... .--..- ...... -.-... ...... 8mN. ..... - 8n. a.“ 3 ....-. ... - ...... ....-. $8. 82. GS. ...; a --..-. ...... -...-. ..---. 8mN. ..--.- 83. 886 N; no ............ -..--. - ..... SS. 38. :8. #5. N; v .-..-- ............ -..-.. 8mN. -..--. 82. ...... ”.0 No ........................ 88. 33. Sam. 88. 9o v ............ .--.-- -..... 8mN. -...-. oomN. --..-. «.o 3 .................. ---... .28. 83. 3mm. 38. To v ...... ...-.. ..... - 8mN. ...--- 8mN. 8cm. ...... ed 8 .................. $8. $8. 3e... 89.. 38. ed c ...... c ........... oomN. ...... comm. ooomd ...... To. mm .................. came. momfi. 3.3. mmmn. $8.0 v.9. v .................. 8mN. ...... 8mN. ............ m8. 8 .................. 88. 88. 83. ...... 9... v .................. 83. 8%. ...... .... N4. 3. ............ 28. $8. 82. 25.. ...... N4. v ............ -..... 8mN.o 88. 888 ............ 0.7 m... ............ 88. 88. 88. 3.3.. ...... .2. N ............ 88m. ...... 83.8 .................. 9N- ON ............ 88. ---... 83. 83. ...... 9N. H ............ 884 .................. .-.... ...... TN. 2 ............ $36 326 82. 28. ...... ...N- o . ............................................... m.N. q ........................ 8&6 8K6 ...... ”N. o ................................................ N.T o ................................................ N5. m.“ m5 m.m m.v m.m m.N m4 m6 . m4. m6 m.m m.v m.n m.~ m.~ m.o hum» w ms mmuma um .uH .xun: nmufiu um .uH .xnn: mmagu a. o.N . wane» o; . .mufi a . n a . 2v :3: 1. 3.393 N :5“ £583 nut. .8 $323..5 ”was 383398 NESEqu £55m ...N ~32 2:. 5;: ”8:533 N NE .3383 nut .3 gurus... was fixfiggm 8838.50 ~35“. .nN ~35 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED 56 .u add ROM MOON H Du «mu—V0 uhu “Saws—UH autum HOW mLUflMI mmNHU Or:- uuuoz .m -N ...cw 900w .— 09 ~m=¢0 0km m£umch nwuw how miuvwi mmd~0 ”Fr ”0902 u. m a»: a a mix .-.- :6. 2o. So. «2. a. 2m. 8N. NS. ”.3 be m -- 8o. 20. m8. m2. ”8. 8m. 31 8o. ".3 a n o .--- ...... -.--- - ..... ii: 9.» o --.-- - --- ..--. ---- -.--- ---- ...... 9n o --.- 2-. . --.- NA o - --:- ---.- ...... --.- ---- ~.m o - -- -.-- ----- - --- ad 0 ----. ----.- .---- ........................ --.-- .2 m ----- ...... - ..... 884 ---.- YN o ............ ----- ---. .---- -----. .---- ...... v.~ m ...... ---. 83. 83. ...... o.~ H .----- .-.--- -.-.- - ........... 88; ---- o." a ...... ---. 8:. $3. 38. o; 2 --.- --.-. ---.-- - ..... 88. 82. ---- a; mo --.- - .- :- 38. 3:. ms... 83. «A ”N --: ---- -.-- :2. 33. 2%. 33. ~._ «2 ..... - 1-.-- $8. $8. 32. 33. $3. 23. m... S I...- --.- ---- SS. ”2:. «8... SS: 33. m... E ---- ------ $8. $3. $3. 23.. :2. «2o. ...o i ............ --.-- 38. 5:. EN... :3. ---- Yo a: ...... 38. :8. $2. $2. $2. $2. 88. 0.0 mm ----- - :1: 33. 22. 32. 33. $2. ”:3. o6 :: ------ 28. $8. 38. 3:. 23. $86 ...... ...c- 5 28. :3. 2:. 3%. :2. $85 ...... fo- mmH ------ $8. 3:. 3:. 32. 83. ...... 38.0 ”5. $ ------ £3. 3:. «SN. 33.. 83. ...... - ---- ma- wS ...... 38. SE. SS. 33, 22. it: ------ N;- 3 ---- $8. 8:. can. So... 83. ............ «.7 3 SS. «moo. 33. SS. 3%. $3. ............ o;- S 88. 38. :2. Son. :5. $8. ............ o;- 3 - $8. 3%. $8. $2. $85 --..-- c.~- fi ----- S3. 2:. 2.8. 3:. 385 ----- ...... O.”- S 835 --. 835 83.. 88. ------ -.--- TN- 2 $35 ...... 83.0 RS: 33: ------ ----. ------ v."- .. ------ 83.0 ------ 886 8%. --.- ---. .--- M:- N ------------ --.- 88.0 88. --- ---.- -.--- mg- H ------------ --.--- -- 88; 1-.-- ------------ fin- H ----- --- ----- - --.- 88; ----------- - ----- . N5- _ --- ------------------ 884 --.. -..--- ----- 9m- — ------ ------------------ 88; ------------------ 9m- m: m.“ m6 m.m m; ma Wu m4 We m; We m.m m... m.n m.~ m4 m6 . .s c as .iu: 3.8 a .8” Jun: mmfiu N. 0.7 u a.- u sun» . .5. ..s m a». a: . . .N. A. m 31»: . . C: :3: A m 3 3 a 533:3 m3» «8 Sings.“ 3e: mafia-£32m 35.5% 335m on 3.5 3H :3: A m 3 a m 3&5“ m3» «6 “5.3.33. 32. Rwfisfifiu 3.83.6.6“. Susan 2 02:. .m :m you Sam A 3 12.50 yum mfiucfl noun how mfg: mmflu 2:. ”302 .u :n you use a 3 =33 PS 939:: noum how 2:5: 320 05 "uuoz n. a a a ..8. us. So. ~:. 03. an. 31 03. "if inn :1 :5. m8. :6. ..2. 2m. 9:. 31 «8. wins :5» o ------------ -.-. ----- - -- - o.» o .-..-. .- .-..- --- --- ------ -.-- 9m o .--- -.--. .-.-.. ~.n o ------ ..-- ----- - ------ . .. . ------ --.- ~.m o -..-- ..- - .- - m.~ o :- -1... .- . ...-. -..- -.-- ..-.. M: o ----- - ...-- -.-- --.. .-.- v.~ o .-..-. -.--. -..-- ------------- ..-..- -..-. ---. ...N A .--.- -.-.. ...... .-...- ..-.. --..- 88; ..... - c.~ a --.. .--- --.- .---. --- o.~ n .--.. -...- --.. --..- -.-... $8. 3.9.. o." o :1: -.. - .-:. 1-... -. -- o.— h -1-.- -. .. -.-. :2. 53. N; m ------ -.-.- .-....- ----------- - 32.: N; 2 ..-..- 2m. 39.. SS. m5 5 ------------ . ..-. ...... Ram. 33. .-..-. a... 2 --.- ------ 3:. 3w... 3:. ...o m ...... 1:: 2-... :2. 28. 8mm. -..-- «to S ------ .--.. 22. 23. 33. 3:. 82. --.- ed a ----- - ------ 2:. :2. 32. 3mm. ...... ....-. ed on --- ------ $2. $2. 88. :3. 226 -.--- To- m ------ --- --5 -.. - 33. ”~26 --- ------ Y? a $3. 38. 3mm. 35. 3:6 -.-... --- as- w ------ 82. ----- r 83. SS. ------------ 1-... m5. 2 ----- - :2... 2:. 33. 3mm. .--. ------ --- ~4- n --- 2:6 ...... Sfi. :5: ------------ --- ~4- 2 835 ----- - 83. 88. 88. --- -.-- ------ 0.7 o 33.9 ----------- - 98m. 33. ...... -.-- --... 0.7 n ............ 53. $2. 8%. -. -.. --.- 9? v - -.- ------ comb. 82.0 ------ -.-- o.- v ------ --- 835 Son. 835 ------------ TN. n ----------- - 88; ------ .--- -.-.- ...... TN- ~ ------ -...- 884 ------------ -.-- m.- ~ . ----------------- 88; ------------------ -.-.- ”4- o ...... --- ------------ - ----- --.- ------ N6. o --.- ----------- - --- ----------------------- - «.m- o ...... -.-.- -.-- ------------ 1-... .2... ------ .3- o ---. ------------ ---. -..-- ------ --.-. ------ cg..- mé. m.o m.m mé m.n m.~ m4 m6 . mK mic m.m mé m.n m.N m4 m5 3.: -m» .u ”.2 Era» a As is: $29 a sun :2“: 32o YT - Wham» m.- u mun» a m n . T: :3: nmerA—muv N :xm «gum—x3 nmuw a? EBQSC. has 33393:“ NUSmuEEg 395% .3 «.33. SM :35 Awfwimé N 2km 5.39.3 ARE k0 $350.5.“ was: hum—:fiaofim Nanci—«v.89 35% RN 03a... 57 SUPPLEMENTAL DATA TABLES .u :a «C uoou — On 3:00 0.3 239.3 noum how 25:: mmwau 2:. "302 .m :a you uoou — cu Esau oua mfiucfi 03m now 933: mun? 2:. "302 N. a . . a .... :8. m8. .3. :8. SN. SN. :3. :3. 0330 .2: .... NS. So. “no. NZ. mNN. mg. RN. So. 1.3g 1; 0 .----- --- -- .--... ...... ed 0 -.-- - ...... . . .----.. ed ~ .----.. -..--. - - - 00004 N6 a .----. ............ .- -- 00004 N.n .— ---... ----.. ------ --..-- ------ -..--- 0000." -----. n.~ fl uuuuuuuuuuuu --.--- .-.--- 0000." --.--. a.” 0.— .----- .----. --.--. ------ .-.--. .....- 000m. 000m. v.~ m - ||||| .----. --.--. -.-..-- ..--- .-.--. 000m. 3:. ...N BN ---.-. ------ :2. none. NNNN. 0.N ow ...... .-..-. ------ .--..- .--.-. vmz. ~03. 3:. 0.~ an .-- . mama. 03m. Ewa. a; 0n --- .-.-.. ..-..- -----. :fi. swam. :mN. o.— m: ------ vunc. vnnm. 30m. 05%. «J 0: ............ :mc. 90:. 2mm. .33. N.“ 2,-- --- --- --.--- none. 33. 3”“... 33... 3:. m6 Rm ----- . ------ 3.8. $8. :3. mean. 2:. wowo. ”.0 8v ............ 5100. neg. mug. Amen. «Ann. unno. v.0 can ............ ch00. nmwo. mama. 30¢. ”awn. onmc. ¢.0 Na? :00. «0:0. 030. 0NNN. mmmn. BOVN. :00. 0.0 N: ...... nN00. 030. mnoo. m:N. 0:“... eku. 0~m0. 0.0 mnv .-.--- muco. nth. 03:. mnNN. onhn. mood. «0—0. v.0- N: 3.00. nnvo. :2. «.33. Don. mmmq. 0:0. v.0. mom 5N00. $00. 05.0. N05. mug. m0~n. n08. “N000 0.0- 3% 5:0. :00. ova. mnmm. omnm. :vo. «~00.0 0.0- mww ..... .- cos. . 3:. 33. 94.3.. 5:. 008. ------ NJ. 03 ~30. «00—. 3.3. mag. awn-4. «30.0 ...... ~4- sa 300.0 33. ~03. mnmn. Nam. 22. 300.0 ------ 0.7 v: :m0. 502. 3:”. mean. Nmoc. ............ 0.7 an --- 0:0” 8:. vnmv. EN». mmmo. ---- ---- o.- S 386 :8. 3.2. new... $3. 23. - ----- o.- a ---. 5.3 o 3:. 2mm. :3. ZS. --- ------ in- w~ ------ 23.0 8:. :2. 3%. 53.0 ------------ Yu- m --.... ...... 32... 33... 3.3. 9.2... -.-.- ------ «.7 w ............ SS... 83.. 8:. .................. m."- m ------ --- --.- ..---- 884 -- --- Na- m ----------------- - 3.3.0 33. ------ --- --- ~6- ~ ...... -----. ----- - ..---- 00004 -----. o.n- H .............................. o.m- . mK m6 m.m m.v m.m m.N m4 m.0 - . . . . . h a m h m o m m m g m 0 . .5 an» whss mum». a .f :5: .38 N. 9° - .m-n» Yo- - mum» Q .N. ... m . a 0“ a n3 :3: n m 0— H0 a _ a .9393“ mm: 0.0 5305C. and... Rmuvauafinm ~§U5mv§ 335m .5 035... A3 :3: fime‘m—muv a. 20$ Jig» mm: k0 5323.» mug mumsmfivnohm Nugmuufig 395m. .8 032. .m :m you uoom .— 3 :53 .95 «Aug: nuum you £33: 330 m5. ”302 .n 3» new uoow ~ av :13 o: mcuuzofi noun you $53: was? 2:. "302 N. a a a -..- so. So. «.8. m2. EN. m3. SN. ”3. 03f a5 ...- N8. 2... 8o. 02. Sf 03. 2? VS. H.233 .1; o .-.... ...... -- . ---..- .-...- -..... ----- - ..... - 9n o ...--- .-.-.- .---.- ...... ...-.- --..- .-.-. -.-.. 9... H - -....- -.---. ..-. - ........... . -.--.. 884 N.» a .-.-i .-.... ...... ...... ...-.- ---- 88A ..-.-- fin H --.-.- ...... ..-..- -....- -...-. 884 ..-... mg g .-...- -...-- ---.-. ...--. ...--- -----. 88; -..--- m.~ a ............ ..-.-. ..-..- .-...- ....-. SS. 82. ...N m -..-. .---. ..---- ..--.. -...-. 82. 88. ..---. in 2 -...-. ... .--..- -.-.-- 22. ER. $3. .2 2 --..-- -----. ---.-- $2. $8. ....-- o.~ co --- ------ -- - --.- $8. our... 33. 003. 0; 0m ------ -- - --- 00:. 00mm. 002. a... 5— --- -- - I: - 38. $8. 3.3.. mama. :3. ~A NE ------ .--.-- --- $3. mmon. 33.. ~30. N4 vm~ .-.-.. ------ once. 38. 3:. Be... mic. 33.. md 03 ------ --- 3.8. 88. 33.. 3.3. 3.8. 0.0 «mm ------ - ----- 88. ammo. 83. N8... «8n. ammo. To men ----- . £8. £8. :3. RS. 32. 33. ...o chm ...... “N00. 2:0. hmoo. oath. omwm. .33. «55. 0.0 vm~ .--.-- :00. v0~0. 0000. anN. ave... mom-u. come. 0.0 mmn mu00. 0N00. n36. noun. omen. Nmnm. mm:. n20. «.0. EN .300. use. 0090. mm“. 2mm. mmoN. 20:. 3.00. v.0- nm~ 300.0 ~08. muse. mmnu. nnmm. nn-. H30. 0000.0 0.0. 9H 300. 020. ova. mmNN. v03. «n2. wzo. «000.0 0.0. aNN ----- - 33. $2. $3. 32.. 2:. $85 .1--. ~.T 8H ...... 83. 1.1. 33. SD: 3:. 385 --.- - ~4- ~2 ...... 3.8. 22. :3. 3mm. 33. -.--..- ...... 04. ”2 £8. 33. Sou. 2:. Sun. 33. ..- 0.7 E ----- . 33. £2. 33. NS”. 28. ............ 9“. S ..---- 38. 2.2. 2?: 53. 3.8. ------ ----. o.~. R ...... :8. $25 3.3. 35. 285 -..-.- - ..... ...N. mu 835 ...... 836 83. 82. 835 ..--. ...--. ...N. n ...... $3... -.-- 336 $3. ...... -.-- ---. m.~. m ...... 826 --..-. 83.0 88. ..-.-. ...... 9? N ........................ 88.— .--- .--.-. ...... ~.n- u -.--. ...... .---.. ...... 884 .1-.- - ----- ”....- fi ------------ ---. --.--. 884 ------------ .1... 9m- ” -..--. ----- . ..-... ..-.-- 884 ........... . 9m. 0.... m.» m6 m.m m4. m.m m.N m4 m0 9 m “:4. mé v20 m.m m4 mi” m.~ m.~ m-0 . .E . n m . .5 am .mH :73: mus—u am An Jenn: 3:0 0 n. We. . snna «.7 . Sun» a . n . A3 :3: nwm$3mé a :5“ .930ng amt. kc .810wa 365 333320 Nugmumuxou 395% .Nm 03a... T: :3: Awmfijmuu N 2km $50.3“ A3» 0.3 E325.» and: mfiéfiungn Nuxcmvuigu 395m An 32:. STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED 58 .u :n ...o.« 33 H an :33 ohm mgumcwfi noun new E33: 320 2:. 6qu .m :n he uoow m 3 3:50 9: 33:3 nuum .5» £33: «3.3 2: “302 N. a mix . N. 8 a»? .... ....-. ...... So. :3. m2. m3: «.3. 23. d3 5 m .... ...... 8° :9 m8. m3. mom. SN. 08. 63 3 m o ............ ----- ..... . ------ ----------- - ------ 8m 0 ----------- - ------------------------------ ------ 8n « .--.-- --..-. --.--. --..-- uuuuu . ooooé N.m A . -..--. . uuuuu - .--.-. ------ ..-..-- oocoé ~.n H ------ ..---- .-.-.. ..---- ----- - 884 ”.... H ------------ .- .- ..-..- ------------ Soc.— m.~ S .--.-- ------ 88. 82. TN 3 . ---.-- . ------ ------ ooom. 83. ...N >~ nnnnnn ----.. .-.--. --.--. .----- onno. ommm. vnav. o.~ hm uuuuu - -.---. -.---- -.--.. ------ Ema. omnw. mnmn. o.~ E ------------------ ----.- ----- Sac. mod... 38. o; E ------ ..--.- --.--- . ----------- :8. vmov. mflm. o; S -.--.- .....- --.-.. ------------ 32. 2mm. :3. N4 ”2 ----------------------------- - 3.3. meme. 89.. NA 3 ------------ ...--. ------ 88. 3:. «N3: $2.0 m6 SN ........... - ammo. m$~. 1mm. 82. m5 5 ------------ - .--. ------ $2. 33. Yo mo~ -...-. wmoo. 22. m2... 3:. 38.0 To 2 -..-- ------------ Ema. 3.3. :3. ..-..- c6 #2 ------------------ 88. EN. $3. $2. ------ ad 3 ------------ 38. mnmc. 33. 8mm. ------ ...o. m: ...... ES. :2. ZR. 8?. $8. ------ Yo- «m ------------ 38. 33. 33. 8mm. . ..... m6- 9: ------ Eva. mmom. «m3. 33. 336 m...- R ----- - ------ Cmo. nvmm. vmov. NEN. .-...- ~4- ~: ------ 33. m2». 3;. ~m-. ------ NJ. 3 ------------ 056.0 nnnn. on”... 32. 0.7 E. ------ $2. 2%. $3.. 23. ------------ 0.7 n .................. 32.0 3:. 22.0 -.-.-. ------ o.- E ------ ... -. $85 33. 3mm. 32. ------------ o.~. N ------ --.... ------------ coco.“ ----------- . ----- . Ta. 2 ------ $3.0 ------ 3mm. 3mm. 33. -...-. ...... TN. 0 .................. ..---- ...... ..--.- -..--. ...... m.~. m ............ 88.8 88. 88. .... . M:- o ...... .-.--. ...... .-..-. .-.--. -----. N.n- H .................. .-..-. ...... ooceé ...... NA..- 8 .................. ...... ............ ..--.. ------ ca- H ........................ 884 .................. 9m- m.» m6 m.m m.v m.m m.~ m4 m-o wmuhw .53»: m n m o vim m.v m.m m.~ m4 m6 Henna» um up. SE: «was am as Jim: mum—o u. a; . wmua» N; . $.58» 5. . a a» mix a... m 31»? A3 :3: Thim— 8V m amxumwa Auum Kn rovuuxé. Ewes mafimfiagm Nugmuufirou 39.3m- .wm 032. SM :55 n. m .m_ Hg n £593N mm: .xo gmguxé. wads Emmmauagm Nu=omuufi§6 38.5m- Km 32m... .m :a you uoou H a... Has—us PS gum—.3 noun how usual mmgu 2:. “302 .m :a new uocm H o» :33 u: «cum—EH noun .3.“ 953) mum? 2:. €qu . .u a .5: a m =3 ..-. .....- So. Be. So. m2. 8%. 8%. m: h E_ 5 .... 28. 88. ms. So. ...8. EN. EN. 2:. Hi .5 m c .-.... ........... . ...--- ...... --..-- ...... ---..- 9m o -...-. --.... ..... . ..--.. . ..... -.-..- ...... -..-.- 0.” a ...-.- ..---. ..--.. ...--- ...--- .--... ..--.- ooocé ~.n — ...--. ........... . ..--..- ..-.-- -1.-.- cocoé N...” A .-..-. -----. ...... ...... .-..-- -..--. ..-..- 88; .3 a ....-. .-..-. ....-- ...... 88.. .2 2 --..-. .----- .-.--- ---.-- ..-..- :3. 88. 88. fi~ S --.... ...... ...... -....- 82. 83. ...~ 2 --.--- . ..... ---.-. ------ ...... :8. $8. ummu. 9N 2 ------ ----.- ---..- ...... .----- 2:. 83. 3mm. o.~ 2 .-..-. ...... .....- ...... -..... 82. N3... 2%. o; 2 .-.-.. ...... ..---- .....- ...-.. 32. EE. 3:. o; 2: --.-.. ........... . ------ 38. SS. 88. mod... N; 2: -..... -----. 23. $2. $3.. 33. N; Sn ...... ...--. .---.. ...... mono. 2:. :31 Amon- wé mNn ---... ..--.- Nova. ovow. mam. 3:. w... «on ............ :8. 88. $3. $3. 3.2. Z. 2;. ...... 28. 28. 3:. £2. 8:. #2. Yo omm ............ $8. 83. ”EN. 83. 2:. 88.0 ad :6 ........... - $8. 83. man. 3:. £8. 88. o.o a: ............ $8. 33. 5.8. 88. £2. ..... . ed. :8 ...... £8. 88. 2:. 28. 8:. :2. £85 «.8. Nnu ...... Roe. ammo. 2m”. Marm- oomm. :vc. m-o- 5mm omooé Once. cmmo. comm. vam. 3mm. vomo. --..-- m6- m2 ...... $8. 38. RS. 33. $8. 28. ~4- oou 88. «m3. 3:. RS. 32. 3.8. ...... «.7 0: 5:0. Bug. coVN. ouhv. mum". mwood - ..... v.7 mg ...... was. mnha. mvmw. 59$. Nana. mmood .--.-- 92—- S ............ $2. 3:. :5: 3:. ...... ..-... 9N. : ...... 28. $8. 83. 3:. :2. ..-... ....-. o.~- 8 ...... £36 33. 8:. N3”. 22. . ..... ..-..- Yu. 8 ...... :2. 33. SE. 3.8. ...... .-.--- ...~. 8 ............ 326 835 8:. 83. ...... .--... m.~. w ....-. 82.9 83.0 83. 8mm. .-...- ...... m.~. m .-.-.. ... .. ...... ...--. 2.3. $86 ...... ~.n. m .... ..-... - .. $3. $3.0 ...... -..... «.m- A ...... ...... .... . 88; ...... ..-..- ...... .3. a ..-... ..-. 884 ...... -..-.- .-. .- Y...- n.» ”...-5 m6 m.m m.v m.» m.N mé m6 H NJ. mK m6 m.m m.~ min m6 . .E "N . .E um a“. .u—hw: mmuHU ”a n n .F a a». :x , N. L m a». :x . A3 :3: “$73 .3 n £593“ mm“; a? Kuwuugxk mud: muflmfingm Nugmunfirg 385m. mm 3an n2 :5: a.» .3 5 A 2393» m3» kc .5395.“ and: whfimndngm Nugmumwxg 3&5». .3 03m... 59 .m :u how uoow _ 3 :53 an 239.3 noum how 233: mmflu 3:. "302 .m Zn .Su Sow N 3 23g 95 mguufi: 39m you 2:3: 3m? 2:. “3oz n— N. ---- ------ --.--. --.--- .----- ------ m2. m2. .--. ........... . ...... 35. «.3. NS. 3.... :o. A. Emu; 2%» a .-.... ...-.- ...... ---... ...... ...... o .................. . ..... . ........... ...... ...... o.” H . -.v. -.-.-- n ..... --.--. u ........... . ...... . ..... . ........... ----.- ococé N.n a . -u .v ---- . ------ -----. A -.---- .. -.-:-- ...... -.-..- cog." m.~ g ....-. .....- .--... ...... --.... N . .... .. - ........... 8cm. 8%. YN a -..... ..-... ..... . --..-. ...... ...-.. ...... N ...... ...... ...... ....-. ........... . 88. 88.0 9N _ .----- ..... . --.--. ........... . ....... N ...... ...... -..... ..... . ...... ...... 884 ...... a; N . --.- ..---. -..-.. ..... . ...... 884 N ...... --.... . ..... . ..... 08m. 88. .....- N; N - ...- -. --- ...... -..- . . .- - 884 N ...... .-.... ... .. ...... 08m. 83. ...... ”5 N .....- ...-.- --.... ............ 88; N ..... . ...... ...... ...... 8cm. 8%. ...... «6 N .....- - ..... ...--. ..... - ...... 884 N .............................. 88:0 ...... as N . .. -.. ...... 82: N ........................ 8am. ............ «.o. H ............ ...... 88; ...... N .--... .-.-.. 08m. ...... ...... me. A .............................. 884 ...... N ...... . ..... 886 ........... - N4- 0 .......................................... H ...... 88; ...-.- 9—. o . ................. .----- ------ ...... o .............................. ..---- ............ o.~- o . ...... --.--- .................. o ....................... . ...... ------ ...... TN- 9 ........... . ------ ------ .................. o ...... . .......... . --..-. . ........... m.~- o .......................................... o ............ ....-. ...... ...... ..-... ...... Na- o .......................................... o .................. ------ ............ ..... 9m- S . . . . . . . . mK m6 rim m.v m.m m.~ m4 m6 E a; mu mo mm mv mm mN m~ mo «:53». L . ... . . w aw .mH is: mmgu um .mu Jim: 33... m. M NJ. u .mmuma m.~ u .m" Q Q . mm 3N =3: nwm.mm_mHv a. :xn 65¢me m3» «3 £03056“ was flimaunoxn NEBEWSAS 395m .mc 39...... A“: :35 awm.wm_muv » 13* £593“ 06%.. k0 «63.»:an and: fifiwnuaosm Nuromunfiroa 395m .3 032. D L m N .m :5 how uoow g 3 Hanan v.5 mzumzua Auum .Sm mfg: mmEu 2:. "302 .m :m new acou _ cu i=3 0: 23:3 noun how 233: 329 9:. ”302 m . . N. N. m ..-. ...... m8. NS. 3N. m3. RN. N8. “.333 ixm .... ..--.. .--... So. 26. SN. 2.... SN. to. as}; 3S W a ...... --.--- ------ ------ ---.-- ------ ---.-- ...... o.m o ........... . ---... --.--- .-.--- -----. ...... -----. 0.». U N .-.... ...... .-..-. ..-... ...... ...... ...--. 88; NA _ ............ ...-.. 884 N.n S N -.-... ....-. .....- ...-.- -..... ...... ...... 884 m.N H . ..... .-.-.. ...-.- --..-. -..-. 88; M: 3 .-...- ...... ---... .-.-.. ---... ocoN. coon. TN 3 .--... --.--. ...... coon. 83. TN 3 ------ --..-- ------ -----. -- ... SS. 83. o.~ mm .----- ---... ..---- .----- ------ ono. 2.3. 3.3. o.~ 2 ............ .---.. ...... . .-. 23. 8%. $1 04 Nm ...... . ...................... . mofl. 33. N03. . o; 3 ...... .----- -----. ------ ------ oom~. ommw. 3.26 N; N... . .......... . ----.- 3:. ”we... win. NA 2 ...... ..---. ..-... 38. SN... mNS: -..-.- as o... 83. 33. $8. m6 3 ............ 33. mNom. oomN. ...... ...o mN .-.. . .... 2.2. 23.. m2»: mvmod To 3 ............ .....- -..-.- ES. 32.. $36 ...-.. o... S .................. Ewe. --. mmum. NNNN. ...... 9.0 3 .................. mama. amen. mmvm. ............ ...o. E .................. 32. 3.3. 9.3.. 05.06 -..-.- To. 3 .................. gen. 83.. coon. ............ m6. 2 -.-... ..-... owmo. anw. Swv. 33. m6. 0 ............ coco. $3. $3.0 ............ NJ. m ............ 3.3. :2. v3}. ~.T m ...... 336 mmmmé .................. 0.7 m ............ 835 83.0 8e... 83. YT o ................. . ...... ..-... .................. o.N. N 88. 88.0 ............ 9N. o ........... . ........... - ........................ TN. A 884 ....-. ...... ...... «N. a .............................. 9N. a .... ...... m.N. o ...... -... ........................ N.m. o ...... . ........... Na. a . . . ...... .. ...... 9m. o ................................................ 9m. mK m6 m.m m.v min m.~ m4 m.o vmlmm aw: mK m6 m.m m.v m.n m.~ m4 mé d . a £5: 32”. a .mu 3:5: 3.5 N. N. Tm n .Mnna o.~ n .Muaa Q . Q . no“ :3: KENS—may a :5“ ..mfimxuN mu; kn Ewuurxk anus mufiwaenoam Magnfiufizou 39.3w .3. 3mm... 3" :3: flywmfimuw » :xm ..afimfiu “Sum k0 rowuoxxk wads mafiwfluagm Nuromfiufigu ngm .3 03m... STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED 60 .m Sm how one H a» :23 mun mcumcou noun now 233: mum—u 2:. "302 .u :u now “com H 3 353 0.3 msuucua noun you 9533 mun: m5. "302 a q u. R. ...- 8o. m8. m3. 2m. :m. 08. 323$ :5 .... :5. one. 3”. 3m. 2». So. H.233 :.E o .--... - ..... -..--. -...-. ...... m.~ a -...-. ..... - -..... -....- .-.... m.~ o -..-.. -..--- -...-. -...-. ...N o ..-... ..-.-. . ..... ...... YN o - .. - .-. - - .-- -..... -...-. o.~ o -...-. ....-. ...... ...--. -..... o.~ m :.... -..-.. 2:. out. :2. a; n ...... ..... - ..---. ...... 884 0; fl ...-.. .-.-.. ...... 82. SS. --.... «4 2 ...... ...... ..--.- 28. SE. «4 no .-..-. ...... ES. 22. 3mm. 38. m5 ..N - .... :8. SR. :2. :3. ”.0 m3 ...... ...... :3. 23. It: 38. 3° 8 .-.... ..... - 38. 3mm. 23. $3. I. m2 ..... - 38. 38. $3. :3. as... as S -..... 23. 82. aim. m3". 236 .3 m2 ...... 33. Sum. Sm... 8:. ...... Yo. an ...... £3. :3. 2mm. 3:. .. To. 2: ...: 32. 3%. :S. 32. ma. 3 386 3:. 3:. 32. 3.8. - .... 9o. 3 ...... 38. $2.. 28. 335 NA. NN ...... 3:. one». 3:. 3.8.0 ...... ~.T mm :35 3:: :3. 3:. -:.. 9T 2 one»: 33. 335 ............ a... m ...... 3mm... 3mm. 2:... ...... o.~. n 33.0 $8. .-.... .. o.~. _ ...... ...: 884 -..... ..-... ...? H --. .. -..-.. 88.— - ed- a .................. .. .. - -.. ..--.. ”2. o ...... ...... .................. w.~. o ...... ...... ...... N6. o ............ -...-. ..... - ............ N? o .-.... ..-..- -..... .-..-. 9m. 0 ............ ...-.- .................. 9n. we Wm m... min m.~ m4 m6 m a... m.m mé Wm m.~ m4 m5 . .E m" .~ . .E um .aH Jun: mmnau n . A: :3: AmeimHV M :xm .2398» AR; .bu 53.33% was EEHBGQKR ”983%:qu 39.8w. HQ». o.~. I H. . m a». :x C— :3: A. :...»; 3 m $59.3 m3." «3 §meu§k ”was mun—NEUQDE Nuxcmuufirg 33.8% .mv .3 032. 03:. .u :n how uoow A 3 333 93 “5:3 me: how £23: mum—u 2:. "uuoz .m :a you uoow ~ 3 :23 a: 23?: Elm you 233: 320 2:. “302 . N. u a»: . u n ...: So 5. an. own. 08. “.3 3 m ...... .-.... m3. m2. 3.. ...... wigs ix“ 0 ............ .-..-. ...... ...... ...... w.~ o ...-.. ..... . .--.-- ..-... .-..-. ...--. m.~ o .-.-.- ...... ...... .................. ...N a . ..... ...... ...... ...... .-..-. ...... ...u o -...-. -....- .-...- ...... ...... .-...- o.~ o ...... -.-.-. ---.-. O.” g ...-.. ...... ...... ...... 884 . .-.. a; o ...... ...... . a; a ....-- -...-. .-..-. ...... 884 3 o .-...- ..-..- -..--. ....-. -...-. N; v ...... ....-- 83. 8mm. ”.9 A ...... ...... ...... ..... - .-..-. ma » .-.... :3. 39.. ...o N ...... ..... . ...... 88.9 ...... I. c .-.... .-.... 83. 83. .-..-. .-.... o.o A ............ 884 .................. as n ............ SS. :2. 33. ...... To. o ........................ .....- ...... «a. h ...... 2:. $3. 3:. £3. -..... as. o .................................... 2.. n ...... $2. $3. 83. 32.9 ...... N;. o --..-. ...... ..-... ...... -..... «.7 n ..-... and £26 £86 ...... ...... ..:. o ..-..- ....-. .. .. a4. a .................. -.-... .-.... ...... 9? o ...... ....................... - ...... o.~. a .-.... ...... . ..... .-.... ...-.. ...... YN. o .................................... YN. o .-.... ............ .-.... ...... ...... ”.m. c ................. . ...... ...... ...... ”4. o ..-... ...... ... . .. ... N”. o ........... . . ..-... ~.m. o ..--.- .... . ...... . 0;... o ...... ... .. -- ..- .--.-- 9.”. N...» m.m m.v m.m m.N wi— m.o m.m m.v m.n m.N m.— m.c . é a Ufa» N. a... :1» um . 8 Jun: 339 um .uH Jim: was n. ...N. . $.n» M3- . fine» A :3 RE: $5.5 . w m a m a» ma: 2 a _ .3 3» :3 33 K0 $30.39* was: muflmauagm Ngonfiufiroc 393% .3 v.32. A: :35 A. m_ .5 m .mxumxfi mu: K0 gwuo§k anus mummmnuagm Ngmufiurg 39.8w .9 3an 61 SUPPLEMENTALDATATABLES .u :a now uoow H 3 fiasco ops 23:: nuum you 233: 320 on... ”Mac: .m :n how uoow a 3 gave 0: m5?»— moum how 953: was: 2:. ”032 N. m a»: u a a ll :8. to. m:. SN. 03. v3. “.2 5 m -.- 8o. e8. 3;. 2n. 2... m2. 3: i :5“ c ---. ----- - -.-- --.--- .2 o ---.-- -.-- ---- --.-- ti: .2 fl --.-- ----- - .- ..... - 88.— ...N o --- ----- --.-- .-.-: ----.- -- «.N m --- .---- ---- ---- 83. 88. .3 q ----- --.--- ii: .2... 8?. 8mm. o.~ mm 1-... .1... .---- $3. E2. :3. o; 3 --- ---- ---- 23. 88. one... a; a: ----. ----- $8. 23. 3mm. 33.. N4 m3 :1: ----- £8. 38. 3%. 32:. N; 2.. -.- ----- 38. SB. 3%. 22. m... Sn 1-... .1... 28. 33. was. 32. .3 one --- --.. SS. 32. $3. 23. 3° 3% -.--. 28. 38. 53. 3%. :fi. v... 2K 2.... $8. 38. 8mm. :3. 3:. ca 80 ..... - $8. 9.8. 8%. 39.. $8. o.o an. ---- $8. 3:. 3.3. NR». $3. 2.. MS 28. 3:. 3m... :3: :8. To- um 88. 38. :2. 28. :2. 285 m5- Sv ES. ER. 33. 82. 885 m5. m: 38.9 :3. 8:. N2... 32. ..-- ~.T m: 38. 2:. $3. 82. - ..... :- afi --- SS. 8:. $3. 38. ---- 0.7 «2 3:. «mm... 32. 33. --- 0.7 3. ----- :3. £8. 22. $3... 2-... 9T 3 886 8:. 83. SS. 886 -----. o.~. 2 --.- $56 3.8. :3. --.- -1.-- .IN- 2 ------ $3.0 35. Row... .2... ------ .3. N ...... 885 88.0 ...... .2- ~ -.---- 884 ----- ............ m.“- o null-n nan-u: nun-n: nun-u: 3v... N6: o :3: uuuuuuuuuuuu N.nu o ---.- 1...- 1... --.-- --.-- 9”. o --.- ...... -.--- --- 9n- m.m mé ma m.~ m4 m6 5.» m.m w... W... m.~ m4 m5 . .s c .5“ :5: ans a .mp. :25... 3a; u. as . .mk» Yo- . Mug» 5 L a a» m»: . N. a a a» mix a: :3: A. :- .a_ .3 m S398N muuw Ho Evanga. made Emfiaunem Nugnfinfigo 33.3.0 .8 033. :— ==5 n.>...\~._ Nu m $§m§~ muum K0 :OUEOSC. 23E numNu-auagm Nugmumfigu uugm .3 035... .m 1; .Su uoom A 3 :35 9:“ 25:3 nwum you £53: mum? 2:. 5qu .a :5 new uoou A 3 133 ohm 2:5: an: wow “53: $53 2:. E32 a N. a So. So. 02. an. NS. ”8- “$13 :$ -- 3°. 26. m2. Sm. 3n. :5. H.293 15» c .nu-n- in--- ---- ---- ---- m.~ o --- ---- uni- --- - w.~ c ------ ---- -- - -.»-- v.~ o -1- - In--- ------ - u--- v.u N --- --- --- ---- coco.” .--- o.~ u -----u --- uu-a-u --- coco.» o.~ on nun-u: unv|n| until- @000. 506%. hooN. a; an lull»: 1-1:-.. nrlnln moan. 0:6. mung. o; o: :un: nun-n. ~29. mvmo. 030. 02K. NA ha u-uuu. :1qu Nag. zon- omwo. N." now 7.-.: nnoo. whoa. 00:. ~30. on:- m.o m: uuuuu nun-n. ammo. meN. nMNn. 536. m6 v; 1...: «woo. omwo. was. :3: swan. v.9 0mm n-u-n- nun-u. acme. ooov. owom. come. «.9 mac Soc. 500. $00. Emv. knew. :wo- 9.0 an uuuuuu mmoo. cnoH. KEV. NNmm. Nmme. 0.0 owe :55. ~08. mum". omnv. NmnN- camo- v.o- mmw uuuuuu Sue. ommm. «nee. vmmw. mnmo. v.0- oov avoo. omwo. Nown. oMNV. VmON. nus. m6. mmN u:- mnma. mvnn. mwcm. mama. #590, m6- oBN :86 aux-o. NNNV. moan. wao. :00. NJ: 9: 3:. unuv. :on. :00. :3. ~4- n: --.- «SN. 83. $3. $35 286 0.7 cm 2%. 3w»: 3:. 3.85 285 04- aN ...... $8, 3:: 33. ------------ 9? MN 3:. --- :1: o.- a ------ 85. 835 335 ------ :2: «.N. v 835 ------------ Yu- H -.-- 884 ------ -.-- --.-. -.- - 9T . o - . --- ...... n.- o .1... --.- - .-- .-.- ...... ~.n. o -.-- ..-.- --.- N.”- o 1-... ------ .. .7 2-.-. --- ------ 0.? o .1... ----------------------- . ------ 0-m- m.m mJ. m.m m.~ m; m6 m.m m4 m.n w.~ m4 m.o hum». aw .uu .xus: mmuuu um .mH .xkuz mmmau Ana =2mu a . mgr-jun“ ~ :3 . m5- “ bra» 3393 mm: Kc 5.395% SEE mud-Nwauaoxm Nurowuufixau £95m. .mv a~AMH N. WT - $3» a A: :3: AnaLEQHV a. lung $59.3 mm»; .80 :om-«SSK anus myfimflgofln Nugmuufiroo 3%.». .5 01...:- STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED 62 .m In new uoom H 3 333 PS 23ng noun you £23: 330 2:. ”ouoz n m. :1 .-.... So. NE. Rm. :3. 2:. ”.333 :xm o ---..- ...... --. .--.. a.~ H -..--- .....- .....- ...... 884 ...N m .-.... ..-..- ..... - ...-.. 88. 8%. o.~ Na -.-... ...-.- -..... $8. $3. 33. «.4 8w ..--.- ..--- --..-. 88. 8mm. 8mm. «4 new ..n---- 0.50. ono. MNNm. ZNV. «.0 men ----- cho. 3:. n35. 3.2. v6 mmw ..... - ...... 33. Si”. 83. 33. ed SN ...... omco. mvao. :3. go». omood «io- mmm --- moco. mag. mmmo. MmNN. --- m6. ao . ...... mono. wmwm. mme. 3:. ...... NA- mv I}--- «.35. -~v. mwmv. vvvcé ...... o4- : --.-- :2. Sam. --.-.- o.~. v ...... oomNd ooomd ...... ...N. o ............ m.- o . 1.... .. .. ............ «.m. o 2--.. .1... .-..-- ............ 9n- w... m.m mg. Wm m.~ m4 m5 m . .E u m C .ma i5: mmflu N. N; u sun» .N. L m a» mix AS :3: a h .3 3 a $59.3 QR; kc Eur-ESQ wmuE mumNmAuaPR “axovfiwrg 39.8% .nm 03m... .m :u now uoow _ ca 2230 w: 2:5: nuum you 9333 mmfiu 2:. ”302 .m :a how :8 a 3 15g 93 939:: “Fun “om £533 mm“? 2:. "302 . . u m a»: N. a a»: ...- ...... :o ~S 1%. Sn. SN. 3: E a .... 08. ms. 81 am. con. mi. 3: 5 m c ..-... .1... 1.... ...-.. .--.. ...... M; o -....- ...... --.... ....- ..-..- --... w.~ H .-.... ...... ..-.-. .-..-. ...... 884 YN H .-.-.. ...-.. ...-.. .-.-.. -..... 884 «.N m .-.- - -- - ...-- 88. 8o». o.~ m .-..-- -.-. coo... 88. o.~ No ---. mmmw. Sm“. o; ow -.- SS. 3:. 3:. o; 03 -.-... 1-.-- 88. 82. 3.8. N; m2 «3°. 33.. 83. N; Re --.- . .--- :8. 33. $5.. 35.. m6 m3. -.--- mnoo. menm. mmmm. m6 3.. ------ -..-.. $8. $2. 3%. 32. «.0 m3 .--..- .-.- 82. $3.. $3.. To m3. ...... ...--. 38. 34m. $3. 28. ed 5.... ..--.. --.-- :8. 3mm. moom. 2.2. o... o: -- - 359. N20. was. mmon. avmo. v.0. chm --.. mwoo. mnmo. «woe. mmov. 0:0. «.0- v~m ------ mus. mmz. Aim. 923. 150.0 m6. $3. Nwooé mus. voom. hwmm. movu. NNoo.o ad- NoN .u...- 035. 03m. $3.. 3:. ------ N4. m: ----- - mug. coon. mfiv. $2. ------ NA- nm ------ ~36. mmam. :wm. mnco. ..--:- @4- m: -..-- «20. once. 33.... 80°. --- 0.7 wN --- :2. mumm. mwmn. :oH. o.- 9m --.-. mmz. Rom. mace. name. --- o.~u w :1: om-.o ”.mhmd omnm. om~1c «..N- a - ::.o cmmmd NNNN. 3.2.20 --.- «ZN- H .--. ----------- - ooooé ..... a m.- H ---.. --- ------ coco; ------ ---a m.N- o ...... .-... ...... 1---. ............ Na. 0 ..-.-. ..... - ...... ...-. ............ Na- o ...-.- ............ --.- ...... .--.- 0.? o ............ --.... ...... .-..-. ...... 9m. .5:an m.m mé m.m m.N min m6 sham m.m m.v m.n m.~ m4 m6 . N um ma Jim: mmgu uu .9». Jim: anus n. ma . .m-n» Tc . nmna» n. a m . Q . n: :3: A. aim—m3 » :xm afluwrmm mmuo k0 €333.“ was muflmaunoam Nuruwfifigu 33.3% .3 032. A: :3: awfurfimxv » :xm 853:3 muuw kc 5.305% 35: EEEdQEQ Nugmunirob 395m. Am 2an 63 .u :a new noow a on 1:50 0.: mzumcua noun how 333: mmgu 2:. "ouaz Q , .... ..--.. .--... - ..... ...--. So. mmm. ”.5393 :5» o ---.-. ..... . ...-.. ...... .....- m.~ _ -..... .-.-.. ...-.- . ..... 88.— «d a -...-- ...... ...... ...... 884 o.~ a ...... ...-.. -..... 884 a; H ...... ...... 884 N; A ....-- 88; ...... as H ---... ..-... . ..... 884 ...-.. To a ..... . . ........... ...... ...... ed a ...-.. --.... .--..- . ........... Yo. o .-.-.. ...... ...... 9o. 0 ...... ...... ...... N4. 9 .................. ...... ...... cg. o ..-.-. ............ .-.... ...... c.~. o ....-. ...-.- ..... . ........... . YT o ...-.- ...... . ........... ...... m.~. o --.... ...... -..... ............ Nd. c ....-. ...... .................. 9.... “:9 m.m m.v m.n m.N m.~ m.° . .5 am .nH Jun: mmgu v.” . $5” a m. . A: :3: n. m.§_mfiv N :5“ afiuwrfl mmuw kc :efiuxé. was mumNu—nunoxm Nagmuufirg 393m. ,3 ~32. SUPPLEMENTAL DATA TABLES .m :u now uoom _ on :13 aka mfmcg muum you 253: 320 2:. uuuoz .m :5 new uoom H an 133 a: mzumnou an: how 223: mum? 2:. “aucz Q Q N. N. a x 1.. ...... ...... 36. «on. 8v. :.1 H.233 :xm .... - ..... as. 2:. R»: an. 1:. “.3 3 .1 n o -..... -.---- . ..... --..-- ---.-- ...-.. a.“ o -..-.. .-..-. m.~ A .....- ...... . ..... .--... --.... coco; <.~ H .....- ...... ....-. 884 ...N m ..-... .----. .. .-- 88; 9N m ...... -..-.- ..-... ...... 83. SS. o.~ 2 ...... .-.--. -..-.- 33. o; S ....-. ...... .....- SS. 83. $8. a; 2 --.-.. -..... -..... 83.. N.” mm -. ... ...... £8. 23.. 33. N; 2 -..--- ...... -..--. $36 ad mm -.-... :8. $3. at”. a... : -.. . --...- ...... .-.... ....o E 28. $3. 3:. :2. Yo S ...... --.-.. -..... 83. 88. -..... 9o on -...-. ...... 88. 8mm. 2mm. $3.9 as m --.... ...... .--.-- 8:. SS. ..... - ...o. S . ..... 28. $:. 83.. 82. . . Tc. a ............ SS. 8:. 83. ..-... m5. 3 ..... . :5. $3.. 3:. ...... as. m ---... ...... econ. 88. 836 -..-.. NA- 3 9.8. N03. 33. E26 ....-. ~.T n - ..... -- .- :3. $8. ...... -..... 0.7 fl ...... 2mm. 33., ...... ..-... 0.7 a ...... :.: 22. $3. ..... ...... o.~- h ...... E5. :35 ............ 9? ~ ...... --.... 885 88.9 .....- ...... ...N. N 836 .................. TN. 0 ....-- ...... ....-- ...... ..-..- ..... . m.~. o . ............ .. m.~. o ..- .. ..-... - ..... «.m. o ............ --.... ...... Na. c ...... 0.7 o ............ ...-.. - ..... ..-.-. ...... 06. m.m m.v m.m m.~ m4 m6 ”FMIMN m.m m... m4 m6 um . a. .xum: mmuau N. o.~ . .muna Q Q A: :3: nwmumflmxv N. lam £593“ m3» %0 .3395.“ and: mumensnEm Nugonzufigu 3%.» .mm as: A: :3: Amem—mav a. :xm amxumrug mmum kc Swans...“ 23.: auflmfluagm Nerowumwroo 395% .«m 032. 64 Table S7. Canditianal mama and variances of «tap lengths; Eula-11‘, ID fly] and firIXIXE 'Vi' I'D-yd] (Run 4A) STOCHASTIC ANALYSIS OF' PARTICLE MOVEMENT OVER A DUNE BED -0.8 -0.4 0.0 0.4 0.8 1.2 1.6 2.0 2.4 4.057 2.842 3.150 3.647 1.417 1.235 .101 3.620 3.385 3.239 3.207 3.195 2.965 2.768 ---------- 1.918 1.529 1.021 .839 941 .360 .395 ---------- 3.476 3.409 3.459 3.357 3.309 3.184 3.199 5.346 ----- 1.768 1.684 1.357 1.156 .966 .743 .985 ---------- ‘2 0 ----- 4.214 3.771 3.407 3.261 3 133 3.102 3.036 2.935 2.909 3.000 4.308 ----- ‘ ----- 1.409 .986 1.133 1.074 1.069 .872 .783 .665 .557 .707 1.503 ----- _1 6 ----- 3.417 3.239 3.173 3.076 2.989 2.956 2.898 2.804 2.773 2.820 3.485 ----- ' ----- 1.665 1.175 1 176 .991 974 .779 .714 .599 .555 .559 .987 ----- -1 2 2.861 2.704 2.848 3.018 3.136 3.063 2.976 2.854 2.808 2.785 2.717 2.710 2.733 3 175 ----- ‘ ---------- .766 1.009 1.125 1.439 1.390 1.210 1.046 .872 .806 .726 .682 .669 1.055 ‘0 8 ----- 2.611 2.478 2.588 2.747 2.785 2.653 2.583 2 505 2.498 2.507 2.466 2.503 2.562 2 879 ----- ' ----- .054 .392 .620 .942 1 137 1.145 1.063 887 .759 .683 .623 .603 .669 1.155 ----- _0 4 ----- 2.082 2.434 2.506 2.589 2.519 2.455 2.336 2.180 2.170 2.197 2.205 2.275 2.360 2.604 ----- ‘ ----- .098 .836 .796 1.251 1.229 1 197 1.001 .925 .772 .656 .602 .563 .653 1.225 ----- 0 0 ----- 1.892 2.557 2.300 2.400 2.274 2.221 2.083 1.929 1.797 1.843 1.916 2.006 2.081 ' ---------- .609 .584 1.011 1.069 1.084 .950 .854 .813 .666 .581 .542 .540 1 296 ----- 0 4 ---------- 2.312 2 289 2.365 2.159 2.054 1.903 1 728 1.595 1.437 1.521 1.661 1.774 ' ---------- .943 .401 .985 .986 .910 771 .723 .652 .637 .495 .435 .429 0 8 ----- 1.692 2.133 2.257 2.108 1.933 1 775 1 604 1.439 1.306 1.127 1.244 1.394 ‘ ----- .010 .892 1.076 1.091 .837 .690 634 .550 .485 .416 .367 .357 1 2 2 279 2.188 1.903 1.756 1.581 1.373 1.203 1.036 .870 .968 1.508 ----- ' 755 1.127 .889 626 .608 .501 .429 .313 .301 .297 1.035 ----- 1 6 ————— 1.430 1.518 1 779 1.687 1.558 1.503 1.288 1.142 .932 .812 .595 1 081 ----- ' ---------- .241 1.693 .942 539 .646 .424 .397 .273 .275 .270 829 ----- 2 0 ............... 1.670 1.377 1.375 1.086 .954 .869 .785 .687 .460 ----- ‘ .................... 000 .084 .120 .125 .128 .132 .147 203 ----- 2-4 III: III: I: I: I: III: III: I: I: I: III: E[X'YD=yj] ----- 1.194 2.172 2.397 2.577 2.486 2.384 2.257 2.119 2.025 1.972 1.938 1.983 2 056 2.458 ----- ----- 1.396 1.232 .884 1.376 1.415 1.277 1.141 1 084 1.035 1.008 .935 .892 .915 1.671 ----- Vex-[XI ID = yj] Note: Upper values are the means, in feet, and lower values are the variancés, in feet squared. SUPPLEMENTAL DATA TABLES Table 58.00nd11tional means and variances of step length; ELYIYE-yi, YD-yj] and V?r[XI1’E-yi, XD-yJ] (Run 16) -3.2 -2.8 -2.4 -2.0 -1.6 -1.2 -0.8 -0.4 0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 -3.6 _______________ _3 2 ----, ———————————————— 3.442 3.379 3.318 3.496 3.594 3.427 3.102 3.042 ' .' ---------------------------------- .115 .164 .091 .026 .051 2 8 ---------- 4.600 4.516 3.852 4.551 4.447 4.293 4.142 4.027 3.833 3.696 ' ' -------------------- .678 1.068 1.179 .930 .832 .868 1.047 .983 _2 4 ---------- 4.334 4.415 4.548 4.612 4.621 4.480 4.363 4.217 4.063 3.905 ' ---------- .032 .467 1.059 .979 1.077 1.053 1.001 .869 .892 .913 _2 0 ---------- 4.185 4.477 4.623 4.626 4.514 4.437 4.399 4.223 4.083 3.931 4.012 ----- ' ---------- .067 .545 .809 .760 .699 .752 .904 .746 .728 .716 .656 ----- _1 6 4.646 4.438 4.524 4.369 4.327 4.261 4.075 3.981 3.834 3.987 3.851 3.816 4.430 4.379 ----- ‘ 1.396 1.074 1.032 .931 .950 1.067 .953 .829 .790 .733 .620 1.163 .858 ---------- _1 2 4.513 4.295 4.269 4.113 4.039 3.942 3.783 3.694 3.562 3.685 3.689 3.568 3.952 3.243 2.391 ' 1.076 1.038 1.079 .960 .918 .978 .880 .748 .705 .678 .665 1.074 .842 .904 ————— _0 8 4.020 3.961 3.878 3.726 3.632 3.553 3.399 3.356 3.287 3.474 3.481 3.519 3.580 2.933 2.112 ' .837 .982 1.077 1.131 1.100 1.082 1.012 .833 .783 .747 .609 .840 .507 .824 ----- _04 3.197 4.391 3.823 3.633 3.608 3.443 3.294 3.188 3.079 2.936 2.927 2.917 3.072 3.116 3.108 3.129 2.698 1.913 ' ----- 3.638 1.307 1.236 1.108 1.069 1.121 1.126 1.131 1.059 .869 .762 .656 .524 .603 .483 .733 ----- 0 0 2.917 3.802 3.384 3.076 3.132 2.969 2.851 2.731 2.617 2.446 2.436 2.457 2.638 2.674 2.582 2.657 2.459 1.726 ' ----- 2.159 .958 1.092 1.020 1.042 1.028 1.021 1.050 1.032 .851 .714 .585 .462 .584 .502 .611 ----- 04 2.534 3.273 2.893 2.727 2.769 2.692 2.559 2.437 2.316 2.155 1.963 1.960 2.174 2.262 2.296 2.421 2.219 1.574 ' ----- 1.598 .860 .696 .737 .745 .745 .760 .756 .733 .766 .675 .516 .411 .498 .424 .433 ----- 0 8 2.266 2.844 2.413 2.317 2.285 2.300 2.207 2.104 1.987 1.846 1.718 1.519 1.684 1.824 1.938 2.116 1.999 1.409 ' ----- 1.072 .739 .576 .599 .591 .629 .617 .582 .558 .551 .612 .465 .367 .459 .358 .337 ————— 12 1.728 1.574 1.431 2.067 2.127 2.058 1.966 1.844 1.713 1.598 1.498 1.388 1.175 1.358 1.544 1.788 1.769 1.197 ‘ ---------- .213 .601 .428 .481 .527 .510 .468 .434 .427 .410 .416 .349 .433 .312 .308 ----- 1 6 --------------- 2.125 1.808 1.780 1.687 1.628 1.481 1.385 1.297 1.215 1.114 .946 1.195 1.438 1.512 .998 ' --------------- .041 .130 .311 .373 .371 .342 .335 .338 .340 .340 .383 .390 .252 .223 ----- 2 0 1.269 1.198 1.109 .960 1.054 1.257 .814 ' .234 .232 .237 .276 .183 .139 ----- 2 4 1.014 .949 .879 .802 .672 .923 .652 ' .083 .083 .083 .084 .096 .017 ----- Z 8 .936 885 832 .780 727 675 496 3 2 ............... 971 875 777 .681 671 .565 512 .460 .407 355 176 E[X|YD=yj] 2.073 2.512 3.008 2.924 3.097 3.049 2.927 2.825 2.716 2.567 2.473 2.402 2.517 2.500 2.516 2.550 2.308 1.495 Var[X|YD = yj] 1.463 3.476 2.290 1.635 1.530 1.512 1.499 1.509 1.547 1.477 1.432 1.381 1.392 1.234 1.252 1.409 1.164 .368 Note: Upper values are the means, in feet, and lower values are the variances, in feet squared. ! 65 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED Table 59. Conditional mm and variancea of atop length,- EIXIIE'yi' ID-yj] and firIXIYE lyi, YD-yj] (Run 17) .2 0 --------------- 4.118 3.699 3.441 3.319 3.285 3.104 3.051 3.000 ----- ' --------------- .395 .335 .725 .575 .596 .546 .531 .542 ----- —1 6 3.693 3.686 3.471 3.320 3.210 3.076 3.056 2.983 2.932 2.979 ‘ .320 .824 .751 .623 .565 .546 .475 .421 .327 .208 -1 2 3.369 3.408 3.117 3.008 2.922 2.818 2.816 2.769 2.704 2.707 ' .700 .632 .611 .629 .589 .552 .477 .453 .411 .416 _0 s ---------- 2.656 3.013 3.017 2.790 2.662 2.574 2.499 2.476 2.465 2.441 2.411 ' ---------- 1.053 .935 .733 .638 .681 .610 .530 .452 .408 .342 .330 -0 4 —————————— 2.516 2.772 2.709 2.461 2.354 2.232 2.141 2.129 2.151 2.187 2.219 1.998 ----- ' —————————— .833 .664 .613 .583 .616 .608 .529 .445 .383 .354 .407 .287 ----- 0 0 ----- 3.396 2.878 2.523 2.379 2.153 2.041 1.934 1.792 1.765 1.814 1.877 1.942 1.798 ---------- ' ---------- .223 .563 .522 .496 .516 .522 .502 .435 .351 .312 .358 .220 ---------- 04 2.123 2.091 1.896 1.763 1.652 1.513 1.377 1.421 1.521 1.625 1.527 1.624 ----- ' .457 .370 .360 .398 .405 .382 .403 .318 .260 .288 .181 ---------- o 8 1.636 1.888 1.846 1.696 1.573 1.464 1.332 1.219 1.069 1.135 1.291 1.396 1.234 ----- ' .439 .238 .237 .245 .289 .279 .257 .256 .272 .241 .247 .182 ---------- 12 ---------- 1.817 1.584 1.621 1.526 1.400 1.280 1.146 1.048 .941 .786 .891 .984 .967 »»»»» ' --------------- .090 .220 .253 .245 .248 .214 .210 .203 .212 .202 .147 —————————— 16 ---------- 1.226 1.285 1.520 1.540 1.289 1.190 1.038 .945 .841 .725 .545 .639 .753 ----- ' --------------- .106 .248 .254 .155 .183 .174 .168 .161 .157 .161 .133 —————————— 2 0 1 366 1 088 948 866 .793 .714 629 494 546 ————— ' 053 108 150 148 147 139 123 102 ---------- 2 4 ........................................ 941 796 682 .577 494 .411 328 »»»»» 2-3 III: I: :3: I: III: III: I: I: I: I: I: III: E[x|yD=yj] ----- .989 2.311 2.473 2.440 2.271 2.153 2.042 1.925 1.866 1.841 1.847 1.899 1.777 .506 ----- v§r[x]yD=yj] ----- 1.815 1.104 1.069 1.010 .845 .861 .849 .811 .818 .797 .762 .765 .606 .424 ----- Note: Upper values are the means, in feet, and lower values are the variances, in feet Squared. SUPPLEMENTAL DATA TABLES Table 60. Estimates of pamters describing two-paraneter gama distribution for conditional ”a? lengthy (Run 44) 2.0 2.4 --------------- 2.438 2.006 2.551 36.109 25.170 18.010 10.871 -——--— ----—- ------ --------------- 7.407 5.700 8.034 131.689 89.329 61.956 24.818 -—---— -—--—- ------ ---------- ' 2.657 1.730 1.337 2.214 3.172 3.322 3.395 8.236 7.003 _____- —————————— 11.036 6.804 6.832 7.494 10.275 12.253 10.343 24.420 19.397 ------- 9.748 24.356 1.195 2.683 1.950 1.966 2.024 2.549 2.904 3.425 4.285 3.248 —--—-- >—---- ------- 35.444 78.305 5.654 10.795 7.029 6.834 6.901 8.817 9.749 11.335 13.644 10.389 -----— ---—-- 3.929 2.991 3.825 3.007 3.036 2.931 3.557 3.877 4.414 5.222 4.243 2.866 ------ 15.120 12.603 14.422 10.245 9.901 9.182 11.035 11.772 12.954 15.193 12.730 12.348 ------ 5.617 2.643 2.052 2.756 2.698 3.104 3.069 3.795 4.059 4.681 4.996 5.045 3.531 15.998 7.780 7.012 8.929 8.561 9.548 9.173 11.217 11.762 13.126 13.855 14.226 12.305 3.530 2.822 2.683 2.179 2.204 2.460 2.728 3.220 3.405 3.742 3.974 4.085 3.009 ------ 9.545 8.039 8.096 6.834 6.750 7.319 7.787 9.042 9.623 10.168 10.768 11.328 9.555 ------ 48.352 6.321 4.174 2.916 2.449 2.317 2.430 2.824 3.291 3.671 3.958 4.151 3.830 2.493 ------ 126.247 15.664 10.803 8.011 6.822 6.147 6.276 7.074 8.221 9.202 9.761 10.390 9.811 7.176 ------ 21.245 2.911 3.148 2.070 2.050 2.051 2.334 2.357 2.811 3.349 3.663 4.041 3.614 2.126 44.232 7.086 7.889 5.358 5.163 5.035 5.451 5.138 6.100 7.358 8.076 9.193 8 529 5.535 3.938 2.374 2.127 2.049 2.193 2.259 2.210 2.767 3.298 3.701 3.854 1.813 ----- 9.058 5.697 4.837 4.551 4.567 4.357 3.972 5.100 6.319 7.424 8.020 4.261 ----- 2.452 5.078 2.401 2.190 2.257 2.468 2.390 2.446 2.256 3.073 3.818 4.135 1.751 ----- ------ 5.668 13.066 5.678 4.727 4.636 4.697 4.130 3.902 3.242 4.674 6.342 7.336 3.654 »- -—- 2.391 2.098 1.932 2.309 2.572 2.530 2.616 2.693 2.709 3.390 3.905 1.654 ----- 5.101 4.734 4.073 4.464 4.566 4.058 3.765 3.517 3.053 4.217 5.443 3.025 ----- ------ 3.018 1.941 2.171 2.805 2.600 2.741 2.804 3.310 2.890 3.259 ------ 6.879 4.248 4.190 4.926 4.111 3.763 3.373 3.429 2.515 3.155 ................... 6.299 1.051 1.791 2.891 2.327 3.038 2.877 3.414 2.953 2.204 1. ................... 9.562 1.869 3.021 4.503 3.497 3.913 3.285 3.182 2.398 1.311 1. ————————————— 16.369 9.050 7.632 6.739 5.947 4.673 2.266 ------------------------ 22.507 9.323 7.231 5.900 4.663 3.211 1.042 __..- Note: Upper values are k , and lower values are r , . 1.0.14 1.17.17 67 68 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED Table 61. Estimates of pmnetera ducnlbing m-pmter gm distribution for conditional step lengths (Run 16) -3.6 -3.2 -2.8 -2.4 -2.0 -l.6 —1.2 -o.8 -0.4 0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 ~3-6 III III: III: III: a” II: III: I I: III: _2 8 ---------- 5.681 4.261 3.772 ————— ' ---------- 21.885 19.393 16.773 ----- _2 4 ----- 135.437 9.454 4.295 4.711 4.291 ————— ' ————— 586.986 41.739 19.532 21.727 19.827 """ _2 0 ----- 62.463 8.215 5.714 6.087 6.458 5.900 4.866 5.661 5.608 5.490 6.116 18.045 ' ----- 216.406 36.777 26.418 28.158 29.150 26.179 21.406 23.906 22.900 21.582 24.537' 64.800 -1 6 ---------- 2.394 3.328 4.132 4.384 4.693 4.555 3.993 4.276 4.802 4.853 5.439 6.211 3.281 5.163 ---------- ‘ ---------- 11.129 15,462 18.339 19.832 20.503 19.708 17.016 17.424 19.117 18.607 21.686 23.920 12.521 22.873 ---------- _1 2 3.142 4.194 4.138 3.956 4.284 4.400 4.031 4.299 4.938 5.052 5.435 5.547 3.322 4.694 3.587 ----- ' 13.795 18.928 17.772 16.890 17.622 17.771 15.889 16.263 18.243 17.997 20.028 20.464 11.853 18.549 11.634 ----- _0 8 3.690 4.803 4.034 3.601 3.294 3.302 3.284 3.359 4.029 4.198 4.651 5.716 4.189 7.061 3.559 ‘ 14.462 19.308 15.977 13.964 12.275 11.992 11.667 11.416 13.521 13.799 16.156 19.897 14.742 25.279 10.440 ----- -0 4 2.925 2.939 3.256 3.221 2.938 2.831 2.722 2.772 3.368 3.828 4.683 5.946 5.154 6.478 3.681 ----- ‘ 11.182 10.678 11.749 11.089 9.679 9.026 8.382 8.140 9.859 11.166 14.386 18.529 16.019 20.270 9.931 ----- 0 0 3.532 2.817 3.070 2.849 2.773 2.675 2.492 2.370 2.862 3.441 4.509 5.788 4.421 5.293 4.024 ----- ' 11.954 8.665 9.617 8.460 7.907 7.305 6.523 5.797 6.973 8.455 11.896 15.477 11.416 14.063 9.896 ----- o 4 3.364 3.918 3.757 3.613 3.435 3.206 3.063 2.940 2.563 2.904 4.213 5.504 4.610 5.710 5.125 ----- ' 9.732 10.685 10.403 9.727 8.790 7.814 7.095 6.336 5.030 5.691 9.159 12.449 10.586 13.824 11.372 ----- 0 8 3.265 4.022 3.815 3.892 3.509 3.410 3.414 3.308 3.118 2.482 3.622 4.970 4.222 5.911 . ' 7.879 9.320 8.716 8.951 7.744 7.175 6.784 6.107 5.357 3.770 6.099 9.065 8.183 12.507 11.858 ----- 1 2 ---------- 6.718 3.439 4.970 4.279 3.730 3.616 3.660 3.682 3.508 3.385 2.824 3.891 3.566 5.731 5.744 ----- ' ---------- 9.614 7.109 10.570 8.805 7.334 6.667 6.270 5.884 5.255 4.699 3.319 5.284 5.506 10.247 10.160 ----- 1 6 .......... 13.908 5.723 4.523 4.388 4.330 4.134 3.837 3.574 3.276 2.470 3.064 5.706 6.780 ----- ' ---------- 25.145 10.188 7.630 7.144 6.413 5.726 4.977 4.342 3.650 2.336 3.662 8.206 10.252 ----- 2 0 ................. 11.319 5.784 7.319 6.227 6.116 5.776 5.423 5.164 4.679 3.478 5.760 9.043 ----- ' ----------------- 20.498 9.938 12.266 9.384 8.679 7.740 6.882 6.186 5.189 3.339 6.070 11.367 ----- 2 4 19.639 ----- 14.600 12.857 12.217 11.434 10.590 9.548 7.000 ‘ 27.769 ————— 17.053 13.886 12.387 10.851 9.309 7.657 4.704 2.8 II II III: III: III... III: 3.2 """""""""""" Note: Upper values are k! , and lower values are r 11!!! 1,8,2} Table 62. SUPPLEMENTAL DATA TABLES Estimates of parameters describing Wo-pamnttar gamma distribution for conditional step lengths (Run 17) -1.6 -1.2 -0.8 -0 4 0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 ---------------- 10.359 —- --- —- ——---- -- --— --— -- ----- ----- ---------------- 36.804 .- .-. —— -—---- -- --- ... -- --—-— ----— ------ 3.809 8.836 8.241 7.700 6.331 5.013 6.530 9.808 9.729 -<-—— --—-- ------ 15.335 33.338 27.640 24.594 20.203 15.280 20.934 35.113 32.185 -—--— ----- _2 0 ---------- 10.425 11.042 4.746 5.772 5.512 5.685 5.746 5.535 6.645 8.953 11.706 ' ---------- 42.931 40.844 16.332 19.158 18.106 17.646 17.530 16.605 20.134 28.855 33.573 —1 6 8.015 11.541 4.473 4.622 5.329 5.681 5.634 6.434 7.086 8.966 14.322 10.660 ---------- ‘ ---------- 29.296 42.620 16.488 16.042 17.692 18.237 17.329 19.661 21.136 26.289 42.666 28.068 ---------- _1 2 ---------- 2.551 4.813 5.392 5.101 4.782 4.961 5.105 5.904 6.113 6.579 6.507. 3.562 ---------- ' ---------- 7.820 16.214 18.377 15.901 14.385 14.496 14.386 16.624 16.926 17.790 17.615 8.959 ---------- -0 8 ---------- 2.522 3.222 4.116 4.373 3.909 4.220 4.715 5.478 6.042 7.137 7.306 5.777 ---------- ' ---------- 6.699 9.709 12.418 12.201 10.406 10.861 11.783 13.563 14.893 17.422 17.615 13.051 ---------- -0 4 ---------- 3.020 4.175 4.419 4.221 3.821 3.671 4.047 4.784 5.616 6.178 5.452 6.962 ---------- ' ---------- 7.599 11.572 11.972 10.388 8.996 8.194 8.665 10.186 12.080 13.511 12.098 13.909 ---------- 0 0 ---------- 12.906 4.481 4.557 4.341 3.955 3.705 3.570 4.057 5.168 6.016 5.425 ‘ ---------- 37.143 11.306 10.842 9.346 8.073 7.165 6.397 7.161 9.375 11.292 10.535 0 4 2.186 6.195 4.646 5.651 5.267 4.430 4.079 3.961 3.417 4.468 5.850 5.642 ‘ ----- 4.614 13.201 9.862 11.817 9.986 7.809 6.739 5.993 4.705 6.350 8.898 9.169 0 8 ---------- 3.727 7.933 7.789 6.922 5.443 5.247 5.183 4.762 3.930 4.710 5.227 ' ---------- 6.097 14.977 14.379 11.740 8.562 7.682 6.904 5.805 4.201 5.345 6.748 1 2 ................ 17.600 7.368 6.032 5.714 5.161 5.355 4.990 4.635 3.708 4.411 ' ................ 27.878 11.944 9.204 8.000 6.606 6.137 5.230 4.362 2.914 4.327 1 6 ---------- 12.123 6.129 6.063 8.316 6.503 5.966 5.625 5.224 4.618 3.385 ' ----- 15.578 9.316 9.337 10.719 7.738 6.192 5.316 4.393 3.348 1.845 2 0 .......... . ---------------- 10.074 6.320 5.851 5.394 5.137 5.114 ' .......................... 10.960 5.991 5.067 4.278 3.668 3.216 24 III: 2.8 ---------- Note: Upper values are 1‘1 11,1] , and lower values are r 1.8.11" 69 70 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED Table 63. Results of goodness of fit teat for conditional atep lengths (Run 4A) YE=yi = .0 8 YEP-V7; e 0.0 YEW-i = 0.3 Y =y D 1 '7 k ,—l Goodness k , Goodness k , Goodness 1,y,y r1 ,3/ m. . of 3/ 1,y,y r1 7 m. . of 1,y.y r y m. . of it'1 “my “'7 Fit Test— ft-l 'y-y “'7 Fit Test ft-l "-‘W ""7 Fit Test —1.6 2.449 6.822 128 :2 < x: 2.127 4.837 146 :2 < :8: 1.932 4.073 60 x2 < r: -1.2 2.317 5.147 217 :2 < I: 2.049 4.551 263 .232 < x: 2.309 4.464 124 22 < z: —0.8 2.430 6.276 336 x2 < x: 2.193 4.567 440 x2 < x: 2.257 4.566 213 12 < at: .0.4 2.824 7.074 423 x2 < x: 2.259 4.357 603 :2 < I: 2.530 4.048 306 22 < z: 0.0 3.291 8.221 457 12 < as: 2.210 3.972 725 z2 < I: 2.616 3.765 403 11 < x: 0.4 3.671 9.202 420 12 < I: 2.767 5.100 666 :2 < :0: 2.693 3.517 455 62 < x: 0.8 3.958 9.761 307 :2 < x: 3.298 6.319 468 1‘2 < x: 2.709 3.053 470 x2 < :0: 1.2 4.151 10.390 181 12 < .2: 3.701 7.424 253 :2 < I: 3 390 4 217 284 x2 < x: 1.6 3.830 9.811 48 :2 < x: 3.854 8.020 56 x2 < x: 3 905 5.443 67 x2 < x: y E[X|YE=y. YD=y'] 3/ (EIXIYfI/x I’D=y'])2 3/ , = —,\—— r , = T"—‘—‘ :2 = critical chi-square value lvlf’y Var[XI.Y :31, ll =y'] 144,14 Var-[XII =y, Y =y'] at a significant level of E D E D 0 05 Table 64. Results of goodness of fit test for conditional step lengths (Run 16) YE=yi = -0.8 YE=yi = o o YE=yi = 0.8 y0=yj 1/ k ,— Goodness k , Goodness k , Goodness 1,y,y 141 .E/ m. . of 3/ 1,y,y r1 . m. . of 1,y,y rl , m. . 0 ft” ’y’y “J Fit Test— ft‘l 'y’y "J Fit Test ft-l 4* "-7 Fit Test -1.6 3.601 13.964 155 :02 < x: 2.849 8.460 184 :2 < at: 3.892 8.951 124 :2 < :8: -1.2 3.294 12.275 239 12 < x: 2.773 7.907 294 :2 < :4: 3.509 7.744 200 :2 < z: -0.8 3.302 11.992 293 :2 < :0: 2.675 7.305 376 x2 < x: 3.410 7.175 254 :2 < x: —0.4 3.284 11.667 341 :22 < x: 2.492 6.523 442 :2 < I: 3.414 6.784 297 x2 < I: 0.0 3.359 11.416 365 x2 < 120 2.370 5.797 482 22 < r: 3.308 6.107 317 :2 < 3:: 0.4 4.029 13.521 337 62 < x: 2.363 6.973 444 :2 < 2:: 3.118 5.357 325 x2 < z: 0.8 4.198 13.799 272 :02 < I: 3.441 8.455 350 x2 < I: 2.482 3.770 331 :2 < I: 1.2 4.651 16.156 146 x7 < I: 4.509 11.896 197 :22 < x: 3.622 6.099 207 02 < x: 1.6 5.716 19.897 52 x2 < x: 5.788 15.477 77 3:2 < a: 4.970 9.065 83 :2 < x: y y 3/ €lewa rD=y'] k] 1 = —-—‘——A _ _ I 431111 Val-[XIYE-y, I’D—y ] A 2 (EIXIYE=y. YD=y'1) r = A——————— 141441, Var[){[l/E=y, YD=y'] x: = critical chi-square value at a significant level of 0.05 Table 65. Results of goodness of fit test for conditional step lengths [Run 17) YE=yi = -0.8 YE=yi = 0.0 YE=yi = 0.8 Y =y . D '7 ,l/ Goodness k , Goodness , Goodness 1.9.9 y 1.14.11 1.ny r1 , m. . of 3/ r1 , mi . of r} , m1. . of ft" .y,y 1"7 Fit Test— ft'1 'y'y "7 Fit Test ft“ ,y,y "7 Fit Test -l.6 4.116 12.418 109 :2 < :5: 4.557 10.842 129 x2 > I: 7.789 14.379 68 :2 < z: —1.2 4.373 12.201 259 :2 > x: 4.341 9.346 301 12 < :0: 6.922 11.740 159 x2 < 3:: -0.8 3.909 10.406 406 :2 > x: 3.955 8.073 493 2:2 < x: 5.443 8.562 267 :2 < 2:: -0.4 4.220 10.861 487 =2 < I: 3.705 7.165 632 12 < x: 5.247 7.682 357 3:2 < z: 0.0 4.715 11.783 502 2:2 < x: 3.570 6.397 704 x2 > x: 5.183 6.904 411 :2 < x: 0.4 5.478 13.563 449 x2 < 2: 4.057 7.161 647 :2 > as: 4.762 5.805 425 x2 < z: 0.8 6.042 14.893 324 4:2 < x: 5.168 9.375 469 x2 < at: 3.930 4.201 427 x2 < x: 1.2 7.137 17.422 155 :2 < x: 6.016 11.292 235 :52 < x: 4.710 5.345 247 12 < x: 1.6 7.306 17.615 53 :2 < I: 5.425 10.535 80 :2 < I: 5.227 6.748 85 :2 < I: _1_/ 2/ 3/ Emma. 4347') k = —.___ l-M' \fa‘rlxlrfy. YD=y'1 " 2 (Elxlrfy, Ifw) 1.8.3' _ v3}[x|yE=y, YD=y'] as: = critical chi-square value at a significant level of 0.05 71 ..... i.-. ---.. --... .--.. 1.-. 9n 1... 3m. .1: m3; --- - N.n ..... «3; l-.. .1: wané 1.-. . ”N .m ZN New ~08 N 3 i=3 0.: “5:3 no: New 233: 330 2... "302 «NV. 8.3 .1: 8m. ommN l-.- «N N. N 08. SN; N34 NR; cGN SYN 9N , 30.8 n .N. .NE SN. SN; m8. m3; SmN 3.: a; o m.N NoN. Nan; N3. N3: NSN Ba; N.N N2. :3; m3. E: NSN an; a... 9 8°“ 8°. Nso. o8. S... n3. NtN . . . . . . . 3 25 So. So. 8n. 5.. NZ. o.N m; N? N as N as N E N NN... N N a mum. 8... So. N2. NNN: 8n. SN. N: NS. 53 m3; mNN: NSN mNoN ed can So. So. N8. N3. 2:. 5:. N.N N: N o8. N8. N2. 8N: Nam. 8N. n... 3». N34 :8; N2: SNN mNNN Yo- . . . . . . . - o3; ace. 3.. N2. SN. 8m. 02. v... o N n3 N m m N NmN N a a 1.; m m N: N . N . . . N3." N8. N8. 2N. RN. 8... «2. ed 3.. a3; RN; NNN N NNm N van N N N- exam :5. ”No. 3;. Sn. NS. 21 .2? So; N3; 3; SEN ax: SYN c4. amok N8” n3. 2:. aNN: NS. ”No. ”6. 36 N 8° NS. NE. N2. 2:. :5. N4. a8; 3m; EN; EN 3o...) NNmN 9N. . . . . . . . - So 8°. «8. mNN. NNN: NE. 2o. 0.7 x: N m3 N :3 :n N «N... N Nan N v N . . . . . . . NNN So. So. oNN. 3n. 0:. NS. ON. 2» N 8N N NNN N a: o m8 m NNN N N. N- 3 o8. Sc. :N. an. 3m. 98. TN. ..... 036 can; ---.. NSN a: N5. m 83. 89° 3nd «.2... N36 89° «N. ..... n3; ..... -.--. mNoN .1: ed. .. N n. .N... u. 9». mé ma 9N m4 m5 2 =3. 3 5a 5 =3. 2 =2. 2 :3. SN :3. N. 5w .m .N d u :2»: 3.5 N. N. :3 m N: ASE»??? um .N$“NNN_:M umN.NwNwMNNNmEm N. m a»: N: 55 N. 3 3 n "BmewN m3» .3 .5395 was ENNSSNENN NSSNNNEE 39$ .3 033 :arEé 2a Sufi: 33.33%“. kc Euugfim :33 2?.ng m3“ k0 8.9%: N39 :83 Hagan—£350 kc «633565 .mo gnu... .m :m Now uoom H on 33.0 own rim—N3 noun you 2.33: mm»? NE... "302 SUPPLEMENTAL DATA TABLES N. N N: . . N812 u . .Eww .n . :u now uoom .— 3 Hugo 0.5 “sum—N: moan now n53: mm“: 2F .30: o 8°. 8o. :8. 8°. 08. 8°. 8°. 8c. 0.” NN 8c. 25. So. So. 25. 2:. SN. N8. Na N5. N. N NN 08. 8o. 8... 3o. NNN. NNm. NE. :o. ”N ”3.8 . .sw w .5 8o. 08. 2... NS. 3N. m3. NnN. N2. «N o --.-- --.-- i--- ..--- 1N SN 80. So. 3... m8. 8N. Non. «RN. NNo. 9N 3 23. 08. n8. m2. 9:. NNN. o.N ”No 08. 08. So. :3. SN. mNm. SN. 98. 9N 2m o8. 8... 8... N8. "8. SN. .3 at; 8... N8. N8. m8. EN. 2;. SN. 08. N4 5; 8°. o8. N8. 2o. 33. NNN. N.N NS.N 8... N8. «N8. N8. m2. 3N. 8N. m2. ”a 2m; 08. 25. m8. N2. N8. NmN. SN. 2:. N... 08; go. 80. 8... ms. 93. SN. 8v. NmN. 3. NS.“ o8. so. So. So. 39 NmN. :q. NNNN. .2. 2a; :5. m8. N8. N8. EN. SN. «2. :3. Q... So.” 8o. :5. 3o. .3. N8. 3N. 3.. SN. as mmN.N N8. es. N8. N:. NNN. 3N. NmN. 08. To. :8; N8. N8. m8. .8. m3. 3N. m2. «3. .1? N3; :5. NS. NS. mNN. NNN. 8N. SN. NB. 3.. 8...; N8. N8. woo. N8. oNN. mNN. 03. NS. as. N3; N8. 29 3o. o2. NNN. 2n. NNN. «N9. N4- N2; so. :0. 8o. 3N. MEN. 2”. SN. N8. 04. 2: N8. N8. 08. N8. 02. NmN. ON... :8. N.N. 02 N8. m8. m8. «mo. ,5. 8n. 2:. 3°. .2. mom 3°. 3°. m3. m3. 3N. Sn. :2. 80. ON. Nan 8o. «8. ms. :o. NNN. NNN. ”an. as. o.N‘ n: :5. N8. 8o. NE. SN. N3. 5. as. N..N- EN 08. So. So. So. mfl. «on. 8,. «8. YN- 8 N8. mNc. 3o. 34. NNN. SN. 3N. vmo. N3. 3 89° o8... 89° 305 5.0 was 3:. 83. ”N. 2 8o. o8. m8. 3N. 8n. NmN. mNN. 08. Na- 0 o8.o 89° 89° 08.0 NmNd 2m... mNNd o8... 5... N5, N “N 3. m.m mé ma .N m; We NNNN . .s . u N N. N mN ma m.m m.« m.» m.N 3 3. N. d .NH :5: “38 . .sw Nun» um .NH :5: was N. a a _NN N5 :3: 2: E N m .9333 nut we 8:25“ 23: ENNNQNNENN Ngowfimgoo 33.3 .8 032. n Ga :35 awn—m3 :me Nexumrg mm»; kc .6393.“ was: nuwNnNacagm Ngofiufitg 398m. .5 3n: .u 3a you uoou a 3 3:30 0.: «59.3 noum how 23:: an»? 2F "30: --- So. '8. :1 m3. an. m2. 25$ ---- --- -.-- --- - - -- mé - -- 2-... £8. 28. TN :8. $8. .38. £8. ed -.-- --- 38. SS. 38. 88. c." --- --- $8. ammo. 88. 88. NJ ---. --- 88. $8. 3.8. 38. a... ---.- was. 3.8. 88. 83. To 38. £8. 33. 28. #8. =5 -..- 88. 28. 28. 28. $8. «6- -.-- 88. 83. £8. 33. 88. n.0- --- 38. $8. :3. 58. R8. ~4- ---- 88. SS. 88. B8. 0.7 --- 88.0 $8. $8. 88. o.~. --- --- 38. Soc. 38. v.- :1: 886 886 886 u.”- --.- -- - --.- --- N.»- mé Wm mé a...” m.~ m4 m6 um .mu Jug: «36 n . E .55 333 » km .Sfiuaomwm kc :38:me V5 @533 mam kn 533.5% was huufimauoosm «5% 383m. .2 03¢... .u :u new 33 u 3 13.8 a: 33:3 noun .Sw “53: m3? 2:. "302 'wu.s. Government Printing Office: 1977-777-102/13 STOCHASTIC ANALYSIS OF PARTICLE MOVEMENT OVER A DUNE BED 72 u x .m :a 3m uoou a 3 :55 a: 93:3 noun new 253: m3? 2:. uuuoz So. woo. ono. no". ANN. «on. «3. who. a .3 m n-«uun .uuu-u --- Inn-nu --c autun- u-u-n- run-nu o.~ «co. doc. .30. muo. og- oou. Hue. mna. nmhvkm - goo. 38. 38. goo. Zoo. Nooo. Vé uu-nnu --- unlo- nun-.. ii: in -u-as- nooo. AZoo. mmoo. oooo. 9.8. :8. o.~ --- oooo. :8. Zoo. :8. wnoo. mnoo. o.N -- - mooo. omoo. oNHo. $3. :8- hmoo. o; nan- 38. mooo. Goo. ohoo. mmoo. nmoo. o..— uuuv-n No8. has. 098. mono. smuo. omNo. mmoo. N.— u--- u-nv-s Hooo. goo. ocoo. chao- owNo. obs. N.“ --- nooo- muoo. :8. acne. movo. ammo. undo. no 1:..- u--. Nooo. 03o. 0mg. ~o~o. novo. mono. m6 -- - oooo. 38. «:0. 83. £3. £3. mouo. v.o n-nn- «coo. mooo. "moo. nn—o. mono. vvoo. Eno- To Soo- Soc. :8. ago. :No. mono. mnno. 23. co 1--- Nooo. oooo. 38. avg. «one. mowo. mnno. o.o Nooo. :8. mvoo. “=0. ammo. vono. :no. 38. «to- «So. Nooo. mooo. nvoo. omuo. oovo. coco. 33. Yo- ..nooo- 20o. ovoo. mus. mVuo. mono. ammo. mnoo. m6: Soc. mooo. ouoo. 023. ”ms. mvno. ammo. 03o. no: 88. mooo. :8. :8. 03o. 020. :3. :8. NA- nooo. Nooo. Zoo. ”moo. 32.... $8. 33. goo. ~4- oooo. mooo. nNoo. «moo. nmoo. o—Ho. mnoo. mooo. 0...”. 3.36 nooo. was. nnoo. nooo. Nwmo. 3.3. mmoo. w.“- gooo. mooo. :oo. 38. Goo. mmoo. snoo. «80. o.- --uu 386 coco-o goo. Goo. mooo. mooo. mooo. o.- oooo. goo. oooo. goo. nuoo. cnoo. :oo. wooo. e.- ------ ---- ---- «ooo. aooo. ouoo. mNoo. oooo.o v4... good 336 nooo. goo. muoo. 38. «So. nooo. w.~u ----- - ---- --- oooo.o oooo.o oooo.o oooo.o n-uuu- m.- n-u-u- --- mooo.o moooo :85 good noooo ooooo N.n- I}--- 1-.-- --:- -u-uu --- -nlu- ---- .u---- N.»- mK m6 .m.m m... m.n m.N m4 m.o mK m6 m.m m4 win win m4 m.o R. Q . on .n . 8 .un £5. 3.5 a .3. Jun: ":8 N. .u 3: . a n nix c: :3: A.» 3 m §$E&% 9:. 55 2. 3 m .5.»..»&% kn romuuafim EB 3&me mu: kc gmuufifi. a3... nufiwfiaozm “$8.“. 395m. .: 032. Kc €325? 98 38:3 mm: a? 5393“» :3: EEEEEA «50m. 398m. .on n32. Intrusive Rocks Northeast of Steamboat Springs, Park Range, Colorado By GEORGE L. SNYDER with a section on GEOCHRONOLOGY By CARL E. HEDGE GEOLOGICAL SURVEY PROFESSIONAL PAPER 1041 Delineation of intrusive rock types in a previously unstudied area of Colorado and comparison with related rocks elsewhere in Colorado and Wjioming UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON: 1978 UNITED STATES DEPARTMENT OF THE INTERIOR CECIL D. ANDRUS, Secretary GEOLOGICAL SURVEY H. William Menard, Director Library of Congress Cataloging in Publication Data Snyder, George Leonard, 1927- Intrusive rocks northeast of Steamboat Springs, Park Range, Colorado. (Geological Survey Professional Paper 104!) Bibliography: p. 39 Supt. of Docs. no.: I 19.16:1041 1. Intrusions (Geology)—Colorado—Steamboat Springs region. 2. Geology, Stratigraphic—Pre-Cambrian. 3. Geology, Stratigraphic—Tertiary. 4. Geology——Colorado—Steamboat Springs region. 5. Intrusions (Geology)—The West. 1. Hedge, Carl E. [1. Title. III. Series: United States Geological Survey Professional Paper 1 04 1 . QE611.869 551.8’8 77—608250 For sale by the Superintendent of Documents, US. Government Printing Office Washington, DC. 20402 Stock Number 024-001—03 131 -1 CONTENTS Page Page Abstract ................................................ 1 Geochronology by C. E. Hedge ............................ 17 Introduction and acknowledgments ....................... 2 Petrography and chemistry of intrusive rocks .............. 18 General geology ......................................... 2 Tertiary intrusives ................................... 19 Precambrian intrusive rocks .............................. 5 Tertiary dikes versus Precambrian dikes .............. 23 Mount Ethel pluton .................................. 5 Northwest versus northeast porphyry dikes ............ 24 Granodiorite and diorite ......................... 8 Pegmatites — are they related to a particular magma Quartz monzonite porphyry of Rocky Peak ........ 8 series? ............................................ 24 Quartz monzonite of Roxy Ann Lake .............. 11 Mount Ethel and Buffalo Pass plutons, similarities and Fine-grained granite ............................. 12 differences ........................................ 25 Leucogranite .................................... 12 Accessory fluorite ............................... 26 Buffalo Pass pluton .................................. 12 Economic implications ........................... 32 Quartz monzonite and granodiorite augen gneiss of Correlation of Park Range rocks with other igneous rocks Buffalo Mountain .............................. 13 of Colorado and nearby Wyoming ................... 33 Equigranular quartz monzonite gneiss ............ 13 Sherman--Silver Plume dichotomy ................ 36 Fine-grained porphyry dikes .......................... 13 Mafic dikes ..................................... 38 Pegmatites .......................................... 15 References cited ......................................... 39 Small mafic and ultramafic intrusives ................ 16 ILLUSTRATIONS Page FIGURE 1. Index map to localities mentioned in text ......................................................................... 3 2. Photograph of North Fork of Fish Creek looking towards the Mount Zirkel Wilderness .............................. 4 3. Geologic map and section of part of the northern Park Range ..................................................... 6 4-—8. Photographs: 4. From summit of Mount Ethel showing granite and quartz monzonite in cirque headwall ..................... 8 5. Contacts between the Mount Ethel pluton and its country rocks ............................................ 9 6. Contact relationships within the Mount Ethel pluton ...................................................... 10 7. Igneous flow structure in Mount Ethel plutonic rocks ...................................................... 11 8. Quartz monzonite and granodiorite augen gneiss of Buffalo Mountain cut by quartz latite dike ............... 13 9. Geologic map of part of northern Park Range showing dikes, pegmatites, and ultramafic intrusives .................. 14 10. Drawings of relations between Mount Ethel plutonic variants and fine-grained porphyry dikes ....................... 15 11. Photograph of peg'matites on Lost Ranger Peak .................................................................. 16 12-—14. Rb-Sr isochron plots: 12. Mount Ethel pluton ..................................................................................... 17 13. Buffalo Pass pluton ..................................................................................... 18 14. Feldspathic biotite gneiss ............................................................................... 18 15. Rittmann comparison of northern Colorado Tertiary igneous rocks ................................................ 24 16. Plots of normative minerals of Park Range Precambrian plutonic rocks compared with experimental and statistical data 27 17. Map showing locations of accessory fluorite in rocks of part of the northern Park Range ............................ 29 18. Photomicrographs of fluorite, Mount Ethel pluton . . . . . . . . . . . . . . . . . .' .............................................. 30 19. Diagram showing assigned temporal relations of Precambrian geologic events, Colorado and Wyoming ............... 34 20. Ternary diagrams comparing Park Range rocks with other Colorado plutonic rocks ................................. 37 TABLES Page TABLE 1. Rb~Sr data for Mount Ethel pluton ............................................................................... 17 2. Rb-Sr data for Buffalo Pass pluton .............................................................................. 18 3. Rb-Sr data for feldspathic biotite gneiss samples ................................................................. 18 4. Petrographic summary of Park Range rocks ..................................................................... 19 5. Chemical, modal, and spectrographic analyses of Park Range rocks ............................................ In pocket 6. Locations and normative mineral constituents of analyzed rocks .................................................. 20 7. Fluorite in Park Range rocks ................................................................................... 28 III INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO By GEORGE L. SNYDER ABSTRACT Major Precambrian and minor Tertiary intrusive rocks northeast of Steamboat Springs in the Park Range between 40°30’ and 40°45' N. lat. are described and compared with related rocks elsewhere in Colorado and Wyoming. The Precambrian intrusives were emplaced in a sequence of high-grade interlayered felsic gneisses, amphibo- lites, and pelitic schists of sedimentary and volcanic origin. These rocks are cut by a major northeast-trending Precambrian shear zone where mainly left lateral movement of 1/2 to 1 mile is certain. Cumulative movement of many miles is possible. The Precambrian intrusives consist of a batholith, the Mount Ethel pluton, a smaller Buffalo Pass pluton, and small dikes or lenses of fine-grained porph- yry, pegrnatites, and ultramafics. The Mount Ethel pluton is an oval shaped body 7 miles wide by about 40 miles long (shown by geophysical data to extend beneath younger sediments in North Park). Outer batholithic contacts are sharp and dip steeply outward at about 85°. Five mappable internal variants consist, in order of decreasing age, of granodiorite, quartz monzonite porphyry of Rocky Peak, quartz monzonite of Roxy Ann Lake, granite and quartz monzonite, andleucogranite. Internal con- tacts between these plutonic variants are sharp, and evidence of li- quid-solid relationships abounds; despite this, all rocks except the granodiorite contribute to an Rb-Sr whole-rock isochron indicating emplacement about 1.4 b.y. (billion years) ago. The most important variants volumetrically are: the quartz monzonite porphyry of Rocky Peak, which forms an irregular 2-mile-thick carapace or mapped band around the west edge of the pluton and is lithologically similar to nearby Sherman Granite, and the quartz monzonite of Roxy Ann Lake, which forms most of the rest of the pluton and is lithologically similar to Silver Plume Granite. An apparent Sherman—Silver Plume dichotomy with similar rock types and similar relative ages is noted throughout Colorado plutons of that age. The Buffalo Pass pluton consists of the quartz monzonite and gra- nodiorite augen gneiss of Buffalo Mountain and equigranular quartz monzonite gneiss. Internal contacts are not exposed. These rocks contribute to an Rb-Sr whole-rock isochron indicating syntectonic emplacement 1.7-—1.8 b.y. ago, essentially the same as the metamorphism of the felsic gneiss wallrocks in the area of this report, and of rocks of Boulder Creek age elsewhere in Colorado. The fine-grained porphyry dikes cut the Buffalo Pass pluton, the ultramafics, and some pegmatites. The dikes are within the age range of the Mount Ethel pluton and are older than the mylonite and shear zones. They occur in both an older northwest-trending and a somewhat younger northeast-trending set but do not appear to change compositionally from one set to the other. Regional con- siderations indicate that they were emplaced between about 1.1 and 1.5 b.y. ago, a time when intermediate to mafic dikes were commonly emplaced throughout Colorado, Wyoming, and southwestern Mon- tana. The pegmatite and ultramafic bodies are not dated directly, but clustering of many pegmatites outside the contacts of the Mount Ethel pluton may indicate a genetic relation of the pegmatites to the Mount Ethel rocks. Fluorite is a common accessory mineral in the rocks of the Mount Ethel pluton; it has not been observed in this area in the petrographically similar rocks of the Buffalo Pass pluton. Fluorite was precipitated most abundantly from the Precambrian magma that formed the quartz monzonite of Roxy Ann Lake. In 70 percent of these rocks fluorite is observed in amounts as great as 2 percent and is successively less abundant in both older and younger plutonic phases. Textural evidence indicates that, although most fluorite is in- tergrown with and contemporaneous with other magmatic minerals, some fluorite is associated with alteration minerals in a manner demonstrating its mobility since its initial deposition. Five areas of economic or potentially economic Tertiary fluorite veins in the North Park area occur in rocks equivalent to the quartz monzonite of Roxy Ann Lake, the unit containing the most interstitial Precambrian fluorite. It seems possible that Tertiary solutions may have redistributed Precambrian fluorite into joints or breccia systems open in the Tertiary, and, if so, the mapped faults and mapped areas of most abundant interstitial fluorite may serve as guides to new economic vein deposits. Twenty-eight chemical, modal, and spectrographic analyses of 24 Park Range igneous rocks are compared with each other as well as with 185 analyses of Colorado igneous rocks of similar age from other sources. The comparison indicates that the Park Range igneous rocks are typical of others of the same age, that Colorado rocks of different ages overlap in composition, but that rocks low in normative orthoclase may be restricted to the l.7-b.y. rocks; whereas Tertiary rocks are comparatively high in alkalis and low in silica. The se- quence of intrusion of the Precambrian plutonic rocks, from mafic to silicic compositions, is similar to that predicted by experimental feldspar-quartz-water systems. The sequence may have resulted from magmatic differentiation at a minimum depth of 11 miles and at temperatures of 705° -740°C. However, disproportionately large volumes of the most differentiated rocks at the level of present ero- sion indicate that differentiation from more mafic rocks by any mechanism must have taken place at deeper levels and that subse- quent upward transportation and differential concentration of silicic derivatives took place. There are two types of fine-grained or glassy Miocene or Pliocene intrusives in the area studied in this report: a dark olivine porphyry and a lighter colored intermediate rock whose mutual contacts are not exposed. In this region dark basalts have been previously shown, asserted, or assumed to be younger than light-colored rocks that have been called trachytes or rhyolites. A review of the field, chemical, and temporal data, however, indicates that this is an oversimplifica- tion. Basaltic rocks have been erupted at several times in the Terti- ary in conjunction with other rocks of mafic, intermediate, silicic, and subsilicic compositions. Available K-Ar dates in this area sug- gest that rocks of the compositional range olivine basalt through an- desite and trachyte to rhyolite are all 10—11 million years old. 1 2 INTRUSIVE ROCKS NORTHEAST 0F STEAMBOAT SPRINGS, PARK RANGE, COLORADO INTRODUCTION AND ACKNOWLEDGMENTS The area of this report in the Park Range of Colorado lies along the Continental Divide from Buffalo Pass north into the southern half of the Mount Zirkel Wilderness (figs. 1, 2). The terrain is one of rugged mountains that have more than a mile of vertical relief. Drainage is by the headwaters of the Elk and Yampa Rivers on the west and by the North Platte River on the east. The area was covered by an icecap in the Pleistocene Epoch (Atwood, 1937). Vegetation ranges from prairie grassland at the lowest elevations through discontinuous successive zones of scrub oak, aspen, lodgepole pine, and spruce to alpine meadows above timberline. Credit for the geologic mapping is given in figure 3. Snyder was in charge of the field mapping and field in- terpretation. Credit for the chemical, modal and semi- quantitative spectrographic analyses is given in table 5. Carl E. Hedge was in charge of performing the radio- metric measurements and interpretations. Zell E. Peterman gave assistance with a computer program for calculating the norms from the chemical analyses. GENERAL GEOLOGY The Park Range owes its topography and relief to tectonic uplift and to resistance to erosion of Pre— cambrian crystalline rocks, mainly metavolcanic and metasedimentary felsic gneisses, amphibolites, and mica schists, and intrusive into them, quartz mon- zonites and granites with which this report is mainly concerned. The sillimanite-grade metamorphic rocks have been extensively folded and faulted during many episodes of deformation, but few of the structures have been accurately delineated or appraised, owing to the lack of detailed mapping and the paucity of regionally traceable stratigraphic units. In the central part of the area the intrusive rocks form an exposed pluton 7 miles wide and more than 17 miles long, whose major axis lies in a northeasterly direction, and several smaller bodies. The crystalline rocks are overlain by rare Paleozoic, ex- tensive Mesozoic, and scattered Cenozoic continental and marine sediments, as well as by Pleistocene tills and outwash deposits (fig. 3). All the bedrock units have been broken by faults. The Precambrian rocks in the southern part of the area have been mylonitized locally and displaced regionally along a pervasive series of northeast-trending shear zones (figs. 2, 3). The relative offset along various shears of this zone is mainly left lateral with minor right lateral movement as shown by drag of the con- tacts of two pelitic schist units and one quartz monzonite body and as shown by the offset of lithologic units including a series of vertical porphyry dikes per- pendicular to the shear zone (porphyry dikes shown diagrammatically in fig. 9, not shown in fig. 3). The nonsymmetrical nature of the drag of the contacts of the pelitic schist of Soda Mountain demonstrates that movement on the different mylonite planes was episodic rather than simultaneous. For example, the ex- treme nonsymmetrical drag southwest of Soda Moun- tain shows that the south contact of the schist of Soda Mountain was not moving as a unit with the north con- tact. The south contact appears to have been drag folded in response to movements along the nearest shear zone southeast of the schist; whereas the north contact was dragged in response to earlier or later movements along the shear zones parallel to and inter- secting the north contact. If all the drag has taken place without significant differential stretching of the folded units, the minimum amount of shearing move- ment needed to cause the dragged contacts would be roughly equal to the sum of the lengths of the short sides of the drag folds. Differential stretching might be counteracted by continuous sliding on fault planes without further drag. However, it is not possible to assess the relative importance of these opposite effects. Notwithstanding, it is interesting to compare the ap- parent displacement for different units, as in the following table: Amount of apparent displacement Unit Left lateral Right lateral (miles) (miles) Porphyry dikes .................... 0.4 0 Quartz monzonite augen gneiss of Buffalo Mountain ................ 1.5 0.4 Schist and gneiss of Lake Dinosaur . 4.2 0 Schist of Soda Mountain ........... 8.7 1.9 Total‘ ...................... 14.8 2.3 1 12.5 miles effectively left lateral. In this table the amount of displacement given for the vertical porphyry dikes and for the quartz monzonite augen gneiss of Buffalo Mountain is the amount of documentable apparent offset measurable along a fault plane; this must be a minimum as other offsets indi- cated by sheared porphyry dike inclusions within mylonite have not been estimated. In contrast, the amount of displacement given for the granodiorite and diorite and for the schist of Soda Mountain is mainly the sum of the lengths of the short limbs of their drag- ged contacts. Two factors are evident: (1) there is a positional or geographic effect— most of the minor right lateral sense of movement has taken place in the GENERAL GEOLOGY 112' 102° “0'._ 103- . ____1o4° T- "—‘-‘[—"——"_- —‘_" 9a 0 ,. Burtooth Mts o “’4. Guy/Ore, ’W ‘1; W "*0 Y o M ‘9, ’a 9 ,- . 9v an,“ Mt: 1’0 6‘ 0/ 6‘9 % °Centennial ‘2, 9:; CHEYENNE ' 4' o --— ..______ 1’7 o‘1},"JelmMo_I_.m_taiL___ 0,9 % Lu °Virginia Dale Harms Paakx‘o g 13%?! 6'; g°Log Cabin I Elkhead Mts x / \ g, < . 5 RABBIT ' Mayhem Milne!» K EARS ”.23...” p... Steamboat" 0 RANGE Springs 0 Hot Sulfur 3‘ Springs e H'd ' 66% "Pia“? oc‘l, Pslllve' g DENVER 71:74ume° m °Morrison , e ’1, LL m 7 i LYA 0 R A D 0 9, ‘8, “ers‘ctt'tflgsf‘ ’Lo 6‘ 38' ‘ ~-————J-—___l__- _i_ __ __ J____-_ _——J--—-——""“ ’- 0 so 100 150 200 MILES I i . . L l 1 | J l—r r I I i I i I o 50 100 150 200 KILOMETERS FIGURE 1. — Localities mentioned in text. east-central and northeastern part of the zone; and (2) there may also be a time effect inasmuch as the porph- yry dikes, which are the youngest because they cut the quartz monzonite augen gneiss of Buffalo Mountain and the schist of Soda Mountain, have been displaced least; the augen gneiss, which cuts the granodiorite and diorite and schist of Soda Mountain, is displaced more; and the latter two units are displaced most. The porpho 4 INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO Mt Zirkel / FIGURE 2. — View north-northeast from North Fork of Fish Creek to Mount Zirkel Wilderness. From foreground to middle distance the northeastward-striking alinement of the topography is controlled partly by northeastward-striking gneisses and schists and partly by northeastward-striking mylonites of the Soda Creek-Fish Creek mylonite zone. Rocks exposed in this zone are felsic gneisses, amphibolites, and mylonites along the north wall of the North Fork yry dikes may in fact occupy extensional joints formed either actually or latently along planes about perpen- dicular to the fault movement direction during early shear movements. At any rate, the apparent cumula- tive effect of all the offsets in the Precambrian rocks is about 12.5 miles in a left lateral sense. Faulting was renewed in this area at the close of the Mesozoic, was continued through the Eocene, and was renewed yet again at the end of the Miocene. A11 Mesozoic and Eocene formations are found faulted against the Precambrian generally along steep to shallow east-dipping thrusts on the west side of the Park Range and along steep west-dipping reverse faults on the east side of the range. Near the heads of Soda Creek and Newcomb Creek, north and northeast of Buffalo Pass, Mesozoic fault movement characterized by brecciation has taken place along the old Pre- cambrian mylonite zones previously described. Breccia is also present along other faults cutting only Pre- Mt Ethel / Buffalo Pass of Fish Creek, older sheared quartz monzonite of Buffalo Mountain complexly interlayered with other rocks on the plateau above Fish Creek, and pelitic schist on Soda Mountain. The rocks exposed beyond Soda Mountain to the snowy peaks of the Mount Zirkel Wilderness in the distance and also on Sheep Mountain are mainly unsheared granite and quartz monzonite of the younger Mount Ethel pluton. cambrian rocks and may represent post-Mesozoic movement, but this is difficult to prove in the absence of transected Mesozoic rocks in contact with the faults. (Farther south brecciation related to northeast-trend- ing shear zones has been related to late Precambrian time (Tweto and Sims, 1963, p. 991, 1006).) Many non- faulted contacts between lowermost Paleozoic or Triassic rocks and Precambrian rocks exist, and several large and small synclines of mantling sedi- ments are present within the area of Precambrian rocks forming the Park Range; the largest one extends from Big Creek to Hinman Park. Post-Miocene faults are documented in the southwest, northwest, and northeast corners of the area. In the southwest and northwest the Browns Park Formation has been offset as much as 1,000 feet; in the northeast post-Miocene fluorite veins (Steven, 1957, p. 337; 1960, p. 376, 397; Hail, 1965, p. 118) have been emplaced along some of the youngest faults. The northeastern fault that is PRECAMBRIAN INTRUSIVE ROCKS 5 shown extending east out of the area south of Roaring Fork is probably continuous with the post-Miocene Spring Creek fault of Kinney (Behrendt and others, 1969, p. 1525, figs. 1, 2; Behrendt and Popenoe, 1969, fig. 2). In addition, abundant conglomerates and uncon- formities in the Mesozoic and Tertiary stratigraphic section show that the Park Range was a positive area subject to repeated upwarpings during the Mesozoic and Tertiary. Other times of faulting for which there is no direct evidence here have been documented just out- side of this area. PRECAMBRIAN INTRUSIVE ROCKS The Precambrian intrusive rocks in this part of the Park Range fall naturally into five or more variants of one large complex body and four groups of smaller bodies. Textures of the more felsic representatives tend to be hypidiomorphic-granular or gneissic-granular, whereas many of the more mafic rocks are diabasic; rocks of any composition may be porphyritic (table 6). The rock groups are described below in order of their areal importance because it is not possible to be certain of the exact time relations either among all the groups or among all the variants within each group. After the section on radiometric ages, the present state of geochronologic knowledge, including both critical field relations and new radiometric data, is summarized. A complex, elongate batholith, here called the Mount Ethel pluton, dominates the area of exposed Pre- cambrian rocks; it extends from the west flank of the range north of Copper Ridge northeasterly for 15-19 miles and disappears beneath the Mesozoic and Terti- ary rocks of North Park. It reappears in North Park on Delaney Butte and Sheep Mountain (fig. 2) (Hail, 1965, pls. 2, 3). A similar appearing granitic intrusive has been mapped in detail by Steven (1954, 1957, 1960) beyond the northeast corner of North Park in the Medicine Bow Mountains. On the basis of geophysical measurements, Behrendt, Popenoe, and Mattick (1969, p. 1523) speculated that the Mount Ethel pluton “pro- bably extends northeast beneath the North Park basin and connects with granitic rocks in the Medicine Bow Range.” If this is true, the Mount Ethel pluton would be 6—7 miles wide by about 40 miles long. This report will be concerned mainly with the segment within the Park Range. The four groups of smaller Precambrian intrusive bodies are as follows: 1. A small pluton extending about 10 miles through Buffalo Mountain across Buffalo Pass to the east margin of the range, here called the Buffalo Pass pluton, and numerous tiny separate intrusives most of which are within 2 miles of the Buffalo Pass pluton. 2. Fine-grained porphyry dikes, too small to be resolved in figure 3, but located diagrammatically in figure 9. 3. Pegmatites, coarse-grained granitic intrusives, also shown in figure 9. 4. Medium- to coarse-grained mafic and ultramafic in- trusive bodies, several of which are shown in figure 9. Uppermost Miocene or Pliocene (post-Browns Park) intrusive rocks occur locally as thin dikes or sills only in the northwest quarter of this area but are not mapped separately for this report. These rocks are all fine grained, locally glassy or porphyritic, and intermediate in composition (tables 4, 5, 6). All are on the eastern fringe of, and should be considered part of, the Elkhead Mountain eruptive field (fig. 15). MOUNT ETHEL PLUTON Rocks of the Mount Ethel pluton are remarkably homogeneous and generally sparsely jointed, and form prominent ledges, canyon walls or cirque headwalls (fig. 4) where the topography is steep, or form large rounded bosses or tors where the terrain is gentler. Quartz, pink and white feldspar, and biotite are visible in all hand specimens, and hornblende and muscovite were observed locally. Except for isolated flat-topped nunataks (a good example is shown in fig. 53), all the area within the Mount Ethel pluton has been glaciated, and the plutonic rocks have contributed numerous large joint-block erratics to the glacial tills lying beyond the boundaries of the pluton itself. The outer contact of the pluton against its country rocks generally is planar or slightly curved and dips outward at 80° -90°. Complex convolutions or reen- trants are present locally, as along the southeast side of the pluton from the head of Soda Creek to the head of Newcomb Creek. Many excellent exposures demonstr- ate the persistence of this steep dip along many miles of contact and in varied topography (fig. 5). Where the contact is convoluted, however, it is irregularly oriented and shows inclusion or apophysis walls dipping from 0° —90° or changing from one to the other within a few hundred feet. Flow structures within the pluton are generally fairly steep but are less prominent in the center of the pluton than near the contact. On the basis of magnetic and gravity anomalies and gradients over the Mount Ethel pluton along the Conti- nental Divide, Behrendt, Popenoe, and Mattick (1969, p. 1532, 1533, fig. 6) have postulated a thick platelike pluton whose contacts dip gently outward. This struc- tural interpretation is not supported by the field facts. Figure 3 shows the line of their profile A—A’. This profile crosses the northern contact east of Pristine Lake where their interpretive cross section shows a 6 INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO W W'. 33 x) 215mm L O 1 2 3 4 MI 5 Les \/ 0 1 2 3 4 5 6 7 8 KILOMETERS FIGURE 3. —Geologic map and section of part of the northern Park Range, Colorado. Section A—A’ is the line of profile used by Behrendt, Popenoe, and Mattick (1969, pl. 1). Geology is mainly by George L. Snyder, 1965-67 and 1975, Warren B. Hamilton, 1965, Fredric Hoffman, 1966, and Ronald L. Bonewitz, 1967. Mesozoic and younger geology along the east margin of the map is mainly from W. J. Hail, Jr. (1965, 1968). contact dipping 35° outward for about 1 mile and then upper shallow-dipping part of the postulated contact 80° outward for about 2 miles; the profile crosses the probably is incorrect for the following reasons: southern contact northeast of Soda Mountain where 1, The profile, intended to be nearly perpendicular to they show a contact dipping 10° outward for about 1 the map trace of the contact, makes an angle of mile and then 70° outward for another 11/2 miles. The 35° with the trace of the northern contact; PRECAMBRIAN INTRUSIVE ROCKS EXPLANATION Browns Park Formation and post-Browns Park porphyry intrusives TERTIARY Coalmont Formation Mainly Mesozoic and minor Paleozoic sedimentary rocks PAL E0 ZOIC Unsheared quartz monzonite and granite Y PRECAMBRIAN Y Ya, aplite and leucogranite Yg', fine-grained granite Yra, medium-grained biotite granite and quartz monzonite ofRoxy Ann Lake er, coarse-grained biotite-homblende quartz monzonite porphyry of Rocky Peak Yd, medium-grained hornblende-biotin granodiorite and diorite Sheared quartz monzonite Xm, medium-grained biotite granite and quartz monzonite gneiss Xb, coarse-grained biotin-hornblende quartz monzonite and granodiorite augen gneiss of Buffalo Mountain xax X9 xu W PRECAMBRIAN X Gneisses and schists (includes pegmatite) XE, biofite gneiss, hornblende gneiss, and amphibolite Xp, mainly pelitic schist Xs, pelitic schist of Soda Mountain Xd, biotite schist and gneiss of Lake Dinosaur 2. The southern contact is intensely convoluted where the profile crosses it; large inclusions of country rock lie just inside the contact and large granitic apophyses occur just outside of it; 3. Within a distance of half a mile outside the north- ern contact very large pegmatites constitute as much as half the volume of the amphibolitic country rock (figs. 50, 11); AND MESO ZOIC Contact Fault Strike and direction of Clip of bedding GB Horizontal ._L Inclined _p_ Overturned Strike and direction of dip of foliation + Inclined + Vertical .631 Sample locality / 8 Photograph or diagram showing figure number in this report Arrow indicates direction of view a: / I , , 409‘5,106 .2 30 45 106°37 30’ 2 V l- s s s " N 4" 4‘ u m “'9 § « z .n < a. .x ’V 'o _ m _. - g is § 4; a .. u 0, \ *° \ a >- E 37’30” x ‘- EN 30 g g n“ N V «a '0 n 2 $15" 48:? j E < q. ¢ < I Q 3 40°30’ INDEX TO OUADRANGLES COVERED BY GEOLOGIC MAP .Area of this report Slllmbol: Spnnp .DENVER COLORADO 0 200 MILES 0 200 KILOMETERS 4. The southern end of the profile crosses a completely different body of granitic rocks, which may ex- pand tremendously in volume within a few miles beneath the surface. All the above observations were not known to Behrendt, Popenoe, and Mattick, and so a new in- terpretation of the geophysical data can now be made. Five mappable variants occur within the Mount 8 INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO FIGURE 4. —View south from summit of Mount Ethel showing typical granite and quartz monzonite exposed in cirque headwall in center of Mount Ethel pluton. Scale shown by man (arrow). Ethel pluton. All of these variants possess knife-sharp contacts against other internal variants of the pluton. Invariably, these contacts give clear, unambiguous evi- dence that one variant originally was solid and one 1i- quid and that one was older and one younger. Further- more, these relations are consistent from area to area. Figure 6 shows typical observed internal contact rela- tions. Despite these clear liquid-solid relationships and the time span implied for the cooling of at least five magmatic variants, no thin, chilled contact margins of one variant against another have been observed (but see the discussion on p. 13), and contact hornfelses are very rare. Apparently, the temperature of the entire region was elevated throughout the intrusion of the Mount Ethel pluton. Reaction rims exist around gra- nitic inclusions but are rare. Local margins of horn- blende- or biotite-rich rock at the edges of inclusions in- dicate that felsic constituents have been selectively removed and added to the magma in contact with the inclusion. The rarity of these margins may indicate that either most inclusions did not form reaction rims or they were not being selectively dissolved at this level of the magma chamber. Flow structures, alined platy feldspar phenocrysts or tabular inclusions, are promi- nent in some variants of the Mount Ethel pluton and are rare in others (fig. 7). The time trend of composition is from mafic to felsic. The five variants, in order of age from oldest to youngest, are granodiorite and diorite, quartz monzonite porphyry of Rocky Peak, quartz mon- zonite of Roxy Ann Lake, granite and quartz monzo- nite, and leucogranite. They are described in this order. GRANODIORXTE AND DIORITE The granodiorite and diorite unit, the oldest recog- nized variant of the Mount Ethel pluton, consists of uniform tough dark-gray medium-grained subdiabasic hornblende-biotite granodiorite and diorite. The rocks of this unit always occur as inclusions in younger units, never demonstrably in their original position of solidifi- cation and never in contact with the metamorphic country rocks. The granodiorite inclusions are con- tained in or cut by the quartz monzonite porphyry of Rocky Peak, the biotite granite and quartz monzonite of Roxy Ann Lake, and the aplite and leucogranite in the southwest corner of the pluton and are contained in a small stock of fine-grained granite in the northeast corner of the pluton. The one inclusion large enough to show in figure 3 occurs in the Gunn Creek drainage in the southwest corner of the pluton and is completely surrounded by quartz monzonite porphyry of Rocky Peak. This inclusion measures about one-third of a mile wide by 1 mile long, and it is cut by a few narrow dikes of thr quartz monzonite porphyry of Rocky Peak (fig. 6A) and the granite and quartz monzonite of Roxy Ann Lake. An irregular zone, as much as 1 mile wide, of the quartz monzonite porphyry of Rocky Peak around this large inclusion has a higher proportion of hornblende than is normal for the porphyry elsewhere. QUARTZ MONZONITE PORPHYRY OF ROCKY PEAK The quartz monzonite porphyry of Rocky Peak con- sists of uniform, red and reddish-gray, medium- to coarse-grained granular biotite-hornblende quartz monzonite containing tabular euhedral microcline phenocrysts as much as several inches long (fig. 7A). It is exposed in boss and tor topography and locally weathers to granule-covered griis knobs. This porphyry forms the southwest, west, and, locally, the north con- tact of the Mount Ethel pluton and is distributed as an irregular 2-mile-thick carapace on the west end of the pluton. The quartz monzonite porphyry is included in and intruded by dikes and complex apophyses of the quartz monzonite of Roxy Ann Lake (figs. 6B, 6E) and . by dikes of leucogranite and fine-grained granite. The contact between the quartz monzonite porphyry of Rocky Peak and the granite and quartz monzonite of Roxy Ann Lake is particularly irregular with apophyses of the latter being found in the former as much as 1 mile from the main contact and with inclusions of the former found as much as 3 miles into the latter. The quartz monzonite porphyry of Rocky Peak rarely comes PRECAMBRIAN INTRUSIVE ROCKS 9 FIGURE 5.—0rientation of the contact between the Mount Ethel pluton and its country rocks. A. Geologist pointing to contact between quartz monzonite porph- yry of Rocky Peak (er) and interlayered amphibolite and felsic biotite gneiss (Xg) north of the South Fork of Mad Creek just east of where it joins Mad Creek. Contact, at ex- tended finger, is nearly vertical or steeply outward dipping (85° in nearby stream exposure). Country rocks to left of finger dip shallower than contact. Quartz monzonite porph- yry to right of finger is relatively homogeneous but has flow structure parallel to the contact and is not composed of alter- nating mafic and felsic layers as it might appear from the drip stains on the vertical cliffs. B. Contact between layered amphibolite (Xg) and quartz monzonite of Roxy Ann Lake (Yra) exposed on the southeast side of The Dome. Contact dips about 80" out- in contact with the fine-grained granite found mainly west of Luna Lake. However, it is possible that some of the smaller detached dikes mapped as the Roxy Ann Lake unit are really misidentified fine-grained granite. Because of the abundance of microcline phenocrysts and the contrast in size between them and their , ‘wg'm‘m"‘F0rk} —_ * _ .i ‘ , .' I: W171?" ‘2‘ ward. An aplitic apophysis (Ya) high on The Dome cuts the amphibolite layering at a shallow angle. C. Looking south from the head of Wolverine Basin across in- terlayered amphibolite (Xg) and pegmatite to a contact with quartz monzonite of Roxy Ann Lake (Yra) that just cuts into the head of the glacial cirque above Pristine Lake. Trace of the contact across the cirque headwall is indicative of its steep dip. White outcrops to left of Pristine Lake are pegmatite. D. Looking west at contact between quartz monzonite of Roxy Ann Lake (Yra) and hornblende gneiss (Xg) on the north side of Roaring Fork of Red Canyon. Contact, which is here about perpendicular to the layering of the hornblende gneiss (not prominent in photograph), dips 85° outward. Two nearly horizontal quartz monzonite apophyses (Yra) cut the hornblende gneiss. groundmass, flow structures, generally nearly vertical, are prominent everywhere within the quartz monzonite of Rocky Peak (figs. 5A, 6A, 6B, 60, 7A). Although most of the few granodiorite inclusions are in the quartz monzonite porphyry of Rocky Peak, the latter unit con- tains very few inclusions of metamorphic country rock 10 INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO FIGURE 6. — Contact relationships within the Mount Ethel pluton. A. Dike of quartz monzonite porphyry of Rocky Peak (er) cut- ting tonalite along the western contact of the granodiorite (Yd) in the north fork of Gunn Creek. Large, fluidally ar- ranged microcline phenocrysts in quartz monzonite porph- yry are parallel to dike contacts. B,E. Irregular contact at 9,400-foot elevation in the east fork of Gunn Creek where fine-grained quartz monzonite of Roxy Ann Lake (Yra) intrudes coarse-grained quartz monzonite porphyry of Rocky Peak (er). Prominent foliation of quartz monzonite porphyry (er) is truncated at contact and quartz monzonite of Roxy Ann Lake (Yra) contains local large microcline xenocrysts. C. Closeup of dike contact at 10,040—foot elevation on west side of Summit Park near head of Hot Spring Creek. Very fine grained granite dike (Yg) intrudes fine-grained quartz monzonite of Roxy Ann Lake (Yra). D. Inclusion swarm of medium-grained quartz monzonite of Roxy Ann Lake (Yra) in fine-grained granite body (Yg) from 10,500-foot elevation half a mile northwest of the west end of Luna Lake. PRECAMBRIAN INTRUSIVE ROCKS 11 FIGURE 7. —— Igneous flow structure in Mount Ethel plutonic rocks. A. Typically excellent flow structure in quartz monzonite porph- yry of Rocky Peak near outer contact of pluton three- fourths of a mile north-northwest of Rocky Peak. Foliation plane, shown by large microcline phenocrysts raised above the general rock surface by weathering, is parallel to plane of hammer (as well as parallel to contact plane not shown). and only a few dikes of the quartz monzonite porphyry penetrate country rock. This contrasts with the Roxy Ann Lake unit and may indicate that, while the latter was emplaced largely by piecemeal stoping of relatively small blocks, the porphyry was emplaced largely by block-caving or cauldron subsidence of relatively large blocks. QUARTZ MONZONITE OF ROXY ANN LAKE The quartz monzonite of Roxy Ann Lake is the most extensive variant of the Mount Ethel pluton and oc- cupies nearly the entire width of the pluton throughout much of its length. This variant is intruded by numer- ous dikes and small plutons of fine-grained granite and by rare dikes of leucogranite. The rocks are gray to red- dish-gray, fine-, medium-, and coarse-grained granular biotite granite and quartz monzonite with rare to very rare tabular euhedral phenocrysts of microcline as much as twice the size of the groundmass minerals. Because of the scarcity of large phenocrysts, flow folia- tion is generally difficult to discern, but in the best ex- posures (for example, fig. 73) vertical flow foliation is visible. The quartz monzonite of Roxy Ann Lake com- monly is exposed in fresh roches moutonnees. In Red Canyon, however, where the rocks have been broken by Tertiary faults and have been extensively mineralized B. Flow structure in quartz monzonite of Roxy Ann Lake in out- crop at 10,480-foot elevation on ridge east of Rosa Lake in the southwest corner of the Mount Ethel quadrangle. Folia- tion as shown by orientation of sparse large feldspar crystals is parallel to pencil. Although this foliation exam— ple is exceptionally good for the quartz monzonite of Roxy Ann Lake, it is still distinctly less pronounced than that typical for the quartz monzonite porphyry of Rocky Peak. and altered, hematitic grfissy badland cliffs predomi- nate. The grain size of the quartz monzonite of Roxy Ann Lake increases gradually from west to east in the western one-fourth or one-third of its area and then is fairly uniform throughout the rest of the pluton. Apophyses of this quartz monzonite unit cutting the western carapace of the quartz monzonite porphyry of Rocky Peak are generally fine grained; the west- central area of this unit (from Soda Creek to the South, Middle, and North Forks of Mad Creek) is generally medium grained, whereas north of Luna Lake and east of the Continental Divide the unit tends to be coarser grained. Compare rocks of the unit in figures GB, 60, and 6E from the western part of the area with those in figure 78 from somewhat farther east and, in turn, with rocks in figure 6D still farther east. This decrease in grain size westward may result from cooling by the large masses of porphyry and gneiss included within the western part of the quartz monzonite unit. There are a few areas both east and west of the Continental Divide near the southern margin of the pluton where the quartz monzonite of Roxy Ann Lake may be a multiple or composite unit. Here, several outcrops dis- play sharp contacts between a coarse-grained and a medium-grained variant, and it is possible that another small pluton exists here, perhaps equivalent to the fme- 1 2 INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO grained granite. However, it has not been possible to map such a pluton. FINE-GRAINED GRANITE Pinkish-white to pinkish-gray fine-grained granite forms many dikes and small plutons in the area bet- ween Lake Margaret and Luna Lake, in the Whalen Creek area, and in the northeast corner of the Mount Ethel pluton. As was mentioned previously, some of the rocks of this unit may be mistakenly mapped as quartz monzonite of Roxy Ann Lake. The cluster of dikes and small plutons in the Lake Margaret —Luna Lake area suggests that the present level of erosion may be just above the top of a single roughly cylindrical stock about 3 miles in diameter (fig. 3, sec. A—A’). This granite is the youngest widespread variant of the Mount Ethel pluton and is cut only by rare dikes of leucogranite. LEUCOGRANITE Pink aplite and fine-grained leucogranite form the youngest, but areally minor, variant of the Mount Ethel pluton. This variant is best developed as a series of north-trending dikes as much as several tens of yards wide and 1 mile long in the southwestern corner of the pluton. A few other dikes of it occur along the north- west margin and in the eastern part of the pluton. BUFFALO PASS PLUTON The Buffalo Pass and smaller related plutons consist of two mappable variants: The areally most important one is well exposed on and near Buffalo Mountain (figs. 2, 8) and will henceforth be referred to as the quartz monzonite and granodiorite augen gneiss of Buffalo Mountain; the other is a medium-grained equigranular quartz monzonite gneiss present in some of the less well exposed parts of the Buffalo Pass pluton and in most of the smaller plutons, especially those intruding the biotite schist and gneiss of Lake Dinosaur. The field relations are suggestive but not conclusive concerning both internal and external age relations of the Buffalo Pass pluton. The augen gneiss of Buffalo Mountain and its accompanying equigranular quartz monzonite gneiss crop out close to each other in several areas, mainly in forested griissy areas at low altitudes, but have not been seen in contact in outcrop or in glacial boulders. So far as is known no rocks of the Buffalo Pass pluton contact any rocks of the Mount Ethel pluton. Therefore, any tentative field conclusions as to the internal or external relative ages of the variants of the Buffalo Pass pluton must rely on in- direct evidence. The several varieties of such evidence follow. 1. The quartz monzonite and granodiorite augen gneiss of Buffalo Mountain is compositionally very similar to the quartz monzonite porphyry of Rocky Peak; the equigranular quartz monzonite variant of the Buffalo Pass pluton is composi- tionally very similar to the quartz monzonite of Roxy Ann Lake. The main difference in hand specimens between these formations is textural, not compositional. The different variants of the Mount Ethel pluton have cooled from an un- disturbed melt; they have relict igneous textures and flow structures (fig. 7) and euhedral pheno- crysts. The different variants of the Buffalo Pass and smaller related plutons, however, completed their crystallization in a tectonic and high-grade metamorphic environment; they have gneissic textures with anhedral augen. 2. The fine-grained porphyry dikes at first glance ap- pear to be the youngest Precambrian rocks. Some of these rocks, which will be described in more detail in the next section, cut most of the major Precambrian units including the Mount Ethel pluton (fig. 10A) and the Buffalo Pass pluton (fig. 8). But incontrovertible evidence has been dis- covered (fig. 103) that some porphyry dikes are older than some Mount Ethel rocks. The north- erly orientation of the Newcomb Creek porphyry dike cut by the quartz monzonite of Roxy Ann Lake is similar to the north-northwesterly orien- tation of the whole porphyry dike series within the Soda Creek —Fish Creek mylonite zone, several of which cut the augen gneiss of Buffalo Mountain (figs. 8, 9). If all dikes of similar orien- tation in this area were intruded at the same time, the largest part of the Mount Ethel pluton would clearly be younger than the largest part of the Buffalo Pass pluton. In this connection it is interesting that Steven (1957, 1960) in the Northgate area (northeast of the area in this report) mapped north-northwesterly trending dacite porphyry dikes short segments of which are “found in xenoliths in the intrusive quartz monzonite stock” (probably equivalent to the quartz monzonite of Roxy Ann Lake) and “are definitely older than the enclosing granitic rock” (Steven, 1957, p. M364, pl. 48; 1960, p. 333, pl. 12). It is even more interesting that the radiometric ages to be described later are signifi- cantly older for the Buffalo Pass pluton than for the Mount Ethel pluton. PRECAMBRIAN INTRUSIVE ROCKS 13 FIGURE 8.—Quartz monzonite and granodiorite augen gneiss of Buffalo Mountain cut by a dark fine-grained quartz latite dike. West (right) contact of 5-foot-wide dike dips 65° W. in foreground; east contact of dike exposed in next outcrop on strike uphill. Dike can be traced with minor offsets 2,000 feet up this hill and down the other side to one of the main Fish Creek mylonite zones where it is truncated and disappears. Behind camera, dike can be traced 1,000 feet with minor offsets to another mylonite zone in a fork of Soda Creek where it is offset 1,200 feet in a left lateral direction. Vehicle in background is parked on Buffalo Pass road near BM 10088. QUARTZ MONZONTTE AND GRANODIORITE AUGEN GNEISS OF BUFFALO MOUNTAIN The largest unit of the Buffalo Pass and related plutons is the quartz monzonite augen gneiss of Buffalo Mountain, a unit consisting of uniform, red and red- dish-gray, medium- to coarse-grained biotite quartz monzonite containing anhedral microcline augen as much as several inches long. Its texture is gneissic gra- nular indicating that it has been extensively sheared and recrystallized during and after its emplacement; subsequently, several portions of the unit west of Buffalo Pass have also been severely mylonitized or crushed without recrystallization. The rock is quite dis- tinct from the interlayered felsic biotite gneiss and amphibolite country rock, but, where its microcline augen are small, it may resemble the accompanying equigranular quartz monzonite. The map pattern of the unit clearly indicates an intrusive origin, particularly on and east of Buffalo Mountain where the unit splits into three or more layers interleaved with the felsic biotite gneiss and amphibolite country rock and east of Buffalo Pass where the unit truncates the boundary between the biotite schist and gneiss of Lake Dinosaur and the biotite gneiss, hornblende gneiss, and amphibolite. EQUIGRANULAR QUARTZ MONZONITE GNEISS Accompanying the quartz monzonite and granodio- rite augen gneiss of Buffalo Mountain in the Buffalo Pass and related plutons is a gray medium-grained equigranular biotite granite and quartz monzonite gneiss. Although the augen gneiss seems to occur as larger bodies and the equigranular gneiss as smaller ones, the two are closely associated in several areas. Unfortunately, in all these areas exposures are so poor that it is not known whether their contacts are grada- tional or abrupt. FINE-GRAINED PORPHYRY DIKES Although the bodies of this unit are too small to be shown to scale on the geologic map of figure 3, their positions and orientations have been diagrammed in figure 9. The rocks consist of light-gray to dark-gray to black, fine-grained, hypabyssal, near-vertical dikes with sparse hornblende or feldspar phenocrysts, usually flow alined with the margins of the dikes. Hand specimens show little metamorphic recrystallization. The dikes are usually planar and range from a few feet to several tens of feet wide and several hundred feet to 1 mile long. One dike (table 5, sample 549), on the south side near the top of Soda Mountain, measures 300 feet wide by 1,000 feet long and is coarser grained than nor- mal and diabasic in texture. Because some of the fine- grained porphyry dikes cut the quartz monzonite of Roxy Ann Lake (fig. 9) and others are cut by it (fig. 103), either two ages of quartz monzonite of Roxy Ann Lake or two ages of porphyry dikes exist. Because the quartz monzonite of Roxy Ann Lake is a much more homogeneous rock unit than the porphyry dikes and is areally self contained, it seems more likely that there are two ages of porphyry dikes even though these can- not yet be distinguished. Figure 9 provides some slight additional justification for this position. Two groups of porphyry dikes may be discernible on the basis of their geographic position and orientation. One group located within or near the Soda Creek-Fish Creek mylonite zone is oriented mainly northwesterly. One of these is clearly intruded by the quartz monzonite of Roxy Ann Lake of the Mount Ethel pluton (fig. IOB) and others are believed to be. Several of the dikes are offset by the mylonite zones. (See the table on p. 2.) The other group of dikes, less areally localized than the previous group, nevertheless, appears mainly in the northwest half of the map and is concentrated near the northwest corner of the Mount Ethel pluton. The trends of the dikes of the latter group range from north through northeast to east. Several of these clearly cut the quartz monzonite porphyry of Rocky Peak (fig. 10A), and the quartz INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORAD ) 2/1“» 42‘ W” 2' 3‘ an; x: \ . O ' Prisr/IEiV. A/ L' 11 A “7*" :- -ger Pk Rwy Ann 5/9 Cr Lyn ....... Lolly Ymo , \ fgoum the! Lawn Late [2‘ La” Morgan! '° Ron O 1 2 3 4 5 MILES 0 1 2 3 4 5 6 7 8 K|LOMETER$ EXPLANATION Tertiary porphyry intrusives and Tertiary, Mesozoic, and Paleozoic sedimentary rocks Younger Precambrian Y rocks of Mount Ethel pluton Older Precambrian Y rocks of Mount Ethel pluton Precambrian X rocks of Buffalo Pass pluton Precambrian X metamorphic wallrocks Contact P Pegrnatite --------- Strike of thin, nearly vertical, fine-grained M Small ultramafic intrusive porphyry dikes \10A Locality of a figure indicated by its number— Arrow indicates direction of view FIGURE 9. — Simplified geologic map of part of the northern Park Range showing location and orientation of near-vertical fine- grained porphyry dikes and location of some pegmatites and ultramafic intrusives. PRECAMBRIAN INTRUSIVE ROCKS monzonite of Roxy Ann Lake and others are believed to do so. How fine-grained dikes could be intruded both before and after the emplacement of a coarse-grained batho- lith poses a problem. The older northwest-trending dikes resemble the younger northeast-trending dikes in degree of crystallinity; therefore, the plutonic magma could not have metamorphosed the northwest dikes sig- nificantly or they would appear different from the younger dikes; this is consistent with the minimal effect that the plutonic magma had on its country rocks (p. 8). The older dikes are thin and uniformly fine grained, perhaps owing to cold wallrocks or rapid loss of volatiles at the time of intrusion. Then the region became warm enough, and cooled slowly enough, to allow five successive plutonic phases to form coarse- grained rock without obvious chilled zones. (Four of these phases were intruded at the same level in the crust.) Furthermore, most superheat (if any), heat of crystallization, heat of solidification, or heat of cooling must have been dissipated upward into rocks now removed instead of outward into rocks now exposed. Then, the area cooled and was intruded by the younger porphyry dikes whose crystallization was also inhibited. All this must have required a neat balance in regional temperature levels or volatile control. PEGMATITES Large to small bodies of very coarsely crystalline white oligoclase-microcline-quartz pegmatite occur throughout the Precambrian rocks but are especially common locally adjacent to the contacts of the Mount Ethel pluton. Most of the pegmatites are unzoned, but a few contain large books of muscovite or biotite that may be in zones. Pegmatites are concentrated along the northeasternmost contact of the Mount Ethel pluton (figs. 5, 11), on and north of Copper Ridge, and in and near the schist of Soda Mountain (fig. 9). Possibly, these pegmatites are all older than any rocks of the Mount Ethel pluton, but the few presumably cross cut- ting contacts are not well exposed and the concentra- tion near the pluton remains suggestive of a relation- ship with the pluton. Possibly, these pegmatites are older than, but were precursors of, the Mount Ethel magma. At Northgate northeast of North Park, re- placement pegmatite is contemporaneous with or younger than quartz monzonite gneiss (Steven, 1957, p. 350), but, though abundant adjacent to intrusive quartz monzonite (Mount Ethel equivalent), it is shown as older than the intrusive quartz monzonite (Steven, 1957, pl. 48). Within the Mount Ethel pluton only a few pegmatites cut units as young as the fine-grained gra- nite. At least one internal pegmatite has been trun- 15 Flow structure in quartz monzonite porphyry of Rocky / FIow-alinod quartz monzonite // ’ ’ porphyry of Rocky Peak an / / microelino , / / a I /‘ //." /» ‘ ~10 FEET Y” A Flow structu Flow structu FIGURE 10.—Relations between Mount Ethel plutonic variants and fine-grained porphyry dikes. A. Sketch of relations shown in cliff exposure of fine-grained trachyandesite dike (Yp) cutting and including quartz monzonite porphyry of Rocky Peak (er) at about 9,650- foot elevation just north of horse trail east of Swamp Park. Xenocrysts of microcline derived from adjacent quartz monzonite prominent in dike near its contact. Sketch drawn partly from field notes. B. Sketch of relations shown in cliff exposure of quartz monzonite of Roxy Ann Lake (Yra), fine-grained porphyry dike (Yp), and felsic biotite gneiss (Xg) on north side of Newcomb Creek. Note that, while flow structure and con- tact of porphyry dike clearly crosscut layering of gneiss, flow structure and contact of quartz monzonite equally clearly crosscut both of these. Sketch drawn from field notes. cated by granite and quartz monzonite of Roxy Ann Lake at a quartz monzonite porphyry of Rocky Peak contact. Internal pegmatites are volumetrically minis- 16 INTRUSIVE ROCKS NORTHEAST 0F STEAMBOAT SPRINGS, PARK RANGE, COLORADO FIGURE 1 1. — Giant pegmatites in northeast face of Lost Ranger Peak near contact of Mount Ethel pluton. White cliffs, ledges, and talus in distance are complexly shaped pegmatites in dark amphibolite sequence of uniform dip. Frost-wedged boulders in foreground are quartz monzonite of Roxy Ann Lake; contact of Mount Ethel pluton passes between clumps of brush in left middle distance. cule and, although they imply that pegmatites formed over a long time, they probably do not contribute much to the understanding of source and time of origin of most pegmatites. SMALL MAFIC AND ULTRAMAFIC INTRUSIVES Numerous small mafic and ultramafic intrusives oc— cur throughout the metavolcanics and metasedimen— tary rocks. Most of these intrusives are amphibolitic, similar to the metavolcanic amphibolites in both ap- pearance and mineralogy, and they are thouroughly metamorphosed so that no trace of either original mineralogy or original texture is preserved. The bodies are mainly a few tens or hundreds of feet in diameter, and no attempt has been made to map them separately. Possibly, most represent feeder dikes for the now- metamorphosed lava flows they occur with, but some, perhaps many, may represent postdepositional pre- metamorphic intrusive sequences. A few of the mafic and ultramafic intrusives, by virtue of either a preser- ved original texture or an unusual lithology, are worth further mention. Medium- to coarse-grained speckled hypersthene metagabbro with relict igneous texture forms a promi- nent series of ledges above 10,200 feet elevation on the north end of the long ridge between the head of the South Fork of the Elk River and Bear Canyon in the northwest corner of the Mount Ethel quadrangle. The contact with the surrounding meta-amphibolites was not observed. The most noticeable feature of the rock is the clear diabasic texture. Nodules of medium- to coarse-grained dark dunite in GEOCHRONOLOGY 1 7 a bright hematite-red soil cap the east side of the 9,035- foot knoll south of the head of Morgan Creek. As no large outcrops are present anywhere in the vicinity, the dunite fragments must be weathering from a subcrop. This dunite locality is nearly on strike with the pre- viously mentioned hypersthene metagabbro, and the two rocks may be genetically as well as spatially rel- ated. In the field float fragments are dark grayish brown with light-green to white raised lumps (coronas) scattered sporadically over their surfaces. The rock may be of some economic interest since it contains a trace of platinum and palladium. (See table 5, sample 116.) A chromite-containing pargasite (?) peridotite oc- cupies several small knolls on the ridge top near triangulation station “Spring” north of Fish Creek in the southeast corner of the Rocky Peak quadrangle. The contacts of this body are not exposed. Fine-grained porphyry dikes (some of which have been shown to overlap the Mount Ethel pluton in age) cut two of the ultramafic bodies, the chromite-bearing peridotite near triangulation station “Spring” and a diopside hornblendite near the head of the Roaring Fork of Red Canyon. Hail (1968, p. 7) reported a dike of black hornblen- dite cutting quartz monzonite about 15 miles south of the southeast corner of the area of this report. This quartz monzonite is probably like the Buffalo Pass type of this report. Thus, tenuous arguments could be raised to demon- strate that the small mafic and ultramafic intrusives were between the Buffalo Pass pluton and the Mount Ethel pluton in age, but much more work needs to be done before the position in time of the small mafic and ultramafic intrusives is accurately known. GEOCHRONOLOGY By Carl E Hedge As part of this study, the Mount Ethel and Buffalo Pass plutons and one of the metamorphic wallrock units were dated by whole-rock Rb-Sr methods. The same analytical procedures were used as those described by Peterman, Hedge, and Braddock (1968) and Peterman, Doe, and Bartel (in US. Geological Survey, 1967, p. B181--B186). Rb-Sr analytical data for the Mount Ethel pluton are presented in table 1 and shown on an isochron plot in figure 12. The Rb-Sr ratios of the subunits of the Mount Ethel pluton increase with silica content. The ratios of quartz monzonite porphyry of Rocky Peak, the quartz monzonite of Roxy Ann Lake, and the fine-grained gra- nite define a single line on the isochron plot. This colin- earity indicates that these three subunits were all TABLE 1. — Rb-Sr data for Mount Ethel pluton 5313:)?” p13?“ 5;; Rb87/Sr86 Sr87/Sr86 Quartz monzonite porphyry of Rocky Peak 520 1138 1307 1.300 0.7295 525 122 382 .928 .7211 Quartz monzonite of Roxy Ann Lake 523 1184 1196 2.734 0.7570 526 176 263 1.940 .7416 527 171 283 1.759 .7393 Granite and quartz monzonite 521 263 124 6.21 1 0.8273 522 271 100 7.976 .8724 Leucogrnnite 528 1211 120.8 31.08 1.3008 ‘ Concentrations are by X-ray fluorescence; others are by isotope dilution. 0.88 I I I I I l 5221 0.86 _ Age = 1,470 i 50 m.y. _ (Sra7/Sr“ )0 = 0.7021 0.84 _ 0.82 m g 0.80 — a” a 0.78 _ 0.76 _ 0.74 _ 0.72 _ 0.70 I I l l 1 | I l 0 1 2 3 4 5 6 7 8 9 Rba'I/sras FIGURE 12. —Rb-Sr isochron plot for samples of the Mount Ethel pluton. Sample 528 is not plotted — see text. emplaced within a short period of geologic time — pro- bably less than 10 or 20 million years (m.y.). Because of its extremely high Rb37/Sr36, the leucogranite sample (528) is not plotted in figure 12. The sample gives an ap- parent age of 1,370: 40 m.y., which is somewhat younger than the other subunits of the Mount Ethel pluton. The significance of this age difference is ques- tionable, however, because this sample is slightly weathered. Incipient weathering also may be the reason that'some of the other samples deviate from the isochron slightly more than expected only from analyti- cal uncertainty. Assigning the Mount Ethel pluton to the time inter- val of 1,370 to 1,47 0 m.y. makes it part of a major period of plutonism widely recognized throughout the Pre- cambrian of Colorado from the Front Range to extreme 18 western Colorado. (See Peterman and Hedge, 1967, and Hedge and others, 1968.) The age of the Mount Ethel pluton is not significantly different from that of the Sherman Granite or Silver Plume Granite of the north- ern Front Range to which it has distinct lithological similarities. The Buffalo Pass pluton gives an age of 1,8001—100 m.y. (table 2, fig. 13). Again the two subunits are con- temporaneous within analytical uncertainty, and again time-equivalent similar rocks are present in the Pre- cambrian of much of Colorado. In the Front Range rocks of 1,700—1,800 m.y. age are gneissic granodiorite (Boulder Creek Granodiorite) and associated quartz monzonite (Peterman and Hedge, 1967). Augen gneisses about 1,750 m.y. age are common in the southern Front Range and in the Mosquito Range (Hutchinson and Hedge, 1967). Three samples of feldspathic biotite gneiss were analyzed to determine the age of the metamorphic TABLE 2. — Rb-Sr data for Buffalo Pass pluton Sample Rb Srt No. ppm PPm Quartz monzonite and granodiorite augen gneiss of Buffalo Mountain Rb87/Sr86 3,375,86 544 1114 1429 0.771 0.7214 554 197.4 1475 .594 .7163 555 189.9 1416 .626 .7179 Quartz monzonite gneiss 550 142 272 1.518 0.7413 553 125 136 2.689 .7694 ‘ Concentrations are by X-ray fluorescence; others are by isotope dilution. 0.78 l Age=1,8001100 m.y. 553 (Sr‘7/Sr“)o =0.7022 0.76 d g 0.74 _ u“: 0.72 — 0.70 0 3 Rb" ISr“ FIGURE 13. —Rb-Sr isochron plot for samples of the Buffalo Pass pluton. INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO TABLE 3. — Rb-Sr data for feldspathic biotite gneiss samples $31399 FRI)?“ 13;: Rb87/Sr86 Sr87/Sr86 631 88.0 76.6 3.347 0.7806 593 122 107 3.316 .7790 616 86.7 192 1.313 .7316 0.80 I 1’ Age=1,6901100 m.y. (SIa7 /Sl’“)o = 0.7006 0.78 _ 0.76 — E“ 33 0.74 _ 0.72 _ l I I | 0.70 0 1 3 4 2 Rb" /Sr86 FIGURE 14.—Rb-Sr isochron plot for samples of the feldspathic biotite gneiss. rocks into which the plutons were intruded (table 3, fig. 14, petrography in table 5). They give an age of 1,6901100 m.y. Since the gneisses must be older than the Buffalo Pass pluton which intrudes them and their analytical uncertainties clearly overlap, the only meaningful age assignment is that both units are ap- proximately 1,700—1,800 m.y. old. Rocks significantly older than 1,800 m.y. have not been found in Colorado, but Hills, Gast, Houston, and Swainbank (1968) deline- ated a boundary in the Medicine Bow Mountains of southern Wyoming between the 1,700- to 1,800-m.y.-old basement typical of Colorado and the approximately 2,500-m.y.-old basement which appears to occur over most of Wyoming. This age province boundary has not been found in the Park Range, but it must lie to the north of the area of this study. PETROGRAPHY AND CHEMISTRY OF INTRUSIVE ROCKS Data on the mineralogy and petography (tables 4, 5) and chemistry (tables 5, 6) are available for most in- trusive rocks discussed in this report. These data can be PETROGRAPHY AND CHEMISTRY OF INTRUSIVE ROCKS studied, averaged, and combined in many ways. Certain combinations, involving known or potential genetic groupings of the rocks, are worthy of further emphasis. The emphasis will be on whether the petrographic and chemical data contribute to the understanding of the known or potential genetic groupings of the rocks, and the genetic groupings that will be discussed further are as follows (generally from youngest to oldest, or from smallest to largest): 1. Tertiary intrusives (are there two kinds?); Tertiary dikes versus Precambrian dikes; Northwest versus northeast porphyry dikes; Pegmatites (are they related to a particular magma series?); Mount Ethel and Buffalo Pass plutons, similarities and differences, including (a) accessory fluorite, (b) economic implications; Correlation of Park Range rocks with other igneous rocks of Colorado and Wyoming, including (a) Sherman —Silver Plume dichotomy, (b) mafic dikes. 2. 3. 4 TERTIARY INTRUSIVES Early workers in the Elkhead Mountains along the west side of the Park Range were convinced that there were two series of post-Cretaceous eruptives: an early 19 light-colored acid porphyry and a late dark basalt. Later work in this and adjacent areas has proved that there is more than one age of basalt and that there is really a continuum of compositions from basalt to rhyolite rather than a strictly bimodal assemblage. However, the variation of magma composition with time is as yet incompletely understood, and more data are needed. Within the area of the present report two compositionally distinctive Tertiary intrusive rocks are present, but their relative ages are not known. Also, the spread in their compositions is less than half as great as the compositional spread of the continuum as a whole (fig. 15). Analyzed examples of these two rocks are pro- vided in table 5: sample 22 is from a light-colored Ritt- mann trachyandesite sill, and sample 19 is from its feeder dike, from within the area shown as Browns Park Formation and post-Browns Park porphyry in- trusives in figure 3 east of Clark; sample 226 is from an olivine-containing Rittmann dark latite dike, one of a series cutting Mesozoic sedimentary rocks and Pre- cambrian crystalline rocks south of Clark. Since the late 19th century, geologic explorers have generally recognized two series of Tertiary igneous rocks in the Elkhead region west of the northern Park Range: light rocks frequently called trachytes or rhyolite porphyries, and dark basalts. Generally, the TABLE 4. — Petrographic summary of primary constituents of metamorphic and igneous rocks of Park Range northeast of Steamboat Springs, Colorado [Key to constituent abundance: A, abundant in all rocks observed; A, abundant in most rocks observed; a, abundant in a few rocks, not observed in most; C, common in small amounts in all rocks observed; C, common in small amounts in most rocks observed; c, present in small amounts in a few rocks, not observed in most rocks; . . ., study of 540 thin sections] not observed. Tabulation based on Constituents i a 33 3 o g E 3 5 Metamorphic and igneous rocks E 3 5 o :3 g E a: s a s = ’2‘ 5 =~ s 6» 2 i 3 ‘§-§~E'E§ss§§5~~2§to~3 2: g'i ,5; o .. a .. ._ _. _. u gaggs-ezssséfigsaagessgssééags: ~ 2" " "‘ ;:: __ a ~- .— i- _ “a assoSooosms=sssasasssao<5s Tertiaryintrusives ...................................... A c C c a c A C . c c C c . c c a c . Fine-grainedporphyrydikes ............................. A a A c C A c C C C c c c C . Pegmatite .............................................. A A A C c c c C c c c c c Mount Ethel pluton: Leucogranite ......................................... A A A C . . C . . C c c c c c . c_ . Graniteandquartzmonzonite .......................... A A A C . c . A . . C c c C C C c . C . l QuartzmonzoniteofRoxyAnnLake .................... A A ‘A C . c . A . . C c C C C C C c . C . QuartzmonzoniteporphyryofRockyPeak .............. A A A c r l 1 c . A c C c C C C C c c . C . Granodiorite ........................................... A A 1A L.. A ,. A -. . A . . C c C A C C . c . C . 7 Buffalo Pass pluton: Quartz monzonite and granodiorite augen gneissofBuffaloMountain ........................... A A A c A . A . . C c C C C C . . C . Quartzmonzonitegneiss ............................... A A A c A . C c C C C C . c C . Mafic intrusives ........................................ a c . A c A . A c c c c Countryrocks: Felsicgneisstoamphibolite ............................ A a A c 4. C a A . C c C C c c c a . C . Peliticschistandgneiss .......................... A a A A iC ,.. a. 1. .‘. c C A a c c , c c a c c c c 20 INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO TABLE 6. -— Locations and normative [Located in same order as in table 5. Normative Sample No. ““3““ 528 521 522 527 523 526 347 437 157 525 520 551 544 Q ........ 30.87 31.62 32.68 26.97 27.29 27.38 30.98 22.69 24.85 26.38 27.77 11.53 13.53 or . . . . . . . 33.84 30.61 30.16 27.99 28.49 30.38 28.49 25.49 28.83 29.22 27.27 16.50 30.66 ab 32.14 29.10 28.99 30.78 29.94 29.15 30.36 31.14 29.36 28.84 30.46 28.33 30.62 an ....... 1.59 3.95 3.25 7.70 5.67 6.78 5.15 10.87 8.71 7.79 6.73 20.10 14.77 ne ....... di. . .,. 2.23 .51 C ........ .69 1.04 1.57 .90 1.60 1.09 1.07 .38 .50 .94 1.07 . . . . . . hy ....... .12 1.84 1.38 2.42 3.31 2.54 1.75 3.66 2.63 2.19 2.86 9.49 6.17 01 ........ mt ....... .19 .72 .88 1.67 1.53 1.25 1.16 3.08 2.78 2.52 1.81 5.55 1.81 cm ....... . . . hm ...... .34 i1 ........ .09 .26 .24 .64 .76 .56 .41 1.28 1.15 .91 .84 2.57 .90 ap ....... .03 .13 .03 .20 .23 .20 .16 .53 .43 .30 .33 1.24 .33 cc ....... .02 .02 .07 .07 .02 .05 .05 .02 .05 .09 .11 fr ...... . .12 .30 .30 .21 .54 .30 .09 .48 .36 .39 .28 .67 .14 hl ........ . . . .02 .04 .02 .02 .02 . . . .04 .04 .04 .02 .05 .04 (H2017 . .08 .36 .46 .48 .56 .29 .35 .29 .31 .43 .50 1.50 .39 Sample locality and description Sample Map unit Elevation Locality Sample No. (fig. 3) Section T. N. R. W. (ft) description description 528 er SEA 8 7 84 8,750 4,150 ft S. 78" E. of summit of Rocky Peak, Rocky Hypidiomorphic-granular texture with Peak 71/2-minute quadrangle. planes of sheared grains. 521 Yg NEW 25 8 84 10,060 200 ft south of southernmost point on shore of Hypidiomorphic-granular texture. Lake Margaret, southeast corner of Floyd Peak 71/2-minute quadrangle. 522 Yg (Unsurveyed) 8 83 10,600 3,950 ft N. 82° W. of west end of Luna Lake, Do. Mount Ethel 7‘/z-minute quadrangle. 527 Yra (Unsurveyed) 7 84 9,800 1,900 ft S. 73° E. of summit of hill 10,033 near Hypidiomorphic-granular, interlocking head of east fork of Gunn Creek, Rocky Peak texture. 71/2-minute quadrangle. 523 Yra (Unsurveyed) 8 83 11,140 6,750 ft N. 74.5° W. of summit of Mount Ethel, Hypidiomorphic-granular texture. Mount Ethel 71/2-minute quadrangle. 526 Yra NE 1/4 36 8 84 9,460 East of the South Fork of Mad Creek, 4,550 ft. S. Do. 78" W. of summit of Horse Thief Peak, northwest corner of the Buffalo Pass 7‘/z- minute quadrangle. 347 Yra (Unsurveyed) 8 83 9,920 East of the South Fork of Mad Creek, 1 mile N. Do. 43° W. of summit of Horse Thief Peak near boundary between Mount Ethel and Buffalo Pass 7‘/z-minute quadrangles. 437 er NW 1/4 22 8 84 9,350 On north side of hill, 5,550 ft N._57° E. of junction Hypidiomorphic-granular, subporphyritic of North and Middle Forks of Mad Creek. Floyd to porphyritic texture. Peak 7l/z-minute quadrangle. 157 er NW'A 8 7 84 9,100 2,150 ft N. 29° E. of summit of Rocky Peak, Rocky Hypidiomorphic-granular, subporphyritic Peak 71/2-minute quadrangle. texture. 525 er SW'A 14 8 84 9,520 On trail 1,950 ft. S. 55° W. of summit of hill Hy'pidiomorphic-granular, porphyritic 10,129, about 11/: miles east of Swamp Park, texture. Floyd Peak 71/z-minute quadrangle. 520 er SEW. 32 8 84 7,920 1,400 ft. N. 78" E. of summit of hill 8,151, Hypidiomorphic-granular, subporphyritic northeast of the junction of Mad Creek and the texture. South Fork of Mad Creek, Rocky Peak 7%- minute quadrangle. 551 Yd SE'/. 21 7 84 7,680 In tributary of Gunn Creek 5,500 ft S. 67 .5° W. of Hypidiomorphic-granular, subdiabasic peak 9,262, Rocky Peak 7‘/2-minute texture. quadrangle. 544 Kb (Unsurveyed) 7 83 10,088 North side of Buffalo Pass road at benchmark Mortar texture. 10088, Buffalo Pass 7'/z-minute quadrangle. PETROGRAPHY AND CHEMISTRY OF INTRUSIVE ROCKS mineral constituents of analyzed rocks leaders. . . ., indicate not determined] 21 Sample No. — Continued 555 554 550 553 549 524 545 1 16 631 593 616 226 22 19 18.86 17.25 27.39 33.81 2.41 10.79 17.00 . . . . . . . . 5.40 7.35 27.49 29.72 32.50 31.39 12.19 17.28 21.98 .30 . . . 18.78 13.38 14.87 30.30 30.30 25.11 25.74 27.32 30.15 31.42 2.61 . . . 23.59 36.11 35.48 11.76 13.35 5.87 3.75 22.42 17.16 12.80 11.70 13.75 15.73 15.29 .. . .32 . . . . . . 6.82 2.72 . . . . . 16.72 10.64 7.40 .64 1.78 1.58 .56 .18 5.69 5.22 3.31 2.22 19.91 9.38 5.49 29.79 . .. 8.55 11.26 30.77 11.57 2.13 1.97 1.04 .28 2.56 5.63 4.99 16.96 5.54 4.13 3.60 .29 . . . . .87 .79 .74 .29 3.12 3.40 2.34 .35 3.18 2.10 1.77 .33 .30 .26 .10 1.23 1.79 1.33 .05 1.58 .91 1.08 .75 .20 .23 .39 .09 .02 .52 .16 .16 .09 .94 .14 .12 .12 .04 .36 .67 .72 . . . .04 .04 .04 .02 .07 .07 .05 . . . .. . .97 .40 .64 .39 1.41 .78 .62 6.84 2.12 .40 .99 Sample locality and description — Continued Sample Map unit Elevation Locality Sample No. (fig. 3) Section T. N. R. W. (ft) description description 555 Xb (Unsurveyed) 6 or 7 83 9,870 3,900 ft N. 17rE'. of summit ol'BuiTalo Mountain, Mortar texture. Buffalo Pass 7‘/z-minute quadrangle. 554 Kb (Unsurveyed) 6 or 7 83 10,160 Along powerline 3,250 ft S. 15" W. of benchmark Do. 10088, Buffalo Pass 71/2-minute quadrangle. 550 Xm (Unsurveyed) 7 83 10,540 Along powerline 6,600 ft S. 68° E. of benchmark Hypidiomorphic-granular, quartz-mosaic, 10300 near Buffalo Pass, Buffalo Pass 71/2- subporphyritic texture. minute quadrangle. 553 Km (Unsurveyed) 6 83 10,030 Along road 3,750 ft S. 37“ E. of lake Dinosaur, Hypidiomorphic-granular, quartz-mosaic Buffalo Pass 71/2-minute quadrangle. texture. 549 Fine-grained (Unsurveyed) 7 83 10,540 1,250 ft S. 61° W. of Soda triangulation station, Megadiabasic texture. porphyry dike. Buffalo Pass 7‘lz-minute quadrangle. 524 . .do ..... SWV. 14 8 84 9,640 North of trail 1,650 ft S. 39° W. of the summit of Diabasic, porphyritic texture. hill 10,129 about 1% miles east of Swamp Park, Floyd Peak 7‘/z-minute quadrangle. 545 . .do ..... (Unsurveyed) 7 83 10,070 Near Buffalo Pass road 300 ft. N. 68° W. of Do. benchmark 10088, Buffalo Pass 7‘/z-minute quadrangle. l 16 Small ultramafic NWV. 28 9 84 9,005 In bald spot of red soil just east of top of hill 9,035 Myrmekitic, poikilitic, corona texture. intrusive. about 5/4 mile east-southeast of Wapiti Ranch, Floyd Peak 71/z-minute quadrangle. 631 Feldspathic (Unsurveyed) 9 82 9,300 3,000 ft N. 66° W. of west junction of Grizzly- Granular, interlocking texture. biotite gneiss, Helena trail with Lone Pine Creek road, Xg. V northwest corner of Pitchpine Mountain 7‘/2< minute quadrangle. 593 . .do ..... (Unsurveyed) 8 82 9,370 North of Beaver Creek 2,400 ft S. 36.5° E. of peak Gneissic, granular texture. 10,225, northwest corner of Teal Lake 7%- minute quadrangle. 616 . .do ..... (Unsurveyed) 7 83 9,925 Near top of small knoll 3,250 ft N. 50° E. of east Do. end of Round Mountain Lake, Buffalo Pass 71/2- minute quadrangle. 226 Tertiary SE 1/4 16 8 85 7,980 On linear knob 950 ft N. 71.5° W. of hill 8,920 on Diabasic, porphyritic, xenocrystic texture. intrusive. Moon Hill, Clark 7'/z-minute quadrangle. 22 . .do ..... SEW. 36 9 85 8,930 4,600 ft due east of Greenville mine near summit Trachytic, porphyritic, vesicular texture. of hill 8,934, Floyd Peak 71/2-minute quadrangle. 19 . .do ..... NW 1/4 36 9 85 8,590 4,100 ft N. 54° E. of Greenville mine on south side Hyalotrachytic, xenocrystic texture. of knob 8,629. Floyd Peak 7‘/2-minute quadrangle. 22 basalts have been considered to be younger than the porphyries, and specific intersecting field relationships reported over the years, although rare of mention, would tend to confirm this. Tertiary igneous rocks of two distinct types, rhyolitic and basaltic, were men- tioned by Gale (1906, p. 29, 30) in the Hahns Peak area, but no statement of relative age of these two types was provided. In the Yampa coal field Fenneman and Gale (1906, p. 32) spoke of two fairly well defined groups of post-Cretaceous eruptives, and “basalt” is shown younger than “acid porphyry” on their plate 1; but no intersections were mapped. George and Crawford (1909, p. 211,212) noted: “While there are several varieties of porphyry in the (Hahns Peak) district, they may, in the main, be considered closely related phases of a series rather than distinct types. They include rhyolite, dacite, latite, andesite, and quartz basalt.” However, they later (p. 212, 213) spoke of several ages of rhyolite and also noted (p. 215, plate in pocket) a quartz basalt dike cutting rhyolite porphyry southwest of Columbine. This mafic dike is mapped as two dikes by Barnwell (1955, p. 48, pl. 5), both cutting rhyolite porphyry throughout their length; but Segerstrom and Young (1972, pl. 1) recognized only intermediate porph— yry throughout part of this same area. An olivine basalt dike has been reported as cutting a rhyolite porphyry laccolith in the crest of the Tow Creek anticline bet- ween Milner and Pilot Knob (Crawford and others, 1920, p. 36-39, pl. 3; Coffin and others, 1924, p. 57 —58, pls. 1, 3), and this is the basis in this area for the joint conclusion that there were two periods of Tertiary ig- neous activity. However, the critical relationships in the Tow Creek crest have been mapped somewhat differently by Bass, Eby, and Wood (1955, pl. 19) and by Buffler (1967, pls. 1, 2). Christensen (1942, p. 29—32, 55—61, 172-174, geologic map) reported two distinct series of volcanics in the Elkhead Mountains, all said to be of “Pliocene (post-Browns Park)” age. First erup- tions were of intermediate composition, mainly light- colored latites and trachytes, with some trachyan- desites. Following a period of erosion, olivine basalt flows were poured out (perhaps at several times), followed by trachybasalts, lamprophyres, and analcite- bearing rocks. “Definitely younger” (Christensen, 1942, p. 45) lamprophyre dikes are shown cutting trachyte flows on Meaden Peak. In his overview of the Tertiary geology of the Elkhead region, Buffler (1967, p. 95 -100) reported that most of the volcanic rocks can be subdivided into two general groups: light-colored inter- mediate porphyritic trachytes and dark basalts and lamprophyres. Buffler stated: “The age relationship between the intermediate rocks and the basic rocks is still not clear, as the two rock types were never ob- served directly associated With each other.” Because INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO the Browns Park Formation, which on Sand Mountain underlies intermediate rocks, is at the same elevation as basic intrusions just to the south, but contains no locally derived basic volcanic debris, Buffler concluded that the intermediate and basic rocks here “are con- temporaneous or more likely that the basic rocks are somewhat younger.” Buffler also reviewed other evi- dence indicating several late periods of mafic extrusion along the Colorado-Wyoming State line at the northern extremity of the Elkhead Mountains. Segerstrom and Young (1972, p. 38, 40, 41) reported a dike of dark alkali trachyte or quartz andesite cutting the altered rhyolite at Hahns Peak. Hahns Peak itself is reported to be the site of a volcano where rhyolite porphyry with a central breccia pipe (Segerstrom and Young, 1972, cross sec. A—A ’; Bowes and others, 1968, p. 179) was ex— truded in immediate post-Browns Park time (Segerstrom and Kirby, 1969, p. B19—B22). Southeast of the area of this report in the north- western Rabbit Ears Range a full spectrum of extrusive and intrusive rocks from basalt to rhyolite is present (Hail, 1968, p. 50—84, pls. 1, 2, 3). The trend may run generally from rhyolite to basalt with time but data are insufficient to prove this; the youngest rock may be the basanite dike that cuts trachyandesite on Diamond Mountain. Izett (1966, p. B42 -B46; 1968, 21 —55) has mapped a Pliocene(?) basalt overlying a complex series of mafic to silicic extrusives and intrusives ranging in age from Cretaceous to Miocene (?) in the Hot Sulphur Springs area. Laboratory data on composition and radiometric ages are as yet insufficient to give the exact composi- tion-versus-time sequence of igneous rocks in this part of Colorado. However, indications are that relations of both composition and timing are more complex than some early, relatively simple models would have had us believe. First let us consider composition: Analyses of 28 eruptive rocks from the Park Range and Elkhead Mountains (including the three average analyses of Tertiary intrusives from table 5) are compared in figure 15 with analyses of 53 eruptives from the Rabbit Ears Range (mostly within a 7-mile radius of Diamond Mountain) and with 20 analyses from Izett’s Hot Sulphur Springs area using Rittmann’s (1952) para- meters, as commonly applied to fine-grained intrusive and extrusive rocks. The three fields of analyses shown are quite similar to one another; in fact, their areas of overlap are pronounced and this may be more impor- tant than the areas of individuality. The points upon which the fields of figure 15 are based are scattered randomly but are more or less continuously scattered throughout the areas shown, and the rocks represented are a wide range of basalts, trachybasalts, trachyan- PETROGRAPHY AND CHEMISTRY OF INTRUSIVE ROCKS 23 desites, trachytes, latites, quartz latites, rhyodacites, and rhyolites. Although saturated or slightly under- saturated basalts may be the youngest rocks, from the compositional data it seems clear that the other rocks form a continuum of all common compositions includ- ing basalts. Do the radiometric ages confirm that the basalts are the youngest rocks? Not at all, although there are some uncertainties and although many critical areas still need to be radiometrically dated. In the Elkhead Moun— tains 10 K-Ar dates are available on biotite, sanidine, or whole rock from both basalts and felsite porphyries (McDowell, 1966; Buffler, 1967; Segerstrom and Young, 1972). The average of five felsites, two from Hahns Peak, is 9.3103 m.y. But two authors think two of these dates are too young, possibly due to argon leakage (Buffler, 1967, p. 102; Segerstrom and Young, 1972, p. 40), and this might make the actual average age 10 or 11 m.y. This compares with Buffler’s range of 9.5— 11.1105 m.y. for two basalts and a lamprophyre and Segerstrom and Young’s two dates of 10.7 and 11.51 0.4 m.y. on the intermediate dike cutting the porphyry of Hahns Peak. Although only one of these dated rocks (the Hahns Peak dike) has also been analyzed chemically, the K-Ar dates clearly indicate that Elkhead Mountains magmas of diverse composi- tion were active over a span of time not greater than several million years near the Miocene-Pliocene bound- ary. This is about the same time, 9.7105 m.y. ago, that basalts were pouring out on Grand Mesa (U.S. Geol. Survey, 1966, p. A81) and, 10 m.y. ago, on the White River uplift (Mutschler and Larson, 1969). The latter event began much earlier, 21 m.y. ago, than anything yet dated in the Elkhead Mountains, about the same time as the two dated groups of basaltic volcanics at Yarmony Mountain south of Steamboat Springs (24511.0 and 21.5110 m.y., Strangway and others, 1969; York and others, 1971) and the rhyolitic airfall ash (24.8108 m.y.) in the lowermost part of the Browns Park Formation 25 miles northwest of Maybell (Izett and others, 1970, p. C150). No radiometric ages are available from the Tow Creek, Columbine, or Meaden Peak dikes or from the rocks they cut. Naeser, Izett, and White (1973, p. 498) reported late Oligocene to early Miocene intrusive activity (23 —30 m.y.) from zir- con fission track ages on 15 effusive rocks from the Rabbit Ears Range, Trail Mountain quadrangle, Green Mountain Reservoir area, and Red Mountain area. Izett (1966, p. B45; 1968, p. 35—37; see also Taylor and others, 1968, p. 42) reported K-Ar sanidine dates of 2913 and 3313 m.y. on quartz rhyolite tuffs from Hideaway Park and northwest of Hot Sulphur Springs, respectively. In the Cameron Pass area east of North Park, Corbett (1968, p. 6, 28) has mapped mafic Eocene volcanics that are overlain and are intruded by silicic volcanics providing K—Ar dates of 27 and 28 m.y. TERTIARY DIKES VERSUS PRECAMBRIAN DIKES Some Park Range Tertiary dikes can be mistaken for some Precambrian intrusives, notably the fine-grained porphyry dikes or the small ultramafic intrusives. Some authors have described rocks in the vicinity as Tertiary that may well be Precambrian; others may be tempted to make the reverse mistake. Field diagnosis of the relative age of some intrusives is obvious while diag- nosis of others may be difficult; petrographic study of the difficult ones can often lead to an unequivocal deci- sion. Intrusives that cut rocks younger than Pre- cambrian are obviously younger than Precambrian. Other features, such as a glassy groundmass, vesicularity, alteration products related to near-surface oxidation, or a certain “look” of feldspar or olivine phenocrysts, quartz xenocrysts, or a flow-structured groundmass, may prove a Tertiary origin; being cut by younger Precambrian rock (for example, fig. IOB) is clear evidence of Precambrian origin. But some in- trusives display none of these features, they cut only Precambrian rocks, are not known to be cut by other Precambrian rocks, and are not capable of being or have not been dated radiometrically. These rocks are problems in outcrop or hand specimen, but table 4 i1- lustrates several useful microscopic criteria. Some Ter- tiary intrusives studied differ from the Precambrian ones in containing high temperature sanidine and cristobalite as well as minor groundmass glass or vesi- cles not visible in hand specimen and also a dis- tinctively zoned plagioclase. Tertiary mafic minerals are generally fresh or unaltered or, if partly altered, the alteration is distinctive, as olivine to reddish iddingsite rather than colorless serpentine or chlorite. Pre- cambrian porphyry dikes and small ultramafic in- trusives on the other hand never contain high tem- perature minerals, glass, vesicles, or the same kind of zoned plagioclase. Olivine and pyroxene are absent from the fine-grained porphyry dikes, and hornblende is green or blue rather than brown or brownish green. Olivine and pyroxene are present, even abundant, in some of the ultramafic intrusives, but as yet there are no known Tertiary ultramafic intrusives in this area. Chlorite and epidote alteration is very conspicuous in many Precambrian intrusives, much more so than in any Tertiary intrusive. Chemically, little difference exists between Tertiary and Precambrian intrusives although a few minor ele- ments may have statistically significant variations. Lanthanum and fluorine may be higher, and nickel lower, in Precambrian porphyry dikes as a group. And 24 the ultramafic intrusives certainly tend to be higher in chromium, palladium, and platinum and lower in stron- tium, yttrium, and zirconium than the average rock of any other group. NORTHWEST VERSUS NORTHEAST PORPHYRY DIKES Despite repeated attempts, no consistent difference in lithology, petrography, or chemistry between the groups of northwest-trending and northeast-trending Precambrian porphyry dikes has been discovered. ( In table 5, field Nos. 549 and 545 represent the northwest- 1.0 0.9 0.8 0.7 0.6 .2 0.5 0.4 0.3 Alk, IN PERCENT 0.2 0.1 1.0 0.9 0.8 0.7 0.6 INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO trending dikes, and No. 524 is northeast trending.) In fact, nearby dikes with the same trend locally differ more in the degree of their alteration or recrystalliza- tion than the difference between averages for each dike set. The emplacement of either dike set may have taken place over a considerable period of time. PEGMATITES— ARE THEY RELATED TO A PARTICULAR MAGMA SERIES? The few available salient facts about whether most pegmatites are related to a particular magma series follow. 0 —0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 an $02 SIo, Alk FM ,1 9 INDEX DIAGRAM SHOWING 0-5 0 LOCATION OF FIGURED 0.4 0.3 0.2 0.1 I | I I I I I l I o Alk A FM _0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 / SA '19 (50A VOA \ an C D EXPLANATION 0 Park Range Tertiary intrusives Elkhead Mountains Tertiary eruptives (:3 Diamond Mountain area Tertiary eruptives in) Hot Sulphur Springs area Tertiary eruptives PETROGRAPHY AND CHEMISTRY OF INTRUSIVE ROCKS 25 1. The mineral composition, which shows minor varia- tions in the grain size and abundance of muscovite, biotite, and garnet, is monotonously similar throughout the area and, in fact, is simi- lar to that of most other pegmatites in other crystalline terranes elsewhere. Chemical com- position, not available here, would be expected to be equally unrevealing. 2. Many pegmatites are concentrated in swarms adja- cent to the contacts of the Mount Ethel pluton (figs. 50, 9, 11), suggesting that these pegmatites may in some way be related to this magma series. { FIGURE 15 (facing page).—Comparison of northern Colorado Ter- tiary igneous rocks using Rittmann chemical parameters. Graphs A, B, and C from Rittmann (1952, figs. 4, 5, and 6, respectively) with data field overprints from the Elkhead Moun- tains (Hague and Emmons, 1977, p. 176--179, 2 analyses; Clarke, 1904, p. 187, 1 analysis; Clarke, 1910, p. 118, 1 analysis; Ross, 1926, p. 225, 4 analyses; Segerstrom and Young, 1972, ta- ble 5,! 12 analyses; E. J. Young, 1968, written commun., 5 analyses), Diamond Mountain area in the Rabbit Ears Range (Hague and Emmons, 1877, p. 125, 126, 2 analyses; Hail, 1968, table 3, 51 analyses; also see Washington and Larsen, 1913, p. 450 for one analysis that plots off these graphs), and Hot Sulphur Springs area (Izett, 1968, table 8, 20 analyses). All analyses recalculated to 100 percent after subtraction of volatiles. The Rittmann parameters shown in this figure are Sio2 =s102, Alk =K20 + 1.5 Na20, PM = Fe203 + 1.1 (FeO +Mn0) +2MgO, K o 0.9 A1 0 —Alk _ _2_ _ __.2_3— k ‘ Alk and a“ ‘ 0.9 A1203 +A1k In the Rittmann system of nomenclature, appropriate chemical parameters of fine-grained eruptives are first fitted onto graph A to obtain a letter (with all but one of these analyses, it is either A or B), and then the chemical parameters are fitted onto either graph B for letter A or graph C for letter B to obtain a number. With the letter and the number one obtains the Rittmann name of the rock from a table. (See Rittmann, 1952, for details.) A. Comparison on Alk-to-SiOz graph of Park Range Tertiary in- trusives with other northern Colorado Tertiary eruptives. Rittmann field: A =oversaturated rocks; B =saturated rocks, C and D =undersaturated rocks. Rocks in fields AB, BA, or BC follow special rules to fit in fields A, B, or C. B. Comparison on k-to-an graph of oversaturated (Rittmann field A) northern Colorado Tertiary eruptives. Fields 1, 2, and 3 =various kinds of rhyolites; 4 =quartz latites; 5, 6 =rhyodacites; 7 and most of 8 =dacites. C. Comparison on k-to-an graph of saturated (Rittmann field B) Park Range Tertiary intrusives with other northern Col- orado Tertiary eruptives. Fields 1, 2, and 3 =various kinds of trachytes; 4 =1atites; 5, 6 =trachyandesites for low FM and trachybasalts for high FM (FM requirements vary); 7, 8 =andesites for low FM and basalts for high FM (FM re- quirements vary.) D. Ternary comparison of Si02 , Alk, and FM of Park Range Ter- tiary intrusives with other northern Colorado Tertiary erup- tives. 3. Lead-alpha ages of monazite and xenotime from “pegmatite, Park Range, Routt County, Col- orado” are 1,430 and 1,420 my, respectively (Jaffe and others, 1959, p. 128), apparently in- dicating that some Park Range pegmatites were emplaced contemporaneously with the rocks of Mount Ethel pluton. 4. The K-Ar age of muscovite from pegmatite within felsic gneiss at the Farwell mine 7 miles north of the area of this report is 1,680-1-50 m.y. (Seger- strom and Young, 197 2, table 3). They reported (1972, p. 16) that the felsic gneisses define an Rb- Sr whole-rock isochron of 1,650—1,700 my. and many of these felsic gneisses look similar to some rocks of the Buffalo Pass pluton in the area of this report, suggesting that some pegmatites that cut the older gneisses may be essentially coeval with them. 5. As noted previously in this report, at Northgate, pegmatite swarms near intrusive quartz monzonite of “younger Precambrian” (Mount Ethel equivalent) age are belived to be of “older Precambrian” age (Steven, 1957, pl. 48). Conclusion— Lithologically similar pegmatites are related in time to both major magma series in the area, but the volume related to each series needs further definition. However, the time relationship permits, but does not prove, a common origin. Some or all pegmatites of exactly the same age as nearby plutons may have been derived from the metamorphic country rocks dur- ing high-grade metamorphism coincident with plutonic intrusion. MOUNT ETHEL AND BUFFALO PASS PLUTONS, SIMILARITIES AND DIFFERENCES The Mount Ethel pluton possesses many chemical and mineralogic trends that parallel the time sequence of its mapped units. These trends tend to differ slightly from those in the Buffalo Pass pluton. Some graphical plots indicate that the fine-grained porphyry dikes are more like the Mount Ethel trend than the Buffalo Pass trend. Fluorite appears to be a key accessory mineral in distinguishing the 1,400-m.y.-old Mount Ethel rocks from the 1,800-m.y.-old Buffalo Pass rocks, and the analyzed fine-grained porphyry dikes have a fluorine content comparable to rocks of the Mount Ethel pluton. Within the Mount Ethel pluton the rocks, from oldest to youngest, range from diorite and granodiorite through quartz monzonite (most of the pluton) and gra- nite to leucogranite. The oldest rocks are dark colored and medium to coarse grained while the youngest rocks are light to very light colored and medium to fine grained. Parallel to the time-color-texture trend are several chemical and mineralogical trends: The SiO2 26 INTRUSIVE ROCKS NORTHEAST 0F STEAMBOAT SPRINGS, PARK RANGE, COLORADO increases from about 57 percent to about 75 percent, the total iron-magnesium oxides decrease from greater than 11 percent to less than 1 percent, the alkali oxides increase from about 6 percent to about 9.5 percent, and CaO decreases from about 6 percent to less than 1 per- cent (table 5); the trend is also one of decreasing Sr and increasing Rb (table 1); there is a decline in the anorthite content of the plagioclase parallel to the decrease in CaO and increase in alkali oxides; the se- quence is also characterized by a general increase in potassium feldspar and muscovite and a decrease in biotite, hornblende, and total mafic minerals; quartz and plagioclase tend to be more constant in amount but with many less predictable fluctuations (tables 4, 5). Fluorite is a common accessory mineral that will be dis cussed in some detail in the following section. (See “Ac- cessory fluorite.” ) The Buffalo Pass plutonic “sequence” is one of in- ference inasmuch as the relative ages of the two map- ped units are not known. The quartz monzonite and granodiorite augen gneiss of Buffalo Mountain of the Buffalo Pass pluton is lithologically most similar to the quartz monzonite porphyry of Rocky Peak of the Mount Ethel pluton; similarly, the equigranular quartz monzonite gneiss of the Buffalo Pass pluton is compara- ble to the quartz monzonite of Roxy Ann Lake, or younger rocks, of the Mount Ethel pluton. (See p. 12.) It is possible that many of the statements made about the chemical and mineralogical trends within the Mount Ethel pluton also apply in the same way to this Buffalo Pass plutonic “sequence.” (See tables 2, 4, and 5.) However, there are also subtle to significant differences between the two plutons: The lack of euhedral phenocrysts in the pre-Buffalo Pass or synkinematic Buffalo Pass rocks has already been mentioned. Blue- green hornblende is much more common in Buffalo Pass rocks than in supposedly comparable Mount Ethel rocks, and K20, and hence microcline, is more abun- dant in Buffalo Pass rocks than in Mount Ethel rocks for comparable values of SiOZ. The chemical similarities and differences between the rocks of the two plutons are well illustrated in the normative mineral plots of figure 16. Although the series are near the median of analyzed rocks as a whole (fig, 16D), they still possess some distinctive properties. On the ternary plots (figs. 16A, 168, 160) the Mount Ethel rocks form a linear trend bunched toward the more silicic or alkalic end of the trend. It is interesting that the fine-grained porphyry dikes, which overlap the Mount Ethel plutonic rocks in age, appear to lie on and to extend this trend at the more sodic or calcic end of the spectrum. The Buffalo Pass rocks, especially the Buffalo Mountain rocks, appear to be always more orthoclase rich relative to the other constituents, and hence the plot slightly off the trend of the younger FIGURE 16. (facing page).—Plots of normative minerals of p analyzed Park R'ange Precambrian plutonic rocks compared with experimental and statistical data. A. Quartz-albite-orthoclase diagram showing progressive varia- tion of quartz-feldspar equilibrium boundaries with H20 pressure; pluses indicate isobaric temperature minima for 500, 1,000, 2,000, and 3,000 bars; triangles, isobaric eutec- tics for 5,000 and 10,000 bars (experimental data from Tut- tle and Bowen, 1958, p. 54—56; Luth and others, 1964, p. 765, 766). B. Quartz-p1agioclase-orthoclase diagram showing progressive shift in quartz-feldspar equilibrium boundaries (at 1,000 bars PH 0) with 3, 5, 7.5, and 10 percent anorthite (experi- mental data from James and Hamilton, 1969, p. 118, 120). C. Feldspar diagram showing plagioclase-alkali feldspar equilibrium boundary (dashed line, 5,000 bars PHZO’ Yoder and others, 1957, p. 211; solid line, projected quartz-satur- ated liquidus at 1,000 bars PHZO, James and Hamilton, 1969, p. 123). D. Silica versus differentiation index (for these rocks this is the sum of Q +Ab +Or) compared to a contoured diagram for the 5,000 analyses in Washington’s tables. (See Thornton and Tuttle, 1960, p. 674.) rocks. The field-demonstrated time sequence of the Mount Ethel rocks from mafic to silicic is generally cor- roborated by experimental results in the Ab-Or-SiOz- H20 and Ab-An-Or-HZO systems. Although the experi- mental systems do not contain all the constituents of the natural ones, experimental results show that the rocks could be the product of magmatic differentiation under conditions of general crystal~liquid equilibrium. (But see comment on p. 33.) In figure 16A the trend of the series is consistent with the expectable trend of an Ab-Or-SiO2 liquid toward the isobaric temperature minimum and the quartz-feldspar equilibrium bound- ary at a PHO of 1000—5000 bars, the latter perhaps most likely. zBateman and others (1963, p. D35) have translated 5,000 bars into a probable depth of differen- tiation of at least 11 miles. Those silicic compositions apparently in the quartz field may actually be in the feldspar field because of the influence of anorthite on the boundary as shown by James and Hamilton (1969) and as illustrated in figure 16B. Figure 160 demonstr- ates a progressive decrease in the normative anorthite content of the plagioclase (parallel to that actually pre- sent in the rocks) as the alkali feldspar equilibrium boundary is approached. As most of these rocks have a plagioclase in the oligoclase range in equilibrium with microcline, it is possible that, at 5,000 bars PH 0, such rocks would have crystallized in the temperaturle range from 705°C to 740°C (Yoder and others, 1957, p. 212, 213). ACCESSORY FLUORITE Many Mount Ethel plutonic rocks (more than 50 per- cent of the samples studied) differ from the Buffalo Pass plutonic rocks in containing a minor amount of ac- cessory fluorite, and, because this is a useful local dis- PETROGRAPHY AND CHEMISTRY OF INTRUSIVE ROCKS 27 7.5--—'v‘—" 5—————r——\-——— Or 100 Sioz, IN WEIGHT PERCENT o I r 100 50 o D DIFFERENTIATION INDEX EXPLANATION - Quartz diorite and granodiorite augen gneiss of Buffalo Mountain 6 Aplite and leucogranite O Fine-grained granite 0 Granite and quartz monzonite of Roxy Ann Lake 0 Quartz monzonite porphyry of Rocky Peak - Biotite granite and quartz monzonite gneiss 1.8 billion years 1.4 billion years 0 Granodiorite and diorite ‘3 Fine-grained porphyry dikes criminatory petrographic criterion and also has some thin section are shown in figure 17. Most of the ob- economic significance, these results are considered in served fluorite clearly falls within or close to the outer some detail here. The distribution of fluorite by map boundaries of the Mount Ethel pluton, although afew of unit or similar broadcategory of rock is given in table 7, the fluorite-bearing samples plotted in figure 17 may be and the geographic localities for fluorite observed in from unmapped inclusions or inliers rather than 28 INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO TABLE 7.—Fluorite in Park Range rocks northeast of Steamboat Springs Number of Number of thin Percent per unit thin sections sections with- of thin Park Range rocks with observed out observed sections with fluorite fluorite observed fluorite Tertiary intrusives .......... 2 10 17 Tertiary sediments .......... 1 25 4 Mesozoic sedimentary rocks1 . 0 35 0 Fine-grained porphyry dikesl . 1 32 3 Mylonite and pegmatitel ..... 3 1 1 21 Mount Ethel pluton: Leucogranite ............. 1 4 20 Granite and quartz monzonite .............. 16 1 9 46 Quartz monzonite of Roxy Ann Lake ......... 105 44 70 Quartz monzonite porphyry of Rocky Peak .......... 21 33 39 Granodiorite .............. 0 9 0 Buffalo Pass pluton: Quartz monzonite and granodiorite augen gneiss of Buffalo Mountain ..... 0 13 0 Quartz monzonite gneiss . . . 0 6 0 Mafic intrusives ............. 0 1 5 0 Country rocks: Felsic gneiss to amphibolite 4 79 5 Pelitic schist and gneiss . . . . 2 56 3 Total ................. 21 56 2391 29 1 Because of mineralogy or texture, small amounts of fluorite might have been missed in some thin sections of these rocks. 2 Total number of thin sections here is greater than that given in table 4 because several thin sections have more than one lithology. plutonic rock. These rocks typically contain less than 0.5 percent fluorite, but, locally, there are individual outcrops or large areas that contain from 0.5 to 2 per- cent disseminated fluorite (fig. 17). Table 7 shows that fluorite appears to have been concentrated in rocks near the middle of the Mount Ethel plutonic sequence. Percents of samples containing fluorite in the Mount Ethel and related units, generally from oldest to youngest are the following: granodiorite, none; fine— grained porphyry dikes, 3 percent; quartz monzonite porphyry of Rocky Peak, 39 percent; quartz monzonite of Roxy Ann Lake, 70 percent; granite and quartz monzonite, 46 percent; and leucogranite, 20 percent. These figures compare with an average of 7 percent in all other units; fluorite was not observed in any samples of the Buffalo Pass pluton. On this evidence alone, one of the following two conclusions seems justified: 1. Most of the fluorite is of magmatic origin and was formed by some process of magmatic differentia- tion that culminated near the middle of the Mount Ethel magmatic cycle 1.4 billion years (b.y.) ago; this conclusion is favored in this report; 2. If the fluorite is younger than 1.4 b.y., and perhaps formed as a hydrothermal replacement, then rocks of the Mount Ethel pluton must be a more favorable host than surrounding rocks, and rocks near the middle of the Mount Ethel plutonic se- quence must be progressively more favorable hosts than rocks near either end of the series. Because the rocks of the Mount Ethel pluton are similar compositionally to other rocks outside of the pluton and because the middle rocks are not obviously more favorable hosts, this conclusion is not favored in this report. The chemical evidence suggests a progression in chemical fluorine content through the Mount Ethel plutonic sequence that permits a magmatic origin for the mineral, fluorite. Chemical fluorine is most abun- dant in the most mafic rocks of the Mount Ethel pluton, the granodiorite unit and the fine-grained porphyry dikes, and is least abundant in the leucogranite; inter- mediate amounts are present in the intermediate rocks (table 5). Thus, although there appears to be a progres- sion in chemical fluorine content through the intrusive sequence, this progression is different from that for the content of the mineral, fluorite. This is apparently so because the chemical fluorine seeks mafic minerals in the most mafic rocks and only crystallizes as the mineral, fluorite, when there are not enough mafic minerals to absorb it. From studies of available chemi- cal analyses of mafic minerals it seems likely that most of the Mount Ethel nonfluorite fluorine is contained in biotite with decreasing amounts in hornblende, apatite, and sphene. The petrographic evidence concerning much of the interstitial fluorite also tends to confirm conclusion (1). Nine photographs depicting the texture and habit of in- terstitial fluorite in quartz monzonites and granites of the Mount Ethel pluton are presented in figure 18. Three kinds of fluorite, perhaps representing three ages of fluorite, may be distinguished in these rocks on the basis of texture, habit, and mineral association: (1) Magmatic fluorite. Anhedral fluorite, crystals the same general size as and associated with, in- terstitial to, or including mafic minerals (figs. 18A, 18E), mafic and felsic minerals (figs. 183, 180), or felsic minerals (fig. 18D). This type is volumetrically most important. (2) Magmatic to deuteric fluorite. Fluorite in long thin veins cutting other magmatic minerals but also associated in the veins with other magmatic minerals like microcline, sphene, and epidote (figs. 18H, 181). This type is relatively rare. (3) Fluorite coincident with deuteric or later (some perhaps much later) alteration. Generally anhe- dral, locally euhedral, small grains of fluorite usually within plagioclase partly to completely altered to sericite and muscovite (figs. 180, 18F). This fluorite is common and may be intimately associated with small grains and veinlets of late calcite (not illustrated). PETROGRAPHY AND CHEMISTRY OF INTRUSIVE ROCKS 29 / Pr/‘sl/‘Iw Lake , /7 D l , , . /, . o .._ ' / Sea. 2 . BUFFALO PASS ‘ . . Mtno PLUTON' ’. . 0 Buffalo " '- ' 5361 Pass . - o O ,4 fl O 3.4”93’0 . V Luke ' “My; 99“ \\<>Di/w.mme \ 6;“ a a . . O I J '3"? do. I. A t. 0 Ln; 0 1 2 3 4 5 MILES l—v—Lr—H—TJ—w—ll—w—A—y—jJ O 1 2 3 4 5 6 7 8 KILOMETERS EXPLANATION Post-Precambrian rocks — Mainly Tertiary and Mesozoic sedimentary rocks Precambrian Y rocks of Mount Ethel pluton Xm Precambrian X metamorphic wallrocks . Locality for thin section containing no observed fluorite 0 Locality for thin section containing <0.5 percent observed fluorite 0 Locality for thin section containg >0.5 percent observed fluorite FIGURE 17. — Location of accessory fluorite in rocks of part of the northern Park Range. One or more of the above varieties of fluorite may be tion (fig. 18D), some formed just after sericitization found in the same thin section (for example, figs. 183, (fig. 181), whereas some formed long after sericitization 18]). Note that some fluorite formed before sericitiza- (fig. 18F). 30 INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO 31 PETROGRAPHY AND CHEMISTRY OF INTRUSIVE ROCKS 223$ 25 $23 mo N :nwEMQ—cna 29¢ Eu N ~ anauwoaozm 2095 222:5 2 315 ::< 58m Mo 323.52 5.5:: 2.5 owfloofimi c3522; MESH. 20> m22328éan=E N .Eoflm 913 we awmwafio: 315 ::< zxom «c 32822 32:6 :23 wonmmE $2.2m 29a 250222 MESS". 26> m22m2ooéquEm E .2me 3223 92220.25 E32322“ 032:— mm £31.28 28 .230 32.82% .BEoim 285 320252 he cums 20¢ 3:5 22¢ 58m mo 028282 5.356 2 £9522 23$ 28 2.22 fits 252083 3E2: 255235 .0 .915 «25 min 313 Amok—O Hm :83qu SEA 25 Exam mo 328202 5.8:”. 29¢ 29% $302me ENSEEom £m2m m 2 3252828 SEosE E 285 @28sz we vac: 29¢ 33..— ::< ham Mo 32022: 5.226 2 3322 SE SEofimmE 222:2: 8E3: .N .334 3:52:30 23: 3?..— ::< axom .«o 3235.2 3.2:". 2 aim—aw 28 wmfloommmi 3 335232 BEBE .Q 4—3.5 wag mo “Eon 52% 2: 29¢ 32A 52 58m «0 323202 5.223 22me umfloommfla voifioEwm bfimn a 2 SEQ—E .0 2935 222:5 29c 324 ::< ham no 320292 5.26 2 £9822 2&2 was 2.22 53> 3263mm SE25 333235 .N .N 22 N magnum .823 3 vwanwmoawu :2: 23 mags". 2.5V 5 29a mm :98 mi? 23.5 222:5 29a 322m vmfimpméza 2 22222 2.22 :23 322033 SE25 255235 .< ASPEN .N 6221?. .m ”3228 .G ”vanMqum >529 mew—25» 6mm~uofim_g .m ”338.22 52 ”maEmem2 62 6:22:22 .M ”mu—Eon: “h ”083% .m “332: .m ”mu—Ewan in ”3223 .2513wa Jaw: wwNEEon 223 2 mquvwE .m 3284 ha .543 mnmuhmcaoan 2m £22323 some 2 22 H 22:5 :3 38m .2553 Monum— 2252 we 232.2% 22 m320202 52:6 2 qucsc mo manufioEoManosml .2 352m 32 INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO Although the fluorite has not been dated radiomet- rically, some petrographic evidence confirms the an- cient age of the fluorite. The usually colorless fluorite commonly becomes colored pink to purple to deep pur- ple next to zircon or allanite (for example, fig. 18G), probably owing to radiation damage to the lattice of the fluorite crystal over a long period of time. In this sense these color haloes are similar to pleochroic haloes in biotite, hornblende, or cordierite, except that the pre- sent color haloes are not pleochroic (do not change col- or) in the isotropic fluorite. Some work has been done on dating pleochroic haloes (for example, Hayase, 1954; Deutsch and others, 1956), and analogous efforts might be successful in directly dating the color haloes in these fluorites. Phair and Shimamoto (1952, p. 664) have noted that in Jamestown, Colo., thorium and uranium haloes in fluorite crystals are especially numerous where the fluorite is dark. Other indirect evidence of ancient, probably Precambrian, fluorite comes from a single thin section of brecciated, mylonitized pegmatite (the 0.5 percent fluorite point near the extreme south- western tip of the Mount Ethel pluton, fig. 17). This thin section contains centimeter-size angular pegmatite fragments in an anastomosing network of dark mylonite veins whose individual mineral constituents are so fine grained as to be partly unresolvable with a petrographic microscope. Fluorite veins as much as 0.1 mm wide cut some breccia fragments and are them- selves truncated at the margins of the breccia frag- ments; a younger generation of fluorite in this same rock forms euhedral cubes as much as 0.03 mm wide replacing the mylonite matrix. If the mylonite is com- parable in age (1.2 by) to the nearest directly dated mylonite (Abbott, 1972), then the fluorite veins trun- cated by the mylonite must be >1.2 by old. Fluorite has been observed in some other quartz monzonites of probabaly equivalent age in the vicinity. The quartz monzonites on Sheep Mountain (fig. 2) and Delaney Butte in North Park (Hail, 1965, pls. 2, 3) are an upfaulted continuation of the Mount Ethel pluton; interstitial fluorite has been observed in both places. It is also present in the intrusive quartz monzonite at Northgate (Steven, 1957, pl. 48) that Behrendt, Popenoe, and Mattick (1969, p. 1523) believed may be continuous with the Mount Ethel pluton. Interstitial fluorite has been observed in thin sections of the follow- ing Sherman Granite bodies mapped by Houston and others (1968, pl. 1) in the Medicine Bow Mountains of Wyoming: (1) the large quartz monzonite body southeast of Mountain Home, (2) fine-grained quartz monzonite south of Ring Mountain, and (3) the quartz monzonite body at Sheep Mountain southeast of Cen- tennial, Wyo. The present study has also confirmed fluorite in quartz monzonite of both the Inner and Outer Cap Rock (Eggler, 1968, p. 1551) and the Log Cabin Granite of Kirst (1968) at Virginia Dale north- west of Fort Collins, Colo. Fluorite is a trace constituent in several facies of the St. Kevin Granite, Holy Cross quadrangle, Colorado (Tweto and Pearson, 1964, p. D30). However, the usefulness of fluorite as a criterion for differentiating 1.4-b.y.-old from 1.8-b.y.-old granites may be purely local. Segerstrom and Young reported colorless fluorite from two samples of felsic gneiss and ,one sample of augen gneiss (1972, table 1, p. 11) on the shoulders of Farwell Mountain, either one or both of which may be equivalent to Buffalo Pass rocks. Fluorite has also been observed in the “older granite” on Jelm Mountain, Wyo. (Houston and others, 1968, pl. 1). ECONOMIC IMPLICATIONS Much fluorite was deposited contemporaneously with igneous mineral grains of the Mount Ethel pluton, and it has been demonstrably mobile in subsequent Precambrian time. It seems possible, if not likely, that this Precambrian fluorite was mobile at other Phanerozoic times, for example, in the Laramide orogeny (Steven, 1960, p. 395, 396) or Tertiary time, and, if so, it could have contributed to the economic Ter- tiary fluorite vein deposits in the North Park area. Similar mobilization of Precambrian lead and gold to form Tertiary deposits at Hahns Peak has been postul- ated by Antweiler, Doe, and Delevaux (1972, p. 312, 313) on the basis of isotopic evidence. Fluorite veins have been exploited or prospected re- cently at five localities in the North Park area: two vein zones north of Northgate (Steven, 1960), the Crystal mine (fig. 17), a vein zone from 10,700- to 10,914-foot elevation on the ridge northwest of the Crystal mine, and a vein zone crossing the top of Delaney Butte. All these vein zones occur in rocks mapped as or equivalent to the quartz monzonite of Roxy Ann Lake, or in thin altered Tertiary sediments directly overlying the same quartz monzonite— the quartz monzonite that con- tains most of the interstitial fluorite (table 7). Steven (1960, p. 401) reported that the Northgate veins are richer where they cross the competent quartz monzonite or Tertiary sediments overlying the quartz monzonite, but they are relatively barren where they cross inclusions of hornblende gneiss country rock. All the vein zones strike northwest to northeast and con- tain white, purple, or green, comb-structured, mammill- ary fluorite, except the vein zone on Delaney Butte where it strikes east-west and contains colorless to white euhedral fluorite. The Delaney Butte zone ap- pears to widen westward. Fault breccia reefs are abun- dant in the vicinity of the Crystal mine (fig. 17) and some of these containing significant fluorite may not PETROGRAPHY AND CHEMISTRY OF INTRUSIVE ROCKS 33 yet have been thoroughly prospected: for example, 22.2 percent fluorite was measured in a brecciated, mylonitized quartz monzonite a mile north of the Crystal mine at the 9,250-foot elevation on the south side of an east-northeast-trending spur 450 feet southeast of Spring Creek. The rhyolitic (?) intrusion reported at the Crystal mine (Popenoe and others, 1970, p. 343) has not been confirmed by this investiga- tion, but aplite dikes or quartz monzonite breccia reefs in the vicinity could easily be mistaken for rhyolite. Possibly, new economically recoverable fluorite deposits (for example, see The Denver Post, Sunday, June 11, 1972, p. 76) will be found along fluorite-ce- mented fault breccias in the northeastern part of the Mount Ethel pluton or elsewhere in the pluton where the more abundant intergranular fluorite (fig. 17) could have served as a source for redistribution by Tertiary solutions. CORRELATION OF PARK RANGE ROCKS WITH OTHER IGNEOUS ROCKS OF COLORADO AND NEARBY WYOMING General correlations of intrusive rock suites between separate areas in Colorado and Wyoming have been made possible by many existing radiometric dates. Some refinement of these general correlations can be accomplished with visual (field) observations of lithologies and structures, some with petrographic com- parisons, and fewer with chemical comparisons. Much more comparative work of all kinds could and should be done, especially detailed mapping of field relationships, comparisons of different field areas, and radiometric dating of presently undated or difficultly datable rocks. The state of knowledge in 1974 of the temporal rela- tions of Precambrian geologic events in Colorado and adjacent Wyoming is diagrammed in figure 19. Assigned age relations of various igneous and struc- tural events have been compiled separately for six subareas with an overall summary based first on radiometric dates and second on relative ages of radiometrically undated events with respect to the known dates. Figure 19 shows the three main Precambrian plutonic events now well known in Colorado that pro- duced the rocks of Boulder Creek age (1.7101 b.y.), the rocks of Silver Plume age (1.41011 b.y.), and the rocks of Pikes Peak age (1.05:0.05 b.y.) (Peterman and others, 1968; Hedge, 1970, Stern and others, 1971). The granodiorites and associated quartz monzonites of Boulder Creek age of the Front Range are represented in the Park Range by the quartz monzonites at Buffalo Pass, in the Needle Mountains by the Tenmile, Whitehead, and Bakers Bridge Granites and perhaps the Twilight Gneiss (Barker, 1969). in the Black Can- yon of the Gunnison by the Pitts Meadow Granodiorite (Hansen and Peterman, 1968), and elsewhere in Col- orado and nearby Wyoming by numerous unnamed gra- nitoid rocks. The rocks of the Silver Plume, Longs Peak —St. Vrain, and Log Cabin batholiths of the Front Range (Peterman and others, 1968) fall in the same general time span as the rocks of the Mount Ethel pluton of the Park Range, the Eolus and Trimble Gra- nites of the Needle Mountains (Barker, 1969), the Ver- nal Mesa and Curecanti Quartz Monzonites of the Black Canyon of the Gunnison (Hansen and Peterman, 1968) and the Uncompahgre Plateau (Hedge and others, 1968), the St. Kevin Granite of the Sawatch Range (Pearson and others, 1966), the Sherman Gra- nite of Wyoming and northern Colorado (Peterman and others, 1968; Hills and others, 1968), and other smaller bodies elsewhere in the vicinity. The Pikes Peak batholith with its satellitic plutons of the southern Front Range is the only large billion-year-old batholith of the region (Hutchinson, 1960b; Hedge, 1970). The only rocks of a remotely similar age in the Park Range are small dikes; fine-grained microgranophyre dikes near the west end of Gore Canyon 20 miles south of this area have been dated at 1.13:0.15 b.y. (Rb-Sr whole rock) (Barclay and Hedge, written commun., 1967; Barclay, 1968, p. 71 —75); “granite porphyry” from Slavonia immediately north of this area has furnished zircon that gave a lead-alpha date of 739 my (Jaffe and others, 1959, p. 128). Granites of a 2.4 b.y. age are found northwest of the Mullen Creek —Nash Fork shear zone in Wyoming, but rocks of this age are not known southeast of this shear zone in Wyoming or Colorado (Houston and others, 1968; Hills and others, 1968). Figure 20 depicts the statistical distribution of 107 Colorado plutonic rocks of Boulder Creek and Silver Plume age concentrated near the experimental granite minima (fig. 16). The reader will note that, within the framework of the available chemical analyses, the con- centrations of most analyzed rocks for each affinity are the same but that the range of composition of the rocks of Silver Plume affinity is more restricted. The more potassium feldspar-poor compositions, especially trondhjemitic varieties, are not represented during the 1.4101 b.y. plutonic episode. At least one of these, the Twilight Gneiss is now thought to have originated as volcanic or hypabyssal igneous rock that was meta- morphosed to high grade within 50 or 75 my after its emplacement (Barker, Peterman, and Hildreth, 1969a). Figure 20D indicates that the field of the most highly differentiated rocks coincides approximately with the field of most abundant compositions (Fig. 20A). and that there is a smooth, though somewhat ir- regular, transition to this field from the least differenti- ated rocks. In short, the more differentiated rocks tend um": MOUNTAINS. wvoumc, AND Monrusnu snout RANGE, COLonAno Intrusion 0800:. £83 63000 7 _ II 2:20 00:06 5.030 6:560:20 -Iv. _ 3:35.: 5:50 SE20 SEE—won IIIIII I .. 52:00an .050 3_:o>mml7l.v Toafioztocxx 3:35.05. 020E223 —..n 30:0 2305“. l PARK RANGE, COLORADO Intrusion Excuse, .0 :0_~S:0E_uww union. @553“. EmEQBESuE .m:0_mom 23:58 60:05:53: |I|I|I .IIT 5:03.205: _ 0:21.050 7. 0:050:23 0:55 ‘ 03:20 £30: 65000 2:20 0:0 8:050:20 02539: 5.020 0::20 _ _ 250E000 llII 30:0 239.01 In I AI— 1 2 South of Mullen Creek— Nash Fork shear zone Intrusion _w:__u:>m wmzw Em_:ao_0> .0 :oszoEfiom Ed E w ? m:_v_0u. m m N va 0:230“. _ EmEEoESuE .0550”. 1 $202.00 603023835. 5:02.235 _ 0:0:I850 _ 0:0:I0:mE2:D 80:20 £0me 65000 ill G 9:20 5:030 6:360:20 0:0:IS_:0~:0E 5.030 SE20 I - ESE—won. I | |I|l 30:0 Pinata“. MEDICINE BOW MOUNTAINS, WYOMING 1 North of Mullen Creek— Nash Fork shear zone INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO Intrusion r__A___fl/_2__1F_2___ are queried where uncertain. Boxes show major events; lines show less important events, and they are F252? :0 :0_HS:0E_uow TV T A1 a II-w IEI I121 :_E0u NI N E 33.30”. N 52982? .225 I $8.058 60:32:22). h LUI i 0:0:I:0_um~:20:_s_ ococleozuo 2me205 3:0020Emocto .8330 65900 I 8:20 3.530 6:550:20 o:0:|0::0N:oE 5800 "2:20 0:02:30 wcoclmméu 22.3.0“. — _ q _ {K— _ _ _ LL T {J1 E (I f mmmm WWW... wumem wwwm. m m mama m 1mm» mmmwm mm? w 2 M Y B m R H R R3 34 FIGURE 19. — Assigned temporal relations of Precambrian geologic events in Colorado and adjacent Wyoming. Arrows point in the direction of 35 PETROGRAPHY AND CHEMISTRY OF INTRUSIVE ROCKS MILLION YEARS BEFORE PRESENT 500 ago 1000 Rocks of Pikes Peak _} 2000 2500 COMPOSITE OF COLORADO ANOAOJACENT WVOMING Intrusion Ew.:mo.0> .0 c0.~8:wE..uem 9.6.0“. 95.30”. EmEQSEEwE .mcoBom $0.038 60.355535. :0_Hm~..mt_m:_s. $50 ofiaEwt... $330 .05an 3.5.0 6:850:20 2.:o~:oE 5530 8.520 3:350“. “8:0 >._>r.at_om BLACK CANYON, NEEDLE MOUNTAINS. AND UNCOMPAHGRE UPLIFT, WESTERN AND SOUTHWESTERN COLORADO Intrusion E3529, .0 cotflcwEEum 9:28. 9.3.30”. EminaESwE .8050”. 28.028 60:22:03.2 ococlcozmufisfis. Nmm. I TIT—IV I .mc:oc>moom:w 82:92:. 350 cCOCIOINESHSD $30.0 .8..o££nEmo...t0 .05an 8.5.0 5:550:90 8.55.9: 5:50 3.520 “ESE—won. .2>:00..QE~: 30:0 22.90.... 3.53222 ,m , 5 CENTRAL AND SOUTHEASTERN COLORADO Intrusion Em_:~o.o> .0 :ouficwEiow 9:20“. 05:30". E£cSoES~E .3053. 52038 50:32:03.2 8.55.2055. €02.83. .050 chESHS 330.0 £33 .9300 3.5.0 6:560:20 SEE—BE 62:020.: 3530 3.5583. 60.0.83 62:20 SEER... .E>..Qo.aEm.EoE. “8:0 3.590; luxcgm Al. 4 LARAMIE MOUNTAINS, SAWATCH RANGE AND SOUTHERN FRONT RANGE, WYOMING. AND NORTHERN FRONT RANGE, COLORADO Intrusion Eu_c8_0> .0 cosficwEium 9.66“. 95.30". EmEEoESoE .mcofiom $3.028 .c0:m~_::o.>.>. E03283. 5.35.835 the true age of events that so far can only be dated as “older than” or “younger than” with respect to radiometrically known dates; these dashed in less certain parts of their ranges. Sources of data listed in the table on the following page. 36 INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO Sources of data for figure 19 1,2. Hills, Gast, and Houston (1965, 1967); Hills, Gast, Houston, and Swainbank (1968); Houston and others (1968); Houston, Hills, and Gast (1966). 3. This paper; J. C. Antweiler (written commun., 1965); Ant- weiler (1966); J. C. Antweiler, M. H. Delevaux, and B. R. Doe (written commun., 1968); C. S. V. Barclay and C. E. Hedge (written commun., 1967); Jaffe, Gottfried, Waring, and Worthing (1959, p. 128); Kenneth Segerstrom and C. E. Hedge (written commun., 1970); Segerstrom and Young (1972); U.S. Geological Survey (1968, p. A28). 4. Abbott (1972); Aldrich and others (1957, 1958); Condie (1969a, b); Eggler (1968); Eggler, Larson, and Bradley (1969); Fenton and Faure (1969); Ferris and Krueger (1964); Giletti and Gast (1961); Harrison and Wells (1959); Hedge, Peterman, and Braddock (1967); Hepp (1966); Hills and Armstrong (1971); Klugman (1960); Moench, Harrison, and Sims (1962); Peterman and Hedge (1967); Peterman, Hedge, and Braddock (1968); Smithson and Hodge (1969). 5. Brock and Singewald (1968); Bryant, Miller, and Scott (1971); Doe and Pearson (1969); Gross and Heinrich (1966); Hawley, Huffman, Hamilton, and Rader (1966); Hedge (1970); Hedge, Peterman, Case, and Obradovich (1968); Hutchinson (1960a, b); Hutchinson and Hedge (1968); Lovering and Goddard Gross and Heinrich (1966); Hawley, Huffman, Hamilton, and Rader (1966); Hedge (1970); Hedge, Peterman, Case, and Obradovich (1968); Hutchin- son (1960a, b); Hutchinson and Hedge (1968); Lovering and Goddard (1950, p. 64—65); Pearson, Hedge, Thomas, and Stern (1966); Peterman and Hedge (1967); Stern, Phair, and Newell (1971); Taylor and Sims (1962); Tweto and Sims (1963); Wells (1967); Wetherill and Bickford (1965); Wobus (1969). 6. Barker, Peterman, and Hildreth (1969a, b); Barker, Peter- man, and Marvin (1969); Bickford, Wetherill, Barker, and Lee‘Hu (1969); Cudzilo (1971); Hansen and Peterman (1968); Hedge, Peterman, Case and Obradovich (1968); Mose and Bickford (1969); Silver and Barker (1968); U.S. Geol. Survey (1969, p. A41, A116). 7. Composite of columns 1——6. to be more abundant than the least differentiated rocks, a situation incompatible either with partial crustal melting and rheomorphism or with differentia- tion by fractional crystallization at this or any other single level of the crust. Differentiation from more mafic rocks by either mechanism must have taken place at a deeper crustal level. Subsequently, silicic rocks rose through the crust leaving behind mafic rocks disproportionately increased in amount compared to their initial volume. SHERMAN -SILVER PLUME DICHOTOMY Current radiometric dating techniques are generally unable to discern any differences in the ages of intra- plutonic activity, and so precise temporal correlation of specific magmatic events in separate areas cannot be made. Nevertheless, there are certain petrographic si- milarities or petrogenetic parallels between the rock suites of separate plutons, especially those of Silver Plume age. These are worth noting even though it is not known whether similar rocks are due to parallel evolu- tion of separate batches of magma intruded at slightly different times or to simultaneous magmatic intrusion with synchronous evolution of all plutons. The general time progression from diorite and granodiorite through quartz monzonite to granite and leucogranite evident in the Mount Ethel pluton is also repeated in some other plutons. The most evident similarities are with one of the best studied nearby areas, the Virginia Dale ring dike complex between Fort Collins, Colo. and Laramie, Wyo. (Eggler, 1968; Eggler and others, 1969; Peterman and others, 1968). Here, the sequence of igneous events, generally from outermost to innermost parts of the pluton, is (1) (oldest) hornblende gabbro; (2) diorite, an- desite, and hybrid rocks; (3) coarse-grained Trail Creek Granite of Eggler (1967) including both the main unringed batholith and the outermost ring of the ring complex, in that order, the latter being somewhat more aluminous than the former; (4) coarse-grained and porphyritic quartz monzonite at Cap Rock including an outer incomplete ring and an inner complete ring, in that order; and (5) medium-grained Log Cabin Granite of Kirst (1968), in the center of the pluton. The coarser grained rocks including both nonporphyritic Trail Creek Granite and both porphyritic quartz monzonites at Cap Rock are together called Sherman Granite, and the medium-grained Log Cabin Granite is equated with type Silver Plume Granite. That the Log Cabin and Silver Plume are younger than the Sherman is shown not only by the more central position of the Log Cabin in the ring complex but also by the fact that a north- west-trending porphyritic andesite dike swarm that cuts Sherman in the Virginia Dale area (Eggler, 1968, pl. 1) is cut by Silver Plume Granite in the Big Thompson Canyon east of Estes Park (Hepp, 1966, p. 65). Reported radiometric ages of the Log Cabin, and its equivalent, the Silver Plume, and the Sherman Granite overlap (Log Cabin and Silver Plume range=1.21~1.45 b.y.: Aldrich and others, 1958, p. 1130; Peterman and others, 1968, p. 2290; Stern and others, 1971, p. 1624—1626; Sherman Granite range =1.31—1.44 b.y.: Aldrich and others, 1957, p. 656; Aldrich and others, 1958, p. 1130; Giletti and Gast, 1961, p. 455; Hills and others, 1968, p. 1770; Peterman and others, 1968, p. 2287—2289), and the Log Cabin —Sherman age difference in northern Colorado “appears to be less, and possibly significantly less, than 40 m.y.” (Peterman and others, 1968, p. 2289). In the Mount Ethel pluton the coarse-grained quartz monzonite porphyry of Rocky Peak appears to be the lithologic correlative of the coarse-grained Sherman Granite of the Virginia Dale area, especially of the porphyritic quartz monzonites at Cap Rock, while the medium-grained muscovite-con- taining quartz monzonite of Roxy Ann Lake and PETROGRAPHY AND CHEMISTRY OF INTRUSIVE ROCKS 37 Or Ab 0r FIGURE 20. —Three frequency distributions (A, B, C) and one con- toured differentiation index diagram (D) comparing ternary plots of normative minerals of 107 analyzed and dated Colorado Precambrian plutonic rocks with those of the Park Range. Fre- quency contours = 0-, 4-, 8-, 12-, and > 16- percent per one-per- cent area. The differentiation index diagram was constructed by contouring the differentiation index value for each rock (Q + Ab + Or in every case) plotted at its appropriate location in the ternary diagram. Rocks of 1.4 i 0.1 by. age are restricted to the right side of the dashed line in the appropriate diagrams; whereas rocks of 7 i 0.1 by. age are distributed throughout the area represented. Points are Park Range plutonic rocks detailed younger granites are a close lithologic match for the medium-grained muscovite-containing Log Cabin and Silver Plume Granites. If this lithologic correlation is correct, the Silver Plume correlatives make up by far the volumetrically greatest part of the Mount Ethel pluton, and the Sherman correlatives make up a smaller part — the reverse of the relative Silver Plume and Sherman volumes in the Virginia Dale pluton. Ab+An Ab Or in previous figures. Besides these 20 analyses, data include 19 analyses from southwest Colorado (Barker, 1969), 33 from central and western Colorado (J. E. Case, written commun., 1965; Hansen, 1964; W. R. Hansen, written commun., 1966, 1967; Hansen and Peterman, 1968; Hedge and others, 1968; Pearson and others, 1966; Tweto and Pearson, 1964), and 35 from northern Colorado (Eggler, 1968; Lovering and Tweto, 1953; Peterman and others, 1968; Sims and Gable, 1967; Wells, 1967). Seven other analyses available too late for inclusion in these diagrams are given by Stern, Phair, and Newell (1971, p. 1620). A similar plutonic sequence appears to be present in the 1.3 -1.5-b.y.-old Eolus batholith and related rocks of the Needle Mountains (Barker, 1969; Bickford and others, 1969; Silver and Barker, 1968). Although crosscutting relations are not present between all rock types, the following sequence is determined by the known relations plus many radiometric ages: (1) (possi- bly oldest) Electra Lake Gabbro (olivine gabbro to gra- 38 nodiorite); (2) quartz diorite of Pine River; (3) rocks mapped as Eolus Granite (ranges from granodiorite — oldest — through biotite-hornblende quartz monzonite— most of batholith— to biotite quartz monzonite, biotite granite, biotite-muscovite granite, and alaskite); (4) porphyritic muscovite-containing Trimble Granite (near the center of the batholith); and (5) rhyolite porphyry and aplite dikes. Here, the pre- dominant quartz monzonites might be the equivalent of the quartz monzonite porphyry of Rocky Peak or the Sherman Granite, while the various muscovite—contain- ing granites might be the equivalents of the quartz monzonite of Roxy Ann Lake or the Silver Plume Gra- nite. Elsewhere in Colorado, the split between a slightly older porphyritic Sherman equivalent and a slightly younger, finer grained, muscovite-containing Silver Plume equivalent may also be present. In the Black Ca- nyon of the Gunnison, the older but volumetrically pre- dominant mesokatazonal Vernal Mesa Quartz Monzonite, a very coarse grained porphyritic biotite quartz monzonite and granodiorite, may be a Sherman equivalent while the younger, volumetrically less im- portant, epimesozonal Curecanti Quartz Monzonite, a medium grained biotite-muscovite quartz monzonite or granite in separate plutons, may be a Silver Plume equivalent (Hansen, 1964; 1968; 1971; Hansen and Peterman, 1968). A batholith of Vernal Mesa and younger separate bodies of biotite-muscovite granite are also reported from the Uncompahgre Plateau (Aldrich and others, 1958; Hedge and others, 1968). The fine-grained to coarsely porphyritic biotite- muscovite granite to granodiorite of the St. Kevin batholith may be the Silver Plume equivalent in the Sawatch Range (Tweto and Pearson, 1964; Pearson and others, 1966). Older porphyritic quartz monzonite of Eleven Mile Canyon (Sherman lithologic equivalent?) has been intruded by fine- to medium-grained quartz monzonite tentatively correlated with Silver Plume in the Florissant quadrangle of the southern Front Range (Wobus, 1969). MAFIC DIKES Tabular dikes of late Precambrian age, variously described as “porphyry,” “lamprophyre,” or just “mafic” are present in many areas of Colorado, and in- trusions appear to have occurred over a very long span of time. Some dikes have been dated directly; most can be bracketed indirectly but usually not within precise limits. In a few areas, dikes of two ages can be observed or deduced even though hand specimens of the two types may be indistinguishable from each other. (For example, the porphyry dikes of pre- and post-quartz monzonite of Roxy Ann Lake age described in this report; see also Pearson and others, 1966, p. 1113; Ga- INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO ble, 1968, p. E35; and Hansen and Peterman, 1968, p. CS6 —CS7, for similar phenomena in the Sawatch, Mor- rison, and Black Canyon areas.) Evidence for at least five periods of mafic dike-producing magma intrusion is available in the Medicine Bow Mountains of Wyoming (Houston and others, 1968). When all available data are assembled for the Precambrian of Colorado and adja- cent Wyoming (fig. 19), the following statements seem pertinent: 1. Basalt or diabase dike-producing events were pro- minent north of the Mullen Creek -Nash Fork shear zone (Houston and others, 1968) from about 2.45 to 2.24 b.y. ago. As far as is known, these are not recorded anywhere in Colorado. 2. Porphyry or lamprophyre dike-producing events were prominent somewhere in Colorado from about 1.52 b.y. to probably 1.16 b.y. ago, possibly to 1.05 b.y. ago. No single area had dike activity throughout this time. Some areas as noted ex- perienced two pulses but these double events seem not to have been completely simultaneous in all separate areas. 3. Late mafic dike events occurred. locally after 600 my. ago. 4. Evidence is lacking for mafic dike events during the Precambrian in Colorado and Adjacent Wyoming except at the times summarized above. North and northwest of this area in Wyoming and in adjacent Montana as many as seven ages of basalt in- trusion have been recognized in the range 0.7 to >26 b.y. (Mueller, 1970; Rowan and Mueller, 1971), and the Beartooth Mountains “nonfolded dike set” in the range 1.2 to 1.5 b.y. may be equivalent to the Colorado porph- yry-lamprophyre dike set, the one represented in the Park Range. Prinz (1964) recognized three ages of Pre- cambrian dikes and one age of Tertiary(?) dikes; one of these Tertiary(?) dikes was later dated by K-Ar whole rock at 1.5. b.y. old. (Condie and others, 1969). The lat- ter authors dated many different dikes by K-Ar whole- rock from the Beartooth Mountains of Montana and Wyoming, the Bighorn Mountains, the Owl Creek Mountains, and the Wind River Range of Wyoming and recognized four ages of dikes in the range 0.7 to 2.5 b.y. Other dating techniques and my. too young (Heimlich and Banks, 1968 — Bighorn Mountains; Spall, 1971 — Wind River Range; Rowan and Mueller, 1971 — Bear- tooth Mountains). Rosholt and Peterman (1969) have indicated 1.6 b.y. as a time of extensive diabase dike emplacement in the Granite Mountains of Wyoming. John C. Reed, Jr. (oral and written communs., 1972; Reed and Zartman, 1972, p. 404; 1973, p. 576) reported that a 150-foot diabase dike in the Teton Range of Wyoming was intruded between 1.3 and 2.5 b.y. ago, possibly closer to the younger age (US. Geological Survey, 1971). It may well be, as Condie and others REFERENCES CITED (1969) and Mueller (1970) contended, that dike intru- sion in this part of the earth’s crust was confined to cer- tain definite periods of time between which no dikes were intruded, although future data may extend the known times of dike intrusion. REFERENCES CITED Abbott, J. T., 1972, Rb-Sr study of isotopic redistribution in a Pre- cambrian mylonite-bearing shear zone, northern Front Range, Colorado: Geol. Soc. America Bull., v. 83, no. 2, p. 487 —493. Aldrich, L. T., Wethen'll, G. W., and Davis, G. L., 1957, Occurrence of 1,350 million-year-old granitic rocks in western United States: Geol. Soc. America Bull., v. 68, no. 5, p. 655 —656. Aldrich, L. T., Wetherill, G. W., Davis, G. L., and Tilton, G. R., 1958, Radioactive ages of micas from granitic rocks by Rb-Sr and K-A methods: Am. Geophys. Union Trans, v. 39, no. 6, p. 1124—1134. Antweiler, J. C., 1966, Isotopic evidence of Precambrian episodes of mineralization in Colorado in Abstracts for 1965: Geol. Soc. America Spec. Paper 87, p. 270 -271. Antweiler, J. C., Doe, B. R., and Delevaux, M. H., 1972, Lead isotope and other evidence on the bedrock source of placer gold at Hahns Peak, Colorado: Econ. Geology, v. 67, no. 3, p. 302 —314. Atwood, W. W., Jr., 1937, Records of Pleistocene glaciers in the Medicine Bow and Park Ranges [Wyo.-Colo.]: Jour. Geology, v. 45, no. 2, p. 113—140. Barclay, C. S. V., 1968, Geology of the Gore Canyon-Kremmling area, Grand County, Colorado: US. Geol. Survey open-file rept., 187 p. Barker, Fred, 1969, Precambrian geology of the Needle Mountains, southwestern Colorado: US. Geol. Survey Prof. Paper 644 —A, 35 p. [1970l. Barker, Fred, Peterman, Z. E., and Hildreth, R. A., 1969a, Genesis of the Precambrian Twilight Gneiss, West Needle Mountains, Col- orado— Implications of Rb and Sr [abs]: Geol. Soc. America Abs. with Programs for 1969, pt. 5, p. 5—6. __1969b, A rubidium-strontium study of the Twilight Gneiss, West Needle Mountains, Colorado: Contr. Mineralogy and Petrology, v. 23, no. 4, p. 271 —282. Barker, Fred, Peterman, Z. E., and Marvin, R. F., 1969, Precambrian melasyenite stock of Ute Creek, San Juan Mountains, Colorado [abs]: Geol. Soc. America Abs. with Programs for 1969, pt. 5, p. 6. Barnwell, William. 1955, The geology of the south Hahns Peak dis- trict: Wyoming Univ., M.S. thesis, 91 p. Bass, N. W., Eby, J. B., and Campbell, M. R., 1955, Geology and mineral fuels of parts of Routt and Moffat Counties, Colorado: U.S. Geol. Survey Bull. 1027 —D, p. 143 —250 [1956]. Bateman, P. 0., Clark, L. D., Huber, N. K., Moore, J. G., and Rinehart, C. D., 1963, The Sierra Nevada batholith — a synthesis of recent work across the central part: U.S. Geol. Survey Prof. Paper 414-D, 46 p. Behrendt, J. C., and Popenoe, Peter, 1969, Basement structure con- tour map of North Park-Middle Park basin, Colorado: Am. Assoc. Petroleum Geologists Bull., v. 53, no. 3, p. 678—682. Behrendt, J. C., Popenoe, Peter, and Mattick, R. E., 1969, A geophysi- cal study of North Park and the surrounding ranges, Colorado: Geol. Soc. America Bull., v. 80, no. 8, p, 1523—1538. Bickford, M. E., Wetherill, G. W., Barker, Fred, and Lee-Hu, Chin- Nan, 1969, Precambrian Rb-Sr chronology in the Needle Moun- tains, southwestern Colorado: Jour. Geophys. Research, v. 74, no. 6, p. 1660 -1676. 39 Bowes, W. A., Segerstrom, Kenneth, and Young, E. J ., 1968, Dissemi- nated lead-zinc-silver deposit at Hahns Peak, Routt County, Col- orado — a preliminary report: Prague, Internat. Geol. Cong., 23d, Abstracts, p. 179—180. Brock, M. R., and Singewald, Q. D., 1968, Geologic map of the Mount Tyndall quadrangle, Custer County, Colorado: US. Geol. Survey Geol. Quad. Map GQ —596. Bryant, Bruce, Miller, R. D., and Scott, G. R., 1971, Geologic map of the Indian Hills quadrangle, Jefferson County, Colorado: US. Geol. Survey open-file rept., 59 p. Buffler, R. T., 1967, The Browns Park Formation and its relationship to the late Tertiary geologic history of the Elkhead region, north- western Colorado —south-central Wyoming: Univ. California (Berkeley), Ph. D. thesis, 175 p. Christensen, A. L., 1942, Igneous geology of the Elkhead Mountains, Colorado: Univ. California (Berkeley), Ph. D. thesis, 180 p. Clarke, F. W., 1904, Analyses of rocks from the laboratory of the United States Geological Survey 1880 to 1903: US. Geol. Survey Bull. 228, 375 p. __1910, Analyses of rocks and minerals from the laboratory of the United States Geological Survey 1880 to 1908: US. Geol. Survey Bull. 419, 323 p. Coffin, R. C., Perini, V. C., Jr., and Collins, M. J., 1924, Some anti- clines of western Colorado: Colorado Geol. Survey Bull. 24, 68 p. (repr. with parts of Crawford, Willson, and Perini, 1920). Condie, K. C., 1969a, Petrology and geochemistry of the Laramie batholith and related metamorphic rocks of Precambrian age, eastern Wyoming: Geol. Soc. America Bull., v. 80, no. 1, p. 57 —82. __1969b, Petrology and geochemistry of the Laramie batholith and related metamorphic rocks of Precambrian age, eastern Wyoming— Reply: Geol. Soc. America Bull., v. 80, no. 11, p. 2385 —2386. Condie, K. C., Leech, A. P., and Baadsgaard, H., 1969, Potassium- argon ages of Precambrian mafic dikes in Wyoming: Geol. Soc. America Bull., v. 80, no. 5, p. 899—906. Corbett, M. K., 1968, Tertiary volcanism of the Specimen —Lulu—Iron Mountain area, north-central Colorado, in Cenozoic volcanism in the southern Rocky Mountains: Colorado School Mines Quart, v. 63, no. 3, p. 1—37. Crawford, R. D., Willson, K. M., and Perini, V. C., 1920, Some anti- clines of Routt County, Colorado: Colorado Geol. Survey Bull. 23, 59 p. (Partly repr. in Coffin, Perini, and Collins, 1924.) Cudzilo, T. F., 1971, Regional aspects of radiometric ages of Pre— cambrian rocks in the central Uncompahgre Plateau, west- central Colorado [abs]: Geol. Soc. America Abs. with Programs, v. 3, no. 3, p. 235. Deutsch, S., Hirschberg, D., and Picciotto, E., 1956, Etude quantita- tive des halos pléochro'i'ques— Application ‘a l’estimation de l’age des roches granitiques: Soc. Belge Geologie Bull., v. 65, no. 2, p. 267 —281. Doe, B. R., and Pearson, R. C., 1969, U-Th-Pb chronology of zircons from the St. Kevin Granite, northern Sawatch Range, Colorado: Geol. Soc. America Bull., v. 80, no. 12, p. 2495—2502. Eggler, D. H., 1967, Gravity survey of the Livermore —’I‘ie Siding area, Colorado-Wyoming: The Mountain Geologist, v. 4, no. 3, p. 109—112. .._1968, Virginia Dale Precambrian ring-dike complex, Colorado- Wyoming: Geol. Soc. America Bull., v. 79, no. 11, p. 1545—1564. Eggler, D. H., Larson, E. E., and Bradley, W. C., 1969, Granites, grasses, and the Sherman erosion surface, southern Laramie Range, Colorado-Wyoming: Am. Jour. Sci., v. 267, no. 4, p. 510—522. Fenneman, N. M., and Gale, H. S., 1906, The Yampa coal field, Routt County, Colorado, with a chapter on the character and 40 use of the Yampa coals, by M. R. Campbell: U.S. Geol. Survey Bull. 297, 96 p. Fenton, M. D., and Faure, G., 1969, The strontium isotopic composi- tion of the Laramie Range syenite, Wyoming, and its bearing on petrogenesis: Geol. Soc. America Abs. with Programs for 1969, pt. 6, p. 15—16. Ferris, C. 5., Jr., and Krueger, H W., 1964, new radiogenic dates on ig- neous rocks from the southern Laramie Range, Wyoming: Geol. Soc. America Bull., v. 75, no. 10, p. 1051 —1054. Gable, D. J., 1968, Geology of the crystalline rocks in the western part of the Morrison quadrangle, Jefferson County, Colorado: U.S. Geol. Survey Bull. 1251—E, 45 p. Gale, H. S., 1906, The Hahns Peak gold field, Colorado: U.S. Geol. Survey Bull. 285—A, p. 28—34. George, R. D., and Crawford, R. D., 1909, The Hahns Peak region, Routt County, Colorado— An outline survey: Colorado Geol. Survey 1st Rept. 1908, p. 189-229. Giletti, B. J., and Gast, P. W., 1961, Absolute age of Pre-Cambrian rocks in Wyoming and Montana, in Geochronology of rock systems: New York Acad. Sci. Annals, v. 91, art. 2, p. 454 -458. Gross, E. B., and Heinrich, E. W., 1966, Petrology and mineralogy of the Mount Rosa area, El Paso and Teller Counties, Colorado — [Pt.] 2, Pegmatites: Am. Mineralogist, v. 51, nos. 3—4, p. 299 —323. Haffty, Joseph, and Riley, L. B., 1968, Determination of palladium, platinum, and rhodium in geologic materials by fire assay and emission spectrography: Talanta, v. 15, no. 1, p. 111—117.‘ Hague, Arnold, and Emmons, S. F., 1877, Descriptive geology: U.S. Geol. Explor. 40th Parallel (King), v. 2, 890 p. Hail, W. J ., Jr., 1965, Geology of northwestern North Park, Colorado: U.S. Geol. Survey Bull. 1188, 133 p. ___1968, Geology of southwestern North Park and vicinity, Col- orado: U.S. Geol. Survey Bull. 1257, 119 p. Hansen, W. R., 1964, Curecanti pluton, an unusual intrusive body in the Black Canyon of the Gunnison, Colorado: U.S. Geol. Survey Bull. 1181 —D, 15 p. __1968, Geologic map of the Black Ridge quadrangle, Delta and Montrose Counties, Colorado: U.S. Geol. Survey Geol. Quad. Map GQ —747. _1971, Geologic map of the Black Canyon of the Gunnison River and vicinity, western Colorado: U.S. Geol. Survey Misc. Inv. Ser. Map I—584. Hansen, W. R., and Peterman, Z. E., 1968, Basement-rock geochronology of the Black Canyon of the Gunnison, Colorado, in Geological Survey research 1968: U.S. Geol. Survey Prof. Paper 600-0, p. CSO—C90. Harrison, J. E., and Wells, J. D., 1959, Geology and ore deposits of the Chicago Creek area, Clear Creek County, Colorado: U.S. Geol. Survey Prof. Paper 319, 92 p. Hawley, C. C., Huffman, Claude, Jr., Hamilton, J. C., and Rader L. F., J r., 1966, Geologic and geochemical features of the Redskin Gra- nite and associated rocks, Lake George beryllium area, Colorado, in Geological Survey research 1966: U.S. Geol. Survey Prof. Paper 550 —C, p. 0138—0147. Hayase, Ichikazu, 1954, Relative geologic age measurements on gra- nites by pleochroic haloes and the radioactivity of the minerals in their nuclei: Am. Mineralogist, v. 39, nos. 9, 10, p. 761—772. Hedge, C. E., 1970, Whole-rock Rb-Sr age of the Pikes Peak batholith, Colorado, in Geological Survey research 1970: U.S. Geol. Survey Prof. Paper 700 —B, p. B86—BS9. Hedge, C. E., Peterman, Z. E., and Braddock, W. A., 1967, Age of the major Precambrian regional metamorphism in the northern Front Range, Colorado: Geol. Soc. America Bull., v. 78, no. 4, p. 551 —558. INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO Hedge, C. E., Peterman, Z. E., Case, J. E., and Obradovich, J. D., 1968, Precambrian geochronology of the northwestern Uncompahgre Plateau, Utah and Colorado, in Geological Survey research 1968: U.S. Geol. Survey Prof. Paper 600 —C, p. C91 —C96. Heimlich, R. A., and Banks, P. 0., 1968, Radiometric age determina- tions, Bighorn Mountains, Wyoming: Am. Jour. Sci., v. 266, no. 3, p. 180 —192. Hepp, M. M., 1966, A Precambrian andesite dike swarm in the north. eastern Front Range, Larimer County, Colorado: Boulder, Col- orado Univ., unpub. M.S. thesis, 75 p. Hills, F. A., and Armstrong, R. L., 1971, Rb/Sr and K/Ar geochronology of the Laramie Range, southern Wyoming [abs.]: Geol. Soc. America Abs. with Programs, v. 3, no. 7, p. 599 —600. Hills, F. A., Gast, P. W., and Houston, R. S., 1965, Chronology of some Precambrian igneous and metamorphic events of the Medicine Bow Mountains, Wyoming, in Abstracts for 1964: Geol. Soc. America Spec. Paper 82, p. 92. _1967, Limits on the age of Precambrian metasedimentary rocks in the Medicine Bow Mountains of Wyoming and in nearby areas [abs.], in Geochronology of Precambrian stratified rocks: Geochronology Comm., I.U.G.S., Canada Geol. Survey, and Alberta Univ., Dept. Geology, p. 51 —52. Hills, F. A., Gast, P. W., Houston, R. S., and Swainbank, I. G., 1968, Precambrian geochronology of the Medicine Bow Mountains, southeastern Wyoming: Geol. Soc. America Bull., v. 79, no. 12, p. 1757 —17 84. Houston, R. S., Hills, Alan, and Gast, P. W., 1966, Regional aspects of structure and age of rocks of the Medicine Bow Mountains, Wyoming, in Abstracts for 1965: Geol. Soc. America Spec. Paper 87, p. 287—288. Houston, R. S., and others, 1968, A regional study of rocks of Pre- cambrian age in that part of the Medicine Bow Mountains lying in southeastern Wyoming — with a chapter on the relationship between Precambrian and Laramide structure: Wyoming Geol. Survey Mem. 1, 167 p. Hutchinson, R. M., 1960a, Petrotectonics and petrochemistry of late Precambrian batholiths of central Texas and the north end of Pikes Peak batholith, Colorado: Copenhagen, Internat. Geol. Cong., 21st., Rept., pt. 14, p. 95 —107. _1960b, Structure and petrology of north end of Pikes Peak batholith, Colorado, in Guide to the geology of Colorado: Geol. Soc. America, Rocky Mtn. Assoc. Geologists, and Colorado Sci. Soc., p. 170—180. Hutchinson, R. M., and Hedge, C. E., 1968, Depth-zone emplacement and geochronology of Precambrian plutons, central Colorado Front Range, in Abstracts for 1967: Geol. Soc. America Spec. Paper 115, p. 424—425. Irvine, T. N., and Baragar, W. R. A., 1971, A guide to the chemical classification of the common volcanic rocks: Canadian Jour. Earth Sci., v. 8, no. 5, p. 523-548. Izett, G. A., 1966, Tertiary extrusive volcanic rocks in Middle Park, Grand County, Colorado, in Geological Survey research 1966: U.S. Geol. Survey Prof. Paper 550 —B, p. B42 —B46. _1968, Geology of the Hot Sulphur Springs quadrangle, Grand County, Colorado: U.S. Geol. Survey Prof. Paper 586, 79 p. Izett, G. A., Benson, N. M., and Obradovich, J. D., 1970, K-Ar age of the lower part of the Browns Park Formation, northwestern Col- orado, in Geological Survey research 1970: U.S. Geol. Survey Prof. Paper 700—C, p. 0150—C152. Jaffe, H. W., Gottfried, David, Waring, C. L., Worthing, H. C., 1959, Lead-alpha age determinations of accessory minerals of igneous rocks: U.S. Geol. Survey Bull. 1097 -B, p. 65—148. James, R. S., and Hamilton, D. L., 1969, Phase relations in the system NaAlSi303—KAlSi303~CaAlZSi208—Si02 at 1 kilobar water vapour pressure: Contr. Mineralogy and Petrology, v. 21, p. 1 1 1 -141. REFERENCES CITED 41 Johannsen, Albert, 1939, Introduction, textures, classifications, and glossary, V. 1, of A descriptive petrography of the igneous rocks [2d ed.]: Chicago Univ. Press, 318 p. Kirst, P. W., 1968, Petrology and structural relationships of the Pre- cambrian crystalline rocks of east-central Mummy Range, Col- orado, in Abstracts for 1966: Geol. Soc. America Spec. Paper 101, p. 405 —406. Klugman, M. A., 1960, Laramie anorthosite, in Rocky Mtn. Assoc. Geologists, Guide to the geology of Colorado: Geol. Soc. America, Rocky Mtn. Assoc. Geologists, and Colorado Sci. Soc., p. 223 -227. Invering, T. S., and Goddard, E. N., 1950, Geology and ore deposits of the Front Range, Colorado: US. Geol. Survey Prof. Paper 223, 319 p. [1951]. Levering, T. S., and Tweto, Ogden, 1953, Geology and ore deposits of the Boulder County tungsten district, Colorado: US. Geol. Survey Prof. Paper 245, 199 p. [1954]. Luth, W. C., Jahns, R. H., and Tuttle, 0. F., 1964, The granite system at pressures of 4 to 10 kilobars: Jour. Geophys. Research, v. 69, no. 4, p. 759—773. McDowell, F. W., 1966, Potassium-argon dating of Cordilleran in- trusives: Columbia Univ. (New York), unpub., Ph. D. thesis. Moench, R. H., Harrison, J. E., and Sims, P. K., 1962, Precambrian folding in the Idaho Springs—Central City area, Front Range, Colorado: Geol. Soc. America Bull., v. 73, no. 1, p. 35—58. Mose, D. G., and Bickford, M. E., 1969, Precambrian geochronology in the Unaweep Canyon, west-central Colorado: Jour. Geophys. Research, v. 74, no. 6, p. 1677 —1687. Mueller, P. A., 1970, Secular variations in the mafic rocks of the southern Beartooth Mountains, Montana and Wyoming [abs.]: Geol. Soc. America Abs. with Programs, v. 2, no. 7. p. 632. Mutschler, F. E., and Larson, E. E., 1969, Paleomagnetism as an aid to age classification of mafic intrusives in Colorado: Geol. Soc. America Bull., v. 80, no. 11, p. 2359—2368. Naeser, C. W., Izett, G. A., White, W. H., 1973, Zircon fission-track ages from some Middle Tertiary igneous rocks in northwestern Colorado [abs.]: Geol. Soc. America Abs. with Programs, v. 5, no. 6, p. 498. Pearson, R. 0., Hedge, C. E., Thomas, H. H., and Stern, T. W., 1966, Geochronology of the St. Kevin Granite and neighboring Pre- cambrian rocks, northern Sawatch Range, Colorado: Geol. Soc. America Bull., v. 77, no. 10, p. 1109—1120. Peterman, Z. E., and Hedge, C. E., 1967, Precambrian history of the Front Range and adjacent areas, Colorado [abs.] in Geochronology of Precambrian stratified rocks: Geochronology Comm. I.U.G.S., Canada Geol. Survey, and Alberta Univ., Dept. Geology, p. 77—78. Peterman, Z. E., Hedge, C. E., and Braddock W. A., 1968, Age of Pre- cambrian events in the northeastern Front Range, Colorado: Jour. Geophys. Research, v. 73, no. 6, p. 2277—2296. Phair, George, and Shimamoto, K. 0., 1952, Hydrothermal uranothorite in fluorite breccias from the Blue Jay mine, J ames- town, Boulder County, Colorado: Am. Mineralogist, v. 37, nos. 7, 8, p. 659 -666. Popenoe, Peter, Brinkworth, G. L., Worl, R. G., and Behrendt, J. C., 1970, Geophysical lineations in north-central Colorado and their possible relationship to Tertiary plutonic activity [abs.]: Geol. Soc. America Abs. with Programs, v. 2, no. 5, p. 343. Prinz, Martin, 1964, Geologic evolution of the Beartooth Mountains, Montana and Wyoming— Pt. 5, Mafic dike swarms of the southern Beartooth Mountains: Geol. Soc. America Bull., v. 75, no. 12, p. 1217—1248. Reed, J. 0., Jr., and Zartman, R. E., 1972, Geochronology of Pre- cambrian rocks of the Teton Range, Grand Teton National Park, Wyoming [abs.]: Geol. Soc. America Abs. with Programs, v. 4, no. 6, p. 403—404. _1973, Geochronology of Precambrian rocks of the Teton Range, Wyoming: Geol. Soc. America Bull., v. 84, no. 2, p. 561 -—582. Rittmann, A., 1952, Nomenclature of volcanic rocks — proposed for the use in the catalogue of volcanoes, and key-tables for the determination of volcanic rocks: Bull. Volcanol., ser. 2, v. 12, p. 75-102. Rosholt, J. N., and Peterman, Z. E., 1969, Uranium, thorium, lead systematics in the Granite Mountains, Wyoming [abs.]: Geol. Soc. America Abs. with Programs for 1969, pt. 5, p. 70. Ross, C. S., 1926, A Colorado lamprophyre of the verite type: Am. Jour. Sci., 5th ser., v. 12, no. 69, p. 217—229. Rowan, L. C., and Mueller, P. A., 1971, Relations of folded dikes and Precambrian polyphase deformation, Gardner Lake area, Bear- tooth Mountains, Wyoming: Geol. Soc. America Bull., v. 82, no. 8, 7 p. 2177—2186. Segerstrom, Kenneth, and Kirby, S. H., 1969, Tuffaceous epiclastic breccia and sandstone near Hahns Peak, Colorado, and their genetic implications in Geological Survey research 1969: US. Geol. Survey Prof. Paper 650—B, p. B19—B22. Segerstrom, Kenneth, and Young, E. J ., 1972, General geology of the Hahns Peak and Farwell Mountain quadrangles, Routt County, Colorado, with a discussion of Upper Triassic and pre-Morrison Jurassic rocks, by G. N. Pipiringos: U.S. Geol. Survey Bull. 1349, 63 p. [1973]. Silver, L. T., and Barker, Fred, 1968, Geochronology of Precambrian rocks of the Needle Mountains, southwestern Colorado — Pt. 1, U-Pb zircon results, in Abstracts for 1967: Geol. Soc. America Spec. Paper 115, p. 204—205. Sims, P. K., and Gable, D. J., 1967, Petrology and structure of Pre- cambrian rocks, Central City quadrangle, Colorado: US. Geol. Survey Prof. Paper 554 —E, 56 p. Smithson, S. B., and Hodge, D. S., 1969, Petrology and geochemistry of the Laramie batholith and related metamorphic rocks of Pre- cambrian age, eastern Wyoming— Discussion: Geol. Soc. America Bull., v. 80, no. 11, p. 2383 —2384. Spall, Henry, 1971, Paleomagnetism and K-Ar age of mafic dikes from the Wind River Range, Wyoming: Geol. Soc. America Bull., v. 82, no. 9, p. 2457 —2472. Stern, T. W., Phair, George, and Newell, M. F., 1971, Boulder Creek batholith, Colorado— pt. 2, Isotopic age of emplacement and morphology of zircon: Geol. Soc. America Bull., v. 82, no. 6, p. 1615-1633. Steven, T. A., 1954, Geology of the Northgate fluorspar district, Col- orado: U.S. Geol. Survey Mineral Inv. Field Studies Map MF —13. _1957, Metamorphism and the origin of granitic rocks, North- gate district, Colorado: U.S. Geol. Survey Prof. Paper 274 —M, p. M335 —M377. __1960, Geology and fluorspar deposits, Northgate district, Col- orado: U.S. Geol. Survey Bull. 1082 -F, p. F323—F422 [1961]. Strangway, D. W., Larson, E. E., and York, D., 1969, A middle Terti- ary magnetic transition in northwestern Colorado [abs.]: EOS (Am. Geophys. Union Trans), v. 50, no. 4, p. 131. Streckeisen, A. L., 1967, Classification and nomenclature of igneous rocks — (Final report of an inquiry): Neues Jahrb. Mineralogie Abh., v. 107, no. 2, p. 144—214. Taylor, R. B., and Sims, P. K., 1962, Precambrian gabbro in the central Front Range, Colorado, in Short papers in geology, hy- drology, and topography: U.S. Geol. Survey Prof. Paper 450 —D, p. D118—122. Taylor, R. B., Theobald, P. K., and Izett, G. A., 1968, Mid-Tertiary volcanism in the central Front Range, Colorado, in Cenozoic volcanism in the southern Rocky Mountains: Colorado School Mines Quart., v. 63, no. 3, p. 39 —50. 42 Thornton, C. P., and Tuttle, O. F., 1960, Chemistry of igneous rocks — [Pt.] 1, Differentiation index: Am. Jour. Sci., v. 258, no. 9, p. 664 —684. Tuttle, O. F., and Bowen, N. L., 1958, Origin of granite in the light of experimental studies in the system NaAlSiaos—KAISiaos— Si02—H20: Geol. Soc. America Mem. 74, 153 p. Tweto, Ogden, and Pearson, R. C., 1964, St. Kevin granite, Sawatch Range, Colorado, in Short papers in geology and hydrology: U.S. Geol. Survey Prof. Paper 47 5—D, p. D28 —D32. Tweto, Ogden, and Sims, P. K., 1963, Precambrian ancestry of the Colorado mineral belt: Geol. Soc. America Bull., v. 74, no. 8, p. 991 —1014. U. S. Geological Survey, 1966, Geological Survey research 1966: US. Geol. Survey Prof. Paper 550—A, 385 p. [1967]. _1967, Geological Survey research 1967: US. Geol. Survey Prof. Paper 575—B, p. B181 —B186. _._1968, Geological Survey research 1968: US. Geol. Survey Prof. Paper 600 —A, 371 p. _._1969, Geological Survey research 1969: US. Geol. Survey Prof. Paper 650 —A, 425 p. [1970]. INTRUSIVE ROCKS NORTHEAST OF STEAMBOAT SPRINGS, PARK RANGE, COLORADO _1971, Geological Survey research 1971: US. Geol. Survey Prof. Paper 750—A, 418 p. [1972]. Washington, H. S., and Larsen, E S., 1913, Magnetite basalt from North Park, Colorado: Washington Acad. Sci. Jour., v. 3, no. 17, p. 449 —452. Wells, J. D., 1967, Geology of the Eldorado Springs quadrangle, Boulder and Jefferson Counties, Colorado: US. Geol. Survey Bull. 1221—D, 85 p. Wetherill, G. W., and Bickford, M. E., 1965, Primary and metamorphic Rb-Sr chronology in central Colorado: Jour. Geo- phys. Research, v. 70, no. 18, p. 4669—4686. Wobus, R. A., 1969, Granitic rocks of the Florissant quadrangle, southern Front Range, Colorado [abs]: Geol. Soc. America Abs. with Programs for 1969, pt. 5, p. 90. Yoder, H. S., Stewart, D. B., and Smith J. R., 1957, Ternary feldspars: Carnegie Inst. Washington Year Book 56, p. 206 —214. York, Derek, Strangway, D. W., and Larson, E. E., 1971, Preliminary study of a Tertiary magnetic transition in Colorado: Earth and Planetary Sci. Letters, V. 11, no. 4, p. 333 —338. QUiS. GOVERNMENT PRINTING OFFICE: 1976—677-026/17 GEOLOGICAL SURVEY TABLE 5. — CHEMICAL, MODAL, AND SEMIQUAN TI TA TI VE SPECTROGRAPHIC ANAL YSES FOR SELECTED IGNEOUS AND METAIGNEOUS ROCKS 0F PARK RANGE NORTHEAST OF STEAMBOAT SPRINGS, COLORADO PROFESSIONAL PAPER 1041 TABLE 5 Mount Ethel Pluton Buffalo p855 Pluton _ - . , Fine-grained h d‘k Small mafic intrusive (dunite) Felsic gneiss country rock Tertiary intrusives Rock unit RDCk unit ....... Leucogranite Graflsnazggigyrtz Quartz monzonite of Roxy Ann Lake Quartz monzonite porphyry of Rocky Peak $2322; 3:11:12: ;:?::r;;t;:&:1§r&r;fl:;;e Quartzrrgioszzomte porp y ry 1 es 4 437 157 525 520 551 544 555 554 550 553 549 524 545 116 116 116 (Avg) 631 593 616 226 226 226 (Avg) 22 22 22 (Avg) 19 19 19 (Avg) field No. {$333636 0161650 0161643 0161644 0101649 0166645 01717648 0160640 0101641 0101278 0101647 0101642 0101655 0101651 0101658 0101657 0101654 0101658 0101653 0101646 0101652 0101277 0141158 0101279 0141162 0101276 0141156 0101275 0141155 Laboratory No. CHEMICAL ANALYSES, IN WEIGHT PERCENT [Analysts: (1) George 0. Riddle, 3 715 766; (2) E. Englernan, 7 -18 567; (3) George 0. Riddle, 10 *5 770. For Rittmann (1952) nomenclature, some aphanitic equivalents (in quotes) are given for phaneritic rocks for comparative purposes. Blanks indicate not looked for.l Sio2 ....... 75.40 ...... 73.88 ...... 73.98 ...... 70.62 ...... 70.19 ...... 71.03 ...... 73.22 ....... 67.09 ................. 68.70 ................. 69.42 ........... 70.52 56.74 ..... 64.14 ..... 65.46 ..... 65.80 ..... 70.09 ..... 74.32 ..... 52.83 ..... 56.58 .......... 61.40 ............ 37.16 ...... 36.63 ...... 36.90 ...... 113.3; ...... 48.85 ...... 48.96 ....... 56.40 ..... 5627 ..... 56.34 ..... 58.31 ..... 5729 ..... 57.80 ..... Sio, A1203 ...... 13.68 ...... 13.73 ...... 13.88 ...... 14.78 ...... 14.66 ...... 14.74 ...... 14.04 ....... 15.02 ................. 14.62 ................. 14.68 ........... 14.40 15.83 ..... 16.93 ..... 15.84 ..... 16.17 ..... 14.70 ..... 13.67 ..... 15.74 ..... 15.24 .......... 15.34 ............ 4.64 ...... 5.13 ...... 4.89 ...... . ...... 14.09 ...... 14.03 ....... 15.19 _____ 15.09 _____ 1514 ..... 15.09 ..... 15.04 .... 15.07 ..... A1203 F1220, ...... .47 ...... .50 ...... .61 ...... 1.15 ...... 1.05 ...... .86 ...... .80 ....... 2.12 ................. 1.91 ................. 1.73 ........... 1.24 3.81 ..... 1.25 ..... 1.47 ..... 1.34 ..... .72 ..... .19 ..... 1.76 ..... 3.86 .......... 3.43 ............ 13.29 ...... 13.47 ...... 13.38 ...... 3.73 ...... 3.85 ...... 3.79 ....... 5.41 _____ 5'37 _____ 5.39 IIII 2.47 ..... 2.4. ..... “6 Fe 0 FeO ........ .09 ...... .94 ...... .76 ...... 1.22 ...... 1.80 ...... 1.33 ...... .99 ....... 1.87 ................. 1.47 ................. 1.28 ........... 1.44 .. 4.01 ..... 2.57 ..... 2.25 ..... 2.16 ..... 2.09 ..... 1.06 ..... 8.17 ..... 4.66 .......... 2.92 ............ 5.47 ...... 5.38 ...... 5.43 ...... 5.05 ...... 4.96 ...... 5.01 ....... 2.13 ..... 2.16 ..... 215 ..... 3.43 ..... 3'4. .... 3'44 ..... Fe?) 3 MgO ....... .05 ...... .27 ...... .24 ...... .64 ...... .57 ...... .48 ...... .35 ....... 1.18 ................. .97 ................. .79 ........... .75 .. 3.34 . .. .. 1.35 ..... 1.34 ..... 1.24 ..... .65 ..... .26 ..... 4.87 ..... 3.21 .......... 1.94 ............ 28.65 ...... 28.34 ...... 28.50 ...... 8.18 ...... 8.03 ...... 8.11 ....... 5.37 ..... 5.39 ..... 538 ..... 4.74 ''''' 4.64 ““““ 4.69 ..... M30 0110 ....... .35 ...... 1.10 ...... .89 ...... 1.82 ...... 1.68 ...... 1.72 ...... 1.21 ....... 2.85 ................. 2.27 ................. 2.02 ........... 1.76 5.81 ..... 3.44 ..... 3.07 ..... 3.13 ..... 1.50 ..... 1.06 ..... 7.17 ..... 5.60 .......... 4.12 ............ 2.40 ...... 2.41 ...... 2.41 ...... :92] ...... 7.91 ...... 7.92 ....... 6'39 ..... 6.48 ..... 6.44 ..... 5'82 ..... 6.22 ..... 6.02 _____ CaO Na,0 ...... 3.79 ...... 3.45 ...... 3.42 ...... 3.65 ...... 3.55 ...... 3.43 ...... 3.58 ....... 3.70 ................. 3.50 ................. 3.43 ........... 3.60. 3.36 ..... 3.65 ..... 3.61 ..... 3.61 ..... 3.00 ..... 3.05 ..... 3.26 ..... 3.58 .......... 3.73 ............ .31 ...... .28 ...... .30 ...... 3.49 ...... 3.36 ...... 3.42 ....... 4.28 _____ 4'19 ..... 424 ..... 4'17 ..... 4.12 .... 4.15 ..... N320 K20 ........ 5.71 ...... 5.17 ...... 5.09 ...... 4.72 ...... 4.80 ...... 5.12 ...... 4.81 ....... 4.29 ................. 4.86 ................. 4.93 ........... 4.60 _ 2.78 ..... 5.17 ..... 4.64 ..... 5.01 ..... 5.48 ..... 5.29 ..... 2.06 ..... 2.91 .......... 3.71 ............ .04 ...... .05 ...... .05 ...... . ...... 3.11 ...... 3.15 ....... 2'24 ..... 2'26 ..... 2.25 ..... 2'47 ..... 2.50 . 249 ''''' K20 H20+ ...... .08 ...... .36 ...... .46 ...... .48 ...... .56 ...... .29 ...... .35 ....... .29 ................. 31 ................. .43 ........... .50 . 1.49 ..... .39 ..... .97 ..... .40 ..... .64 ..... .39 ..... 1.41 ..... .78 .......... .62 ............ 6,58 ...... 6.72 ...... 6.65 ...... 2.12 ...... 2.07 ...... 2.10 ....... .37 ..... .42 ..... .40 _ 1.11 84 . _98 . H20“ H20' ...... .03 ...... .03 ...... .08 ...... .07 ...... .04 ...... .09 ...... .09 ....... .05 ................. 06 ................. .05 ........... .05 . .09 ..... .03 ..... .03 ..... .01 ..... .08 ..... .04 ..... .03 ..... .09 .......... .09 ............ .62 ...... .98 ...... .80 ...... .35 ...... .59 ...... .47 ....... .30 ..... .36 _____ .33 ..... .55 ..... 30 ..... .68 ..... H20— TiO, ....... .05 ...... .14 ...... .13 ...... .34 ...... .40 ...... .30 ...... .22 ....... .67 ................. 61 ................. .48 ........... .44 .. .. 1.35 ..... .47 ..... .46 ..... .42 ..... .39 ..... .15 ..... 1.64 ..... 1.78 .......... 1.23 ............ .18 ...... .18 ...... .18 ...... 1.70 ...... 1.62 ...... 1.66 ....... 1.09 ..... 1.10 ..... 1.10 ..... .93 ..... .90 ..... .92 ..... T102 13205 ....... .01 ...... .05 ...... .02 ...... .09 ...... .10 ...... .08 ...... .07 ....... .23 ................. 19 ................. .13 ........... .14 . . .. .52 ..... .14 ..... .14 ..... .13 ..... .11 ..... .04 ..... .52 ..... .75 .......... .56 ............ .02 ...... .02 ...... .02 ...... .67 ...... .65 ...... .66 ....... .38 ..... .38 ..... .38 ..... .33 ..... .56 ..... .45 ..... P205 MnO ....... .01 ...... .04 ...... .05 ...... .04 ...... .05 ...... .04 ...... .04 ....... .06 ................. .04 ................. .04 ........... .04 .... .11 ..... .07 ..... .09 ..... .06 ..... .04 ..... .02 ..... .15 ..... 11 .......... .08 ............ .25 ...... .25 ...... .25 ...... .13 ...... .13 ...... .13 ....... .11 ..... .11 ..... .11 ..... .10 ..... .1) ..... .10 ..... Mno CO2 ........ .00 ...... .01 ...... .00 ...... .01 ...... .03 ...... .03 ...... .01 ....... 02 ................. .02 ................. .01 ........... .02 . . . . .04 ..... .05 ..... .33 ..... .09 ..... .10 ..... .17 ..... .04 ..... 01 .......... .23 ............ .07 ...... .07 ...... .07 ...... .05 ...... .08 ...... .07 ....... .01 ..... .06 ..... .04 ..... .25 ..... .54 ..... .41 ..... COz Cr203 ...... .19 ...... Orzo; NiO ........ .06 ...... 1010 C1 ......... .00 ...... .01 ...... .02 ...... .01 ...... .01 ...... .01 ...... .00 ....... .02 ................. 02 ................. .02 ........... .01 .. .03 ..... .02 ..... .02 ..... .02 ..... .02 ..... .01 ..... 04 ..... .04 .......... .03 ............ .09 ...... .09 ...... .01 . . .01 ....... .02 ..... .02 ..... .03 ..... .03 ..... Cl F .......... .01 ...... .15 ...... .15 ...... .11 ...... .27 ...... .15 ...... .05 ....... .25 ................. 19 ................. .20 ........... .15 .. .37 ..... .08 ..... .08 ..... .07 ..... .04 ..... .02 ..... 22 ..... .39 .......... .40 ............ .01 ...... .01 ...... .13.. .13 ....... .06 ..... .06 ..... .07 ..... .07 ..... F Subtotal .. 99.73 ...... 99.83 ...... 99.78 ...... 99.75 ...... 99.76 ...... 99.70 ...... 99.83 ....... 99.71 ................. 99.74 ................. 99.64 ........... 99.66 . 99.68 ..... 99.75 ..... 99.80 ..... 99.66 ..... 99.65 ..... 99.74 ..... 99.91 ..... 99.59 .......... 99.83 ............ 100.03 ...... 99.93 ...... 99.74 . . 99.62 ....... 99.75 ..... 99.77 ..... 99.87 ..... 99.76 ..... Subtotal w .00 ...... .06 ...... .06 ...... .05 ...... .11 ...... .06 ...... .02 ....... ' 11 ................. .10 ................. .08 ........... .06 . .16 ..... .03 ..... .03 ..... .03 ..... .02 ..... .01 ..... .10 ..... 17 .......... .18 ............ .02 ...... .02 ...... .05. .05 ....... .03 ..... .03 ..... .04 ..... .04 ..... Less 0 Total . . . 99.73 ...... 99.77 ...... 99.72 ...... 99.70 ...... 99.65 ...... 99.64 ...... 99.81 ....... 99.60 ................. 99.64 ................. 99.56 ........... 99.60 99.52 ..... 99.72 ..... 99.77 ..... 99.63 ..... 99.63 ..... 99.73 ..... 99.81 ..... 99.42 .......... 99.65 ............ 100.01 ...... 99.91 ...... 99.91 ...... 99.69 ...... 99.30 ...... 99.57 ....... 99.72 ..... 99.64 ..... 99.74 ..... 99.83 ..... 99.4' ..... 99.72 ..... Total Analyst (2) ......... (2) ......... (2) ......... (2) ......... (2) ......... (2) ......... (2) ......... (2) ..................... (1) ..................... (2) ............... (2) ........ (2) ......... (2) ........ (2) ........ (2) ........ (2) ........ (2) ........ (2) ........ (2) ............. (2) ............... (1) ......... (3) ......... (1) ......... (3) ......... <1) -------- (3) --------- (1) --------- (3) ~~~~~~~~~ new“ A A r V 0 . , . . . Johannsen Leucogranite. Adamellite. Adamellite. Adamellite. Adamellite. Adamellite. Leuco- Adamellite. Granodiorite. Adamellite. Adamellite. Granite Adamellite. Granodiorite. Adamellite. Adamellite. Adamellite. Adamellite. Leuco- Meladacite. Dacite. Trachyandesite. Kimberlite .............................. Leuco quartz Quartz latite. Dac1te. Trachyte andeSIte or Mugearlte ............ Ande31te .............................. Dac1te(?) ............................. Johannsen (1950) adamellite. adamellite. latite ..... (1950) nomen- nomenclature. clature. . I Rittmann “Alkali “Rhyolite.” “Rhyolite.” ”Quartz latite" .......... “Trachy- Olivine Trachyandesite. Quartz latite. “Picritic “Picritic “Picritic Dark latlte. Dark latite. Trachy- Trachy- Trachy- Trachy- Trachy- Trachy- thtmann (1952) (1952) rhyolite." andesitel’ andesine basalt.” basalt." basalt." andesite. andesite. andesite. andesite. andesite. andesite. nomenclature. nomen- trachybasalt. clature. _ . ' ' _ , . Streckeisen Granite. Granite. Granite. Granite. Granite. Granite. Granite. Granite. Granodiorite. Granite. Granite. Syenite. Granite. Granodiorite. Monzonite. Granite. Monzonite. Granite. Granite. Andesine Andesite. Latite andesite. Amphibole peridotite ..................... Rhyodacite- Rhyodamte- Quartz Latlte ................................... Latite-andes1te ........................ Latrte-andesxte ........................ Streckelsen (1967) quartz quartz quartz andesite. (1967) nomen— basalt. latite. latite. nomenclature. clature. ' Irvine and Calc- Cale-alkaline Cale-alkaline "Picrite Alkaline Calc-alkaline Cale-alkaline Irvme and Baragar alkaline andesite. andesite. basalt.” (potassic) andesite. andesite. Baragar (1971) (1971) basalt. trachybasalt. nomenclature. nomenclature. MODAL (POINT-COUNT) ANALYSES, IN VOLUME PERCENT [Petrographic data compiled by G. L. Snyder, 1970. Figures in reading material are volume percent also, except NORMS, which are weight percent] A r \r A N Number of 1,518 3,016 2,998 2,905 2,897 2,813 3,001 2,797 2,270 6,440 2,243 3,003 2,887 2,195 2.483 2,777 2,705 2,824 1,441 2,135 2,534 2,619 3,243 2,095 2,734 2,761 1,442 1,482 1,447 Number of points. pomts. Plagioclase 22.33 Anl2 33.81 An16 25.59 An“, 36.62 An21 41.03 Ann 33.58 An13 29.50 An24 36.85 An23 43.13 Ann 26.33 An23 33.08 An27 24.22 An24 altered to 41.10 An25 36.15 A026 39.33 An26 34.03 An24 28.00 An25 27.52 An25 28.05 An15 39.16 An40 49.74 An28 (1.35 61.62 Anas (5.03 0.06 An78 nearly all altered to kaolin ..... 29.99 Anl7 35.25 Ann 53.59 An15 Plagioclase. altered to altered to altered to altered to altered to (An22 altered to altered to altered to altered to altered to 4.59 sericite and altered to altered to altered to altered to altered to altered to altered to altered to large large phenocrysts altered to altered t9 altered 90 0.13 5.20 3.37 5.84 7.72 center to 2.40 sericite 3.18 3.09 3.70 1.62 0.20 coarse 12.50 14.71 0.90 1.86 0.27 3.85 0.56 1.77 phenocrysts, altered to 1.89 0.19 sericite 0.71 seric1te. 0.07 SGI‘lCite sericite. sericite, sericite, sericite, sericite, An4 rim) and 0.51 sericite sericite sericite, sericite, muscovite. sericite, sericite, sericite. sericite. sericite, sericite, sericite sericite 44.71 small sericite, <0.07 and <009 and 0-13 0.19 kaolin, 0.07 kaolin, 0.06 kaolin, 0.20 kaolin, altered to coarse and 0.55 and 0.48 <0.08 <0.08 <0.07 0.15 coarse and 0.21 0.28 and 0.49 and 0.14 groundmass large muscovite, kaolin. coarse . . . . . . . . . ‘ and 0.51 and 0.45 and 0.13 0.52 coarse 2.89 muscovite. epidote. epidote. kaolin, and kaolin, and coarse muscovite, epidote. kaolin, coarse epidote. crystals, 1.30 0.14 epidote, muscowte. 52.02 unresolved mesostaSIS Consisting mamly 57.42 consmtlng of: 1) 55.59 ohgoclase or 17-07 recognizable 91881001886. 2.07 as large. coarse coarse coarse muscovite, sericite, 0.25 0.33 muscovite, and 3.98 0.42 coarse muscovite. large xenocrysts) 56.59 small 0f andesme mlCWllteS P1113 signlficant andesme groundmass lathes and 2) 1-93 fretted An" xenocrysts or unstable muscovite. muscovite. muscovite. and 0.07 0.34 coarse epidote. epidote. 0.07 epidote. muscovite, 2.38 altered groundinass POtaSSI‘Jm f8111813911” and POSSlbly minor albite 01” anorthoclase plates growmg into phenocrysts, 15-00 as gr oundmass epidote. muscovite, epidote, and 0.33 mainly to sericite crystals altered nepheline. (NORM contains 46.76 An”, vesicles and grading locally into crystals. 65.92 crystal mush where and 0.14 and 0.46 epidote. and epidote. to 0.84 sericite). 18.81 or, 2.90 ne, and 0.30 C) ............. groundmass. (NORM contains 51.86 A1130, individual minerals not resolvable epidote. carbonate. 13.36 or.). probably largely plagioclase. (NORM contains 50.74 An”, 14.91 01:). Potassium 46.91 30.15 35.33 28.71 26.75 32.15 37.54 25.17 17.24 24.01 25.50 52.49 23.68 10.82 26.45 29.05 40.44 36.05 29.42 0 ........ 5.06 microcline 8.03 orthoclase 0 .................................... 35.93 . 20.11 2.75. . Potassium feldspar microcline. microcline. microcline. microcline. microcline. microcline. microcline. microcline. microcline. microcline microcline microcline microcline. microcline. orthoclase microcline. microcline. microcline. microcline. (3.89 microcline. microcline. microcline. feldspar. altered to altered to includes 0.20 to groundmass, 1.17 0.15 <0.08 carbonate veins microcline. xenocrysts). sericite. sericite. and 0.06 muscovite veins. Quartz ..... 28.19 ...... 29.00 ...... 30.07 ...... 26.42 ...... 22.71 ...... 28.63 ...... 28.42 ....... 31.25 ...... 28.14 ...... 38.68 ...... 31.05 ...... 17.55 ............ 25.06 ...... 18.91 ...... 16.12 ..... 21.92 ..... 15.22 ..... 26.55 ..... 35.39 ..... 8.05 ....... 6.62 ............ 5.52 ............. 0 .................................... 32.85 ....... 35,40 ....... 32.47 ....... 1.73 xenocrystic material consisting of 1.45 0.30 cristobalite crystals projecting into 0.90 large xenocrysts altered (melted) to Quartz quartz cores (altered to 0.62 sericite and cavities. (NORM contains 5.44 Q.). 0.07 brown glass inner react1on rim and <0.07 zeolite or albite) surrounded by either pyroxene outer reaction rim (percentage single reaction rims of clinopyroxene or triple given below). reaction rims of 0.07 inner clinopyroxene, 0.21 medial aegirine and sanidine, and outer clinopyroxene. Outer clinopyroxene percentages given below. ........... Muscovite .. 2.24 ........ 1.35 ........ 2.27 ........ 0.39 ........ 0.72 ........ 0.55 ........ 0.07 ........ 0 ......... 0 ......... 0 ......... 0 ......... 0 ................ <0.07 ..... 0 .......... 0 I ......... 0 ......... 0 ......... 0 ......... 2.91 ....... 0 ......... 0 ............... 0.07 ............. 0 .................................... 0.29 ........ 0.15 ........ 0 ........... 0 .................................... 0 ..................................... 0 ...................................... Muscovite. Unknown(p0ds <0.07 ...... 0 ........... <0.07 ...... 0 ........... 0 ........... 0 ........... 0 ........... 0 ......... 0 ......... 0 ......... 0 ......... 0 ................ 0 ......... 0.07 ....... 0 ......... 0 ......... 0 ......... O ......... <0.07 ..... 0 ......... 0 ............... 0 ................ 0 .................................... 0 ........... 0 ........... <0.07 0 .................................... 0 ..................................... 0 ..................................... Unkn0wn (POdS in biotite). 1n biotite). Prehniteipods 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0.14 (plus veins 0 ........... 0 ......... 0 ......... 0 ......... 0 ......... 0 ................ 0 ......... 0 .......... 0.07 ....... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ............... 0 ................ 0 ......................................... 0 ........... 0 ........... 0 ........... 0 .................................... 0 ..................................... 0 ..................................... Prehnite' (pOdS in biotite). . . in microcline). inblotlte). Olivine ..... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ............ 0 ........... 0 ......... 0 ......... 0 ......... 0 ......... 0 ................ 0 ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... O ......... 0 ............... 0 ................ 75.54 chrysolite (Mg75Fe25) altered to a mixture 0 ........... 0 ........... 0 ........... 8.88 colorless olivine near chrysolite altered to 6.94 large olivines altered to 6.25 red 042(7) completely altered '30 pyroxene 0-21 Ollvme. of yellowish serpentine and goethite (19.64) 4.72 yellowish nontronite(?). (10.40 olivine iddingsite. (No olivine in NORM) and magnetite-ilmenite 0.21. (N0 olivine and then further to a mixture of chlorite in NORMJ -------------------------- 1n NORM-)- (31.57) and magnetite (11.22). ........... . . . Ortho- 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ......... 0 ......... 0 ......... 0 ......... 0 ................ 0 ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ............... 0 ................ 7.13 bronzite (Mgngen) 3.24 in isolated grains 0 ........... 0 ........... 0 ........... 0 .................................... 1.06 elongate crystals prOJectmgmto 0 ------------------------------------- OrthOPyroxene- pyroxene in olivine, 3.89 in outermost corona rings cavities, altered to 0.15 limonite on rims. between olivine and either diopside or pargasite(?), altered to serpentine or chlorite (0.59 and pargasite(?) (0.25). . . . . Clino- O ........... O ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ......... 0 ......... 0 ......... 0 ......... 0 ................ O ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ............... 0 ................ 0.34 diopside (MggoFem) <0.03 in isolated 0 ........... O ........... 0 ........... 27.11 yellowish clinopyroxene with hourglass 29.86 (equidimensional) ................ 4.70 consisting of 4.22 in groundmass and Clinopyroxene. pyroxene grains in olivine, 0.34 in discontinuous twins, 26.42 in groundmass, 0.69 as outer 0.48 as outer reaction rims around quartz corona rings between bronzite and reaction rims around quartz xenocrysts. xenocrysts. pargasite(?) ............................. . Amphibole 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0.07 green 048 green <0.02 green 0.17 green 0 ............... 0 ........ 8.90 blue- 3.80 blue- 3.35 blue- 2.87 blue- 0 ........ 0 ........ 27.83 blue— 8.44 (0.13 large 0 .............. 13.75 pargasite(?) (Mg93Fe7) 3.27 as a 0 ........... 0.21 green 6.07 blue- 0 .................................... 0 ..................................... 3.66 olive-brown hornblende, mostly Amphlbole. hornblende. hornblende. hornblende. hornblende. green green green green green brown myrmekitic intergrowth with spinel in hornblende. green separate crystals, some as rims on hornblende ferro— ferro- ferro- hornblende hornblende; 8.25 corona centers, 3.05 in a clear corona ring ferrohas- pyroxene. altered to hasting- hastingsite hasting- to light- small green between the spinel myrmekite and either tingsite 0.87 light site. altered to site. green hornblende) diopside or bronzite, 7.18 as isolated grains altered to green to 007 light- actinolite. altered to 0.06 in olivine, 0.25 altered to chlorite ......... 0.20 colorless green chlorite. chlorite. actinolite— actinolite~ tremolite. tremolite, <0.07 chlorite. Garnet ..... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ......... 0 ......... 0 ......... 0 ......... 0 ................ 0 ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ............... 0 ................ O .................................... 0.66 altered to 0 ........... 0 ........... 0 .................................... 0 ..................................... 0 ..................................... Garnet. 0.19 chlorite. . . Biotite . . . . 0.13 ...... 4.47 altered to 4.71 altered to 5.85 altered to 6.98 altered to 4.32 altered to 3.84 altered to 4.60 altered 6.67 altered 5.59 altered 6.94 altered 4.78 altered to 0.58 7.26 altered 12.63 altered 12.27 altered 7.69 altered 9.16 altered 8.73 altered 4.09 altered 21.78 altered 16.55 altered to 15.30 altered to 0 .................................... 0.09 altered to 7.49 altered to 3.82 altered to 3.88 .................................. 0 ..................................... 0 ------------------------------------- Biotite. 0.52 1.44 0.40 1.66 0.21 0.25 to <0.07 to <0.08 to 0.32 to 0.17 chlorite, 0.06 to 1.25 to 5.42 to 0.14 to 6.89 to 0.27 to 1.76 to 0.62 to 0.14 1.56 Chlorite and 0.19 chlorite. <0.09 0.34 0.27 chlorite and chlorite and chlorite, chlorite and chlorite. chlorite and chlorite. chlorite, chlorite, chlorite, sphene, <0.06 chlorite chlorite, chlorite. chlorite, chlorite chlorite chlorite chlorite. 0.06 sphene. chlorite. . . chlorite. . . chlorite. . . <0.06 0.13 <0.07 0.27 0.06 <0.08 0.06 <0.08 muscovite, and and 0.66 1.08 0.40 and 0.07 and 0.26 and 0.14 sphene. sphene. sphene, and sphene. sphene. sphene, sphene, sphene, 0.06 epidote. sphene. sphene, sphene, sphene. sphene. sphene. 0.13 and <0.08 and 0.02 and <0.08 and 0.51 and 0.20 epidote. epidote. kaolin. kaolin. epidote. epidote. Chlorite . . . . 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ......... 0 ......... (0.02 ..... <0.08 ..... 0 ................ 0 ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ............... 0 ................ 0 .................................... 0 ........... 0 ........... 0 ........... 0 ..................................... 0 ..................................... 0 ..................................... Chlorite. Magnetite- 0,13 ,,,,,,,, 0.26 ,,,,,,,, 0.26 ,,,,,,,, 0.40 ........ 0.60 ........ 0,42 ........ 0.38 ........ 0.82 ....... 1.45 ....... 1.81 ....... 1.25 ....... 0.32 ............. 0.20 ....... 2.60 ....... 0.14 ....... 0.67 ....... 0.68 ....... 0.20 ....... 0.07 ....... 0.84 ....... 3.38 ............ 4.35 ............. 0.56 .................................. 0.19 ........ 0.62 ........ 0.27 ........ 6.24 .................................. 4.27 .................................. 0.48 .................................. Magnetite- ilmenite. ilmenite. Pyrite ...... <0.07 altered <0.06 altered 0 ......... O ......... 0 ......... 0 ......... 0 ......... 0.07 altered to 0.08 altered to 0.13 altered to <0.08 altered 0 ................ 0.07 altered to <0.07 partly 0 ........ 0 ........ 0.07 ...... 0 ........ 0 ........ 0 ........ 0.19 altered to 0.13 0.25 altered to 0.06 0.06 altered to hematite ................ 0 ........... 0 ........... 0 ........... 0 ..................................... 0 ..................................... 0 ..................................... Pyrite. to hematite. to hematite. hematite. <0.08 0.1 1 to hematite. hematite. altered to hematite. hematite. hematite. hematite. hematite. Sphene . . . 0.07 altered to 0 ......... 0 ......... 0.47 ...... 0.27 ...... 0 ......... 0 ......... 0.69 altered to 0.96 altered to 2.25 altered to 1.59 altered to 0.39 ............. 1.58 ....... 2.96 (zoned). 0.36 ...... 1.01 (zoned) 0.62 altered 0.27 altered <0.07 . . . . 0.70 ...... 5.56 ........... 2.46 ........... 0 .................................... 0 ........... 0.28 altered to 0.27 ........ 0 .................................... 0 ..................................... 0 ..................................... Sphene. leucoxene. 0.14 <0.08 0.29 0.25 altered to to 0.14 to 0.20 0,07 leucoxene. leucoxene. leucoxene. leucoxene. 0.07 leucoxene. leucoxene. leucoxene. leucoxene. Epidote ..... 0 ........... 0.19 ........ 0.26 ........ 0.87 ........ <0.07 ...... 0 ........... 0 ........... 0.21 ....... 1.37 ....... 0.61 ....... 0.17 ....... <0.06 ........... 0.13 ....... 5.42 ....... 1.01 ....... 1.34 ....... 2.19 ....... 0.07 ....... 0 ......... 0.28 ....... 1.75 ............ 1.10 ............ 0 .................................... <0.09 ...... 0.07 ........ 0.14 ........ 0 .................................... 0.15 .................................. 0 ..................................... Epidote. Allanite .. .. 0 ........... <0.06 ...... 0.07 ........ <0.07 ...... 0.47 ........ 0.07 ........ 0.19 ........ <0.07 ..... <0.08 ..... 0.06 ....... <0.08 ..... 0.06 ............. 0.59 ....... 0.29 ....... <0.07 ..... <0.07 ..... 0.21 ....... 0.20 ....... 0.07 ....... 0 ......... 0 ............... 0.26 ............ 0 .................................... <0.09 ...... 0.14 ........ <0.07 ...... 0 .................................... 0 ..................................... O ..................................... Allanite. Apatite ..... 0 ........... 0.06 ........ <0.07 ...... 0.20 ........ 0.07 ........ 0.07 ........ 0.06 ........ 0.27 ....... 0.40 ....... 0.40 ....... 0.25 ....... 0.19 ............. 0.20 ....... 1.16 ....... 0.22 ....... 0.47 ....... 0.41 ....... 0.13 ....... <0.07 ..... 0.84 ....... 0.19 ............ 1.04 ............ 0 .................................... 0 ........... 0.14 ........ 0.20 ........ 0 .................................... 0 ..................................... 0.07 .................................. Apatite. Zircon ...... 0 ........... <0.06 ...... 0 ........... <0.07 ...... 0.07 ........ <0.07 ...... <0.06 ...... <0.07 ..... <0.08 ..... 0.11 (zoned) <0.08 ..... <0.06 ,,,,,,,,,,, 0.13 ,,,,,,, 0,07 ,,,,,,, <0.07 <0.07 ..... <0.07 ..... 0 ......... <0.07 0 ......... 0 ............... 0 ............... 0 .................................... <0.09 ...... <0.07 <0.07 ..... 0 .................................... 0 ..................................... 0 ..................................... Zircon. (zoned). (monazite?) (zoned). . .. Fluorite . . . . 0 ........... 0.71 ........ 1.44 (purple 0.07 ........ 0.33 ........ 0.07 ........ 0 ........... 0 ......... 0 ......... c ......... 0 ......... <0.06 ........... 0 ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... 0 .......... 0 ......... 0 ............... 0 ................ 0 .................................... 0 ........... 0.14 ........ 0.07 ........ 0 .................................... 0 ..................................... 0 ..................................... Fluorite. near allanite). Carbonate -- 0 ........... 0 ........... 0 ............ <0.07 ...... 0 ........... 0 ........... 0 ........... O ......... 0.08 ....... 0 ......... 0 ......... <0.06 ........... 0 ......... 0 .......... 0.23 ....... 0.47 (some in 0.13 ....... 0.28 (halfin 0 ......... 0.07 ....... 0.07 ............ 0 ................. 0 .................................... 0 ........... 0 ........... 0.35 ........ 0 .................................... 0 ..................................... 0 ..................................... Carbonate. veins). .... veins). .... Spinel ........ 0 ........... 0 ........... O ........... 0 ........... O ........... 0 ........... O ........... 0 ......... 0 ......... 0 ......... 0 ......... 0 ................ 0 ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ............... 0 ................ 2.28 green spinel or ceylonite (t Mg80 Fem), 0 ,,,,,,,,,, 0 .......... 0 .......... 0 .................................... 0 ..................................... 0 ..................................... Spinel, 1.63 in myrmekite with pargasite(?), and 0.65 in isolated grains. Chromite ..... O ........... 0 ........... 0 ........... O ........... 0 ........... 0 ........... 0 ........... 0 ......... 0 ......... 0 ......... O ......... 0 ................ 0 ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ............... 0 ................ 0.25 .................................. 0 ........... 0 ........... 0 ........... 0 .................................... 0 ..................................... 0 ..................................... Chromite. Vein ....... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ......... 0 ......... 0 ......... 0 ......... 0 ................ 0 ......... 0 .......... 0 ........ 0 ........ 0 ........ 0 ........ 0 ........ 0 ........ 2.45 (0.13 chlorite, 0 .............. 0.03 (chalcedony) ..................... 0 ........... 0 ........... 0 ........... 0.14 (carbonate) ...................... 0 ..................................... 0 ..................................... Vein. 1.94 epidote, 0.19 carbonate, 0.13 quartz, and 0.06 magnetite- ilmenite). Vesicle ..... 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 .................................... 11.40 (not included with mineral percentages). 0 ..................................... Vesicle. Glass ...... 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 ........... 0 ........... 0 .................................... 0 ..................................... 6.50 .................................. Glass. \ J\ j , V V SEMIQUANTITATIVE SPECTROGRAPHIC ANALYSES, IN WEIGHT PERCENT [Analysts: (1) J. C. Hamilton, 1 520 566', (2) G.W.Sears, Jr., 3 13.457; (3) B. Wayne Lanthorn,1 58570; (4) L. B. Riley, W. D. Cross, L. B. Breeden, Joseph Haffty, and Harriet Neiman, 12 *8 569 (Pd, Pt, and Rh determined by method described by Haffty and Riley, 1968). Results, reported in weight percent, are to be identified with geometric brackets whose boundaries are 1.2, 0.83, 0.56, 0.38, 0.26, 0.18, 0.12, and others, but results are reported arbitrarily as midpoints of these brackets, 1.0, 0.7, 0.5, 0.3, 0.2, 0.15, 01, etc. The precision ofa reported value is ap- proximately plus or minus one bracket at 68 percent, or two brackets at 95 percent confidence. Symbols used are: 0, not detected or at limit of detection; L, detected, but below limit of determination or below value shown; (blank) = not looked for; ‘, usual limits of determinations do not apply owing to use of dilution techniques; +, Lithium present but could not be further evaluated at date of these analyses; <, less than, but usual detectabilities do not apply. Looked for but not found: Ag, As, Au, Bi, Cd, Eu, Ge, Hf, Hg, In, Pr, Re, Sb, Sm, Sn, Ta, Te, Th, T1, U, w, and Zn] Field No. ...... 528 521 522 527 523 526 347 437 157 525 520 551 544 555 554 550 553 549 524 545 116 116 116 631 593 616 226 226 226 22 22 22 19 19 19 Field No. Laboratory No. 0101650 0101643 0101644 0101649 0101645 0101648 0101640 0101641 ‘D101278 0101647 0101642 0101655 0101651 0101658 0101657 0101654 0101656 0101653 0101646 0101652 '0101277 0141158 D141158 “0101279 0141162 0141162 "D101276 0141156 0141156 '0101275 0141155 0141155 Laboratory No. B .......... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ....................... 0 ....................... 0 ................ 0 ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ............... 0 ................ 0 ........... 0 ........... 0 ........... L ........... 0 .......... 0 .......... 0 .......... L .......... B Ba ......... .005 ....... 1 ......... .05 ........ 2 ......... .15 ........ 2 ......... .1 ......... .2 ..................... .15 .................... .2 ............... .15 ....... .2 ......... .15 ....... .1 ........ .1 ........ .2 ........ .1 ........ 1 ........ .2 ............. .2 ............... .002 ....... .005 ....... .5 ......... 3 ......... .15 ....... .15 ....... .2 ......... .2 ......... Ba Be ......... .0003 ...... .0003 ...... .0003 ...... .0003 ...... 0003 ...... .0002 ...... .0003 ...... .0002 .................. 0 ....................... .0002 ........... .0002 .00015 .00015 .0002 .00015 .00015 . . . 0001 .0001 .0002 .......... .00015 .......... 0 ........... 0 ........... 0 ........... .0002 ...... 0 .......... .00015 0 .......... .00015 Be Ce ......... 0 ........... 0 ........... 0 ........... .02 ........ .03 ........ .02 ........ 0 ........... .03 .................... .02 .................... .03 ............. .02 ....... .02 ....... 0 ......... 0 ......... 0 ......... .03 ....... .02 ....... .015 ..... .03 ............ .03 ............. 0 ........... 0 ........... .02 ........ .03 ........ 0 .......... .015 ...... 0 .......... .015 ...... Ce C0 ......... 0 ........... 0 ........... 0 ........... .0005 ...... 0005 ...... 0007 ...... 0003 ...... .001 ................... .001 ................... .0005 ........... .0005 .002 ...... .001 ..... .0007 .0007 .0005 . . . . 0 ......... .003 ..... .002 ........... .001 ............ .015 ....... .015 ....... .005 ....... 005 ....... .003 ...... .005 ...... .003 ...... .003 ...... Co Cr ......... .0001 ...... .0002 ...... 0002 ...... .001 ....... .0005 ...... 0005 ...... 0005 ...... .002 ................... .002 ................... .0007 ........... .0015 .003 ...... .003 ..... .003 ..... .002 ..... .0007 .0001 .01 ....... .007 ........... .0015 ........... .1 ......... .1 ......... .03 ........ .02 ........ .03 ....... .02 ....... .02 ....... .015 ...... Cr Cu ......... 0 ........... 0 ........... .0001 ...... .0003 ...... 0003 ...... 0001 ...... 0003 ...... .001 ................... .002 ................... .0005 ........... .0015 .007 ...... .0003 .0001 0 ......... .0003 0 ......... .003 ..... .002 ........... .005 ............ .007 ....... .01 ........ .007 ....... .005 ....... .003 ...... .003 ...... .003 ...... .003 ...... Cu Ga ......... .005 ....... .003 ....... .003 ....... .003 ....... .003 ....... .003 ....... .003 ....... .003 ................... .005 ................... .003 ............ .003 ...... .002 ...... .002 ..... .002 ..... .002 ..... .002 ..... .002 ..... .003 ..... .003 ........... .003 ............ .001 ....... .001 ....... .005 ....... .003 ....... .005 ...... .003 ...... .003 ...... .003 ...... Ga La ......... 0 ,,,,,,,,,,, .005 ....... .005 ....... .015 ....... .015 ....... .01 ........ .007 ....... .03 .................... .01 .................... .015 ............ .015 ...... .015 ...... .007 ..... .003 ..... .005 ..... .02 ....... .015 ..... .01 ...... .015 ........... ,02 ............. 0 ,,,,,,,,,,, 0 ,,,,,,,,,,, .01 ,,,,,,,, .015 ....... .005 ...... .007 ...... .003 ...... .007 ...... La L1 .......... 0 ........... + ........... + ........... + ........... + ........... + ........... + ........... + ....................... 0 ....................... 0 ................ + ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ............... 0 ................ 0 ........... 0 ........... 0 ........... 0 ........... 0 .......... 0 .......... 0 .......... 0 .......... Li Mo ......... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... .0003 .................. 0 ....................... .0003 ........... 0 ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... O ......... <.0005 <.0005 ......... .0003 ........... .0005 ...... .001 ....... 0 ........... .0005 ...... 0 .......... .0005 ..... 0 .......... .0005 ..... Mo Nb ......... .005 ....... .001 ....... .001 ....... .001 ....... .001 ....... .001 ....... .001 ....... .001 ................... .002 ................... .001 ............ .001 ...... .001 ...... .001 ..... .001 ..... .001 ..... .001 ..... .001 ..... .001 ..... .001 ........... .001 ............ O ........... L.002 ....... .002 ....... .002 ....... .0015 ..... .002 ...... .0015 ..... .0015 ..... Nb N1 ......... .0003 ...... 0 ........... 0 ........... .0003 ...... 0 ........... .0003 ...... 0 ........... .0007 .................. .001 ................... .0005 ........... .0005 .003 ...... .0007 .0007 .0007 .0003 0 ......... .005 ..... .003 ........... .001 ............ .03 ........ .07 ........ .02 ........ .03 ........ .015 ...... .015 ...... .01 ....... .015 ...... Ni Pb ......... .003 ....... .005 ....... .005 ....... .003 ....... .005 ....... .003 ....... .003 ....... .002 ................... .003 ................... .002 ............ .002 ...... .001 ...... .002 ..... .002 ..... .002 ..... .002 ..... .005 ..... .0015 .001 ........... .0015 ........... 0 ........... 0 ........... .003 ....... .003 ....... .002 ...... .002 ...... .003 ...... .003 ....... Pb Pd ......... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ....................... 0 ....................... 0 ................ 0 ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ............... 0 ................ 0 ........... 0 ........... 0.014 ....... 0 ........... 0 ........... (0.004 ....... 0 .......... 0 .......... <0.004 0 .......... 0 .......... <0.004 Pd Pt ......... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ........... 0 ....................... 0 ....................... 0 ................ 0 ......... 0 .......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ......... 0 ............... 0 ................ 0 ........... 0 ........... 0.014 ....... 0 ........... 0 ........... <.010 ........ 0 .......... 0 .......... <.010 ..... o .......... 0 .......... <.010 ..... Pt Rh ......... <.005 ...... <.005 ........ <.005 ..... <.005 ..... Rh Sc ......... 0 ........... .0005 ...... 0005 ...... .0005 ...... .0005 ...... .0005 ...... .0005 ...... .001 ................... .0007 .................. .0005 ........... .0005 .002 ...... .0015 .0007 .0007 .0005 0 ......... .002 ..... .002 ........... .0015 ........... .001 ....... .0015 ...... .0015 ...... .002 ....... .003 ...... .002 ...... .0015 ..... .002 ...... Sc Sr ......... .002 ....... .015 ....... .007 ....... .03 ........ .02 ........ .02 ........ .015 ....... .05 .................... .05 .................... .03 ............. .03 ....... .05 ....... .07 ....... .05 ....... .05 ....... .03 ....... .01 ....... .05 ....... 1 ............. .07 ............. .007 ....... .007 ....... .3 ......... .2 ......... .15 ....... .15 ....... .15 ....... .15 ....... 51‘ V .......... .001 ....... .0015 ...... .001 ....... .005 ....... .003 ....... 003 ....... .002 ....... .01 .................... .005 ................... .007 ............ .005 ...... .015 ...... .007 ..... .005 ..... .005 ..... .005 ..... .0007 .02 ....... 02 ............ .01 ............. .01 ........ 01 ........ .03 ........ .02 ........ .015 ...... .015 ...... .015 ...... 015 ...... V Y .......... .002 ....... .002 ....... .005 ....... .0015 ...... 005 ....... 001 ....... .003 ....... .002 ................... .0015 .................. .003 ............ .0015 .003 . . .... .003 ..... .003 ..... .002 ..... .002 ..... .0015 .003 ..... .003 ........... .003 ............ 0 ........... 0 ........... .002 ....... .003 ....... .002 ...... .003 ...... .0015 ..... .002 ...... Y Yb ......... .0002 ...... 0003 ...... 0005 ...... .0001 ...... 0003 ...... 0001 ...... .0003 ...... .0002 .................. 00015 ................. .0003 ........... .00015 <.0003 .0003 .0003 .0002 .0002 .0001 <.0003 <.0003 ......... .0002 ........... .0002 ...... .0002 ...... .00015 .0002 ..... .00015 0002 ..... Yb Zr .......... 007 ....... 01 ........ .003 ....... .02 ........ .03 ........ 02 ........ .015 ....... 015 ................... 01 .................... .015 ............ .015 ...... .015 ...... .007 ..... 015 ..... 007 ..... .02 ....... .01 ....... .02 ....... 03 ............ .03 ............. 0 ........... 0 ........... .01 ........ .02 ........ .005 ...... 015 ...... .007 ...... .015 ...... Zr Looked for only when La or Ce found. Looked for only when La or Ce found ...... Nd ......... 0 ........... 0 ........... 0 ........... .007 ....... .015 ....... .007 ....... 0 ........... .015 ................... .015 ................... .015 ............ .007 ...... .015 ...... .007 ..... 0 ......... 0 ......... .015 ..... .01 ....... .007 ..... .015 ........... .015 ............ 0 ........... .015 -------- .015 ....... .007 ...... .007 ...... .007 ...... 0 ........... Nd Analyst (2) ......... (2) ......... (2) ......... (2) ......... (2) ......... (2) ......... (2) ......... (2) ..................... (1) ..................... (2) ............... (2) ........ (2) ......... (2) ........ (2) ........ (2) ........ (2) ........ (2) ........ (2) ........ (2) ............. (2) ............... (1) ......... (3) ......... (4) ......... (1) ......... (3) ......... (4) ........... (1) ......... (3) ......... (4) ......... (1) ......... (3) .......... (4) ......... Analyst. nus. GOVERNMENT PRINTING OFFICE: 1978—677-026/I7 WW “397 m w _ éjfimgffigf $5 Hg ,3 , Geology of the Decaturville Impact Structure, Missouri By T. W. OFFIELD and H. A. POHN GEOLOGICAL SURVEY PROFESSIONAL PAPER 1042 Work done in behalf of National Aeronautics and Space Administration A detailed description of stratigraphy, style of deformation, and shock features at the Decaturville structure. Origin of the structure by impact is indicated UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1979 UNITED STATES DEPARTMENT OF THE INTERIOR CECIL D. ANDRUS, Secretary GEOLOGICAL SURVEY H. William Menard, Director Library of Congress Cataloging in Publication Data Offield, Terry W. Geology of the Decaturville impact structure, Missouri. (Geological Survey Professional Paper 1042) Bibliography: p. 46 Supt. of Docs. no.: I l9.16:1042 l . Cryptoexplosion structures —Missouri—Decaturville region. 2. Geology —Missouri—Decaturville region. I. Pohn, Howard A., joint author. II. Title. III. Series: United States Geological Survey Professional Paper 1042. QE613.5.U5035 551.8 77—608119 For sale by the Superintendent of Documents, US. Government Printing Office Washington, DC. 20402 Stock Number 024-001-03166-3 CONTENTS Page Regional paleogeographic implications .............................................. 10 Structure ............................................. . 11 Deformation outside the structure .. 11 Ring—fault boundary ................................................................... 13 Depressed zone ........................................................................... 15 Exotic blocks in the depressed zone ............... 16 Central uplift .............. . ............... 17 Geophysical data ............................................................................... 22 Shock-related deformation features ................................................... 26 Monolithologic breccias ................................................. Mixed breccias ............................................................... Shatter cones .............................................................................. Planar elements in quartz ........................................................... 30 Asterism in dolomite .......... 34 Kink bands in mica 34 Shock-envelope configuration .................................................... 34 Age of structure and depth of erosion ............................................... 35 Origin of the structure ..................... . 37 Endogenetic origin ..... .. 37 Exogenetic origin ........................................................................ 39 Interpreted crater profile and postcrater disruption ................... 41 Implications for other astroblemes .................................................... 43 References cited ................................................................................. 46 ILLUSTRATIONS Page Abstract ............................................................................................. 1 Introduction .......................................................... 1 Previous work ................................................................. 2 Acknowledgments . . 4 Stratigraphy ....................................... 4 Precambrian rocks ............... 6 Cambrian System ....................................................................... 6 Lamotte Sandstone ............................................................. 6 Bonneterre Dolomite . 6 Davis Formation ................................................................. 7 Derby and Doe Run Dolomites . 7 Potosi Dolomite ........................... . 7 Eminence Dolomite ................... 7 Ordovician System ..................................................................... 7 Gasconade Dolomite ........................................................... 7 Gunter Sandstone Member .......... 7 Upper part ................................ 8 Roubidoux Dolomite ....... 8 Jefferson City Dolomite ...................................................... 8 Kimmswick Limestone ........................................................ 9 Noix(?) Oolite .......................................... 10 Leemon Formation . ....................... IO Silurian System ................. lO Bainbridge Limestone ................................................... .. 10 PLATE 1, Geologic map of Decaturville structure, Missouri ........... 2. Block diagrams showing structural detail, Decaturville impact structure, Missouri Page In pocket In pocket FIGURE 1. Index map of southern Missouri, showing location of Decaturville structure, outline of area shown in figure 2, and features mentioned in text ..................................................................................................................................................... 2 2. Frame 1073-16224 by Landsat, showing Decaturville region ................................................................... 3 3. High-altitude aerial photograph showing circular patterns of Decaturville structure ......................................................... 4 4. Column showing lithologies and estimated thicknesses of formations present in Decaturville structure ........................... 5 5. Structure contour map on the base of upper part of the Gasconade Dolomite .................................... 12 6. Map of isopleths showing vertical displacements relative to prestructure configuration ....................................... l4 7. Photograph showing drill hole numbers and locations at center of structure, and diagrams of sections cut in drill holes around pegmatite locality and sulfide pit ................................................................................................................................ 20 8. Pegmatite area showing juxtaposition of blocks of different formations at center of structure ......................................... 22 9. Sulfide pit showing juxtaposition of blocks of different formations and mixed breccia .............. 22 10. Aeromagnetic map, Decaturville region .......................................................................................... 23 ll. Gravity map, Decaturville region ................................................................................................. 24 12. Gravity map, central area of Decaturville structure ........................................................................................................... 25 III lV FIGURE 13-18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. CONTENTS Photographs: 13. Monolithologic breccia, Eminence Dolomite 14. Core samples of mixed breccia .................................................................................................................................. 15. Core samples of mixed breccia showing alined grains in flow laminae and fragments drawn out or shaped by flowage ............................................................................................................................................................... 16. Mixed breccia at sulfide pit in center of structure .................................................................................................... l7. Sandstone from Lamotte Sandstone emplaced between block of Potosi Dolomite and mixed breccia in sulfide pit l8. Shatter-coned Derby and Doc Run Dolomites ......................................................................................................... Stereographic projection of shatter-cone striations .......................... Photomicrograph showing rhombohedral fractures in quartz grain of sandstone .............................................................. Photomicrograph of decorated planar features 1n quartz grains from mixed breccia, spaced 9-60pm and oriented parallel to i 0001} .......................................................................................................................................................................... Photomicrograph of mixed breccia, showing quartz grain with dominant planar features spaced 1-3 Wand oriented parallel to {1013} ............................................................................................ . ................................................................ Histograms showing frequency and orientation of quartz planar fractures ................................................................. Photomicrograph of kink-banded muscovite grain surrounded by strained quartz in pegmatite ....................................... Cross section showing inferred shock-pressure envelope in undisrupted strata, movement of data points by disruption, and inferred crater positions .............................................................................................................................................. Map showing areas of volcanism and unusual structural disturbance near the 38th parallel .............. Maps of mineralized areas, igneous intrusions, and “explosion structures” in zones trending east-west Schematic cross sections showing inferred time sequence of structural movements during cratering event ....................... Page 26 27 27 28 28 29 30 31 31 32 33 34 35 38 39 44 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI By T. W. OFFIELD and H. A. POHN ABSTRACT The Decaturville structure of central Missouri is ascribed to meteorite or comet impact on the basis ofstructural style and presence of shock features. The intense deformation of rocks in a circular area 5.5 km across is in marked contrast to the surrounding flat-lying strata ofthe Ozark Plateau. A normal fault bounds the structure which consists of a central uplift and a surrounding structural depression. In the central uplift, Cambrian sandstone, dolomite, and shale normally at least 300 m deep are exposed and isolated blocks of Precambrian basement granite pegmatite and schist are 540 m above their normal position. Convoluted strata in the uplift have strike lengths 25-30 percent longer than the perimeters on which the strata lie, indicating that inward movement and crowding of beds accompanied the upward movement. The inward movement involved folding and thrusting and was succeeded by adjustments on steep faults with both upward and downward displacements of as much as 150 m. The depressed zone around the uplift is characterized by thrusts inward and outward relative to its center and by steep faults that formed both before and after the thrusting. Shock features include monolithologic and mixed breccias, shatter cones, planar features in quartz, and intense intragranular deformation. Monolithologic breccias, formed by successive dilation and crushing of individual beds without mixing of adjacent beds, are common throughout the structure. Shatter cones occur at the center of the structure in a circular area about 450 m across. The shatter-coned rocks form a capping layer over a megabreccia column containing blocks as much as several tens of meters in size. These blocks are from Cambrian formations that make up the bottom 240 m of the 540—m disturbed sequence and are not in original stratigraphic order. A fine-grained matrix of mixed breccia around the megabreccia blocks contains quartz grains with close-spaced planar features, including some parallel to {1013}, probably indicating shock pressures as great as 60-100 kilobars. intensity of shock effects in the structure decreases downward; the basement rocks do not contain positively identified shock features. The deepest point of shock—energy release is not well defined but was not deeper than about 360 m in the disturbed section. Because the Decaturville structure is one of several features, volcanic or of uncertain explosive origin, that lie in general alinement across southern Illinois, Missouri, and Kansas, it commonly has been ascribed to origin by endogenetic explosion. A subterranean explosion, however, would result in dilation and outward dislocation rather than centripetal movement of beds in the central uplift and would not explain shock deformation at pressures far too great to be sustained beneath a few hundred meters of strata. Decrease of shock deformation downward also implies energy applied from above rather than below. In addition, the structure has many similarities to known impact structures and is unlike any proved endogenetic structures. The structure is of considerable paleogeographic interest because in the depressed zone it contains the only known occurrences of the Silurian Bainbridge Limestone and the OrdovicianNoixQ) Oolite,Leemon Forma- tion, and Kimmswick Limestone in that part of Missouri. Exposures of Cambrian formations found in the uplift are also unique in central Missouri. Blocks of units normally 189-540 m deep lie on the locally youngest formations in the ring depression and are interpreted as crater ejecta, or fallback, indicating that original ground surface is present in parts of the depression. On this basis, poststructure erosion in the area is estimated at no more than 50 m. Because such a small amount of erosion could not have destroyed a crater estimated to be 3,300 m in diameter and 540 m deep, we believe that the crater was destroyed by the immediate inward movement of beds that formed the central uplift. This inward movement was a response to passage ofthe shock wave, and it produced the ring fault and structural depression as necessary concomitants to the development of the central uplift (volumes of the uplift and depression are virtually equal). As the crater closed, material spalling from the walls was trapped to form the megabreccia mass at the center of the structure. This explanation may apply to other astroblemes where erosion is not thought to be great but where a central peak is present without a topographic crater. The age of the Decaturville impact structure is not definitely known but is almost certainly post-Pennsylvanian, and may be younger than Cretaceous. INTRODUCTION The Decaturville structure is located on the Ozark Plateau of central Missouri, along State Highway 5 about 13 km south of Camdenton and Lake of the Ozarks (fig. 1). The structure has about the same elevation and relief as the surrounding plateau; however, it is marked by a ring of low, wooded hills inside a nearly complete circle of low unwood- ed ground (fig. 2). The US. Geological Survey topographic maps of the Stoutland (1933) and Macks Creek (1934) 15-minute quadrangles with green forest overprint clearly show the circular pattern. Streams drain radially outward from the center of the area, a pattern in conspicuous contrast to the generally rectilinear, trellislike stream network of the surrounding countryside. The circular pattern of wooded and cleared ground and the radial drainage also are apparent in a high-altitude aerial photograph (fig. 3). The circular area, 5.6-6.4 km in diameter, is one of intense deformation in fault contact with the little disturbed, flat-lying strata of the Ozark Plateau. The structure consists of a central uplift surrounded by a structurally depressed zone. The regional setting, style or type of structural deformation, and shock-metamorphic features of the Decaturville structure are characteristic of more than 60 similar, roughly circular features known around the world 1 2 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI 39°95° 94° 93° 92° 91° 90° I, 789° 1 l " "f7 # A 7 W l / "\ N l / l 51/ St Louis 1 l / l Lake of the ILLINOIS Ozarks Q l g l ‘ T\7 \\ 38° l ‘0 | Camdenton i g oRolla ' O DECATU RVILLE I STRUCTURE ‘l I OLebanon ' M l s s / 0 U R l l oSp’ringfield ' KArea shown in 37° __ figure 2 ARKANSAS 50 100 KILDMETEHS 0 P—H—H—LTAFT . f? ,7 0 50 MILES FIGURE l.—lndex map of southern Missouri, showing location of Decaturville structure, outline of area shown in figure 2. and features mentioned in text. (15 in the United States), many originally thought to be produced by volcanism which caused an explosion but left no trace of volcanic activity. The name “cryptovolcanic” was applied to the group by Bucher in 1936, extending usage of the term applied by Branco and Fraas (1905) to the Steinheim structure in Germany. Boon and Albritton(l936) suggested that the structures might be of extraterrestrial (impact) origin. Dietz (1946) recommended the term “cryp- toexplosion” as noncommittal with respect to impact or volcanic origin. The detailed study of the Decaturville structure was done during 1968-70 as part of an investigation of possible lunar-crater analogs. Objectives of the study were to establish criteria (such as shock metamorphism and struc- tural style) by which impact structures could be distinguish- ed from endogenetic explosion structures, and to gain insight into the cratering process. PREVIOUS WORK Decaturville has been an area of continuing geologic interest since the report of Shumard (1873, p. 220), who noted that a few pounds of galena had been taken from pits some 600 feet (183 m) from an “igneous dyke of granite.” Winslow (1894) referred to the Wheeler mine near a pegmatite dike surrounded by intensely disturbed rocks in an area more than 2 miles (3.2 km) wide; he presented a sketch map of structure around the mine and suggested a post-Pennsylvanian age for the intrusion and disturbance. In 1905, Shepard (p. 114~1 17) suggested two periods ofuplift to explain the deformation and commented that the pegmatite area was surrounded by “an irregularly broken circle of low hills* **making the whole area resemble somewhat the contour of an old crater.” He also reported that the pegmatite was first discovered in float blocks and was excavated in 1869 by lead prospectors. Shepard (1905), Ruhl (1904), and Dake and Bridge (1927) ascribed the deforma- tion to uplift accompanying intrusion of the pegmatite. Buehler and McQueen (1933) cited Dake’s observation that the disturbed area measured approximately 5.5 miles (8.8 km) east-west and 3.5 miles (5.6 km) north-south. Bucher (1936, p. 1071) also believed the pegmatite to be intrusive, dated the structure as latest Cambrian to earliest Ordovi- cian, and categorized the structure as one of the “cryp- tovolcanic” group. PREVIOUS WORK \ 40 MILES t—fieeeee—x—Q U 40 KILOMETERS FIGURE 2.7Frame 107346224 taken4 October 1972 by Landsat (formerly Earth Resources Technology Satellite) showing Decaturville region. All these early authors reported certain stratigraphic identifications or relationships not borne out by the present study. Boon and Albritton (1936) noted the “cryp— tovolcanic” association, and suggested that these unusual features, including Decaturville, might be impact structures. Tarr, in 1938, and also in a discussion of Bucher’s 1936 paper (Bucher, p. 1083). expressed belief that the disturbance involved a buried hill of Precambrian granite. This opinion was shared by Tolman and Landes ( I939) because ofthe lack of contact metamorphism and because the overlying strata probably were too thin to permit slow enough cooling of an intrusive to form a coarse-grained pegmatite. In 1962, 4 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI FIGURE 3. AHigh-altitude aerial photograph showing circular patterns and radial drainage of Decaturville structure; x marks center of structure; D, village of Decaturville. US. Army photograph. June I970. radiometric age determinations by Tilton, Wetherill, and Davis confirmed a Precambrian age for the pegmatite exposed at the surface. Kiilsgaard, Heyl, and Brock (1963) pointed out that the Decaturville disturbance and the very similar Crooked Creek structure 113 km to the east are probably located at intersections of regional tectonic lines, and they proposed subterranean gaseous explosions as the mechanism of disturbance. The idea was expanded by Snyder and Ger- demann (1965), who noted that the Decaturville, Crooked Creek, and Weaubleau disturbed areas, and five volcanic occurrences, some of which involved explosive processes, lie on an apparent east-west line that extends from southern Illinois to eastern Kansas. They postulated that volcanic gas explosions formed the Decaturville and Crooked Creek structures at intersections of an east-west zone of crustal weakness and local tectonic lines. These “alined” structures were the subject of a field trip in 1965 and described in the trip guidebook (Snyder and others, 1965). Other papers in this same period by Amstutz (1959, 1964a, 1964b; Krishnaswamy and Amstutz, 1960; Amstutz and Zimmer- man, 1966; Zimmerman and Amstutz, 1965) ascribed the Decaturville structure variously to rejuvenation of ring dike or polygonal fractures “typical for the primordial crust of the earth,” to diapirism, and to mud volcanism. Shoemaker and Eggleton (1961) presented an impact origin for the structure, and Dietz later (1963) noted that shatter cones at Decaturville were indicative of intense shock and that therefore the structure probably was an impact feature. Mineral exploration of the structure in the late 19th century reportedly consisted of four shafts, from 3.7-l6.2 m deep, sunk by the property owner, Mr. Harry Wheeler. About the end of World War 11, some churn drilling and pitting was done in the center and at the outer edge by the St. Louis Smelting and Refining Co. and the Athletic Mining and Smelting Co. Since 1955, the central area has been owned or controlled by the Ozark Exploration Co. This company obtained more than 5,500 m of core from about 60 drill holes near the center of the uplift, and over a thousand meters more was taken in drilling the Gunter Sandstone Member of the Gasconade Dolomite, a sandstone bed that rings the uplift. Pits were opened in an area of sulfide-rich breccia (known as the “sulfide pit”) and around the pegmatite. Many assays of sulfidic core samples also were obtained by the Ozark Exploration Co. ACKNOWLEDGMENTS The study was done in behalf of the National Aeronautics and Space Administration under contract 160—75-01-43-10. We are particularly grateful to Mr. H. B. Hart, president of Ozark Exploration Co., for arranging access to much of the property on which the structure lies and for permitting use of drill core and privately held information without which this work would not have been complete. Consultations with M. R. Dence, Dominion Observatory of Canada, and con- sultations in the field with D. J. Roddy and D. J. Milton, US. Geological Survey, were helpful in making com- parisons with other similar structures. P. E. Gerdemann, St. Joseph Lead Co., provided information on local stratigraphy and subsurface data. We thank the staff of the Missouri Geological Survey, and especially Mrs. M. H. McCracken, for making available well-log and other stratigraphic information on central Missouri. Ground surveys of the gravity and magnetic pattern of the area were conducted by J. D. Hendricks, US. Geological Survey. We also thank R. W. Paul, who assisted in fieldwork ( 1968), and C. E. Nichols, who provided a detailed stratigraphic section of the Jefferson City Dolomite outside the structure. Fission-track studies bearing on the possible age of the structure were made by C. W. Naeser, US. Geological Survey. The topographic base map was prepared by J. L. Derick, G. M. Nakata, and Raymond Jordan, US. Geological Survey. STRATIGRAPHY The Decaturville structure is underlain by a sequence of rocks approximately 540 m thick that ranges in age from Precambrian to Silurian. The Upper Cambrian and Lower Ordovician sequence of Missouri apparently is complete from the Lamotte Sandstone through the Jefferson City Dolomite, and all the formations (with the possible excep— tion of the Cambrian Bonneterre Dolomite) are exposed, at least in part, at the surface. The lithologies of the formations in the Decaturville stratigraphic sequence are shown in figure 4. Potosi Dolomite €10 Band Doe olom Ites Cdd Derb Run Davis Formation CAMBRIAN Bonneterre Dolomite Lamotte Sandstone Cl PRECAMBRIAN Basement SILURlAN OR DOVICIAN CAMBRIAN x. w ... i: D O STRATIGRAPHY Kimmswick Limestone, Noix(?) Oolite, Leemon Formation1(0rdovician) and Bainbridge Limestone (Silurian) E 03 IO :‘éu 806 338 to 0 fl .3 E 2 0 ol— so 7 o .‘9 n 3 o n: 6 5 .313 E 2 o 4 Do: go N C 8 (II (D (D 3; age Emo «:2 (I) 3 2 Eminence Dolomite Ce EXPLANATION Dolomite Limestone Sandstone Shale Calcareous shale 2g Schist t \ _’ 5/ 1 Granite pegmatite / l ’ A 4 . Massive chert E Cryptozoon chert Glauconite 101‘ Thompson and Satterfield, 1975 FEET METERS 0 l] 100 50 VERTICAL SCALE FIGURE 4.—Column showing lithologies and estimated thicknesses of Cambrian, Ordovician, and Silurian formations present in Decaturville structure; marker beds indicated by numbers. Base of section lower left; top of section upper right. 6 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI Extensive core drilling at the center of the structure has recovered more than 5,500 m of core in 60 holes. Cores from the center of the structure reveal a chaotic mixture of rocks but all the formations from the top ofthe Precambrian basement through the Cambrian Potosi Dolomite have been recognized. The difficulty of solving the complex structural relationships at Decaturville has been increased by the discontinuous nature of outcrops, by lateral variations in lithology of individual beds, and by low—angle thrust faults that make detailed measurements of thick stratigraphic sections all but impossible. Several marker beds that were generally continuous or showed evidence of predictable lateral variation were used to solve these structural problems. Subsequent stratigraphic descriptions pay special attention to the lithology and color of these marker beds. Less detailed descriptions of the intervening beds con- tribute to a general understanding of the gross lithologies of the formations. PRECAMBRIAN ROCKS Precambrian basement rocks, normally about 540 m below the surface, are represented in outcrop by a single block exposed near the center of the structure (map sector L—l3, pl. 1). The block measures about [5 by 21 m in lateral extent and is continuous to a depth of 15 m in drill core. Most of this large block consists of granite pegmatite which displays a general pattern of lithologic zonation. In the poorly exposed center of the block, boulders lying on the surface indicate a quartz core a meter or two across. Around the core is a zone 3-6 m wide of coarse pegmatite containing quartz and microcline crystals as much as 5 cm long, intergrown to produce a graphic texture. This pegmatite locally contains as much as 5 percent muscovite. Outward from the coarse pegmatite is a much more disrupted zone of schistose and granulated micaceous pegmatite, containing accessory amounts of black tourmaline. The Rb/Sr age of muscovite from the pegmatite was determined by Tilton, Wetherill, and Davis (1962) to be l,445 m.y. A K40/Ar40 age for the same muscovite was 1,290 my, somewhat less, probably because the mica was slightly altered. Muscovite schist at least 0.9 m thick is exposed in contact with the pegmatite in a trench on the northwest side of the pegmatite block. The schist contains accessory apatite and black tourmaline. Tourmaline crystals are oriented to form a lineation that is deformed by pervasive rounded to kinklike crenulations typically 0.3 cm in amplitude. Along other margins the pegmatite is in contact with displaced blocks of the Potosi, Derby, Doe Run, Davis, and Lamotte For- mations. A few feet below the main pegmatite block, another 4.3 m of pegmatite was cut in a drill hole 31 m deep, and 1.8 m of schist was found. Other drill holes near the center of the structure cut a few small slivers of schist or pegmatite at various depths. CAMBRIAN SYSTEM LAMOTTE SANDSTONE The outcrops of the Lamotte Sandstone in the structure are limited to one block 2 m long, to thin sand rinds that surround two blocks of Potosi Dolomite in the sulfide pit (map sector M-IZ, pl. 1), and to tightly cemented blocks adjacent to the pegmatite pod (map sector L-l3). In addition, occurrences are numerous in drill cores. The thickest section is more than 37 min a core that was started in Lamotte covered by soil. Unfortunately, the section shows no identifiable dip. Estimates made from drill holes nearest the structure (southeast of Lebanon, more than 32 km away) indicate a total thickness of 90 m for the formation. Locally, the Lamotte is a tightly cemented bimodal sandstone, composed of coarse, light- to medium—gray, subrounded to rounded frosted quartz grains in an extreme- ly fine grained nearly white matrix of quartz. Layering is marked by the distribution of the coarse grains. Some very fine grained parts of the Lamotte are seen in thin section to consist of angular quartz grains with about 25-30 percent feldspar. In core, the sandstone appears to be very clean; but in outcrop, the formation weathers yellow and shows some ferruginous cement. In the eastern part of the State, where the Lamotte crops out, all the basal units are made up of a conglomerate of cobbles of Precambrian granite and schist in a sand matrix (Weller and St. Clair, I928, p. 34-37; Dake, 1930). This particular lithologic character is not now seen in outcrop or in drill core on the structure; but in 1956 when the pegmatite pit was newly opened, T. E. Mullens (oral c0mmun., 1974) observed several centimeters of conglomerate containing fist—sized pebbles of basement rock directly overlying the pegmatite. The conglomerate seems to grade laterally into schist, but the two rocks may be tectonically intermixed. BONNETERRE DOLOMITE No outcrops of unquestionable Bonneterre Dolomite have been identified, although it would seem reasonable to assume that some rock fragments in the mixed breccia in the sulfide pit are Bonneterre. In core, the Bonneterre has been identified as being a moderately glauconitic, medium— to fine-grained, light- to dove-gray dolomite that has a pale-olive—green cast when wet. The glauconite occurs as small pockets of pelletal material. Occasional thin shale partings are seen in the core sections. Small cavities lined with dolomite rhombs are characteristic. Two drill holes at the sulfide pit penetrate a tilted block of Bonneterre. One core shows a continuous section of 43 m of Bonneterre, and although the formation is overlain by the Davis Formation and is underlain by Lamotte Sandstone, it U44 STRATIGRAPHY 7 is by no means definite that the section is not faulted. Three wells southeast of Lebanon show the total thickness of the Bonneterre to be 53 m. DAVIS FORMATION The Davis Formation is the most easily identified formation, both in outcrop and in drill core. Parts of the formation are seen as huge blocks as well as fragments incorporated in the mixed breccia in the sulfide pit, and as a single small block beside the pegmatite pod. In core as in outcrop, the Davis is composed of interbedded light- to medium-gray, fine- to medium-grained limestone or limy dolomite and light-grayish-green limy siltstone and shale. The dolomite weathers pinkish tan and commonly contains lineations caused by alinement of millimeter-sized grains of glauconite. The shale and siltstone occur as beds from 0.3 cm to 15 m in thickness. A few angular grains of feldspar occur in some of the siltstone layers. The greatest thickness found in core section is 41 m, and although the hole was started in the Davis Formation, it is felt that this core represents most of the formation. Thickness of the Davis in drill holes southeast of Lebanon is 40-43 m. DERBY AND DOE RUN DOLOMITES The Derby and Doe Run Dolomites crop out as an ellipse 400 by 460 m at the center ofthe structure. These formations are extensively represented in drill cores, including one core from map sector 1-19, well outside the central outcrop area. From 30 to 40 percent of the materials in the cores from the center has been identified as belonging to the Derby and Doe Run. Where the Derby and Doe Run crop out, they are white to tan, coarse—grained, massive dolomites that weather to a yellowish tan. Bedding is very rare in the cores and has only been tentatively identified in three outcrops; stylolites occur in a few core sections. No attempt was made to map the Derby and the Doc Run separately, but the lower half of the map unit is composed of medium- to coarse—grained, white to yellowish- or pale-brownish-white dolomite. The upper or Doe Run half of the map unit as seen in core section is composed of medium—gray, fine- to medium-grained dolomite crystals in a very fine grained light-gray matrix. The occurrence of individual grains of glauconite is common to rare, but small pockets of clay and glauconite are common. In rare instances, samples contain a few quartz grains, some as nuclei of dolomite crystals. The thickness of the Derby and Doe Run in the wells southeast of Lebanon is 40—43 m. Because of the paucity of bedding at the structure, no estimate of the thickness is possible. POTOSI DOLOMITE The Potosi Dolomite is poorly exposed in an irregular annulus which surrounds the Derby and Doe Run Dolomites at the center of the structure. The areas between outcrops are characterized by a typical deep-red to red- dish—brown very clayey soil. The formation is a conspicuous part of the breccia in many core sections. One of the most distinctive formations in the section at Decaturville, the Potosi is a poorly bedded, fine- to medium-crystalline, light- to medium-pinkish-brown dolomite. The color is distinctive, but the chief diagnostic feature of the Potosi is quartz druse found in outcrop. The druse consists of microscopic to macroscopic clear to tan crystals on botryoidal surfaces of white to gray quartz. Thin sections show the Potosi to contain minor amounts of detrital quartz and chert, and very sparse microcline. Freshly broken rock has a noticeable bituminous smell. The Potosi has been logged in drill holes 13 km to the northwest and 8-16 km to the south as being 61:6 m thick. EMINENCE DOLOMITE The Eminence Dolomite is the oldest unit in the structure to exhibit exposures of Virtually the whole formation. It typically supports abundant stands of cedar trees. The exposures of Eminence are, in general, good as well as widespread. Occurrences of Eminence in core are rare. The lower half of the formation is composed of very light gray to slightly tannish gray, fine- to medium-crystalline dolomite and dolomite breccia in beds 1—10 cm thick. Most of the breccias consist of light-gray clasts in a yellowish-tan matrix. A single 8-cm bed ofwhite porcelaneous chert (no. 1, fig. 4; pl. 1) is present locally approximately in the middle of the lower half of the Eminence section. The upper half of the Eminence is characterized by more massive and resistant outcrops than the lower half. It is composed of very light- to light-tannish—gray, fine to coarsely crystalline dolomite breccia and sparse unbrec- ciated dolomite in beds 0.15-2 m thick. As in the lower halfof the Eminence, a single bed of white porcelaneous chart (no. 2, fig. 4; pl. 1) as thick as 0.9 m occurs in places approximately in the middle of the section. Distinctive light-tan soils that are formed on both the upper and lower halves of the Eminence are composed of fine-grained dolomite sand. The thickness of the Eminence in drill holes 16 km to the northwest and 16 km to the east is 99:5 111. ORDOVICIAN SYSTEM GASCONADE DOLOMITE GUNTER SANDSTONE MEMBER The Gunter Sandstone Member is the lowermost of the key beds used for detailed mapping of the Decaturville structure (no. 3, fig. 4; pl. 1). This unit occupies a roughly triangular outcrop belt 610-l,220 m from the center of the structure. A huge isolated block of the sandstone also is found at the village of Decaturville, well outside the outcrop 8 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI belt (sectors G-l9, H-l9, pl. 1). In outcrop, the Gunter commonly supports a lush growth of kelly-green moss, and where the sandstone is shallowly buried the unit is marked by a dense growth of poison ivy. The poison ivy is so closely tied to the sandy soil that in many places it is possible to follow the buried sands for one or two hundred meters between outcrops by walking out the poison ivy belt. The Gunter is a quartz sandstone containing a bimodal distribution of rounded, frosted fine- and medium-sized grains. lt weathers brown or buff and is clean white to slightly iron stained on fresh surfaces. The medium—sized sand grains are extremely pitted and have the appearance of miniature translucent golf balls. They make up 5-20 percent of the rock. In several places, notably in G-l9 and 1-1 I, the Gunter is highly fractured, with the fractures silicified enough to be slightly resistant and to weather in raised relief. The thickness of the Gunter Sandstone Member varies from 3 to 8 m within the structure and from 2 to 12m in drill holes and outcrop within a 32—km radius of the structure. UPPER PART The upper part of the Gasconade crops out sparsely in a zone from the Gunter Sandstone Member to a distance of approximately 1.6 km from the center of the structure. The width of this zone is small relative to its thickness as compared to those relationships for most of the other formations in the structure. The upper part of the formation consists principally of medium— to thick-bedded gunmetal-gray dolomite, medium to coarsely crystalline and saccharoidal in texture. Some beds are tectonically brecciated on a fine scale and are composed of typical Gasconade clasts in a tan— to whitish-gray matrix. A bed of cryptozoon chert 0.3-0.9 m thick (no. 4, fig. 4; pl. 1) is approximately 20—30 m below the top of the Gasconade within the structure. It likely corresponds to the cryptozoon ledge described as the top of the lower part of the dolomite sequence of the Gasconade in the Lebanon quadrangle several kilometers to the south (Searight, 1955, p. 7). A second bed of cryptozoon chert, 0.3-0.6 m thick (no. 5, fig. 8; pl. 1), occurs approximately 3 m below the Gasconade-Roubidoux contact southeast ofthe structure. A thin dark-gray chert bed in the same position west is extremely porous and ropy; coral-like rounded masses ofit are abundant in float. Between the west edge of the area shown on plate I and Banks Branch 0.5 km farther west, several outcrops of sandstone as much as 3 m thick were found within the Gasconade. These sandstones do not resemble those seen in the Roubidoux. (See following section.) Moreover, they do not seem to be faulted into the Gasconade. They seem to be similar to the sandstone beds and lenses reported from the Gasconade in Miller County 56 km to the northeast (Ball and Smith, 1903, p. 37). The thickness of the upper part of the Gasconade at the structure is estimated to be 93:2 m as determined from drill holes to the east, north, and south of the structure. ROUBIDOUX DOLOMITE Outcrops of Roubidoux Dolomite are common in a zone from the southwest to the northwest parts ofthe Decaturville structure, but are uncommon in the rest of the mapped area. In addition to lying in normal sequence in an annulus from 1,520 to 2,000 m from the center of the structure, the Roubidoux commonly occurs in fault slices as the formation nearest the ring fault on the west side. Where bedrock is shallowly buried the residual soil is characterized by an abundance of sandstone and chert float, the chert commonly being present as blocks as much as 3 m on a side and 0.3-0.6 m thick. In many areas the Roubidoux Dolomite supports dense stands of scrub oak. The base of the Roubidoux is in general represented by a single chert bed 1.5-3.4 m thick, typically a blocky sedimen- tary breccia but in places laminated and showing algal structures (no. 6, fig. 4). The 12-15 m of section above the basal chert is lithologically indistinguishable from the underlying Gasconade Dolomite. A second chert bed 0.3-0.6 m thick (no. 7) is present about 18 m above the base of the formation. In the section above the second chert the dolomite becomes successively thinner bedded, lighter gray, and more finely crystalline. Some beds or laminae contain numerous sand grains. A single sandstone bed 1.2 m thick (no. 8) occurs approximately 30 m above the base of the Roubidoux. In places on the west side of the structure this sandstone seems to be represented by a mixed sand and chert sedimentary breccia. In hand specimens the fine to medium quartz grains are subangular to subrounded and most are slightly frosted. The individual grains are weakly cemented by iron oxide which gives the sandstone a reddish—tan cast. The upper approximately 6 m of the formation is composed of thin-bedded (1-8 cm), aphanitic, tan marly dolomite (locally termed “cotton rock”) interbedded with beds of tectonically brecciated medium-gray dolomite 8—15 cm thick. Green shale partings also occur at places in the upper part of the Roubidoux. The total thickness of the Roubidoux Dolomite on the structure varies from 31 to 37 m, although our choice of the formation top for mapping purposes may add about 4 m to the unit, relative to other mapped areas. (See following section.) Outside the structure, drill holes and mapping show that the formation is 29-38 m thick. JEFFERSON CITY DOLOMITE The outer zone, ranging in width from 610 to [,220 m from the Roubidoux Dolomite to the ring fault, is underlain mostly by Jefferson City Dolomite, the youngest formation of Early Ordovician age present. Exposures of the Jefferson City are excellent on the west side of the structure. The belt of Jefferson City is the widest in proportion to formation thickness of any of the formations in the structure. STRATIGRAPHY 9 Although the contact ofthe Roubidoux with the Jefferson City has been widely accepted to be the well-known “buhr-stone” chert marker bed and is mapped that way by C. E. Nichols (1973) in an area just south of the structure, we have seen this contact at only one place on the structure. Nichols mapped a bed of sandstone l.5 m thick occurring 1.8 m below the “buhr—stone” chert, and we are confident that this is the same bed that we have mapped as the l.2-m sandstone bed 6 m below the marker that we have selected as the base of the Jefferson City. That marker bed (no. 9, fig. 4; pl. 1), chosen for mapping convenience because of its numerous outcrops and distinctive character, is a platy sandstone, commonly mud cracked and ropy, composed of uniformly fine grained, subangular, clear grains. This sandstone varies in thickness from 8 cm to 0.3 m, is moderately well cemented with iron oxide, and in general has a reddish cast. The section for 23 m above the basal sandstone is a mixture of thin to medium beds of light- to medium—gray-brown, medium-crystalline dolomites, tan, aphanitic marly dolomites (“cotton rock”), and thin- to thick-bedded dolomite breccias. The conspicuous marker sequence of thick-bedded dolomite called “quarry ledge” and mapped some ll m above the base of the formation elsewhere in the region was not positively identified anywhere in the ring zone. Possible occurrences were noted in sectors L-6, G-8, and F-lS. Thin beds (2.5-7.6 cm) of hackly chert breccia and thin dolomite beds containing algal structures occur in a few places. Large ovate nodules of brown chert appear in float. Approximately 23 m above the basal sandstone is a second sandstone bed (no. 10) composed of clean, fine-grained, angular, clear to slightly frosted quartz. At places, the sandstone grades laterally into a sandy chert bed or into a zone 15 cm to 0.3 m thick that is composed ofalternating 2.5- cm beds of aphanitic, white-weathering, whitish-tan dolomite and pink, very fine grained sandstone that is made up of subrounded and slightly frosted grains and covered with black lichen. The differential weathering combined with the alternating light and dark zones results in a banded appearance rather like that of varved glacial clays. The approximately 14 m of section between the second or middle sandstone bed and the third sandstone bed (no. 11) of the Jefferson City is mostly composed of thin-bedded, aphanitic, tan dolomite and dark-grayish- to chocolate-brown dolomite and dolomite breccia. Beds become increasingly more massive upward in the section, and the unit just below the upper sandstone is a bed of pale-reddish~purple cryptozoon dolomite 1.8-2.4 m thick. The third or upper sandstone bed (no. ll) is the most highly variable in both texture and thickness of all the marker sandstone beds in the section. The thickness varies from 0.6 to 3 m (and at map locality H-7 more than 12 m). Lithologically, the fine-grained sandstone varies from dominantly subangular and glassy grains with fewer rounded-to-subrounded frosted and clear grains to mostly subrounded frosted grains mixed with highly comminuted glassy grains. As much as 6 m of section is preservedjust above the upper sandstone bed. This 6 m is predominantly thin bedded dolomite breccia and dolomite beds, overlain by a series of massive beds 0.6-0.9 m thick of grayish-tan medium-crystalline dolomite and brecciated dolomite. On the east side of the structure, many ofthe outcrops in the ring zone are isolated exposures of sandstone. Massive white sandstone, especially where near limestone outcrops, can reasonably be identified as the upper sandstone of the Jefferson City, and the basal sandstone is typical and readily identifiable where it occurs. Many of the other sandstone outcrops, however, have a bimodal grain-size distribution not like the distribution in any ofthe sandstone beds mapped in the Jefferson City on the west side. Numerous large frosted quartz grains or white chert grains are scattered through a fine-grained, angular, clear quartz matrix. These sandstone beds also do not resemble any units below the Jefferson City. They are considered tentatively to be equivalent to the middle sandstone bed of the west side and are mapped as such; but it is possible that some of the outcrops may be Roubidoux sandstone in lithology slightly different than previously observed, and they thus may indicate unmapped fault blocks. KIMMSWICK LIMESTONE Numerous small patches of limestone occur atop the Jefferson City Dolomite in the depressed ring zone around the structure; these are assigned with reasonable certainty on the basis of fossil content to the Kimmswick Limestone, which is of Middle Ordovician age. The limestone is thin bedded, light to medium gray, and generally aphanitic, but at places fine or medium crystalline. It typically weathers light gray with a slight bluish cast. Numerous outcrops are brecciated and contain coarsely crystalline white calcite. Resistant fossil material is evident on weathered surfaces of many of the limestone outcrops. The limestone fauna identified to date (collection E6-4; R. J. Ross, written commun., 1971) consists of brachiopods Paucicrura sp., Roslricellula? sp.,S0werbyellasp., Z ygospira sp., Furcitella? sp., and conodonts (identified by L. A. Wilson) Acontiodus alveolaris Stauffer, Cordylodus sp., Amorphognathus sp., Drepanodus homocurvatus Lindstrom, Polyplacognathus sp., Oistodus inclinatus Branson and Mehl, Ozarkodina cf. 0. ('oncinna Stauffer, Panderodus panderi (Stauffer), Panderodus compressus (Branson and Mehl), Prioniodina sp., Sagirtodontus sp., and Trichonodellaflexa Rhodes. In a few places, the thin-bedded limestone rests confor- mably on a thicker bedded aphanitic to fine crystalline limestone that seems to be much less fossiliferous. The lower part of the limestone section possibly is Plattin Limestone. Exposed thickness of the limestone beds nowhere exceeds 4 m in the Decaturville structure, although in places 10 limestone outcrops occur as much as 9 m below hill crests. This may be explained by structural displacements or by hills having an irregular erosion surface on the Jefferson City and being partly plastered with limestone, but the possibility also exists that the hilltops are underlain by more ofthe limestone or even by unexposed younger formations. NOIX(?) OOLITE One small patch of a very distinctive limestone (the southeasternmost of a group of five limestone patches in sector M-5) has been identified as Late Ordovician in age. The limestone is thin bedded, medium gray, and weathers light brownish gray on a bas-relief surface showing the rock to consist mostly of oolites and fossil hash. It occurs within a few feet laterally of Leemon Formation of Thompson and Satterfield (1975) of Late Ordovician age. The unusual lithologic character of the limestone matches that of the Noix Oolite which elsewhere underlies the Leemon Forma- tion. The Noix and Leemon are part ofthe Edgewood Group of Thompson and Satterfield (1975, p. 77). The limestone “contains most of the elements of the faunule described from the Thebes Sandstone” (collection E2-2; J. W. Huddle, written commun., 1971, revised by J. Repetski, 1975): Spet'imenr idenli'fied Cvrtoniodusflexuosus (Branson and Mehl) ....................... l4 “Dichognalhus“ sp. ............................................................. 2 Drepanoistodus suberectus (Branson and Mehl) type element ................................................................ 4 homocurvatus element ..... ll Panderodus sp. ................................................................... 6 Plectodina furcata (H inde) ozarkodiniform element ............................................... 30 cordylodiform element ............................... l3 prioniodiniform element ............................ 28 trichonodelliform element ............................................ l6 zygognathiform element .............................................. 12 cyrtoniodiform element ............................................... 15 Protopanderodus inscu/ptus (Branson and Mehl) .............. l6 LEEMON FORMATION Next to the Upper Ordovician limestone (Noix (?) Oolite) in sector M—5 are four outcrops of the Upper Ordovician Leemon Formation (newly defined by Thompson and Satterfield, 1975, as part of their Edgewood Group), a brownish-red to purple limestone, partly coarsely con- glomeratic and partly somewhat shaly. The Leemon con- tains fossils that may range from Late Ordovician to Early Silurian age: Dalmanella edgewoodensis, a primitive clorin- did of the genus, Brevilamnulella, and a rostrospiroid (collection E2-1; _A. J. Boucot, written commun., 1971). A coral fragment, identified by W. A. Oliver, Jr. (written commun., 1971) as cf. Pycnolithus sp., is less diagnostic but indicates Late Ordovician to Middle Silurian age. Similar limestone just south of a fault in the northwestern GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI part of locality E-7 probably is Leemon also. The color is more like that typical of the Bainbridge Limestone, but the fauna is of Leemon age. SILURIAN SYSTEM BAINBRIDGE LIMESTONE Thin—bedded cream or tan limestone with a greenish cast similar to part of the Bainbridge Limestone (Ball, 1939) occurs in two outcrops surrounded by outcrops of massive dolomite in sectors L-4 and M-4. Fragmental fossil material from this limestone has been tentatively identified as Middle Silurian (R. W. Paul, written commun., 1969), which is the age of the Bainbridge. Similar limestone forms the northern- most ofthree limestone outcrops in the center of sector U—10. Another three outcrops of fossiliferous limestone, light dove gray, fine to medium crystalline, and weathering pinkish brown, is found in a stream bed on the line between sectors U-9 and U-lO and probably represents another part of the Bainbridge. It contains a fragmentary strongly trilobate pentamerinid brachiopod with characteristics believed to indicate Middle Silurian (Wenlock-Ludlow) age (A. J. Boucot, written commun., 1971). Boucot comments, “The Bainbridge Limestone is the nearest named unit that this material could be assigned to, but shells ofthis type have not previously been reported from the Bainbridge (dolomites in the Mid—Continent do have them)” This limestone also contained an indeterminate stromatoporoid. Two localities in F-8 also appear to be Bainbridge Limestone. A sixth locality, not seen by us, was reported from the northeastern part ofthe ring zone, in a highway construction drill hole made when State Route 5 was moved a short distance (noted in Snyder and Gerdemann, 1965, p. 485). It should be noted that all the limestone outcrops shown on the map, except those specifically described here as Noix(?), Leemon or Bainbridge, very much resemble lithologically the limestone beds paleontologically identified as of Kimmswick age. However, very few of the limestone outcrops were examined paleontologically; therefore, some of them may be unrecognized Silurian limestone or possibly Middle and Upper Ordovician limestones not previously mentioned. Because of the paleogeographic importance of the limestone stratigraphic assignments, a specific study of all the limestone outcrops mapped is highly desirable. REGIONAL PALEOGEOGRAPHIC IMPLICATIONS The Decaturville structure provides interesting paleogeographic information because it exposes formations otherwise seen only on the flanks of the St. Francois Mountains 113 km to the east and because it preserves remnants of other formations that elsewhere are missing across central Missouri. The Upper Cambrian-Lower Or- dovician sequence is different from that described in eastern STRUCTURE Missouri only in relatively minor ways (Koenig, 1967). No conglomerate or arkose is seen in the exposed or cored Lamotte Sandstone, although the basal part may include such phases. Edgewise conglomerate beds were not found in the Davis Formation, nor were sandstone beds. The Derby and Doe Run differ considerably from those formations in described eastern Missouri sections; they are massive to the point of being without bedding, entirely dolomite without any siltstone or shale, and white or mottled two—tone gray. In the Eminence Dolomite, no quartz druse or fossils were found, though these are common elsewhere. The Gunter Member and the upper part ofthe Gasconade, Roubidoux, and the Jefferson City rocks are relatively similar to those in these formations in the surrounding region. A maximum of 43 m of Jefferson City Dolomite was present in the area when the Middle Ordovician Kimmswick Limestone was deposited. The Kimmswick seems principally to have been laid down in lows or on slopes of an erosion surface with about 15 m of relief. It is considerably thinner than where it crops out in eastern Missouri. No rocks of the intervening section (including the Middle Ordovician St. Peter Sandstone) were found, and it is not known if they ever covered the area. A similar unrecorded interval existed after the Kimmswick until latest Ordovician time when the Noix(?) Oolite and the Leemon Formation were deposited. ln- asmuch as the Sexton Creek Limestone of Savage (1909) has not been identified in the area, the Middle Silurian Bainbridge Limestone is assumed to have been deposited disconformably upon the Leemon Formation. Whether these units were continuous or filled only topographic lows is not known, but the Upper Ordovician to Lower Silurian seas and Middle Silurian sea clearly covered this part of the Ozark dome. STRUCTURE The Decaturville structure is a sharply fault bounded, roughly elliptical area of intense and unusual deformation, set within the broad expanse of nearly flat lying Paleozoic sedimentary rocks of the Ozark Plateau. The major axis of the ellipse trends northwest and measures 6.3 km; the minor axis is 5.5 km long. Inside the fault boundary, a structurally depressed ring zone about [.6 km wide surrounds a central domelike uplift about 2.4 km across. Rocks preserved in the ring depression include Ordovician and Silurian limestones, outliers of two formations which have westward pinchouts in eastern Missouri, 135 and 277 km from Decaturville. The central uplift is capped by a formation 305 m above normal stratigraphic position, and isolated blocks of other units have been lifted as much as 540 m. Decaturville affords the westernmost exposure of these formations after they dip beneath younger rocks on the flank of the St. Francois Mountains [45 km to the east. The disturbed area is characterized by complex folds and faults and by brecciation of unusual type. ll DEFORMATION OUTSIDE THE STRUCTURE The structure lies along the crestal zone of a regional upwarp, as can be seen in the Geologic Map of Missouri (McCracken, I961) and in the structure contours on the base of the Gasconade Dolomite (fig. 5). These contour data indicate that the Gasconade in the vicinity of Decaturville has regional dips of less than 2 m per km east and west from the axis of the upwarp. Eleven kilometers northwest along the axial strike from Decaturville, the Eminence Dolomite and perhaps some Potosi Dolomite are exposed near Camdenton in what is called the Proctor anticline. These formations are partly truncated on the west limb of the anticline by the Red Arrow fault. The fault does not apparently continue toward Decaturville, but a small fault that offsets the structure’s ring boundary lies along a direct projection of the Red Arrow fault and has the same sense of displacement. The regional upwarp persists another 24 km northwest of the Ozarks. No direct relationship between the Decaturville structure and this regional fold is apparent. Another slight regional tilt is observed from elevations ofthe base of the Roubidoux Dolomite outside the structure. The Roubidoux seems to dip northward fairly uniformly across the mapped area, at about 4 m per km. A slight eastward component of dip on the Roubidoux base amounts to about 2 m per km from one side of the structure to the other. Most bed dips outside the structure are very low, but isolated outcrops show dips as steep as 32°, particularly 4.8—6.4 km away to the south (near Eldridge) and t0 the north (along County Highway K). The dipping beds do not define folds, and outcrops are too sparse for tracing of possible structural lines. Because most outcrops show flat-lying beds between the structure and these dipping beds, and because a zone of disturbance was not found at this distance all around the structure, we do not believe that the tilted rocks are related to the Decaturville structure. More likely they are evidence of faults related to the deformation episode marked by the Red Arrow fault. An orthogonal drainage pattern outside the structure is typical of much of central and western Missouri and reflects pervasive joint sets trending northeast and northwest. That the axes of the elliptical structure also trend in these directions is probably not coincidence. The drainage pattern and jointing are completely disrupted at the ring boundary of the Decaturville structure, and little evidence would indicate that the joint pattern influenced deformation within the structure or that it subsequently was reimposed across the disturbed area. About 1.6 km west of the structure, and just off the boundaries of the geologic map (pl. 1), a north-trending straight segment of Banks Branch marks a fault of unknown but probably small displacement. Small crossfaults are associated with the east side of this fault, but they cannot be traced eastward far enough to intersect the ring boundary. Nearby, and in the mapped area (sector N-2), sandstone, 12 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI l T. 40 N. 39 PROCTOR ANTICLINE 38 38° — — 0 0 <3 37 DECATURVILLE 36 STRUCTURE 750\ 35 30 EXPLANATION STRUCTURE CONTOURS—Showing 34 elevation of the base of upper part of Q @ Gasconade Dolomite in feet above sea r“ level. Hachured to indicate closed area _ 500 of lower elevation. Contour interval 50 feet (15.2 rn) .389 LOCATION AND ELEVATION, IN FEET, 33 OF WELL—LOG MEASUREMENT _ A 0 20 KILOMETERS P_J_.._I_P l $2" 0 10 MILES N. i l l | l l R. 19 w. 18 17 16 15 14 13 12 R. 11 w. FIGURE 5,—Structure contour map on the base of the upper part of the Gasconade Dolomite. Well-log information from Missouri Geological Survey. probably Roubidoux but possibly Gasconade, occurs along a fault in the Gasconade. A few small folds were seen short distances beyond the structure. Low-amplitude undulations a few feet wide occur in thin-bedded sandstone and chert of the Roubidoux in poor roadside exposures at the south edge of the mapped area (State Route 5, sector 2-18) and just off the southeast corner of the area of plate 1 on County Highway W (near center of sec. 15, T. 36 N., R. 16 W.). One fairly sharp monoclinal flexure, showing asymmetric rotation outward from the structure, was noted in a cliff exposure of Roubidoux Dolomite along a streamcut at sector G-23. Axes of these features are oriented crudely circumferential with respect to the Decaturville structure. The small faults and folds described above cannot be related specifically to the structure, but the close proximity STRUCTURE and type of deformatioanarticularly the folds—suggest that these features were products ofthe main disturbance. In three other places, the relationship is clear: small faults cut the ring fault that bounds the structure (sectors D-7, D-l2, G-5). Gentle dips along the line of the crossfault in D-7 indicate that deformation extends no more than 183 m beyond the ring fault. Broad—scale warping of the plateau sequence just outside the ring fault can be inferred from the isolated outcrops of marker beds. It is not clear how much of the apparent warping just outside the fault was produced when the structure was formed and how much might be due to earlier regional deformation. Elevations of the base of the Jefferson City Dolomite in the zone outside the ring fault indicate that formations south and north of the structure are respectively about 31 m and 76 m below the positions expected from the regional structure contours shown in figure 5. The northward deepening matches the plunge of the regional structure, but either that anticlinal plunge culminated in a low north of the Decaturville structure and then slightly reversed northward toward Camdenton or the Decaturville disturbance produced significant downwarp outside the ring boundary. The probable original dip ofstrata across the area was on the order of 1° , and the disturbance can be inferred to have caused elevation changes in formations of 31 m at the maximum, from one side of the structure to the other. The structurally induced dips outside the structure, therefore, probably were typically 10 to 2°. RING-FAULT BOUNDARY A normal ring fault, with downthrown side toward the center, surrounds the structure and forms the boundary around the area of most intense deformation. As can be seen from outcrop distribution shown on plate 1, this fault can only be inferred along much of its length; in places, the boundary could be a flexure rather than a fault. However, the likelihood of a continuous ring fault is indicated by the abundance of Roubidoux float which in general ends rather abruptly just outside the inferred fault line, by the very appreciable stratigraphic displacement over short distances, and by the general lack of dip in beds outside the line sufficient to account for the displacement across the boundary. The fault is known to be broken by crossfaults in three places on the north and northwest. Probably many other crossfaults have not yet been detected; these might alter the ring-fault trace now shown as a combination of inferred straight and somewhat scalloped segments. We believe, however, that changes in float and breaks in topographic slope permit the general line of the ring fault to be drawn to within :61 m of the actual location even in areas of sparsest outcrops. Displacement on the ring fault can be estimated fairly closely in many places. Generally, Jefferson City Dolomite is l3 downdropped against Roubidoux Dolomite, but in places offset is entirely within the Roubidoux; in only three places of low topography is Gasconade Dolomite exposed against the fault on the footwall. The displacement ranges from 20 to 128 m and varies considerably in short distances. Because of unknown amounts of drag on outside beds and uncertainties in contact elevations and formation thicknesses, the es- timates of displacement are considered accurate to about 6 m. Figure 6 shows generalized amounts of stratigraphic displacement in the structure. On the north side, the ring fault has dropped blocks down 61-92 m. In the northwest, displacements are 61—107 m. An outcrop block in sectors D-7 and E-7 (pl. 1) is down 107 m at the north end and 82 m at the south end. The western sector of the fault shows the maximum known range of displacements, from less than 20 m to more than 122 m, the large variations over small distances owing to the presence of small fault-bounded blocks that dropped independently of each other. At the south edge of L-3, uppermost Jefferson City is juxtaposed against Gasconade, and displacement is 107 m. In another fault block just to the north, Roubidoux lies against Roubidoux along the ring fault, and displacement is no more than 20 m. At one place on the south edge (sector V-12), displacement of 60 m is believed probable. In the southeast and northeast, reliable estimates cannot be made. In two map sectors on the east boundary, N-22 and 0-22, dis- placements are 76 and 95 m. Larger and smaller displacements alternate irregularly around the perimeter. The pattern is clear on the north and west sides where outcrop data provide good control, and it seems on the basis of sparse data also to be clear on the east and south sides. A wavelike sequence of highs and lows marked by displacement is consonant with the type of movements recorded by complex structural features of the central uplift, to be discussed in a subsequent section entitled “Central uplift.” Good outcrops just outside the fault area are sparse, but drag on the footwall strata seems commonly to extend no more than 30—60 m beyond the ring boundary. An exception is seen in sector D-7 where dips of 8° are seen about 180 m outside the fault boundary, but these may relate to a fault that cuts the ring fault. In sector L-3, where fault displace— ment is 107 m, beds are dragged down about 2 m; a resulting tilt is not seen more than 15 m beyond the fault. In a few places at short distances beyond the ring fault, Roubidoux sandstone has been highly fractured, and the fracture network so silicified that it is resistant and stands out in relief on outcrop surfaces. The dip of the fault is known from direct evidence only in one place, M-3, where the fault has the maximum known displacement. At M-3, an exploratory hole drilled about 150 m inside the fault line by the St. Louis Smelting and Refining Co. went through a normal downdropped sequence to 90 m, then through a stratigraphicallyjumbled zone to 150 m, and then crossed into the normal footwall sequence. The wide 14 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI l KILOMETER | I EXPLANATION lSOPLETH—Showing vertical displacement relative to pre— structural configuration. Contour interval of downdropped area (stippled) 50 feet (15.24 m), of uplifted area, 100 feet (30.48 m). Dashed line labeled 0 indicates line of no relative elevation change FIGURE 6.7Map of isopleths showing pattern of vertical displacements in Decaturville structure relative to prestructure configuration. Heavy solid lines, faults. broken zone probably reflects complications from a fault fault is clearly evident at 150 m. This gives the fault an between the hole site and the ring fault, but the main ring average dip of 45° from the surface to a depth of 150 m. A STRUCTURE 15 similar dip is suggested in L-3just to the north by a large slab of fault breccia that dips 45° E. This dip does not fit much of the fault trace shown on plate 1. Most of the trace across the gentle topography suggests a steeply dipping fault at the surface. The curving fault trace is drawn as a smooth, continuous line, which may be misleading because in places the fault may be mislocated by perhaps 60-90 m, and may actually be broken by many small concealed faults. It is believed that most of the fault is steep at the present ground surface, and in profile curves inward below ground to form a continuous bowl-like boundary surface around the entire Decaturville structure. We surmise that the fault dips 60°-75° just below surface and about 30° 150-200 meters lower to give the observed average dip of 45°. Crossfaults that offset the ring fault by 140, 79, and 107 m, are mapped in G-S, E-7, and D—12, respectively. Inasmuch as these faults have displaced beds outside the ring fault little, if any, they are regarded as tear faults that have horizontal displacement. They probably formed simultaneously with the ring fault; and beds inside the ring fault, being more free to move in a different stress environment, are vertically displaced different amounts on opposite sides of the crossfaults. In E-8, for example, Jefferson City marker beds indicate 60 m of downthrow on the ring fault on the north side of the tear fault, and 107 m on the south side, in the present position of the blocks. If the apparent lateral displacement is removed, the difference in vertical dis- placements on opposite sides of the fault is even greater. DEPRESSED ZONE Inside the ring fault lies a ll/z-km-wide belt of mostly Jefferson City beds, extremely complexly deformed and structurally depressed as much as 128 m relative to their normal position outside the structure. Figure 6, contoured to show lines of equal vertical displacement throughout the whole structure, indicates that the depressed zone on the north and west is variously monoclinal, synformal, and antiformal in general cross section. It has a warped monoclinal slope out to the ring fault on the east and south. The depressed zone ends inward at a hinge line, generally within the Roubidoux outcrop belt, which marks the beginning of the central uplift. Roubidoux in normal stratigraphic sequence is exposed by erosion in relative structural highs within the depressed zone, but more commonly it is exposed in windows through thrust plates of Jefferson City or in thrust plates overlying Jefferson City rocks. Complex interrelated folds, steep faults, and particularly thrust faults can be mapped or reasonably inferred throughout the zone in the western half of the structure. The rest of the zone has poor outcrops and may be of equally complex structure, but because only the top 23 m of the rock section generally is recognized in outcrop, we believe that fairly gentle, broad undulations, and small faults and folds characterize this part. In the southern and eastern areas of poor outcrop, small folds typically are asymmetric, with the steeper limb toward the center of the structure; the few inferred faults apparently are steep. For the western sector, outcrop data are good enough to permit construction of block diagrams, shown on plate 2, which provide a means to add inferred structural details that we feel are necessary to explain observed surface complex— ities. Because they represent various styles of deformation that are characteristic of different sectors of the depressed zone, their settings, structures, and movement sequences are described in some detail. Block A displays a shinglelike arrangement of 10 thrust plates. Age relations among the faults are not entirely clear, but faults A and Bapparently formed early with northward displacement (clockwise relative to the center of the struc- ture). The other, later thrust faults all involved displacement northeastward toward the center, commonly raising the Roubidoux to a slight structural high in the depressed ring zone. Lateral displacements on the faults range from about 30 to 90 m, as shown in the displacement of fault A by fault D; and stratigraphic throw is on the order of6 to 30 m. Small faults are numerous and probably formed as local surfaces of adjustment during thrusting. No steep faults were mapped in the block area. Bed dips are gentle to moderate and the general configuration suggests gentle, circumferentially oriented folds associated with, or broken by, the thrusts. One radial fold, overturned northward, is delineated by the sandstone on thrust plate J. Pavement exposures of coarse breccia, believed to mark the sole of plate B, are seen in two places. Block B shows gentle folds, circumferential to the center of the structure, that probably formed and were broken by steep faults before the inward thrust pushed at the south corner of the block. That thrust was overridden by a later thrust which apparently moved outward relative to the center of the structure and southward, and the folds probably were warped in axial trend, plunge, and symmetry by that movement. The late thrust block itself was broken by a fault that has a curving trace. Repetition of the sandstone marker bed suggests that this fault was a surface of slip which allowed part of the late thrust block to lag behind the leading edge of the thrust mass in its upward and southward movement. Block C is a transparent representation of generally circumferential but crosswarped folds and faults in the middle marker sandstone bed (marker no. 10) of the Jefferson City. The folds vary considerably in shape and symmetry, perhaps in part because of block rotations along the faults that break the folds. Near the east end of the diagramed area the sandstones in two outcrop lines separated by a fault are overturned in opposite directions; these may have been connected above the present surface as parts of a mushroomlike box fold, or the steep fault shown l6 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI may actually be an outward thrust that caused the pronounc- ed westward overturning of the fold in the outer block. The folds and faults were overridden by a thrust plate of Roubidoux apparently directed toward the northwest. Block D portrays a particularly well exposed structure and a clear sequence of deformation phases. Circumferential folds, one of which is slightly overturned outward, formed first. Steep faults broke the folds to produce a horst-and-graben pattern. A small thrust seems to have formed after the folds and before the steep faults, but the main thrust movement clearly occurred after the normal faults and involved displacement of the entire mass toward the center of the structure. Pavement outliers of coarse breccia in front of the thrust-plate edges show that the plates originally extended at least short distances farther eastward. Block E displays a gentle dome partly overridden by thrusts. The dome was broken by one steep fault; then inferred inward and outward movements probably involving mostly bedding-plane slip occurred in the gently dipping section on opposite sides of the dome. Fault intersections or truncations indicate that inward thrusting preceded outward thrusting, as in block B. The change from inward to outward movement is an important facet of our interpretation of the total-movement picture given in a later section. Outward slip apparently was the dominant movement and is expressed by four intricate thrust sheets on the side of the dome toward the center of the structure. Coarse breccias mark the thrust soles at the leading edges or in windows through the plates. Block F ShOWS the main ring fault and is characterized by imbricate thrusts in a gently warped eastward-dipping section. Eight outward thrusts can be inferred from available exposures. Deletion and repetition of marker beds varies in the block. The upper sandstone of the Jefferson City Dolomite (marker no. 11) is repeated at the north end, but does not appear in the exposed section at the south end where the thrusts instead repeat the Kimmswick Limestone. Inasmuch as the section below the Kimmswick in two of the plates in this block diagram is thick enough to include the upper sandstone of the Jefferson City, its absence suggests that still other thrusts, not mapped, may be present. The outermost thrust is truncated against the ring fault which is believed to predate the thrust. One steep fault at the south end of the block cuts three thrust plates and itself is cut by a fault that probably changes northward from a reverse fault to a thrust. Block G also shows a series of imbricate thrusts involving outward movement of plates. Of particular interest is the outermost thrust, represented by the upper sandstone of the Roubidoux (no. 8) which is partly eroded and shows the overridden section. The sandstone evidently was a well-lubricated material that flowed in part during its outward translation by thrusting or sliding; at one place it is 12 m thick instead of the normal [.2 m. A small outlier ofthe sandstone shows the former minimum extent of the thrust plate. The section is virtually unfolded; one high-angle reverse fault was mapped and appears to have formed before the thrusts. Block H shows a transparent representation of a very gently folded sequence broken by two outward-directed thrusts. The salient feature is local structure caused by the thrusts in the middle and northern segments of the block. Progressive overturning of the basal sandstone of the Jefferson City indicates the effect of either push or drag by the thrusts that must have extended over that part of the sandstone. The drag folding of beds just beneath one of the thrusts is virtually completely exposed in the area of the northern block segment. In summary, the structural features displayed in the diagramed blocks and elsewhere on the west side permit some generalizations concerning a sequence of deformation phases in that part of the depressed zone. Circumferential folds occur in seven of the eight blocks and radial folds (excluding small crosswarps) in only one block. Steep faults were mapped in four of the eight blocks; in five instances in these blocks they occur with folds which they postdate. In three of the four blocks that contain both steep faults and thrusts, the steep faults are earlier than the thrusts. Ifwe look at the entire west side, it seems clear that two phases of block faulting occurred: One associated with an early phase of folding and before thrusting, and another as a late, postthrusting period involving vertical structual ad- justments along relatively long faults. Outward thrusts occur in seven ofthe eight blocks and inward thrusts in four blocks; in two of three blocks where the outward and inward displacements occur together, inward movement appears to have taken place earlier than outward. Except on the southwest, thrusts inferred to be inward are uncommon, and outward thrusts are dominant. On the north side, the configuration of marker beds indicates fairly open folds around axes oriented both circumferentially and radially with respect to the center of the structure. Folds with radial axes seem more common in this area than elsewhere, and they probably slightly postdate circumferential folds. One inward thrust is inferred, but the area is dominated by late block faults along subradial and circumferential lines. To the east and south, no large vertical or horizontal dislocations are suggested by the marker beds seen in sparse outcrops, but many small faults undoubtedly are present. Dips indicate that gentle principal- ly circumferential folds are the dominant type of structural feature in that part of the depressed zone. EXOTIC BLOCKS IN THE DEPRESSED ZONE Three unusual rock occurrences in the depressed zone on the northeast, near the village of Decaturville (G-l9, H-l9), are believed to have special significance. At the road intersection in the village is an area of outcrop of highly fractured sandstone some 55 by 180 m. Continuity between outcrops is confirmed by drill data. This sandstone is STRUCTURE identified as Gunter Sandstone Member of the Gasconade Dolomite because its bimodal grainsize, coarse, pitted, golf-ball-Iike grains in the large-size fraction, and its unusual grain fracture or cleavage are exactly like those of the Gunter along the main outcrop belt farther into the structure. The sandstone bears no resemblance to any other Paleozoic sandstone that conceivably might once have been present in the area. Available drill cores show that the sandstone extends 27 m below the surface and lies upon Kimmswick Limestone in normal sequence above the Jefferson City Dolomite. The sandstone dips 70° at the surface, and according to the drill data probably dips about 45° overall. The underlying rocks are horizontal. Normal thickness for the Gunter is no more than 8 m, so at least a doubling of thickness by flowing, faulting, and perhaps folding is indicated here. It is clear that at this locality a great tilted block of Gunter sits 171-180 m above its normal stratigraphic position and upon distorted beds at the top of the local stratigraphic section. A second unusual occurrence involves coarse flakes of mica, presumably from a block of displaced Precambrian rock, found in a fresh grave in the Decaturville cemetery (sector G-20) by Mr. H. B. Hart. The find was confirmed by Snyder and Gerdemann (1965). This material lies along the ring fault and, like the sandstone, seems from its general position to be lying on top of the normal stratigraphic sequence. Precambrian rock in this occurrence is about 530 m above its normal position. The third occurrence is known only from drill cores of a hole put down just south of the village of Decaturville in a borrow pit in sector 1-19. Cores were obtained showing about 15 m of Derby and Doe Run Dolomitesjust below the surface. The base of the Derby was not penetrated but inasmuch as both formations here are 305 m above normal position, the Derby is assumed to be a block resting on Jefferson City Dolomite at the top of the local sequence. Blocks like these, markedly above normal position, have been noted in other cryptoexplosion structures and in fact are rather a hallmark of this type of structure. At Sierra Madera, Tex., for example (Wilshire and others, 1972), blocks of shatter-coned Permian rock occur along faults in Cretaceous rock at the edge of the structure. These rocks are about 210-300 m above normal position and 3-4 km laterally outside the envelope of shatter coming in the uplift. The implication of the exotic blocks as to the nature ofthe Decaturville and similar structures will be discussed in another section. CENTRAL UPLIFT The zero-displacement line on figure 6 marks the hinge line between the depressed zone and the central uplift. It occurs stratigraphically generally within the lower half ofthe Roubidoux Dolomite. The hinge line is very irregular, but roughly circular; the diameter of the uplift is about 10 percent greater in the northeast-southwest direction than in 17 other directions (3,350 versus 3,050 m). Except for areas of block faulting, upward displacement increases in a fairly regular fashion toward the center ofthe structure (fig. 6); and successively older formations are exposed, as in a normal domical structure. Thus the Gunter Member is raised 90-120 m, and the Derby and Doe Run at the center have been uplifted about 280-305 In. A zero-displacement line can be located approximately for each formation in reasonable cross sections, and the volumes of material inside and outside of a zero-displace- ment surface can be compared. Using average figures obtained for displacement on eight profiles across the structure, we calculate the total volume of uplifted material to be about 7.4-7.6x10x m3, and the volume of material in the depressed zone to be about 7.4-7.9xlO8 m3. Despite the approximations involved, the agreement of the two volumes is strikingly close. if the well-exposed Eminence Dolomite is representative, formations on the uplift are intricately deformed on a small scale. However, the formations as whole units apparently have relatively gentle dips off the dome. Comparisons of formation thickness to width of outcrop belt indicate that the Gasconade Dolomite has dips that average 10°-30° outward, the Eminence lO°-15°, the Potosi 20°—60°, and the Derby and Doe Run 5°-10°. The Derby and Doc Run Dolomites are intensely deformed and shatter coned, but they must be nearly flat lying in order to cover the central area which is 490 to 610 m across; they are only 43 m thick, and their base is nowhere exposed in the area. Calculated steeper formational dips for the Potosi Dolomite ringing this area may indicate an unusual structural discontinuity, to be discussed in another section, or they may simply indicate that the unit is thinner than supposed in his area. The Gunter Sandstone Member is the only thin marker bed (no. 3) useful for detecting structural offsets in the whole uplift sequence. Its outcrop line is conspicuously lobate and defines numerous gentle folds around axes radial to the center. On the south side of the uplift, several shallow basins are outlined by the Gunter. At about the Gunter outcrop line, a system of apparently steep faults nearly rings the center of the structure. On the east side of the dome the center is displaced upward along these faults, and on the west side it is displaced downward. Outside this fault line, in most places where outcrops permit mapping, block faulting has raised Eminence and Gunter commonly as much as 90-120 m within the Gasconade. Two isolated blocks within Gas- conade in a complex of block faults on the north side (l-l3) consist of uppermost Roubidoux beds downdropped [20-135 m. The large block faults postdate the folding ofthe Gunter and seem to represent a late phase of major vertical structural adjustment. Block [of plate 2 shows folds in the Gunter which vary from broad, open warps to a tight fold overturned outward and which are cut by steep faults with scissor, hingelike, and simple vertical displacements. One area of thrusts has been mapped within this general 18 zone, as illustrated in block J of plate 2. Deformation is very complex, but apparently involved three thrust sheets, partly broken into schuppen during thrusting, and later steeply dipping faults along which small—scale block movements occurred. The thrusts brought Gasconade inward over Eminence along a surface marked by conspicuous breccia, and Gunter inward in two slices over the upper part of Gasconade. A synclinal thrust block outlined by the Gunter Sandstone Member lies separated by erosion from its root thrust just to the west. The inward thrusts and the lobate Gunter line are believed to indicate an important phase of inward movement as the uplift formed. The Gunter lobate configuration also could have been produced by simple radial folding, but the lack of similar folds in beds inside and outside the Gunter suggests that uneven inward movement was more significant in producing the pattern. Lateral translation of beds also is suggested by the departure of the Gunter line from the general circularity seen in other contacts. The Gunter outcrop belt forms a rough somewhat triangular northeast—trending ellipse, with irregular wings produced by block faulting on the northwest and southeast. This ellipticity is more conspicuous than the 10 percent extension of the zero-displacement line of figure 6 noted previously, and is shown in that figure particularly by the 120-, 150-, and 180-m uplift contours. The northeast- trending major axis of the Gunter ellipse measures roughly 2,440 m, compared to the 1,950 m for the northwest-trending axis. The uplift ellipse is oriented perpendicular to the ellipse defined by the ring fault that bound the structure. The total length of Gunter segments (pl. 1) is about 30 percent longer than the perimeter ofthe ellipse or triangle on which the segments lie. This implies a shortening of the perimeter of the Gunter at its original prestructure stratigraphic level by 30 percent, as a result of inward and upward movement during formation of the central uplift. Similar configurations have been reported from cryp- toexplosion structures at Sierra Madera, Tex., Wells Creek Basin, Tenn., Gosses Bluff, Australia, and Vredefort, South Africa (Wilshire and Howard, 1968). As the Gunter typically is thinner within the uplift than it is outside the Decaturville structure, some of the apparent lengthening of it along its present perimeter may be due to thinning rather than to crowding of material into a smaller area. Nevertheless, a minimum of 20 percent shortening of perimeter by inward movement is indicated. Inside the elliptical Gunter zone, most deformation involves small open basins and canoelike folds on a scale of ten to a hundred meters. Many of these small structures apparently are laterally dislocated short distances and are slightly rotated. Numerous small thrusts are suspected but few could be mapped with certainty; lack of marker beds makes movement directions indeterminate. Across the Eminence belt a few steep faults extend (pl. 1); however, these were mapped mostly along lines of bed truncation and breccia lenses rather than on the basis of known displace- GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI ment. A rather unusual type of map ofthe southeastern area of Eminence outcrop was presented by Zimmerman and Amstutz (1965, fig. 4). Their interpretation involved several different periods of movement and different stress fields; and they reported cemented faults cut by later faults as indication of time separation of the fault periods. We examined the same area but could not find the features they showed; our interpretation of the structure there is not nearly as complex as theirs, although folds clearly have various orientations and are broken by thrusts and steeper faults. One out-of—place block was found within the Eminence belt—a small silver of Potosi raised along a steep fault in M-ll. The Potosi and Derby and Doe Run at the center of the uplift are poorly exposed and so massively bedded that in general structure within them could not be determined. As mentioned previously it is thought that the Derby and Doe Run, in particular, are nearly flat lying as a whole. It is clear from drill cores, however, that at least part ofthe central area is characterized by extreme and unusual brecciation and by movement of large blocks which mixed formations on a scale of 200 meters. Near the geometric center of the structure is the outcrop of pegmatite (L-l3). A drill hole (C-2 in fig. 7) through the pegmatite penetrated 15.2 m of pegmatite, then 1.8 m of slightly conglomeratic sandstone (presumably Lamotte), 4.3 m of pegmatite and schist, and then 140 m of mixed breccia with a few blocks of Potosi. Trenching and scraping around the pegmatite has shown the basement block to be about 15 by 20 m in extent. A sheath of muscovite schist perhaps as much as [.8 m thick is exposed on the north side, beyond which is a jumble of blocks of Lamotte Sandstone, Davis Formation, and Derby and Doe Run Dolomites (fig. 8). On the south and west the pegmatite is bounded byjuxtaposed blocks of Davis, Derby and Doe Run, Potosi, and Lamotte. The mixed formations occur on scales of small chips to individual blocks as much as 6 m across. Another outcrop 120 m southwest of the pegmatite outcrop shows a block of shatter-coned Potosi within the Derby and Doe Run country rock. About 360 m farther to the southwest is a large trench called the “sulfide pit” by the Ozark Exploration Co. This trench provides fresh exposures of shattered Potosi, mixed breccia (some of which contains sulfides), and upfaulted blocks of Davis (fig. 9). The trench and surrounding drill holes show the Davis to cover an area of about 90 by 90 m. It is well bedded and nearly horizontal in the open cut but has steeper dips in some ofthe drill holes. One hole shows an apparently complete section of Davis (41 m), but data from other holes suggest that most ofthe Davis may be somewhat thinner (27-30 m). If Bonneterre in the drill holes is not faulted, it shows similarly smaller thicknesses than expected from data outside the structure. This may indicate tectonic thinning on the order of 25-30 percent of original stratigraphic thickness of at least some strata at places within the generally thickened section in the central uplift. One drill hole (C-34 in fig. 7) near the large STRUCTURE Davis block started in Eminence and others (C-43, 52, 54) in Derby and Doe Run, indicating the presence of other blocks out of place within the sulfide-pit area. Outside the pit area, however, red soil, sparse outcrops, and a few drill holes show Potosi at the surface where it should be in its annulus around the center. Surface exposures and drill-hole data (figs. 7, 9) in the area of the sulfide pit reveal a fairly simple picture of blocks of upraised Davis and Derby and Doe Run and, in at least one place, downfaulted Eminence. In general, drill holes show normal stratigraphic sequences in the raised or dropped blocks. The mechanism of upfaulting is not clear, but movements must have been violent and must have been generated from below this stratigraphic level, because fine-scale mixed breccia involving rocks as deep as the Lamotte occurs as a matrix around some blocks in the sulfide pit. A much more complex and remarkable configuration is revealed in 30 drill holes from an area around the pegmatite; the area drilled is about 270 by 360 m and extends southwest from the pegmatite toward the sulfide pit (fig. 7). Of the 30 holes (27 shown in fig. 78), one was started in Precambrian pegmatite, one in Lamotte Sandstone, one in Bonneterre Dolomite, and two in Potosi Dolomite. Other out—of-place blocks undoubtedly exist at the surface, but the remaining 25 holes started in Derby and Doe Run, a fact which together with stream-bed outcrops, shows that these dolomites cover most of the central area in normal stratigraphic position. Below these capping dolomite units, however, the drill holes reveal a chaoticjumble of blocks, some with thicknesses as much as 60 m in core, totally out of sequence as to formation, and, in part, with a matrix of mixed breccia. Thisjumble of blocks fits the term “megabreccia” (Eggleton and Shoemaker, 1961) which commonly has been used for central zones of impact and cryptoexplosion structures. Long blocks in the megabreccia commonly seem to be on end, pointing inward and upward in an acicular pattern. The mixed breccia around the blocks exhibits evidence of flow and shock deformation; it is described in the section on shock features. In three holes near the pegmatite, the mixed breccia is notjust a matrix but makes up most ofthe 162-241 In penetrated in each hole. Small slivers of basement rock were cut in holes 15-37 m from hole C-2 which started in the pegmatite. Lamotte, the next deeper unit, was seen generally in minor amounts from the surface to a depth of 180 m in seven holes. Hole C-8, near the center ofthe structure, cut 37 m of Lamotte at the surface. In hole C-8, Lamotte is 180-200 in above its normal position relative to the Derby and Doe Run cap unit, but is lifted more than 440 m above its normal elevation. Bonneterre was only tentatively identified but is believed to be present in cores from eight holes. Six holes southwest of the pegmatite cut partial sections of Davis; this is a much lower frequency of occurrence than would be expected for the unit directly below the Derby and Doe Run which blanket the surface. Derby and Doe Run are scattered 19 throughout the jumble cut by most of the drill holes and occur at least as deep as the deepest core. Potosi, positively identified in five holes and tentatively in nine others, occurs in slivers and blocks down as much as 210 m relative to its outcrop position above the present elevation of the Derby and Doe Run cap layer. Various Potosi blocks range in structural uplift from 60 to 2l0 m above their prestructure elevations. No formations above the Potosi were recognized in the central area drill cores, although some gray dolomite tentatively assigned to the upper half of Derby and Doe Run might possibly be Eminence. The extreme subsurface complexity shown by cores would be difficult to overstate. Fromjust below the Derby and Doe Run to as deep as 320 m from the surface, it is rare to be able to connect, even tentatively, formations cut in adjacent drill holes even spaced as closely as 15 m. Inasmuch as the holes are limited to a small area, the lateral extent of this megabreccia is not well defined. Analogous intense block brecciation andjumbling of formations, however, are found in holes 275 m from the pegmatite but not at 425 m, in the southwest direction. To the northeast, megabreccia is not present 360 m from the pegmatite. Northwest and southeast of the pegmatite, the area of megabrecciation extends more than 120 m. Thus, this type of brecciation apparently characterizes a zone perhaps as much as 610 m long and at least 270 m wide. Depth ofthe megabreccia mass or lens is greater than 320 m, asjudged from the one hole that went that deep. Certainly the basement and overlying Lamotte were disrupted, but the cores show such a minuscule amount of basement rock and such a small amount of Lamotte relative to its presumed 90-m thickness as to suggest less mixing with other units than that undergone by higher formations. If the stratigraphic pile were undisturbed, then basement would lie about 200 m deep at the center of the uplift. In fact, except for isolated blocks torn loose and lifted, not only does the basement lie below 320 m, but so does the Lamotte. This gives a minimum figure of4l0 m below surface for basement rocks; the lack of Lamotte in the deep hole suggests that it may occur at least 30-60 m deeper, making basement depth likely to be 440-470 m. Normally, basement in the surrounding plateau is 490-520 m below the elevation at the center ofthe structure, so little general uplift of the basement surface under the structure seems indicated. Intensity of formation-mixing within the megabreccia mass is greatest at its apparent center; only there do the blocks of basement occur, and only there is fine-scale mixed breccia the dominant lithology rather than a sparse filling around large blocks. The general pattern seen in the megabreccia suggests that it is of limited lateral extent and is confined under the capping stratum of the uplift center, that it dominantly involved formations above the Lamotte but none higher than Potosi, and that it decreases somewhat in intensity of block dislocation outward from a central focus of disruption. GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI ,2 9'15 Peg matite :_ z ’ _ A» . ': «’5 -_ Approximate limit of, ‘ shatter cones at‘surface Sulfide pit '37 , W I, ‘ AOOFEET 100 METERS FIGURE 7 (above and facing page).—Numbered drill—hole sections and locations at center of Decaturville structure. A, photograph showing drill holes. B, drill-hole sections in pegmatite area. C, drill-hole sections in sulfide-pit area. 21 STRUCTURE CORE MISSING ”H7 42.“: A 707 A A ZADAT Sinai” mm A Bar 4 A _4..H HEAE A E/ 4 A E ..... A A J 2.07 Zr: A _: A E A 7 /_ ainxxxgzt M H H H_U.U_"_E___:_:_§r/m H H H u ..... in u n H H H H 1. M.” WU ,,,,, LAr 457 sass/A. AOOP7 EMT: FNIUZEJEEEIJ7 8:th .... filo T m/ 4.. MA 474:7 _______ 4.21.2733. éZAfi fl/HV [—7 A 4% /47 Aim/4.474.327 4.37 ..... 4 B A NEE FEET METERS 7 0 50 0 E ii 100 3.2;:_2.3.::_.___._.____.__ ...... # 4V..._______11:12:12___:r:_,____________:_ xxxxxx QMIUYLHELLAE: fig I __v_____.____._____Ale_,_~, on U A __:___._.4____.____:::__:__::.;. . .52.}: .3. 22.: xxx-.xunhxxxx _ : wNIO TE.T.;::. wmIUEE.I;.I.:E. 3.0 a _ _ xxxxxxx _ 7 7 _ _ _ xxxxxxxxxx _____1_.1_i_i_ :_:......x ___________:__::_7 é::__::__.. x a :— m e m m n o t .m m ® a D m m m s m F e e .u k. m m m m w. m m w m N D B L P S O m 1% EEEM m L .W. e n m m .n nun E m m m. e e .m .m m m n c D 0 fi 6 I duh m m w m mm a m H i m .mm h .I u m 0 e .. S M S E P D Elu§ .......... UA I§ flay/A. C 22 FIGURE 8.—Pegmatite area (sector L-l3), showingjuxtaposition of blocks of different formations at center of structure; DC. Precambrian pegmatite and schist;€|, Lamotte Sandstone; Cd, Davis Formation; €dd. Derby and Doe Run Dolomites; Cp, Potosi Dolomite;mb, mixed breccia. FIGURE 9.——Sulfide pit (sector M-IZ) showing juxtaposition of blocks of different formations; Cd, Davis Formation;€p, Potosi Dolomite;mb, sulfidic mixed breccia. GEOPHYSICAL DATA The magnetic map of Missouri (Missouri Div. Geol. Survey, 1943), compiled from ground-based readings on 3.2- km centers at a lOO-gamma contour interval, depicts regional lithologic and structural trends in the Precambrian basement. Linear alinements trending northwest, and to a lesser extent northeast, dominate the pattern. More detailed aeromagnetic maps are available for the Decaturville area GEOLOGY OF THE DECATURVILLE IMPACT‘STRUCTURE, MISSOURI and make an interesting comparison with the local geology (fig. 10). The Decaturville structure is centered 4.8 km west of a conspicuous northwest-trending linear magnetic high, which is continuous to the Camdenton area where it corresponds closely to the Proctor anticline. A steep gradient on the west edge of the high underlies the position of the Red Arrow fault near Camdenton and is continuous to the southeast beyond Decaturville. It passes to the northeast of the Decaturville structure, approximately tangent to the ring fault. This suggests that the Red Arrow fault does not intersect the ring fault on the north side, as a simple projection of its surface trace would indicate. A similar and more conspicuous high 10 km farther east is seen on the State magnetic map. It extends for at least 105 km to the northwest but does not seem to correspond to surface geologic features. The two highs end or are interrupted by an east-west break in the magnetic pattern; this break is subtle but may extend west from some 5—8 km south of the Decaturville structure all the way to the Kansas border. Eastward it is not continuous but may be represented by minor breaks in a sharp high southwest of the Crooked Creek structure and across the tight magnetic pattern of the St. Francois Mountains. The aeromagnetic map of Stoutland quadrangle (fig. 10) shoWs a northwest-trending elliptical lOO-gamma low, about 1,520 m long and perhaps 1,220 m wide, centered about 180 m south of the center of the Decaturville structure. No evidence in outcrop or drill core indicates the presence of impact melt breccia like that which was found to have thermo-remanent magnetism that caused lows at the Gosses Bluff astrobleme in Australia (Milton and others, 1972). A probable explanation for the Decaturville low is that the basement was disrupted sufficiently by fracture and breccia- tion when the structure formed so that it now has no coherent remanent magnetic field. This was suggested by Beals, lnnes, and Rottenberg (1963) to be the reason for magnetic lows at various Canadian impact structures. It may be, however, that even with the coincidence of location no special explanation for the low at Decaturville is necessary, because the magnetic maps show small circular or elliptical features to be common in the Stoutland and Macks Creek quadrangles. These may relate to the presence of small masses of granitic rock scattered within a schist area, as is indicated from basement rocks exposed in the Decaturville area. Magnetometer ground traverses across the central highly broken zone of the Decaturville structure revealed no discrete small anomalies that might indicate either meteoritic or volcanic material within the breccia envelope or underlying basement. Gravity measurements were made by J. D. Hendricks, US. Geological Survey, at about 1.6-km intervals along roads in the area to produce a regional map. Terrain correction was not considered necessary because of the low relief in the area, particularly around the structure itself. 38°00' GEOPHYSICAL DATA 45’ 32°40' 92°50I ‘ " " X l \‘1 \L w _ é: [4.9 \\ 23004)} 2400 ‘ o \ T. qr \ S; x U \\\\\\ \‘ W / x 37°50' R. 16W. 5 KILOMETERS | l I 3 MILES EXPLANATION MAGNETIC CONTOURS—Showing total intensity of “5’5 LOCATION OF MEASURED MAXIMUM OR MINIMUM (I INTENSITY IN GAMMAS WITHIN CLOSED HIGH OR magnetic field of the earth in gammas relative to x 09 @ arbitrary datum. Hachured to indicate closed areas CLOSED LOW of lower magnetic intensity. Contour interval 20 270” gammas FIGURE IO.—Aeromagnetic map of Decaturville region showing outline of structure (modified from Missouri Div. Geol. Survey, l:62.500, Macks Creek and Stoutland aeromagnetic quadrangle maps, I962). 24 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI 38°00’ _C_0__ co \ , Area of .I OUTLINE OF DECATURVILLE figure 12 / STRUCTURE 37°50 \/ 5 KILOMETEFIS | 0 i I I I I F . . | 0 3 MILES EXPLANATION —-40.0— GRAVITY CONTOURS—Showing 0.5 0 LOCATION OF MEASUREMENT STATIONS — mGaI intervals relative to arbitrary datum FIGURE II.—Gravity map, Decaturville region; prepared by J. D. Hendricks, US. Geological Survey. Not terrain corrected. GEOPHYSICAL DATA Sulfidé pit GRAVlTY CONTOURS—Showlng 0.1 mGal intervals . relative to arbitrary datum p. 500F§ET ,r' ‘1 V “ Jae METERS FIGURE 12.~Gravity map, central area of Decaturville structure; prepared by J. D. Hendricks, U. S. Geological Survey. Not terrain corrected. Figure ll shows two gentle gravity highs from which a regional gradient extends northward to lower values. The Decaturville structure partly coincides with one of the highs but does not seem to influence the transition from the high to the regional slope, and therefore may have little to do with the presence of the l-mGal high. Inasmuch as most of the rocks in the stratigraphic section have fairly similar den— sities, vertical dislocations of the section associated with the structure should not produce much of a gravity effect. If anything, the brecciation and fracturing of much of the rock inside the ring fault might be expected to cause a slight relative gravity low across the structure. Such an effect is in fact seen in a more detailed map ofthe center of the structure (fig. 12), which shows an elongate gravity low exactly where outcrops and drill core show brecciation to be far greater than anywhere else in the structure. The central zone contains large quantities of polymict breccia having densities of 2.1—2.4 gm/cm3, as compared with the brecciated but more nearly normal dolomites, shale, and sandstone (densities 2.5-2.9 gm/cm3) of the surrounding area. Not enough data exist to permit modeling of the size and overall density ofthe breccia lens to compare with the observed gravity low. However, the shape and the location ofthe gravity low likely are valuable clues to the configuration of the breccia envelope. It is interesting to note that the gravity low has the same shape and trend as the larger oval of uplifted material defined by outcrops of the Gunter Sandstone Member. Within the generally low central zone, the two lowest areas are in places (particularly the sulfide pit) where exposures and drill cores indicate the 26 maximum known amount of sulfidic breccia fillings. Gravity lows in those places suggest a lack of large sulfide concen— trations in the subsurface. The significance of a 0.1-mGal positive anomaly located north of the pegmatite area is not known. SHOCK-RELATED DEFORMATION FEATURES The Decaturville structure contains several types of deformation features not normal for most structurally disturbed areas. These include monolithologic and mixed brecciation, rock granulation and small—scale rhombohedral rock cleavage, shatter cones, and intragranular planar features in quartz, dolomite, calcite, and mica, all believed to indicate extremely intense deformation. MONOLITHOLOGIC BRECCIAS Breccias consisting of clasts and mylonitic matrix derived from single beds occur throughout the structure in most of the fine— to medium-grained carbonate rocks. They are' particularly well developed in the Eminence Dolomite and the finer grained beds of the Roubidoux and Jefferson City Dolomites. Less common in the Gasconade Dolomite, they occur principally in the fine-grained dolomites rather than in the more abundant medium-grained saccharoidal dolomites. The one example of monolithologic breccia outside the structure was found in fine-grained Gasconade Dolomite a few feet beyond the ring fault. Pure fine-grained dolomites of the Jefferson City are brecciated, but the tan marly dolomites are not. Limestone beds above the Jefferson City are locally coarsely brecciated. The Potosi Dolomite contains some brecciated beds of this type but more commonly displays an unusual close-spaced irregularly rhombic cleavage which gives the beds a shattered appearance. Derby and Doe Run Dolomites also show this type of cleavage or intense granulation rather than a clast and matrix breccia development. Monolithologic breccias are nearly absent from rocks of the Davis Formation and Bonneterre Dolomite, and are not found in any of the sandstones throughout the stratigraphic section. Gunter Sandstone Member in places is highly fractured but not visibly brecciated; it is however intensely granulated. Snyder and Gerdemann (1965, p. 484) reported that “grain size analysis of the disaggregated sandstone shows far more -200 mesh material than is normal for this or other lower Paleozoic sandstones in Missouri.” Intensity of brecciation in Eminence beds near the center of the structure does not seem appreciably greater than in Jefferson City or Roubidoux beds in the outer depressed zone, suggesting that the brecciation is not related to distance from the center of general disruption. It should be emphasized that monolithologic breccias are not associated with faults; their occurrence and character clearly indicate that they result from shattering and dilation of individual beds. Even in sequences of beds less than 2.5 cm thick, each GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI FIGURE l3.—Monolithologic breccia, Eminence Dolomite. A, weathered surface, fragments in bas relief. B, sawed surface fragment-matrix relationships. showing brecciated bed maintains perfect stratigraphic coherence; there is no mixing with adjacent beds. In formations such as the Jefferson City a 7.6—cm bed of clean gray dolomite typically will be thoroughly brecciated, whereas the 2.5-cm beds of tan marly dolomite above and below it will show no brecciation whatever. In detail, monolithologic breccias are made up of angular to subrounded fragments that are of normal color for the dolomite unit in which they occur, generally less than 2.5 cm across, and surrounded by a white, cream, or tan matrix. Breccia character is especially clear on sawed surfaces, but fragments are commonly more resistant than the matrix, and weathered outcrop surfaces provide excellent displays ofthc breccias (fig. 13). The lowest intensity of brecciation produces fragments separated by hairline cracks that are filled with extremely fine grained mylonitic material. With SHOCK-RELATED DEFORMATION FEATURES more extensive development, to the point of having a greater percentage of matrix than fragments, the fragments become internally fractured and the matrix becomes a poorly sorted mixture of very small fragments, individual carbonate grains, and mylonite. Fragment-matrix boundaries typically are sharp and clean in thin section, showing little evidence of plucking or fraying of the fragments to produce the matrix. Many fragments apparently underwent rotation, and some alinement of grains near the borders of fragments suggests flow of the matrix; but the lack of mixing of adjacent beds shows that any movement must have been very small. The shattering and dilation that produce the fragmentation probably are followed by both flowage and crushing recompaction to produce the mylonitic matrix. The monolithologic breccias at Decaturville are virtually identical to those described for the Sierra Madera, Tex., impact structure by Wilshire and others (1971), as well as those for other impact structures in carbonate rocks, such as Flynn Creek, Tenn. (Roddy, 1968a), and Wells Creek, Tenn. (Wilson and others, 1968). MIXED BRECCIAS The megabreccia at the center of the structure consists of blocks ranging in largest dimension from one-third to twenty meters, which represent formations from the basement to the Potosi Dolomite. Finer grained mixed breccia consisting of fragments of the same range of formations forms a matrix around many of these blocks. ln drill holes C-2, 4, 10, and 12 at the center (fig. 7), mixed breccia makes up most ofthe 760 m ofcored material. The breccia is composed offragments in the complete range of sizes from clay to cobbles. Fragments of chalky Derby and Doe Run are markedly predominant, though easily recognized glauconitic pieces of Davis are common also (fig. 14). The other formations are poorly represented. Bonneterre, for instance, has not been positive- ly identified in the fragments, but this probably is a problem of lack of recognition of it as small fragments, because Bonneterre is at least sparsely present in the megabreccia blocks. Typically the fragments are white to pale gray (or green and gray for the Davis) and the matrix is light to medium gray; the two-tone color pattern is very con- spicuous. Fragments are very irregular in shape: some are angular with extremely sharp corners; some are very rounded. Some are broken or are plucked at the edges, with the darker fine-grained matrix material separating the pieces of the original single fragment. The matrix is composed of very tiny fragments and of individual grains, part of them microcrystalline. Individual grains are mostly dolomite, but glauconite is common, and quartz may constitute as much as l0 percent of some thin-sectioned samples. A very few ofthe quartz grains display planar features indicative of various intensities of shock. Shatter-cone segments are present in a few fragments of Potosi and Derby and Doe Run within the breccia. 27 FIGURE l4.—Core samples of mixed breccia with white fragments of Derby and Doe Run Dolomites and dark (dark-green) fragments of Davis Formation in a light-gray matrix of comminuted material from several rock units. Top sample from hole C-2, 132 m; bottom sample from hole C-lO, 52 m. Some zones within the core sections of mixed breccia are notable because of their flowage features. The matrix is finely banded and clearly shows folds and swirls and the characteristic of wrapping around intricately sculpted edges of fragments (fig. 15). In places, dolomite fragments surrounded by flow—banded material themselves seem to have been drawn out plastically to become irregular FIGURE 15,—Core samples of mixed breccia showing alined grains in flow laminae and fragments drawn out or shaped by flowage. Top sample from hole C-26, 65 m; bottom sample from hole C-2, 130 m. 28 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI FIGURE l6.w Mixed breccia at center of structure. A, exposure in sulfide pit (hammer for scale). Large block is Lamotte Sandstone in which the dark areas are discolored by sulfidization. Smaller angular fragments mostly dolomite from the Potosi and Derby and Doe Run Dolomites. B, hand specimen of mixed breccia with medium-dark—gray sulfidic matrix, light fragments of Derby and Doe Run Dolomites, and dark fragments ofiron sulfide (arrows). schlieren with ragged flamelike ends. At the center of the structure, a few zones as much as several inches thick in certain cores are conspicuously medium to dark gray. These contain as much as 10 percent sulfide fragments, which are mostly marcasite, with minor chalcopyrite, and traces of galena and sphalerite (fig. 16). They seem to be the same material as the rest ofthe breccia except for the sulfides. In the sulfide pit, most ofthe mixed breccia exposed is this type of dark material, with sulfides in numerous small grains as well as fragments as much as 8 cm across. The sulfide is incorporated in the breccia like other rock fragments, although overgrowths on a few fragments suggest minor deposition or redistribution of iron sulfide after the breccia was emplaced. Numerous fragments are botryoidal or concentrically banded, with sharp broken edges clearly showing that they did not crystallize in the breccia (Zimmerman and Amstutz, 1972). Mixed breccia is rare in core from drill holes at or near the sulfide pit, but this paucity is probably because such material had been removed from the core before we made our examination. This explanation may also apply to the apparent paucity of sulfide material in cores from the center. Sulfide-rich mixed breccia at the sulfide pit fills openings between large blocks of shattered Potosi, the normal country rock at that position. It also partly surrounds large upfaulted blocks of Davis. The Davis blocks are perhaps 60 m vertically up out of place, and indicate relatively large scale jostling of the local units. Granular material such as the mixed breccia was probably capable of moving greater distances than blocks, and of flowing or being injected into openings. An extreme example of granular flow is seen in the sulfide pit where Lamotte Sandstone has flowed along the contacts of two blocks of Potosi and the surrounding mixed breccia to form a laminated rind around the blocks (fig. 17). FIGURE l7.—Sandstone from Lamotte Sandstone (arrows) emplaced by flow between block of Potosi Dolomite (Cp) and mixed breccia(mb) in sulfide pit, Pen for scale. SHOCK-RELATED DEFORMATION FEATU RES This sandstone must have come from at least 150 m stratigraphically below its present position. Fragments in the breccias at the sulfide pit are primarily Potosi, Derby and Doe Run, and Davis. Some fragments have the texture of Derby and Doe Run but are mottled light and dark gray instead of white. These may be from the Eminence but are not typical of the Eminence. They have very small amounts of an extremely fine grained sulfide mineral that might be enough to give either Eminence or Derby and Doe Run an unusual color. This, together with the abundance of sulfide fragments, suggests that sulfide deposition favored the locality of the present sulfide pit over other parts of the area before formation of the structure. Mixed breccia different from the type seen in the central zone is shown in many places on the geologic map, but it apparently formed mainly by crushing, or in part by the filling of dilatant zones, along faults. It consists ofangular to somewhat rounded fragments of the various lithologies available from adjacent formations in a very fine grained matrix ofdolomite that weathers white to buff. For example, these breccias in the Jefferson City Dolomite contain fragments of gray dolomite, tan “cotton rock”, and chert; breccias in the Gasconade Dolomite contain fragments of chert with algal banding distinctive of the Gasconade. The matrix and some of the fragments show crushing but no indication of unusual crystal deformation such as might have been‘produced by shock in this type of breccia. The breccia is found along the traces of several thrust faults, forming pavement exposures that extend (for example, in map sectors N-lO, 0-7) well beyond the present edge of the upper plate and give some clue to the pre-erosion extent of the plate which overlay the thrust-sole breccia. Along steeper faults, the breccia is seen in erosion-resistant wall-like outcrops. Alined outcrops of breccia were used to infer faults in several places. SHATTER CONES Distinctive striated fracture surfaces in the form of cones or segments of cones have been reported from about 20 of the more than 60 known cryptoexplosion structures. At several structures, such as Sierra Madera, Tex. (Wilshire and others, 1972), Gosses Bluff, Australia (Milton and others, I972), the Ries Basin, Germany (M. V. Engelhardt, oral commun., I971), Brent, Canada (Dence, 1968), Nicholson Lake, Canada (Dence and others, 1968), and Sudbury, Ontario (Bray and others, 1966), shatter cones occur in association with clearcut petrographic evidence of shock deformation. Shatter cones have formed in quartz diorite at shock pressures of about 30 kb (kilobars) produced during chemical-explosion cratering experiments (Roddy and Davis, 1969). They have not been found in volcanic explosion environments or tectonically deformed rocks. Dietz as early as 1959 suggested that shatter cones indicate meteorite impact origin for the host structure, and no persuasive evidence to the contrary has yet been produced. 29 FIGURE l8.#Shatter-coned Derby and Doc Run Dolomites from near the pegmatite pit. At Decaturville, well-formed shatter cones (fig. 18) are common in outcrops ofthe Derby and Doe Run ina roughly elliptical area measuring about 420 by 480 m. (See pl. 1.) They do not occur at the present surface in identical dolomite of the Derby and Doe Run where the dolomite extends outside that ellipse in the central area. Cones are not found in the surrounding outcrop belt of Potosi but are present in displaced blocks of Potosi within the Derby and Doe Run area at the center ofthe structure. Fragments and blocks of Derby and Doe Run and Potosi display cone segments in many drill cores. The deepest occurrence noted is in a Derby and Doe Run block at 250 m in drill hole C—27, at the north limit of shatter cones in outcrop. Cones typically are 2.5-5 cm high, but cones as high as 25 cm have been found in mining pits near the pegmatite outcrop. Double cones pointing in opposite directions from a common base, or single cones pointing opposite to the general mass ofcones in an outcrop are not common, but seem more abundant at Decaturville than they are reported to be at other structures. The formation of shatter cones is believed by Johnson and Talbot (1964) to occur where rock inhomogeneities focus compressive stresses induced by passage of a shock-wave front. Another theoretical study (Gash, 1971, p. 34) suggests that cones should form beneath the shock source and that “very high amplitude shock sources are required” in order to provide the necessary tensile stresses. Gash estimated that well-formed cones in a rock with tensile strength of 100 bars would require a minimum source stress of 20-40 kb. Cone 30 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI axes in theory should be oriented perpendicular to an expanding, generally hemispherical wave front, in effect defining radii trom the shock center. Hargraves (1961), and later Manton in more detail (1965), at the Vredefort Ring, South Africa, showed that striations ofapparently randomly oriented cone segments in a single outcrop plot stereographically to define the projection of a cone pointing toward the center ofthe structure. Shatter cones form before the major structural deformations that produce the circular structures, so use of cone axes to define the shock center requires sufficient information to reconstruct the position of shatter-coned beds before the disturbance. This has been done successfully not only at Vredefort, but also at Sudbury (Bray, and others, [966) and Sierra Madera (Howard and Offield, 1968). At these structures, if beds are replaced in their original position, shatter cones point inward and upward to an inferred focus of shock-wave energy release. Shatter cones in a thick stratigraphic section at Gosses Bluff permit similar determination of focus; high beds give a different focus than lower beds, the difference being a measure of greater inward movement ofthe lower beds as the central uplift formed (Milton and others, [972). At Decaturville, striations on cones and cone segments were measured and plotted stereographically at eight localities to determine cone-axis orientations. Despite the intensive post—shatter coning disruption suffered by the FIGURE l9.—~Stereographic projection of shatter-cone striations defining cone with 92° apical angle (rotated to vertical). Of 21 measurements, one lies more than 110° from mean circle. Azimuth and plunge of striations measured in situ (x) and rotated to define cone with vertical axis (~); axis of cone in situ shown by small open square. Derby and Doe Run Dolomites, six ofthe eight plots showed that striations on diversely oriented cone segments gave excellent conical projection (fig. 19). Atotal of 120 striations were used in the 6 plots, and all but 12 fell with a ~+.10° variation in apical angle from the mean cone projection. Apical angles ranged from 79° to 92°, with a mean of 86°. Unfortunately, the orientation of the shatter cones is uncertain because bedding in the Derby and Doe Run is virtually impossible to define. Three localities display surfaces which may be bedding, and rotation of these to horizontal was done by stereographic projection. Cones in these beds then pointed in and down to a focus within the Davis. If the Derby and Doe Run beds have moved inward, as will be argued in the section on origin of the structure, then the downward-pointing cones in their original positions farther out would have defined an even deeper shock focus, probably within the Bonneterre. This is not in accord with the distribution of shock features discussed in the remainder of this section. We therefore believe that bedding in the shatter-coned rocks has not been correctly identified and that these cones cannot be used to determine the stratigraphic level at which the energy release was focused. PLANAR ELEMENTS IN QUARTZ Quartz develops distinctive intragranular features as a result of deformation under high transient pressures, such as occur during passage of a shock wave. Data from shock experiments and observed deformation features around nuclear-explosion sites show a definite progress in develop- ment of planar elements in quartz as intensity of shock deformation increases. At Decaturville, planar elements in quartz include open “cleavage” fractures, closely spaced parallel planes called “planar features,” and possible defor- mation lamellae (permanent strain zones produced by slip and indicative of high shock pressures). Rhombohedral cleavage is common in individual quartz grains of the Gunter Sandstone Member, including the large exotic block of Gunter at the village of Decaturville, The cleavage planes are open, clean fractures that typically appear in a rectilinear network of traces spaced 40-100 mm (micrometers) (fig. 20). They may occur in nwrly every grain in a thin section or in only a few. Cleavage is not found in quartz grains from formations that crop out outside the ring of Gunter. Nor has it been observed in the quartz that occurs as isolated grains or small clusters of grains in the carbonate formations below the Gunter. Rhombohedral cleavage does occur in some grains in blocks of Lamotte Sandstone, not as abundantly as in the Gunter but with finer spacing (4-8 am). The Lamotte, however, has a few grains that show well-developed basal cleavage 4000 I}, a cleavage not seen in the Gunter Sandstone Member. This cleavage is marked by dark traces about 3-5 um wide and spaced at 9-60,um (of the type shown in quartz grains of unidentified stratigraphic position in fig. 21). Inasmuch as basal cleavage requires more bonds broken per SHOCK-RELATED DEFORMATION FEATURES 3] FIGURE 20,—Photomicrograph (120X) showing widely spaced rhom- bohedral fractures in quartz grain of Gunter Sandstone Member of the Gasconade Dolomite. unit area than does rhombohedral cleavage (Fairbairn, 1939), the presence of basal cleavage and the much closer spacing of rhombohedral cleavage in the Lamotte suggest higher strain than was experienced by the Gunter along the present outcrop ring. Small lateral displacements or faults as defined by Carter (1968, p. 457) mark both types ofcleavage in the Lamotte grains, but they are not numerous. In fact, grains with any kind of planar fractures in the Lamotte probably make up no more than 2—3 percent of the 17 thin sections examined. Moreover, it is interesting that only Lamotte samples taken at the center of the structure show cleavage in the quartz; samples from the sulfide pit 490 m away show marked granular flowage on megascopic scale but the grains do not have cleavage. Even at the center, cleavage development is extremely variable. One Lamotte sample from a block in the wall ofthe pegmatite pit contains nearly twice as many grains with cleavage as all other samples combined, and it shows basal cleavage as common as rhombohedral cleavage in nearly all the fractured grains. Neither the pegmatite nor the schist below the Lamotte shows fracturing of quartz; some grains show unusual curving or chevronlike strain extinction bands, but complex- ly sutured boundaries are undisturbed and cleavage is absent. Quartz grains make up a few percent of the matrix in the mixed breccia at the center of the structure. One to two percent of these show closely spaced rhombohedral and basal cleavages, about equally abundant and well developed. In a very few grains, planar elements both parallel and perpendicular to the c-axis are spaced only about l-3,um apart. Other elements in some grains, equally rare and closely spaced, are crystallographically oriented parallel to {10B}. (fig. 22). These closely spaced planes are conspicuous in bright-field illumination and are visible independent of direction of light vibration. They are reasonably good examples of what Carter (1968) defined as “planar features,” but are much less common and are in fewer sets per grain than are reported in studies of several other probably impact structures (for example, Robertson and others, 1968; Short and Bunch, I968; Chao, 1967). Even rarer are grains containing planar elements parallel to {0001 l and { 10l3} and marked by abundant tiny cavities or inclusions. These are only moderately visible in bright-field illumination but are seen equally well in plane and polarized light; their Visibility FIGURE 21.~Photomicrograph (lZOX) of mixed breccia. Two large quartz grains in a carbonate matrix have prominent decorated planar features spaced 9-60pm and oriented parallel to{0001}. Polarized light. 32 varies with the direction of light vibration. These rare features seem to be rather poorly developed examples of what Carter (1968) has called deformation lamellae; they connote intragranular flow. As with the Lamotte, only mixed breccias from the center of the structure contain quartz grains that have the described planar elements. Virtually identical mixed breccia from the sulfide pit does not contain such grains. The stratigraphic origin of the quartz grains in the mixed breccia is not certain. Many of the grains are large and rounded and on that basis alone might have come from the Lamotte, Gunter, or perhaps the Davis. No fragments from higher than the Potosi have been identified in the mixed breccia; thus it does not seem likely that sand grains from a formation as high as the Gunter would be present. The Lamotte breccia fragments that were examined contain grains having basal and rhombohedral cleavage and a few microfaults, but the grains do not have planar features and in particular they do not show planes parallel to lIOT3l. The few quartz grains that do have {10.13} planar features were FIGURE 22.7Photomicrograph (200X) of mixed breccia, showing quartz grain with dominant planar features spaced l-3 pm and oriented parallel to {10%}. Plane light. mostly found in clusters of small angular grains within a single small core sample of mixed breccia. These grains are typical of grains in clusters or laminae in the silty portions of the Davis, although they could possibly be derived from any formation below the Eminence. We suspect that the grains having planar features are in fact from the Davis, because, as mentioned, they are not likely to have come from either the Gunter or the Lamotte, and other intervening formations generally lack quartz. The problem of representative sampling is considerable in trying to assure that grains were seen which indicated the GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI maximum shock—pressure levels attained. Quartz grains probably make up less than 5 percent ofthe examined mixed breccia samples. In samples showing the highest levels of shock found for both the breccia and the Lamotte, only 1 grain in 50 or [00 may contain planar features. The meaning of planar elements in quartz rests largely on statistical arguments in which type and orientation of planes are compared with like data obtained from static and dynamic deformation experiments and from proved impact structures. Figure 23 compares a histogram offrequency and orientation of the planar fractures observed in rocks of the Decaturville structure, with histograms from Meteor Crater, Ariz., from other impact structures, and from shock experiments. Planar elements at Decaturville are similar to those from Meteor Crater and Middlesboro, Ky.; the Decaturville histogram is deficient in the {IO—13} orientation relative to those from Sierra Madera, Tex., Clearwater Lake, Quebec, or from rocks experimentally shocked at 250-300 kb (kilobar) pressures; and the planar elements at Decaturville differ in every category from elements in tectonites or rocks deformed statically at low confining pressure and much lower strain rates. The high frequency of basal elements shown in the histogram of Decaturville samples occurs because cleavage, planar features, and (probable) deformation lamellae are plotted together. Clear— ly, most of the components measured were cleavage. The cited experiments and field studies permit little doubt that the types of planar elements in quartz observed at Decaturville were produced by high transient pressures associated with passage of shock waves. Reasonable es— timates of the shock pressures involved at Decaturville can be made by comparison with results obtained from ex- periments. Studies of granodiorite at the Hardhat nuclear explosion site showed that sets of “cleavage” fractures in quartz began to develop at pressures ofabout 50 kb (Short, I966). A similar pressure can be inferred from laboratory results of Horz (1968). The lowest pressure levels he described were about 50 kb, at which level cleavages developed but only sparsely as compared to those formed at higher pressures. At 52 and 54 kb, with the quartz oriented differently with respect to the shock-wave direction, Horz produced cleavages dominantly on rhombohedral planes, but also parallel and perpendicular to the c-axis. In a third orientation, and at 109 kb, basal cleavage became much more conspicuous. It is reasonable to assume that the lowest level of cleavage development (other than the simple rhombohedral fracturing in the Gunter) observed at Decaturville required on the order of 50 kb. The presence of microfaults associated with cleavages is a further indication that pressures on the order of40 kb were experienced by part of the Lamotte (Christie and others, 1964; Carter, 1968). Planar features and deformation lamellae have been shown to require considerably greater pressures. Carter (1968, p. 470) estimated that basal slip requires a stress difference at 76 kb at 300°C, and Horz (1968) found in SHOCK-RELATED DEFORMATION FEATURES 33 c{ooo1} mum's) "(10m r,z{1o1'1) s{1 12'1} mania} ciooon 1 I . 80 70 60 50 40 30 20 52(101-1} m{101_0} 40?” B “’ 7’ 5 ’f s 7’ 4191's} anon) l 3 J 2 1o f#_m%_‘ 0 so 60 7o 80 90 mum‘s} §{112'2) m{1o1'0} r,z(1o1'1} cloom} whom «(101' 2) §{11i2) ‘ ‘ r,z{1o1'1) so 10 2030 40 5060708090 FIGURE 23.—Histograms showing frequency and orientation of quartz planar fractures. A, Decaturville, 100 grains from Lamotte Sandstone and mixed breccia; B. Meteor Crater, Ariz. (Bunch, 1968); C, Middlesboro, Ky. (Bunch, I968); D, Sierra Madera, Tex. (Wilshire and others. 1972); E, Clearwater Lake. Quebec (Bunch, 1968); F, Shock experiment. 250-300 kb (Bunch, 1968); G, Tectonites ; (Carter and Friedman 1965). Vertical axis, frequency (percent); horizontal axis, angle between c axis and pole to planar element set (degrees). experiments on single crystals that pressure between 100 and We have no good evidence on temperatures associated with 120 kb was required to form planar features parallel to the shock event at Decaturville, except that no melt phase {10T3land {1001 l. The lower the temperature, the greater the has been found and complete thermal annealing of fission pressure required to produce the observed planar features. tracks in the basement rocks probably did not occur. 34 Annealing of tracks in apatite depends on the duration ofthe thermal event, but if we assume that raised temperatures prevailed for less than a year, basement temperature probably did not exceed 150°C (Naeser and Faul, 1969). Ahrens and Gregson (1964) calculated that in sandstone the temperatures associated with various pressures are 229°C at 25.4 kb, 336°C at 52.8 kb and 720°C at 104.5 kb. These data suggest that pressures in the basement rocks probably were less than 20 kb and that transient temperatures in the most highly shocked part of the Lamotte (presumably the upper part) probably were around 300°-350°C. The quartz con- taining planar features parallel to {0001} and {101—3} probably came from strata above the Lamotte and thus may have developed planar features at somewhat less than the 100 kb used in experiments. It seems reasonable to suppose that these features record pressures between 70 and 100 kb. The wide—spaced cleavage seen in the Gunter does not necessarily indicate shock pressures, but it is unusual for rocks deformed under normal tectonic conditions. Such large, open fractures may indicate that the Gunter, less confined than deeper rocks and with more open packing of quartz grains, dilated markedly after passage of the com- pressive shock wave. ASTERISM IN DOLOMITE Permanent damage to crystal lattices is shown by asterism—the scattering of X-ray spot reflections from a single grain when rotated in a Debye-Scherrer camera. The greater the crystal damage, the greater the tendency of spot reflections to merge into single lines such as are seen in X-rays of powders. Line broadening is an additional measure of damage intensity. X-ray examination of single dolomite crystals from shatter cones in the Derby and Doe Run near the center of the Decaturville structure has been reported by Simons and Dachille (1965) and by Dachille, Gigyl, and Simons (1968). One Decaturville sample showed a complete powder arc and extreme line broadening, comparable to those damages in shocked calcite from a nuclear-explosion site and to those in shocked quartz from Meteor Crater, Ariz. (Simons and Dachille, 1965). The intensity of asterism and line broadening were much greater than that seen in shatter-coned material from impact structures at Sudbury, Ontario, Sierra Madera, Tex., and Wells Creek, Tenn. Dachille and others ( 1968, fig. 7) showed that Decaturville samples have X-ray patterns similar to powders that were shocked at about 40 and 90 kb. They correspond in damage intensity to samples from Sierra Madera, Tex. and the Steinheim Basin, Germany. Shatter cones in dolomite may form at pressures as low as 15-20 kb and they are also found at Sierra Madera in dolomite and limestone shocked to levels around 100 kb (Wilshire and others, 1972). Thus, the range of40 to 90 kb suggested by the asterism study is consistent with what is known about the shock envelope in which shatter cones form in dolomite. GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI FIGURE 24.7Photomicrograph (120X) of muscovite grain surrounded by strained quartz in pegmatite, showmg differently oriented kink bands (indicated by arrows). KINK BANDS IN MICA Muscovite grains in the pegmatite display kink bands (fig. 24). As many as three directions of kink axes are seen in a single mica grain. Muscovite in the schist that encloses the pegmatite is almost entirely without kink bands, suggesting that the bands develop better in rocks that cannot yield readily by slip between grains. The kink bands are too few for the kind of statistical measurements described by Horz (1970) or Cummings (1968). They vary considerably in asymmetry and width and cannot be said with certainty to have formed by shock. If the kink bands did in fact form as a result of shock, they could indicate pressure as low as 9 kb (Horz, 1970). SHOCK-ENVELOPE CONFIGURATION Control points for reconstructing the envelope of shock pressure are: (1) Probable 20 kb at the outer limit of shatter cone formation; (2) no more than 10-15 kb in the basement ' rocks; (3) about 50 kb at the top of the Lamotte; (4) 70-100 kb for quartz grains believed to be derived from the Davis; AGE OF STRUCTURE AND DEPTH OF EROSION 35 METERS 800 a a l FEET ______ 2400 700 m m ' — 2200 ,- _ m 2000 500 _ Original ground 397%, ..\E...__,, Present ground surface Jefferson City Dolomite i 1800 500 i if, {T :Roubidoux‘ Dolom SR); : ”7777i” ,. - E 1500 7 T T E Ordovicianfi ~¥Gunter¥18andstone Gasconade Dolomite — 1400 400 7_ Mem3;%1:;7:§e\ill4w wfl‘ ' b” M e 1200 300 7 ........................... Emmehff‘i'f’m'te — e 1000 200 _ V39 Derby and Doe Run Dolomites Cambrian — 0 A; ’——— * O-_L_00 13—", {36’6/ / 6 Davis FormatioIL ¥ 600 *’// :wab Bonneterre Dolomite — 400 100 TZq “(7&7/ ' ”"A " "“ ‘ 2 ,/// 6 Lamotte Sandstone , 200 __.—4- SEA : “i i .E. g SEA LEVEL 1 10-15kb Precambrian rocks LEVEL FIGURE 25.~Cross section showing inferred shock—pressure envelope (dashed lines) in undisrupted strata. Data points shown in present positions and in inferred predisruption positions assuming about 300- -m inward move of beds as central peak formed. Arrows show change In position of data points after inward and upward move. Data points are: _(1) Basement rock not certainly shocked; (2) uppermost Lamotte Sandstone with basal cleavage; (3) mixed breccia quartz grains with {1013Heatures, probably derived from Davis Formation; (4) asterism suggestive of 90 kb; (5) outer limit of shatter cones; (6) Davis and Lamotte blocks upfaulted in sulfide pit (cleavage and planar features lacking); (7) downfaulted block of Eminence Dolomite near sulfide pit; (8) fault block of shattered Potosi Dolomite; (9) cleavage in Gunter Sandstone Member of the Gasconade Dolomite; (10) downfaulted block of Jefferson City Dolomite. Dotted line, inferred profile ofinitial (transient) crater. Vertical and horizontal scales equal. and (5) probable 90 kb for shatter-coned Derby and Doe Run at the center of the structure. Cleavage in the Gunter Sandstone Member and monolithologic breccias beyond that as far out as the ring fault suggest the passage of a decaying shock wave, with subsequent rarefaction, across the whole structure; but these features probably involved pressures of only a few kilobars. Figure 25 shows projected envelopes for various levels of pressure, based on inferred initial positions for the control points in the original stratigraphic pile before the inward and upward movements occurred to form the central uplift. The figure also shows the present locations of the control points and straight-line paths to represent rough vector sums of their postshock displacement. The amount of upward movement is clear; the amount of inward movement is an assumption based on the amount that seems to be required by the present Gunter ring configuration (pl. 1) and by the subsurface data obtained at the center of the structure. This problem is discussed in the section on origin. The stratigraphic level of the focus of energy release is difficult to establish. Maximum pressure levels recorded in any of the rocks preserved at the center are in the Davis and Derby and Doe Run Formations. (Shatter-coned Potosi is present there, but its degree of asterism has not been measured and no quartz fractures have been found.) We have no evidence to rule out the possibility of greater shock pressures in higher strata. If the recorded pressures described here were in fact the maximum obtained at the defined points in these strata, then the shapes of the inferred envelopes in fig. 25 suggest that the focus was in the Potosi or possibly in Derby and Doc Run. (Compare with depiction of shock-wave geometry in Gault and others, 1968, fig. 6.) It is argued in a later section that the Decaturville structure originated by impact. The shock-pressure configurationjust described requires the impacting body to have penetrated some 270-390 m of sedimentary strata. AGE OF STRUCTURE AND DEPTH OF EROSION The Bainbridge Limestone is involved in the deforma- tion, so the Decaturville structure is clearly younger than Middle Silurian. Shepard (1905, p. 116) identified an outcrop of Hannibal Shale and Sac Limestone of Early Mississippian age just north of the village of Decaturville, but we could not find this exposure in our mapping. The Precambrian schist exposed at the center of the structure contains apatite. Fission—track counts from the apatite grains were reported by Offield, Pohn, and Naeser (1970) as indicating the age of structure to be 210:20 m.y., or Triassic. Subsequent analysis by C. W. Naeser (written 36 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI commun., 1973), however, shows a fission-track count corresponding to an apparent age for the schist of I47i30 m.y. Unfortunately, the pertinence of this age—data to the time of formation of the structure is not clear. Precambrian apatite from the St. Francois Mountains of southeastern Missouri, which has not experienced heating other than through shallow burial and regional epeirogenic movement, gives an apparent fission-track age of213i21 m.y. Annealing of fission tracks that gives false apparent ages for Precam— brian rocks of the midcontinent region evidently does not require an unusual thermal event. Possibly heat associated with the formation of the Decaturville structure caused enough annealing of apatite fission tracks so that the apparent age is less than that of Precambrian apatite from the nearby undisturbed St. Francois area. If this is true, the heating event may have caused only partial annealing and moreover could have occurred at any time after the apatite formed, and the fission-track count would thus not indicate the age of the structure. It is likely, however, that the fission—track determination does indicate a maximum age of the structure, that is, Jurassic-Cretaceous time. An additional point bearing on the age of the structure is that the brecciation occurred after the area was mineralized. Zimmerman and Amstutz (1972) illustrated broken grains of microbanded sulfides as part ofthe fragments in the breccias near the center of the structure. The age of mineralization in central Missouri is not certain, but by comparison with stratigraphic dating of similar microbanded sulfide in western Missouri (A. V. Heyl, oral commun., 1974), it is at least post-Pennsylvanian and possibly as young as Cretaceous. From this argument, the structure most probably has a maximum age of Permian, but it may be younger than Cretaceous. When this is considered together with the probable maximum age of Jurassic-Cretaceous suggested by fission-track count, it seems likely that the Decaturville structure formed after Cretaceous time. A post-Pennsylvanian date for the structure, and the regional stratigraphic pattern, permit reasonable inferences to be made concerning the position of original ground surface when the structure formed. The geologic map of Missouri (McCracken and others, 1961) shows outliers of Mississippian strata in a pattern that suggests successive onlap of Kinderhookian, Osagean, and Meramecian Series units around a high area containing Decaturville within the broader Ozark dome. A Meramecian outlier is 26 km southwest of Decaturville, and Osagean outliers are 35 km east and west; these lie on the Jefferson City Dolomite. From the map pattern it cannot be said with certainty that the Decaturville area was overlain by Osagean or Meramecian units, although Schuchert (1943, p. 684) supposed that the Osagean Burlington Limestone completely covered the Ozark dome. If these beds were deposited there, they were apparently stripped away following the uplift and the considerable erosion in Late Mississippian and Early Penn- sylvanian time. Middle Pennsylvanian shale lies upon Jeffer- son City Dolomite, Roubidoux Dolomite, and in one place Gasconade Dolomite on the flanks of the local high around Decaturville; outliers of the shale are found 31 km east, 45 km north, and 56 km west of Decaturville. Pennsylvanian sediments probably covered the dome (Melton, 1931, p. 219), but their thickness can only be estimated. It seems likely that, at a maximum, the Des Moinesian section (about 120 m) would have been present. We believe it more probable that only a part of the Des Moinesian strata was deposited over the center of the locally high area within the Ozark dome. The Ozark area was probably elevated sometime in the Permian, though possibly earlier during warping in Penn- sylvanian time (Van Tuyl, 1918, p. 280). No Permian sediments are known to have been deposited in the region. No Pennsylvanian rocks are preserved in the Decaturville structure, even though they would have directly overlain Ordovician and Silurian limestones found in many places in the depressed zone. This strongly suggests that they were removed as the regional high was eroded before formation of the structure, which hypothesis meshes well with the argument for a Permian or later age of formation. If this hypothesis is accepted, then when the structure formed probably nothing was present above the stratigraphic level of the middle part of the Jefferson City Dolomite, except some patches of Ordovician and Silurian limestones, probably in topographic lows. As discussed in the section on structure, blocks of lower units believed to be ejecta sit upon the Jefferson City or limestone surface—a strong indication that the exposed surface was at that stratigraphic level when the structure formed; and this surface was downdropped and preserved in the outer part of the structure. It follows that the area of flat-lying strata outside the ring fault has been in general eroded just enough for removal of most or all of the Jefferson City. A value typical for erosional lowering of the general plateau level in the immediate vicinity of the structure since the structure was formed is about 45 In. More erosion may have occurred to plane off the central uplift and the crater rim depending on how much those features protruded above the level of the plateau. These considerations suggest that except for removal of the upraised crater rim and central peak, the structure is preserved nearly in its original form rather than being the deeply eroded root of a structure that might have looked very different at a higher pre-erosion level. This point bears importantly on any explanation of origin and on any reconstruction of original appearance and subsequent alterations of the structure. Moreover, the small amount of erosion believed to have taken place after the structure formed suggests that the structure is young—quite possibly Cretaceous or younger. ORIGIN OF THE STRUCTURE 37 ORIGIN OF THE STRUCTURE The Decaturville structure involves such unusual and intensive deformation in a small, sharply bounded, circular area that an explosion-like origin by sudden highly localized release of tremendous energy seems certain. No evidence exists to permit serious consideration of the intrusion of igneous material or a diapiric mass of sedimentary material as possible origins of the structure. No igneous material is involved, and diapirism is ruled out by a stratigraphic configuration that could not permit flowage of strata under the impetus of density inversion. Two hypotheses involving explosion-like origin of the structure are in contention: endogenetic volcanic gas or phreatic explosion) and ex- ogenetic (meteorite or comet impact). An excellent summary of these arguments for cryptoexplosion structures in general has been given by French (1968), and other statements of the two viewpoints have been made by Bucher (1963) and Dietz (1963). Any explanation for the Decaturville structure must account for its major features: I. It is circular and sharply fault bounded, containing a central uplift and a surrounding structural depres- sion. 2. Deformation is shallow, virtually within a 540-m column of sedimentary strata atop a little-disrupted basement. 3. Shock features are apparent and intensity of shock deformation diminishes downward. 4. No evidence of volcanic processes has been found. 5. The central uplift involved large-scale inward and upward movements of strata by thrusting and folding. 6. Blocks of strata (basement to Potosi Dolomite) are mixed, a hundred meters out of place, in a central breccia mass below a capping layer of Derby and Doe Run Dolomites. 7. Exotic blocks of basement rock, Derby and Doe Run, and Gunter sit atop the youngest rocks in the section within the ring depression. We believe that the type and intensity of deformation at Decaturville cannot be explained by a terrestrial volcanic event, but that they are what would be expected from a meteorite or comet impact. The two hypotheses are discuss- ed in detail below. ENDOGENETIC ORIGIN In the absence of any sign of volcanism at Decaturville, the argument for endogenetic origin rests almost entirely on regional structural setting. Snyder and Gerdemann (1965) noted that Decaturville was one of eight features related to volcanism or of unusual structural character that lie nearly in a line along the 38th parallel from Illinois to Kansas (fig. 26). They considered this lineation to indicate a regional extension of the Rough Creek-Shawneetown - Palmer fault zone crossing western Kentucky, southern Illinois, and eastern Missouri (fig. 26). Snyder (1970) later extended the zone from Virginia to central Kansas, including in it eight mineralized areas, ten igneous intrusions, and eight “explo- sion” structures (five of which were among the features described in the 1965 report). He also added a similar lineament to the south, broad enough that features overlapped with features of the 38th parallel lineament. The two “lineaments” merge to form a zone 320 km wide from the Carolinas to Kansas and Oklahoma, which contains 44 areas or features of several different types and ages (fig. 27, modified from Snyder, 1970). A zone near the 38th parallel may be a valid if somewhat nebulous structural line. McCracken (1971) suggested that the line is marked by the Ste. Genevieve fault system, the Palmer fault, and some small east-west faults near Rolla (fig. 26). She cited changes in fold trends and fold asymmetry in Paleozoic and possibly Precambrian rocks and in the distribution of pre-St. Peter rocks, all of which apparently reflect the influence of a 38th-parallel structural line. Also, along some small segments of the line are indications of breaks in regional magnetic trends. Heyl (1967) suggested that a right-lateral wrench fault underlies the line. Of the eight features originally used by Snyder and Gerdemann (1965) to define the lineament, only Crooked Creek and Decaturville are “cryptoexplosion” structures with no evidence of volcanism. Hicks dome appears certainly to have been produced by violent expansion of volcanic gases beneath a stratigraphic cover a thousand meters thick. The Avon diatremes, scattered through an area of 260 km2 rather than on a neat structural line, represent another manifestation of explosive (mafic) volcanism at the surface. At Furnace Creek and Hazel Green there is no doubt about the presence of volcanic materials. The Rose dome area contains three domes (spanning some 32 km transverse to the general east-west line) associated with mafic volcanic rocks. The Weaubleau area is like none of the others; it consists of an area 11 by 5 km intensely broken by thrusts and normal faults, and of adjacent areas of 44 km2 containing unusual breccias. Ages of those features that were dated range from Cambrian to Cretaceous or early Tertiary. Two of the five clearly volcanic features involved ex- plosive processes that produced mixed breccias rather like those at Decaturville and Crooked Creek. The volcanic explosion breccias and the breccias at Weaubleau do not contain quartz that displays features indicative of shock such as have been described in this report and such as have been found at Crooked Creek (French, 1968). Only the structures at Decaturville and Crooked Creek display shatter cones. Superficially, Hicks dome may seem like the two cryp- toexplosion structures, but in detail it is very different. It is a simple elongate domal uplift some 16-19 km across, without a surrounding structural depression and apparently with only a few small extensional faults produced by the uplift process. The Decaturville and Crooked Creek structures 38 I GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI 0 95° 93° 91. 89° 39 I I i \x" ‘ . 7 I .JEFFERSON /) CITY ILLINOIS KANSAS I O I 38" v OROSE DOME WEAUBLEAUO \: oRol'a o I DECATURVILLE HAZEL CROOKED I GREEN CREEK \ M 1 s s o u R r : .Springfield 37° —— —~ I I OKLAHOMA i—“ n —_ __ —— l ‘ A R K A N s A S 36k \ I ‘ i i 200 KILOMETERS T . 100 MILES FIGURE 26.—Map showing areas of volcanism and unusual structural disturbance in apparent alinement near the 38th parallel. (Modified from Snyder and Gerdemann, I965, p. 467.) RC, Rough Creek fault zone; SG, Ste. Genevieve fault; P, Palmer fault; C,Cuba fault; RA. Red Arrow fault. have ring depressions and much more complex deformation involving inward movement and compressional tectonics. Roddy (1968a, p. 307-309) presented a concise discussion of the differences between volcanic-explosion structures and impact structures, focusing particularly on the impossibility in volcanic settings of shallow buildup or containment of the necessary gas volumes and pressures and the lack of horizontal stress to produce the observed folding and thrusting. Snyder and Gerdemann (1965) pointed out that several of the structures along the 38th parallel lie at or near intersections of local structural lines. For example, Hicks dome lies near the intersection of the east-west Rough Creek fault zone and a local anticline. The Furnace Creek volcanic center occurs near the intersection of a local fault and the east-west Palmer fault. The Crooked Creek structure lies at the west end of the Palmer fault where it would be intersected by the Cuba fault if that fault were extended 12.8 km southward (although no evidence for such an extension is known). Decaturville is near the intersection of the east-west lineament with the projected line of the Red Arrow fault, along the west flank of the Proctor anticline. The Red Arrow fault cannot be seen at the surface, but the magnetic map pattern suggests that at basement level it may be tangent to the ring fault on the northeast, some 3 km from the center of the structure. No specific structural loci along the east-west line were suggested by Snyder and Gerdemann (1965) for the Avon, Hazel Green, Weaubleau, or Rose dome features, and so the argument for local tectonic control for the “alined” structures is rather incomplete. The presence of sulfide minerals, generally in minor amounts, also has been considered by various workers (Snyder and Gerdemann, 1965; Kiilsgaard and others, 1963; Krishnaswamy and Amstutz, 1960) as evidence of en- dogenetic origin of the host structures. Mineralization at Hicks dome and Crooked Creek, however, is poststructure in age and merely shows the preference of mineralizing solutions for fractured rock. At Decaturville the sulfide minerals are older than the structure; no great coincidence is called for to have a structure form (by impact or other means) in mineralized ground because sulfide mineral localities are dispersed abundantly in central Missouri (Heyl, 1968). The argument for endogenetic origin of the Decaturville structure (and its near-twin Crooked Creek) thus deals strictly with a series of geologic associations. Existence of a discrete structural line along the 38th parallel is made doubtful by the fact that the features said to define it are of five or possibly six significantly different types, ages, and origins. The fact that five of eight features are Of volcanic association cannot be shown to bear directly on the origin of the remaining three. Moreover, the 38th parallel is within a zone of many different features (fig. 27) through which lines in any direction desired could be drawn. Structural loci for the features are lacking, or depend on projections of faults, or are in fact not at but only near the features. In any event, ORIGIN OF THE STRUCTURE 39 100° 96" 92° 88° 84° 80° I INDIANA ILLINOIS MISSOURI NORTH CAROLINA 35° _ OKLAHOMA ARKANSAS A SOUTH CAROLINA I L I MISSISSIPPI I ALABAMA GEOREA \ 0 500 KILOMEIERS ‘ , o 300 MILES 100° 96° 92° 88° 84° 80° MISSOURI 36° TENNESSEE SOUTH CAROLINA MISSISSIPPI GEORGIA 32° ~ I I I 500 KILOMETERS I 0 I I I fir I 0 300 MILES FIGURE 27.—Maps modified from Snyder (1970) showing mineralized areas, igneous intrusions, and “explosion structures" in what are construed to be two separate zones trending east—west. A, lineament near the 38th parallel, with eight mineralization areas (irregular heavy outlines) numbered Ml-M8, ten igneous intrusions (crosses) numbered Dl-DIO, and eight “explosion structures” (small circles) numbered El-E8. B, Tennessee lineament with ten mineralization areas (irregular heavy outlines) numbered Ml-MIO, three igneous intrusions (crosses) numbered Dl-D3, and five “explosion structures” (small Open circles) numbered El-E5. location of random impacts on regional structures is not explained by a hypothesis of origin for the Decaturville unknown. Meteor Crater, Ariz., is on a local fold within a major area of explosive volcanic vents; Gosses Bluff, Australia (with impact melt and abundant shock features), lies on a regional anticline (Milton and others, 1972); Sierra Madera, Tex. (with melt and shock features), is along the projected trend of structure in the nearby Glass Mountains (Wilshire and Others, 1972) and near the speculated Texas lineament; Flynn Creek, Tenn., lies along the extension of a major fault zone (D. J. Roddy, written commun., 1974). Referring again to the seven Observations which must be structure (see the section “Origin of the structure”),an endogenetic explosion might produce the breccia core and possibly even the presumed ejecta blocks if venting occurred. It is not consistent with the other five listed observations. EXOGENETIC ORIGIN Three features of the structure in particular are strong evidence of impact origin: shallow deformation, shock indicators (with downward lessening of shock intensity), and 40 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI centripetal movement of strata to form the uplift. Drillhole information obtained at the center and the minor amount of basement rock in the breccia indicate that most of the disturbance was confined to the cover sequence of sedimentary rocks, probably little more than 550 m thick. At Crooked Creek, where the sedimentary cover probably was about the same as at Decaturville, the disturbance is not known to have affected basement rock (Snyder and Ger- demann, 1965). This evidently shallow character of deformation is conso- nant with the shock-envelope reconstruction presented in fig. 25. From the shock criteria discussed earlier, we infer that a release of energy sufficient to produce pressures of 60-100 kb at the stratigraphic level of the Derby and Doe Run or Davis occurred at that level or higher. If it was higher, then still greater pressures may have been exerted on materials not now preserved in the structure core. The envelope is reasonably established from the Observed shock features and it indicates that pressures markedly diminished downward, probably to less than 20 kb at the basement surface. Thus, disturbance by endogenetic means requires that gas migrated upward from the basement until it suddenly reached a point within probably 370 m of the surface where it was released instantaneously to generate tremendous pressures. One perspective on such an event is provided by the calculation of 3-kb pressure for the 1956 eruption of the volcano Bezymianny in the Soviet Union, which ejected blocks to a distance of 30 km (Gorshkov, 1959). Explosions produced by gas expansion or instan- taneous conversion of water to steam cannot much exceed that level of pressure. Moreover, it is difficult to understand how pressures from below reaching 60-100 kb could have been contained until the gas reached 370 m from the surface. Lithostatic pressure at that point would have been only about 100 bars; explosive release obviously would have had to occur very much deeper. This point is considered quantitatively by Roddy (1968a, p. 307). The structural geometry, with beds crowded inward and upward to form a central uplift and concomitant develop- ment of a surrounding ring depression, is not like any terrestrial structures except diapirs. However, origin of the Decaturville structure by diapirism is ruled out by the lack of low-density material, too thin a sedimentary sequence, the presence of the breccia core, the shock features, and the blocks best explained as ejecta. Upbulging of strata by a subterranean explosive volcanic-gas expansion, as at Hicks dome, produces dilation that is represented generally by simple upward or downward movements on steeply dipping faults. This is vastly different from the pattern observed at Decaturville and most of the “cryptoexplosion” structures now believed to be impact structures, where horizontal stress components were dominant (Roddy, 1968a, p. 309). The structural pattern resulting from centripetal displacements provides a key argument in favor of impact origin (Wilshire and Howard, 1968) that directly applies to the Decaturville structure where inward thrusting and telescoping of beds are such conspicuous phases of deformation. Uplifts surrounded by depressions, and with structural patterns indicating centripetal movements, were produced in high-explosive cratering tests in Canada (Diehl and Jones, 1967; Roddy and others, 1969). Tracing of marker canisters buried in the cratered ground clearly showed ground movements inward and upward, with folding and faulting of strata, to make the uplift. Some canisters ejected from the crater came from below the floor of the apparent crater. Conspicuous circumferential and radial cracks developed in the ring depression and outside the crater. Roddy (1968b; Roddy and others, 1969) noted that this experiment resulted in forms most directly analogous to those mapped in terrestrial structures when the explosive charges were at the ground surface rather than buried. He suggested, on the basis of this comparison and of calculations of energy relationships, that structures having central uplifts are caused by the impact of low-density bodies, such as comets, that do not deeply penetrate the ground owing to shock vaporization at or near the ground surface. As Wilshire and others (1972) have commented, central uplifts are found in known or suspected terrestrial impact structures at least 2.4—3.2 km wide (at least in sedimentary rocks); on the Moon, where gravity is one~sixth that on Earth, central peaks occur in craters wider than 19 km. Terrestrial structures with central uplifts and ring depressions in layered materials commonly show close similarities in the ratio of maximum stratigraphic displace- ment in the uplift to outer diameter of the depression or final crater form. The main inner depression is used for the calculation in cases where multiple ring structures are present. Values close to 1:10 for uplift: ring diameter ratio are calculated for Sierra Madera, Tex. (Wilshire and others, 1972); Gosses Bluff, Australia (Milton and others, 1972); Flynn Creek, Tenn. (Roddy, 1968a); Wells Creek, Tenn. (Wilson and Stearns, 1968); and Decaturville. Recent work at Steinheim, Germany (D. .I. Roddy, written commun., 1974) indicated a 1:10 relationship for that structure also. Canisters buried 9 m deep were ejected during excavation of a lOO-m crater in one Canadian explosion-cratering experi- ment (D. J. Roddy, oral commun., 1970). Other similar structures, such as Jeptha Knob, Ky.; Serpent Mount, Ohio (Bucher, 1936); and Crooked Creek, Mo. (Hendriks, 1954), have ratios of 1:15 to 1218 as based on the available literature. These structures typically are not well exposed, however, and it is possible that isolated exotic fragments from deep formations were present but not mapped and so the uplift estimate is smaller than the true value. It is also possible that a ratio of 1:10 cannot be attained if the ring depression is not sharply bounded and completely developed, as seems to be true for Crooked Creek and Jeptha Knob. Two remaining features that impact origin of Decaturville must explain are (l) the breccia core capped by a broken and ORIGIN OF THE STRUCTURE 41 crushed but apparently continuous stratum and (2) probable ejecta blocks in the ring depression. A subterranean explo- sion could explain the breccia core but (aside from previous- ly stated arguments against it) does not seem likely to have ejected material through the 370 m of rock overlying the capping bed of Derby and Doe Run. One possible explana- tion may come from high-explosive cratering experiments (D. J. Roddy, oral commun., 1973). Charge detonation at or above the ground surface has been observed to produce uplifts in which cores of jumbled material are developed below uplifted and folded strata which are locally con- tinuous but which are broken by complex faulting. Again it is difficult to explain ejection of material from the base ofthe breccia column, although as cited above, subcrater material has been found in ejecta. The presence of such ejecta implies that either transient openings exist below what becomes the final crater floor or material is somehow drawn up and ejected with great force from below the deepest crater surface. Transient openings seem the more likely explana— tion; this interpretation bears importantly on our attempt to determine the dimensions of the original crater at Decatur- ville. INTERPRETED CRATER PROFILE AND POST- CRATER DISRUPTION Two models of original (transient) impact crater size and shape are considered basically to fit most of the field observations as well as what has been seen in cratering experiments. One model calls for very shallow or no penetration of a low-density impacting body. From crater- ing experiments, this may be expected to result in excavation of a shallow crater with central uplift, possibly containing a disrupted and brecciated core. In this model the observed breccia core has a cap of Derby and Doc Run because these formations would not have been excavated, but merely uplifted. Lateral movement might have occurred during formation of the uplift but likely would have been small as compared to the vertical displacements involved. The crater would have bottomed no deeper than the top of the Derby and Doe Run and thus would have been less than 330 m deep. Brecciation below the Derby and Doe Run would represent the effect of bulking as the uplift formed during relaxation after passage of the shock wave. This explanation accounts well for the observation that except for small amounts of Potosi, all fragments in the breccia core come from Derby and Doe Run and older formations; it also could explain the orientation of large blocks pointing inward and upward. It does not seem to account for basement ejecta, unless jetting of deep material occurred along transient fractures that extended some 180 m below the crater floor. The model also does not appear to accord with the inferred scale of lateral movement of strata, or to explain the inferred shock-envelope configuration. The second model, our preferred model of crater shape, is highly inferential, but it takes account of every pertinent observation. We have no direct evidence on depth of penetration, but from the configuration of probable shock envelopes (fig. 25), the burst point seems likely to have been at about the level of the Potosi, too deep for a low-density impacting body. Thus the cratering mechanics probably were substantially different from those of the shallow-burst cratering experiments. Moreover, it seems clear that the basement must have been adjacent to a free surface, at least transiently. With a probable burst point as deep as Potosi, excavation to the basement at the very center of the crater is not unlikely, and we infer that the instantaneous excavation (or transient cavity) bottomed at or near the basement surface, as shown in fig. 25. This means that the Derby and Doe Run, which now lie completely over the central area, would necessarily have been excavated. Inward movement of about 300 m from all sides is indicated by the structural pattern at the Gunter level; by general comparison with inferences from shatter-cone orientations at Gosses Bluff (Milton and others, 1972), inward movement of deeper strata may have been even greater. We thus infer that the Derby and Doe Run moved inward a minimum of 300 m and closed over the bottom one-third of the crater, now filled with blocks that spalled from the crater walls during the shock unloading phase, and as the strata moved inward. The movement of the Derby and Doe Run and subjacent beds involved bed- ding-plane slip, block displacements, and considerable flowage (as shown by thinning of the Derby and Doc Run and Davis at the sulfide pit and flow deformation in the Lamotte as well as the mixed breccia). Formations below the Derby and Doe Run and the Derby and Doe Run themselves make up most of the breccia core. The Derby and Doc Run probably slid more readily and farther than lower beds because it was moving atop the well-lubricated limy shale beds of the Davis. The blocks overridden by the Derby and Doe Run now form the breccia core, changed in shape from a filling of the lower part of the crater to a tall column by the crushing inward move of the crater walls. This explanation accounts for the observed tendency of long blocks in the megabreccia core to stand nearly on end, pointing inward and upward in an acicular pattern. The inferred 300 m of movement permits us to establish the position of the crater wall at the level of the Derby and Doe Run prior to the inward move (near point 4 in fig. 25). Above that level the crater probably widened sharply as shown in fig. 25. Small inner craters that extend below the main shallower craters are seen at least partly developed in Odessa Crater, Tex. (Shoemaker and Eggleton, 1961) and in a crater produced experimentally in granite (Hb'rz, 1969). Our conjectured initial, transient crater is a shallow bowl with a deeper excavation at its center. The inward move of Derby and Doe Run across the deep inner crater accounts for the absence of blocks from strata above the Derby and Doe Run (except for a minor amount 42 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI of Potosi which probably fell in during the move). Higher strata would have been excavated over the wider area of the main crater bowl and would be present in the central column only as minor pieces of fallback material. Even fallback pieces of upper strata would not be expected to be present if the Derby and Doe Run beds had closed the lower part of the crater before vertically ejected material could fall back. That this is feasible is indicated by the fact that at Gosses Bluff, blocks had risen more than 1,830 m to stand in open space and then toppled or were thrown outward apparently before airborne ejecta landed in the area (Milton and others, 1972). The time required for such dramatic movement in the uplift obviously must be measured in seconds. The most highly shocked material would have been ejected, but some would have been mixed with rubble on the crater floor. This material is now found only in the mixed breccia, tremen- dously diluted by the addition of grains and fragments produced by the grinding together of the large blocks spalled from the crater walls. The inferred profile of the main crater bowl is based primarily on three points of information. A downdropped block of Eminence at point 7 in figure 25 indicates that Eminence was not entirely excavated there and thus must have been below the crater floor. A similar argument can be made to explain the presence of a Jefferson City block at point 10, and this occurrence puts a probable limit on the diameter of the crater at the original ground surface. Additional surmises are that cleavage in the Gunter (point 9 in fig. 25) indicates proximity to a free surface—probably the crater wall—and that the Gunter tectonic style suggests involvement in the crater rim and overturned flap (such as described at Meteor Crater by Shoemaker, 1960). For example, the only reasonable explanation for the presence of the enormous Gunter block at the village of Decaturville, atop rocks that would have been at ground surface, is that the block slid into place as a projection of the overturned flap. The inferred crater shown in figure 25 has a maximum depth of about 550 m and a rim diameter of about 3,000-3,300 m. This crater, produced by nearly instan- taneous excavation, was modified in a matter of seconds by the inward and upward move of strata in the surrounding ground. The physical impetus for such movement is not completely understood, but the movement can be considered roughly as a rebound that followed intense compression by the shock wave. D. J. Roddy (written commun., 1974) indicated that recent large-scale cratering experiments and computer simulations of shock-wave cratering show inward crater-floor movements to be common in surface—burst events. Figure 28 shows a highly schematic time sequence of inferred stages in the development of the structure from transient cavity through postcrater disruption. Instantly after or even during crater excavation, the inward mass displacement involved in the initiation of the central uplift produced the ring fault. The position of such a ring fault in impact structures may depend on the depth to which strata are disturbed. In general, movement of strata downward and inward along the boundary fault produces a ring depression and provides further impetus for the rise of a central peak, and this push-pull mechanism seems to have acted effectively at Decaturville. The inferred linkage of uplift and ring fault may explain the difference between Decaturville and Crook- ed Creek. Uplift must be supported by flow and slippage of strata inward and upward. This movement may not be able to develop fully if the ring depression is incomplete. It seems possible that a configuration like that at Crooked Creek (a central horst with a dome collapsed around it) would occur if initial uplift were unsupported and partial collapse then ensued because a ring fault did not form to provide additional impetus for inward movement of material. The schematic cross sections of figure 28 illustrate an inferred sequence of stages in the development of the Decaturville structure, a sequence which probably occupied a few minutes at most. Present ground surface and geologic features at that surface along or near a northeast-southwest section are portrayed as observed; features shown above and below the ground surface are intended only to convey a general sense of the hypothetical deformation sequence. Section A shows the initial crater or transient cavity, with a rim formed largely by the outward-overturning of a flap of strata from the top part of the excavated sequence. The overturned flap would be much more broken and jumbled than portrayed and would undoubtedly have extended farther outward as a thin discontinuous cover of loose blocks. Ejecta blocks of Precambrian rock and the Derby and Doe Run Dolomites are shown still falling, to represent the two blocks actually present in the northeast part of the structure. The clean, sharp crater represented presumably could not really have existed even momentarily because of fallback and rubble not ejected, and because spalling and centripetal movement of beds would have been taking place even before excavation was complete. Section B shows the beginning of structural adjustments following the crater-excavation phase and passage of the shock wave. A large block of Gunter Sandstone Member of Gasconade Dolomite has slid outward from the crater rim into place at the village of Decaturville. Inward and upward movement of beds closes the inner crater, lifts the rim zone, and modifies the crater profile. The ring fault forms in response to inward movement at the center, and folds and small faults form in the moving beds. Closure of the inner crater involves the crowding and crushing of blocks of all sizes which spalled from various formations in the in- ward-moving crater walls. Sections C and D show the changing crater and structural adjustments as centripetal movement of beds continues. Crushing at the center produces a cylindrical mass of breccia which increases in height because of addition of new material and squeezing from all sides by the inward-moving strata. IMPLICATIONS FOR OTHER ASTROBLEMES 43 The breccia column is overridden by the Derby and Doe Run and Potosi. Lateral movement of beds involved many small thrust faults; these are portrayed in the upper beds where they exist along the present surface profile. Deformation style in the deeper beds probably involved bedding-plane slippage and thickness changes, but many thrusts and normal faults probably occurred also. In section C, steep faults are shown forming along the flanks of an upward bulge which is part of a fold ring surrounding the center of the structure. This faulting raised the large block seen at the sulfide pit; much of the upward impetus, and support for the block once it was raised, was provided by inflow of Lamotte Sandstone. Upfaulting in this area is completed in section D to give the presently observed configuration of blocks. Mixed breccia is shown as being piped or injected from the central column into fault channels in the sulfide-pit area. The centripetal movement of beds in the central peak area is clear and apparently was an early phase in a very complex structural sequence. Beds in the ring depression on the southwest side also moved inward, largely before steep faults and outward thrusts formed. Elsewhere on the west side of the structure in the ring depression, outward thrusting seems to have dominated the other phases of movement. We believe that the outward movement primarily was gravity sliding of beds from the central peak. Some sliding may have occurred as the peak was being lifted, but most probably occurred after most of the peak was formed. In other sectors of the perimeter of the peak, late outward sliding of layers may have been precluded by fault—block topography. Block faulting was the dominant structural phase on the north and south sides and outward thrusts have not been identified in those areas. The push-pull mechanism of ring fault-central uplift has been discussed for lunar craters, where the slump of large masses on crater walls is believed to have contributed energy and mass to the formation of central peaks (Dence, 1968). The final configuration at Decaturville may have been much like that of lunar craters, with blocks unevenly downfaulted inside the ring boundary and a central peak that, as it formed, nearly instantly destroyed the initial crater outlined in figure 25. No direct evidence concerning the impacting body or its direction and angle of impact was obtained in our study. Most of the body presumably would have vaporized at levels above the rocks now preserved at the center. Its azimuth, however, may be indicated by the northeast—southwest axis of rough symmetry in the inner part of the central uplift. This direction is marked by ejected material in the northeast part of the structure and elongation of the gravity map contours. The outcrop and the gravity patterns may merely indicate greater inward movement from the northwest and southeast, possibly influenced by such factors as regional joints. We suspect, however, that the oval shape may relate to the original pattern of crater development. Study of the Gunter ring, for example, shows that the sandstone is dominantly very steep to overturned on the northeast and dips gently on the southwest. This asymmetry in tilting of strata and the distribution of ejecta suggest the possibility of a low-angle impact from the southwest, if small missile-impact structures are valid analogs (Moore, 1969). Overturning of the Gunter on the northwest is consonant with our beliefthat it was part of the overturned flap around the crater rim and helps further to understand the presence of the exotic block at the village of Decaturville. IMPLICATIONS FOR OTHER ASTROBLEMES No other cratering events so far studied in layered rocks are reported to have resulted in beds moving inward over a preserved core of breccia to form a central uplift (although subsurface data are seldom available to reveal such a core). Other facets of our proposed reconstruction, however, may have more general application to other astroblemes and may help to explain some puzzling aspects of published studies. Decaturville had always been considered to be deeply eroded, not on the basis of field evidence but mostly because the presumed crater topography seemed to have been removed. There is, however, no reason to suppose deep erosion. Paleogeographic evidence neither requires nor provides support for there having been more strata atop the presently existing section at the time of impact; certainly no higher strata are now found in the structure. Moreover, the huge block of Gunter seems possible to emplace (atop the highest preserved beds in the section) only as part of the flap of strata overturned outward around the crater to lie in scattered blocks upon the ground surface. An exotic block of Derby and Doe Run and fragments from the basement lie atop the whole section like the Gunter block, but they are from sufficient depth that they must be ejecta. The basement fragments seem to be caught in the ring fault. We have estimated about 45 m of post-structure erosion. Thus the Decaturville structure immediately after its forma- tion would have shown a ring zone dropped as much as 120 m below the surrounding plateau surface. If our reconstructed initial crater profile is approximately correct, the central peak probably was at most 30-60 m higher than today. The raised rim of the destroyed crater stood on the flanks of the central peak as an easily eroded pile of broken rock perhaps as much as 180 m above the plateau level. Erosion at Sierra Madera was conjectured by Wilshire and others (1972) to have amounted to a few thousand feet, and the present surface was considered to show the style of deformation at a deep level in the impact structure. Shatter-coned Permian rocks from, the central peak area are found along ring faults in Cretaceous strata at the present topographic rim, 3.2 km outside the shatter-cone envelope. Wilshire and others (1972) conjectured that such blocks were driven outward and upward along curved faults. We suggest instead that, as with the exotic blocks near the village of GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI SOUTHWEST PRESENT GROUND Sl 4 RING FAULT ‘ W§>m \\\\\\_ A \ / ‘\ \\\\“‘$“ [A \ \, °‘ BLOCK AT /“‘\$¢ ' \ ca SULFIDE PIT \\\\\\“ A.“ !«{{§‘\“;\“€3§i _ 3&41’7 ,“ “ FIGURE 28.—Schematic northeast-southwest cross section showing inferred time sequence of IMPLICATIONS FOR OTHER ASTROBLEMES 45 NORTH EAST DERBY AND DOE PRECAMBRIAN EXPLANAT'ON s4 RUN EJECTA EJECTA ______ 01c -, - Silurian and Ordovician limestones ___—_——r as“ ORDOVICIAN Ojc Jefferson City Dolomite Roubidoux Dolomite Upper part, Gasconade Dolomite Gunter Sandstone Member, Gasconade Dolomite CAMBRIAN Eminence Dolomite DERBY AND DOE GUNTER EXOTIC PRECAMBRIAN Cp Potosi Dolomite EJECTA Derby and Doe Run Dolomites Cd Davis Formation Bonneterre Dolomite Lamotte Sandstone Precambrian granite and schist Mixed breccia of Silurian, Ordovician, Cambrian, and Precambrian rocks CONTACT _L '$_ FAULT—Arrows show direction of relative movement FEET METERS 0 —— 0 3000 — — 1000 0 1000 MEI'ERS l . l . l T . fl 0 3000 FEET NO VERTICAL EXAGGEHATION structural movements during cratering event at Decaturville structure; section A is earliest. 46 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI Decaturville, these blocks are ejecta trapped in faults at or not far below original ground surface as jostling occurred in the ring-boundary zone. This explanation seems to fit well with the presence of highly shocked or even melted mixed breccia material filling cracks and blanketing hillsides in the central uplift. If the present surface of the central uplift is little different from the original surface after the structure formed, then the mixed breccia may well represent material deposited on the crater floor. The highly granulated and partly fluidized breccia would naturally have drained or flowed into cracks between blocks in the rising central peak. Scaling from Decaturville, the transient cavity at Sierra Madera would have been about 7,300 m across; the rim would have been just outside the outermost mapped occurrences of mixed breccia. At Gosses Bluff, the presence of impact melt at the present surface suggests that little post-crater erosion can have taken place; Milton and others(1972) speculate that Mt. Pyroclast, 5 km from the center of the structure, may represent a mass of suevite-like material which was plastered against the crater wall. Other evidence suggests that the central peak formed as ejected material was still falling or moving laterally along the surface, as we now surmise for Decatur- ville and Sierra Madera, and which was observed at the Snowball experimental cratering event (D. J. Roddy, written commun., 1976). It is interesting to note that, if Mt. Pyroclast marks the position of the transient cavity wall, then the cavity diameter was approximately half the 20-22 km diameter of the final disturbed area, essentially as inferred for the ratio of transient cavity to final ring diameter at Decaturville, The implications for the structure of lunar craters have been discussed by Wilshire and others (1972) and Milton and others (1972). They make the reasonable analogy that, as in terrestrial impact structures, peaks in lunar craters consist of rocks brought up from deep below the crater floor. We would only add the thought that the ratio of upward displacement to final ring diameter probably varies con- siderably depending on layering of the excavated rocks and their ability to flow or slip. In lunar terrain of massive basalt layers, that ratio might be much less than 1:10. In terrain where fragmental material is extremely deep and could move by granular flow, the ratio might be significantly larger than 1:10. REFERENCES CITED Ahrens, T. J., and Gregson, V. J., Jr., 1964, Shock compression of crustal rocks—Data for quartz, calcite, and plagioclase rocks: Jour. Geophys. Research, v. 69, no. 22, p. 4839-4874. Amstutz, G. C1, 1959, Polygonal and ring tectonic patterns in the Precambrian and Paleozoic of Missouri, U.S.A.: Eclogae Geol. Helvetiae, v. 52, no. 2, p. 904-913. I964a, Impact, cryptoexplosion, or diapiric movements? (A discus— sion of the origin of polygonal fault patterns in the Precambrian and overlying rocks in Missouri and elsewhere): Kansas Acad. Sci. Trans, v. 67, no. 2, p. 343-356. l964b, Tectonic and petrographic observations on polygonal structures in Missouri, in Geological problems on lunar research: New York Acad. Sci. Annals, v. 123, art. 2, p. 876-894. Amstutz, G. C., and Zimmermann, R. A., 1966, Decaturville sulfide breccia of south—central Missouri—A fossil mud volcano, in Abstracts for 1965: Geol. Soc. America Spec. Paper 87, p. 4. Ball, J. R., 1939, Type section of Bainbridge formation of southeastern Missouri: Am. Assoc. Petroleum Geologists Bull., v. 23, no. 4, p. 595-601. Ball, S. H., and Smith, A. F., 1903, The geology ofMiller County: Missouri Bur. Geology and Mines, 2d ser., v. 1, 207 p. Beals, C. S., Innes, J. M. S., and Rottenberg, J. A., 1963, Fossil meteor craters, in Middlehurst, B. M., and Kuiper, G. P., eds., The moon, meteorites, and comets, v. 4, ofThe solar system: Univ. Chicago Press, p. 235—284. Boon, J. D., and Albritton, C. C., Jr., 1936, Meteorite craters and their possible relationship to “cryptovolcanic structures": Field and Laboratory, v. 5, no. 1, p. 1-9. Branco, W., and Fraas, E., 1905, Das Kryptovulkanische Becken von Steinheim: Akad. Wiss. Berlin, Phys.-Math. K1 Abh., v. I, p. 1-64. Bray, J. G.. and others, I966,_Shatter cones at Sudbury: Jour. Geology, v. 74, no. 2, p. 243-245. Bucher, W. H., 1936, Cryptovolcanic structures in the United States [with discussion]: 16th Internat. Geol. Cong., Washington, DC, 1933, Rept., v. 2, p. 1055-1084. 1963, Cryptoexplosion structures caused from without or from within the Earth? (“astroblemes” or “geoblemes”?): Am. Jour. Sci., v. 261, no. 7, p. 597-649. Buehler, H. A., and McQueen, H. S., 1933, Guidebook, Kansas Geol. Soc., 7th Ann. Field Conf., p. 19. Bunch, T. E., 1968, Some characteristics of selected minerals from craters, in French, B. M., and Short, N. M., eds. Shock metamorphism of natural materials, lst Conf., 1966, Proc.: Baltimore, Md., Mono Book Corp., p. 413-432. Carter, N. L., 1968, Dynamic deformation of quartz, in French, B. M.,and Short, N. M., eds. Shock metamorphism of natural materials, lst Conf., 1966, Proc.: Baltimore, Md., Mono Book Corp., p. 453-474. Carter, N. L., and Friedman, Melvin, 1965, Dynamic analysis of deformed quartz and calcite from the Dry Creek Ridge anticline, Montana: Am. Jour. Sci., v. 263, no. 9, p. 747-785. Chao. E. C. T., 1967, Shock effects in certain rock—forming minerals: Science, v. 156, no. 3772, p. 192-202. Christie, J. M., Heard, H. C., and LaMori, P. N., 1964, Experimental deformation of quartz single crystals at 27 to 30 kilobars confining pressure and 24°C: Am. Jour. Sci., v. 262, no. 1, p. 26—55. Cummings, David, 1968, Shock deformation of biotite around a nuclear explosion, in French, B. M., and Short, N. M., eds., Shock metamorphism of natural materials, lst Conf., 1966, Proc.: Baltimore, Md., Mono Book Corp., p. 211-217. Dachille, Frank, Gig], Paul, and Simons, P. Y., 1968, Experimental and analytical studies of crystalline damage useful for the recognition of impact structures, in French, B. M., and Short, N. M., eds., Shock metamorphism of natural materials, 1st Conf., 1966, Proc.: Baltimore, Md., Mono Book Corp., p. 555-569. Dake, C. L., 1930, The geology of the Potosi and Edgehill quadrangles: Missouri Bur. Geology and Mines Rept., 2d ser., v. 23, 233 p. Dake, C. L., and Bridge, Josiah, 1927, Early diastrophic events in the Ozarks [abs.]: Pan-Am. Geologist, v. 38, no. 2, p. 155-156. Dence, M. R., Innes, M. J. S., and Robertson, P. B., 1968, Recent geological and geophysical studies of Canadian craters, in French, B. M., and Short, N. M., eds., Shock metamorphism ofnatural materials, lst Conf., 1966, Proc.: Baltimore, Md., Mono Book Corp., p. 339-362. Diehl, C. H. H., and Jones, G. H. S., 1967, The Snowball crater, general background information: Defence Research Establishment (Canada) Suffield Expt. Sta., Tch. Note 187, 23 p. REFERENCES CITED Dietz, R. S., 1946, Geological structures possibly related to lunar craters: Pop. Astronomy, v. 54, no. 9, p. 465-467. 1959, Shatter cones in cryptoexplosion structures (meteorite im- pact?): Jour. Geology, v. 67, no. 5, p. 496-505. 1963, Cryptoexplosion structures—A discussion [of paper by W. H. Bucher, 1963]: Am. Jour. Sci., v. 261, no. 7, p. 650-664. Eggleton, R. E., and Shoemaker, E. M., 1961, Breccia at Sierra Madera, Tex., in Short papers in the geologic and hydrologic sciences: U.S. Geol. Survey Prof. Paper 424-D, p. D15l-D153. Fairbairn, H. W., 1939, Correlation of quartz deformation with its crystal structure: Am. Mineralogist, V. 24, no. 6, p. 351—368. French, B. M., 1968, Shock metamorphism as a geological process, in French, B. M., and Short, N. M., eds., Shock metamorphism of natural materials, lst Conf., 1966, Proc.: Baltimore, Md., Mono Book Corp., p. 1—17. Gash, P. J. S., 1971, Dynamic mechanism for the formation of shatter cones: Nature; Phys. Sci., v. 230, no. 2, p. 32-35. Gault, D. E., Quaide, W. L., and Oberbeck, V. R., 1968, Impact cratering mechanics and structures, in French, B. M., and Short, N. M., eds., Shock metamorphism of natural materials, lst Conf., 1966, Proc.: Baltimore, Md., Mono Book Corp., p. 87-99. Gorshkov, G. S., 1959, Gigantic eruption of the volcano Bezymianny: Bull. Volcanology, 2d ser., v. 20, p. 77-109. Hargraves, R. B., 1961, Shatter cones in the rocks of the Vredefort Ring: Geol. Soc. South Africa Trans., v. 64, p. 147—153. Hendriks, H. E., 1954, The geology of the Steelville Quadrangle, Missouri: Missouri Div. Geol. Survey Water Resources Rept. 36, 2d ser., 82 p. Heyl, A. V., 1967, The 38th parallel lineament and its relationship to ore deposits [abs.]: Mining Eng., v. 19, no. 12, p. 33. 1968, Minor epigenetic, diagenetic, and syngenetic sulfide, fluorite, and barite occurrences in the central United States: Econ. Geology, v. 63, no. 6, p. 585-594. Hdrz, Friedrich, 1968, Statistical measurements of deformation structures and refractive indices in experimentally shock loaded quartz, in French, B. M., and Short, N. M., eds., Shock metamorphism of natural materials, lst Conf., 1966, Proc.: Baltimore, Md., Mono Book Corp., p. 243-353. 1969, Structural and mineralogical evaluation of an experimentally produced impact crater in granite: Contr. Mineralogy and Petrology, v. 21, no. 4, p. 365-377. 1970, Static and dynamic origin of kink bands in miCas: Jour. Geophys. Research, v. 75, no. 5, p. 965-977. Howard, K. A., and Offield, T. W., 1968, Shatter cones at Sierra Madera, Texas: Science, v. 162, no. 3850, p. 261-265. Johnson, J. P., and Talbot, R. J., 1964, A theoretical study of the shock wave origin of shatter cones: U.S. Air Force Inst. Tech. School Eng., unpub. thesis, 90 p. Kiilsgaard, T. H., Heyl, A. V., and Brock, M. R., 1963, The Crooked Creek disturbance, southeast Missouri, in Geological Survey research 1962: U.S. Geol. Survey Prof. Paper 450—E, p. E14-E19. Koenig, J. W., 1967, The Ozark uplift and midcontinent Silurian and Devonian Stratigraphy, in Symposium—Silurian-Devonian rocks of Oklahoma and environs: Tulsa Geol. Soc. Digest, v. 35, p. 119-147. Krishnaswamy, D. S., and Amstutz, G. D., 1960, Geology of the Decaturville disturbance in Missouri [abs]: Geol. Soc. America Bull., v. 71, no. 12, pt. 2, p. 1910. McCracken, M. H., 1971, Structural features of Missouri: Missouri Geol. Survey Water Resources Rept. Inv. 49, 77 p. McCracken, M. H., and others, 1961, Geologic map of Missouri: Div. Geol. Survey Water Resources, Rolla, Missouri. Manton, W. 1., 1965, The orientation and origin of shatter cones in the Vredefort ring, in Geological problems in lunar research: New York Acad. Sci. Annals, v. 123, art. 2, p. 1017-1049. Melton, F. A., 1931, Post—Pennsylvanian denudation of the Ozark dome: Am. Jour. Sci., 5th ser., v. 21, p. 214-219. 47 Milton, D. J., and others, 1972, Gosses Bluff impact structure, Australia: Science, v, 175, no. 4027, p. 1199—1207. Missouri Division Geological Survey Water Resources, 1943, Magnetic map of Missouri: Rolla, Missouri. Missouri Division Geological Survey Water Resources, 1962, Aeromagnetic map of Macks Creek and Stoutland Quadrangles: Rolla, Missouri. Moore, H. J., 1969, Subsurface deformation resulting from missile impact, in Geological Survey research 1969: U.S. Geol. Survey Prof. Paper 650-B, p. BlO7-Bll2. Naeser, C. W., and Faul, Henry, 1969, Fission-track annealing in apatite and sphene: Jour. Geophys. Research, v. 74, no. 2, p. 705-710. Nichols, C. E., 1973, Geology of the southern half of the Stoutland, Missouri Quadrangle: Missouri School of Mines, Ph. D. thesis. Offield, T. W., Pohn, H. A., and Naeser, C. W., 1970, The character and origin of the Decaturville, Missouri, cryptoexplosion structure, in Abstracts with programs: Geol. Soc. America, v. 2, no. 7, p. 639. Robertson, P. B., Dence, M. R., and Vos, M. A., 1968, Deformation in rock-forming minerals from Canadian craters, in French, B. M., and Short, N. M., eds., Shock metamorphism of natural materials, lst Conf., 1966, Proc.: Baltimore, Md., Mono Book Corp., p. 433—452. Roddy, D. J., 1968a, The Flynn Creek crater, Tennessee, in French, B. M., and Short, N. M., eds., Shock metamorphism of natural materials, lst Conf., 1966, Proc.: Baltimore, Md., Mono Book Corp., p. 291—322. Roddy, D. J., 1968b, Comet impact and formation of Flynn Creek and other craters with central peaks [abs.]: Am. Geophys. Union Trans., v. 49, no. 1, p. 272. Roddy, D. J., and Davis, L. K., 1969, Shatter cones at TNT explosion craters [abs.]: Am. Geophys. Union Trans., v. 50, no. 4, p. 220. Roddy, D. J., Jones, G. H. S., and Diehl, C. H. H., 1969, Similarities of 100 and 500 ton TNT explosion craters and proposed comet impact craters [abs.]: Am. Geophys. Union Trans., v. 50, no. 4, p. 220. Ruhl, Otto, 1904, Observations at Pegmatite Hill, Camden County, Missouri: Drury College, Bradley Geol. Field Sta. Bull. 1, p. 36-40. Savage, T. E., 1909, The Ordovician and Silurian formations in Alexander County, 111.: Am. Jour. Sci., 4th ser., v. 28, p. 509-519. Schuchert, Charles, 1943, Stratigraphy of the eastern and central Untied States: New York, John Wiley & Sons, 1013 p. Searight, T. K., 1955, The geology of the Lebanon Quadrangle, Missouri: Missouri Div. Geol. Survey Water Resources Rept. Inv. 18, 33 p. Shepard, E. M., 1905, Spring systems of the Decaturville dome, Camden County, Mo.: U.S. Geol. Survey Water Supply Paper 110, p. 113-125. Shoemaker, E. M., 1960, Penetration mechanics of high velocity meteorites, illustrated by Meteor Crater, Arizona: 21st Internat. Geol. Cong, Copenhagen, 1960, Rept., pt. 18, p. 418-434. Shoemaker, E. M., and Eggleton, R. E., l961,Terrestria1features of impact origin, Paper A in Proc. Geophys. Laboratory~Lawrence Radiation Laboratory Cratering Symposium: California Univ., Livermore, Lawrence Radiation Lab. Rept. UCRL-6438, pt. 1, p. A1—A21. Short, N. M., 1966, Shock processes in geology: Jour. Geol. Education, v. 14, no. 4, p. 149-166. Short, N. M., and Bunch, T. E., 1968, A worldwide inventory of features characteristic of rocks associated with presumed meteorite impact craters, in French, B. M., and Short, N. M., eds., Shock metamorphism of natural materials, 1st conf., 1966, Proc.: Baltimore, Md., Mono Book Corp., p. 255-266. Shumard, B. F., 1873, Laclede County, Chapt. 12 of Reports on the geological survey of the State of Missouri, 1855-1871: p. 213—222. Simons, P. Y., and Dachille, Frank, 1965, Shock damage of minerals in shatter cones [abs.]: Geol. Soc. America Ann. Mtg. program, p. 156-157. Snyder, F. G., 1970, Structural lineaments and mineral deposits, eastern United States, in Rausch, D. 0., and Mariacher, B. C., eds., Mining and concentrating of lead and zinc, v. 1, of AIME world symposium on mining and metallurgy of lead and zinc, [Proc.]: Am. Inst. Mining, Metall., and Petroleum Engineers, p. 76-94. 48 GEOLOGY OF THE DECATURVILLE IMPACT STRUCTURE, MISSOURI Snyder, F. G., and Gerdemann, P. E., 1965, Explosive igneous activity along an lIlinois-Missouri-Kansas axis: Am. Jour. Sci., v. 263, no. 6, p. 465-493. Snyder, F. G., and others, I965, Cryptoexplosion structures in Missouri: Geol. Soc. America Field Trip Guidebook, 72 p. Tarr, W. A., 1935, The origin of the Decaturville dome, Camden County, Missouri [abs]: Missouri Acad. Sci. Proc., v. I, p. 99-101. Thompson, T. L., and Satterfield, I. R., 1975, Stratigraphy and conodont biostratigraphy of strata contiguous to the Ordovician—Silurain boundary in eastern Missouri: Missouri Geol. Survey Rept. Inv. 57, pt. 2, p. 61-108. Tilton, G. R., Wetherill, G. W., and Davis, G. L., 1962, Mineral ages from the Wichita and Arbuckle Mountains, Oklahoma, and the St. Francois Mountains, Missouri: Jour. Geophys. Research, v. 67, no. 10, p. 4011-4019. Tolman, Carl, and Landes, K. K., 1939, Igneous rocks of the Mississippi Valley lead-zinc districts: Geol. Soc. America Spec. Paper 24, p. 71—103. Van Tuyl, F. M., 1918, The physical history of the upper Mississippi Valley during the late Paleozoic [abs]: Illinois Acad. Sci. Trans, v. 10, p. 280. Weller, Stuart, and St. Clair, Stuart, 1928, Geology of Ste. Genevieve County, Missouri: Missouri Bur. Geology and Mines Rept., 2d ser., v. 22, 352 p. Wilshire, H. G., and Howard, K. A., 1968, Structural pattern in central uplifts of cryptoexplosion structures as typified by Sierra Madera: Science, v. 162, no. 3850, p. 258-261. Wilshire, H. G., Howard, K. A., and Offield, T. W., 1971, Impact breccias in carbonate rocks, Sierra Madera, Texas: Geol. Soc. America Bull., v. 82, no. 4, p. 1009-1018. Wilshire, H. G., Offield, T. W., Howard, K. A., and Cummings, David, 1972, Geology of the Sierra Madera crytoexplosion structure, Pecos County, Texas: US Geol. Survey Prof. Paper 599-H, 42 p. Wilson, C. W., Jr., Stearns, R. G., Tiedemann, H. A., Wilcox, J. T., and Marsh, Phyllis, 1968, Geology ofthe Wells Creek structure, Tennessee: Tennessee Div. Geology Bull. 68, 236 p. Winslow, Arthur, 1894, Lead and zinc deposits: Missouri Geol. Survey Rept., v. 7, p. 432—434. Zimmermann, R. A., and Amstutz, G. C., 1965, The polygonal ring structure at Deeaturville, Missouri—New tectonic observations: Neues Jahrb. Mineralogie Monatsh., v. 9, no. II, p. 288-307. 1972, The Decaturville sulfide breccia—A Cambro-Ordovician mud volcanoe: Chemie der Erde, V. I7, bd. 3!, p. 253-273. GUS, GOVERNMENT PRINTING OFFICE: 1979—677026/33 PLATE 1 Geology mapped in 1967—89 aus. GOVERNMENT PRINTING OFFICE; l979-677»026/33 PROFESSIONAL PAPER 1042 queried where probable. Bar and on downthrown side 7 ll Dashed where inferred R on upthrown side —Showing crestline ./—t- ‘35: so f ball, where present, THRUST—Dashed where inferred, queried where probable. Sawteeth on upper plate FAULT BRECCIA NORMAL— Inclined Overturned Horizontal REVERSE FORM LINE—Showing trace and apparent dip of bed SYNCLINE—Showing troughline STRIKE AND DIP OF BEDS CONTACT ANTICLINE 5 /6 M ———fi—— OVERTURNED ANTICLINE—Showing crestline and direction of limbs —-fi— OVERTURNED SYNCLINE——Showing troughline and direction of limbs 4000 FEET 1000 METERS r i will“ Hull“ .“0 “H” rlzuflwi ,\,ll\// J ,/ $ ORDOVICIAN > CAMBRIAN PRECAMBRIAN l Lower Ordov1c1an > Upper Cambrian l a Cryptozoon chert bed 0.3~0.9 m and marker bed 3, the Gunter r SCALE 1 7000 CONTOUR INTERVAL 10 FEET NATIONAL GEODETIC VERTICAL DATUM OF 1929 GEOLOGIC MAP OF DECATURVILLE IMPACT STRUCTURE, MISSOURI EXPLANATION t; marker bed 4 Potosi Dolomite Davis Formation Gasconade Dolomite Includes marker bed 5, a Cryptozoon chert layer 0.3—0.6 m thick, about3 m below the Eminence Dolomite Lamotte Sandstone UNCONFORMITY Granite, pegmatlte, and schist PREPARED IN COOPERATION WITH THE NATIONAL AERONAUTICS AND SPACE ADMINISTRATION Doe Run and Derby Dolomites Roubidoux contac Sandstone Member, a single bed 3—8 m thick, at the base of the formation white porcelaneous chert in the lower half of the section thick 20—30 m below the top of the Gasconade; Gasconade approximately in the middle of the section, and marker bed 1, a single 8—cm bed of Includes marker bed 2, a single bed of white porcelaneous chert as thick as 09 m SILURIAN > ORDOVICIAN l Middle Silurian Middle Ordovician > Lower Ordowcran } Upper Ordovician ?) Oolite UNCONFORMITY UNCONFORMITY Kimmswick Limestone Jefferson City Dolomite Includes marker bed 11, a sandstone bed 0.6—3 m thick containing subangular to Roubidoux Dolomite Includes marker bed 8, a sandstone 1.2 m thick, locally a sandy chert layer, about 30 m moderately well located to well located; dark color, marker Ordovician and Silurian limestones are shown in one color on the map. Unlabeled KEY—Light color, covered interval; medium color, outcrop bed, indicated by float between outcrops Bambndge leestone UNCONFORMITY ,. I «that» saw , sat outcrops of limestone are Kimmswick; labels indicate outcrops of Noixl?) Oolite and Leemon Formation (Onl) and Bainbridge Limestone (Sb) mt W the base of the section; and marker bed 9, a platy, mudcracked, ropy sandstone bed 0.08—0.5’ m thick at the base of the formation above the base of the formation; and marker bed 6, a chert breccia layer 1.5—3 m subrounded, glassy or frosted grains, about 37 m above the base of the formation; marker bed 10, a sandstone bed 0.15-0.3' m thick, or locally or chert bed or zone of alternating 2.5—crn layers of white dolomite and pink sandstone, about 23 m above above the base of the formation; marker bed 7, a chert breccia 0.3—0.6 m thick, 18 m thick, at the base of the section Leemon Formation Of Thompson and Satterfield (1975) and Noixl n .A ,iw‘OZ : a 5, lllr.zw§ IA, , M. Ikkcz ugmk APPROXrMA‘E Ml IAN DECL warmly I979 m lines are 305 meters apart on man UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY G. M. Nakara, and R. Jordan, 1958 Topographic compilation by J. L Derick, Grid syste UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY PREPARED IN COOPERATION WITH THE NATIONAL AERONAUTICS AND SPACE ADMINISTRATION PROFESSIONAL PAPER 1042 PLATE 2 EXPLANAT 09 COMPOSITE SECTION SHOWING ALL UNITS IN BLOCK DIAGRAMS + _A_A_A_ H 4% /H . x6"; I Transparent ground with marker beds \ ['4 I‘ §< N ‘ / Ordovician Kimmswick Limestone Ordovician Jefferson City Dolomite, showing marker beds Ordovician Roubidoux Dolomite, show- ing marker beds Ordovician Gasconade Dolomite (09), showing Gunter Sandstone Member (099) Cambrian Eminence Dolomite CONTACT OR STRUCTURAL FORM LINE FAULT NORMAL—Bar and ball on down‘ thrown side THRUST—Sawteeth on upper plate. Thrusts in block A labeled in order of occurrence in movement se— quence: A is earliest, J is latest REVERSE—R on upthrown side FAULT BRECCIA—Can occur in small quantifies throughout section \ Q. 1000 METERS I I I I 3000 FEET O‘—O INDEX OF BLOCK DIAGRAMS Arrows show direction of view 700 <9 #2420) 3% N\ "N“ €56 VERTICAL EXAGGERATION X 2 Transparent ground with marker beds Transparent ground with marker beds BLOCK DIAGRAMS SHOWING STRUCTURAL DETAIL, DECATURVILLE IMPACT STRUCTURE, MISSOURI Block diagrams by H. A. Pohn, 1969 QU.SI GOVERNMENT PRINTING OFFICE: 1979—677-026/33 62637:" fifi {)6 _ ' - 7_ DAYS ‘ v-1043- ‘ ‘ I I L Subsurface Stratigraphy and Geochemistry of Late Quaternary Evap0rites, ‘ Searles Lake, California ‘ ",,‘* ‘ v‘ ‘L ‘ , , .‘GEOLOGICALSURVEY PROFESSIONAL PAPER 1043 !. wY—’w A ~ 5., ‘4‘; RETURN EARTHS ' ‘ Cdrfh TO 1!. cmNces UBRARY‘ “3 gm . A...“ A.— ‘ A; “44‘- v...‘ 1'“ V" w— I' -A' ‘ r ‘Y‘r—w ‘TW Subsurface Stratigraphy and Geochemistry of Late Quaternary Evaporites, I Searles Lake, California I By GEORGE 1. SMITH ‘ With a section on RADIOCARBON AGES OF STRATIGRAPHIC UNITS I By MINZE STUIVER and GEORGE 1. SMITH ‘ GEOLOGICAL SURVEY PROFESSIONAL PAPER 1043‘ Description of the stratigraphic succession of muds and salts deposited by closed-basin takes that occupied Sear/es Valley during late Quaternary time UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON: 19%9 I I I I I I UNITED STATES DEPARTMENT OF THE INTERIOR CECIL D. ANDRUS, Secretary GEOLOGICAL SURVEY H. William Menard, Director Library of Congress Cataloging in Publication Data Smith, George Irving, 1927— Subsurface stratigraphy and geochemistry of late Quaternary evaporites, Searles Lake, California. (Geological Survey Professional Paper 1043) Bibliography: p. 118-122. Includes index. 1. Evaporites—California—Searles Lake. 2. Geology, Stratigraphic—Quaternary. 3. Geochemistry—California~Searles Lake. 4. Geology—Calif0rnia~Searles Lake. I. Title. II. Series: United States. Geological Survey. Professional paper 1043. QE471.15.E8SS8 551.7’9’0979495 77-10704 For sale by the Superintendent of Documents, US. Government Printing Office Washington, DC. 20402 Stock Number 024-001-03163-9 4.; Av A A ‘7‘ ,L V" CONTENTS Page ' l Page Abstract ______________________________ 1 Stratigraphy of the evaporite deposits—Continued \ Introduction ___________________________ 2 Upper Salt —Continued ‘ Discovery and economic development of Searles Lake _ {1 Estimated bulk composition of the unit _____ 64 Geologic environment and history of Searles Lake _ _ (65.5) Overburden Mud _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _‘_ 66 Previous geologic studies of Searles Lake ________ 6 Areal extent and volume _ _ _ _ _ _ _ _ _ _ _ _ _ L 66 Acknowledgments ______________________ 8 Mineral composition and lithology ________ l. 66 Stratigraphy of the evaporite deposits ____________ 8 Chemical composition _______________ l. 68 Mixed Layer _________________________ 13 Radiocarbon ages of stratigraphic units, by Minze Stuiver‘ Areal extent and thickness _____________ 13 and George 1. Smith _________________ ‘_ 68 Mineral composition and lithology _________ 13 Introduction _________________________ 68 Bottom Mud ___________________ 15 Reliability of sampled materials ____________ l- 70 Areal extent and thickness _________ 16 Probable true ages of stratigraphic units _______ l_ 73 Mineral composition and lithology _ _ _ _ 16 Rates of deposition ______________________ L 7g Chemical composition of the Bottom Mud 17 Geochemistry of sedimentation _______________ i- (78/ Lower Salt __________________________ 18 Mud Layers _______________________ l- 79 Areal extent and volume ______________ 20 Saline layers ________________________ ‘_ 82 Saline units ___________________ 20 Phase relations applicable to Searles Lake salts _ _ 83 Mud units ____________________ 20 Salines in the Mixed layers _ _ _ _ _ _ _ _ _ _ _ _‘ _ Mineral composition and lithology _________ 34 Salines in the Bottom Mud _ _ _ _ _ _ _ _ _ _ _ _‘ _ Saline units ___________________ 34 Salines in the Lower Salt, Upper Salt, and Over-1 Mud units ____________________ 37 burden Mud ________________ _L _ Chemical composition ________________ 40 Lower Salt __________________ ~+ _ Chemical analyses of the solids _________ 40 Upper Salt __________________ T _ Chemical analyses of the brines ________ 43 Overburden Mud _________________ Estimated bulk composition of the unit ______ 47 Geochemical influence on shape and thickness of PartingMud ________________________ 47 saltbodies ____________________l_ Areal extent and volume ______________ 47 Source of salt components ___________ 1 _ 98 Mineral composition and lithology _________ 48 Geochemistry of diagenesis _______________ .4 _ 100 Chemical composition ________________ 54 Mud layers _______________________ _, _ 100 Upper Salt __________________________ 56 Salt layers _______________________ 1 _ 106 Areal extent and volume ______________ 56 Reconstructed depositional history _____________ 108 Mineral composition and lithology _________ 59 Correlations with deposits in other areas _ _ _ _ _ _ _ _ _‘ _ 112 Chemical composition ________________ 60 Economic geology _ _ _ __ _ __ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _‘. _ 116 Chemical analyses of the solids _________ 60 References cited ______________________ J, _ 118 Chemical composition of the brines ______ 60 l l l l ILLUSTRATIONS , l Page PLATE 1. Subsurface sections showing relations between stratigraphic units of Searles Lake, California. ________ In pocket 2. Diagrams showing components in Mixed Layer and Bottom Mud of Searles Lake, California __________ in pocket FIGURE 1. Index map showing location of Searles Lake and other Pleistocene lakes ______________________ 4. _ _ 3 2. Graphs showing average temperature and rainfall, Searles Valley __________________________ + _ _ 4 3. Map showing location of core holes ____________________________________________ T _ _ 9 4. Stratigraphic column showing units in Searles Lake _____________________________________ 11 5. Cross section of Searles Valley showing bedrock profile and stratigraphy of upper part of fill _ _ _ _ _ _ _ _ _ _ l. _ _ 12 6. Graph showing weight percent acid-insoluble material in Bottom Mud _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ‘_ _ _ 18 7. Contour map on base of Lower Salt ___________________________________________ L _ _ 19 ‘ III IV CONTENTS FIGURES 8—20. Isopach maps: Page 8. Unit S—l _______________________________________________________________ 21 9. Unit S—2 _______________________________________________________________ 22 10. Unit 8—3 _______________________________________________________________ 23 11. Unit 8—4 _______________________________________________________________ 24 12. Unit 8-5 _______________________________________________________________ 25 13. Unit 8—6 _______________________________________________________________ 26 14. Unit 8—7 _______________________________________________________________ 27 15. Unit M—2 ______________________________________________________________ 28 16. Unit M—3 ______________________________________________________________ 29 17. Unit M—4 ______________________________________________________________ 30 18. Unit M—5 ______________________________________________________________ 31 19. Unit M—6 ______________________________________________________________ 32 20. Unit M—7 ______________________________________________________________ 33 21. Diagrams showing compositions of brines in Lower Salt in relation to phase boundaries ________________ 46 22. Contour map on base of Parting Mud _____________________________________________ 49 23. Isopach map, Parting Mud ___________________________________________________ 50 24. Graphs showing volume percent of components in Parting Mud ______________________________ 52 25. Graph showing percentage of pore water in samples of Parting Mud ___________________________ 54 26. Contour map on base of Upper Salt ______________________________________________ 57 27. Isopach map, Upper Salt _____________________________________________________ 58 28. Diagram showing composition of brines in Upper Salt in relation to phase boundaries _________________ 65 29. Photograph showing polygonal cracks in lake surface ____________________________________ 67 30. Stratigraphic column showing position and age of 1‘C dated samples from cores _____________________ 69 31. Stratigraphic column showing relation between depth and new ”C ages of mud layers in Lower Salt _________ 74 32. Graph showing relation between area, volume, and salinity of Pleistocene Searles Lake ________________ 78 33. Phase diagram of NaHCO3—Na2CO3—Na2804-NaCl—H20 system at 20° C _________________________ 83 34. Graph showing percentages of major minerals in Lower Salt and Upper Salt _______________________ 87 35—39. Phase diagrams showing: 35. Stability fields in NaHCO3—Na2C03—NaZSO4—NaC1—H20 system at 20° C for crystallization of units 8—1 to 8—5 _ _ 87 36. Stability fields in Na2CO3—Na2SO4—NaCl—H20 system at 15° C for crystallization of units S—6 and 8—7 _______ 89 37. Superimposed 5—component and 4-component systems at 20° C ______________________________ 89 38. Stability fields in NaHCO3—Na2COg—Na2804—NaCl—H20 system at 20° C ________________________ 91 39. Stability field of hanksite in NaHC03—Na2CO3—Na2804—NaCl—H20 system at 20° C __________________ 93 40. Graph showing relation between temperature and depth in sediments of Searles Lake _________________ 101 41. Graph showing inferred history of fluctuations in Searles Lake, 0—150.000 years ago __________________ 109 42. Diagram showing possible correlations between the history of Searles Lake and the histories of other lakes and glaciers 112 43. Diagram showing possible correlations between the history of Searles Lake and the histories indicated by other climatically sensitive criteria ________________________________________________ 113 44. Diagram comparing details of Searles Lake fluctuations and Laurentide ice sheet fluctuations ____________ 115 TABLES Page TABLE 1. Nonclastic minerals in the Searles Lake evaporites _____________________________________ 10 2. Chemical analyses of core samples from the Mixed Layer __________________________________ 15 3. Area and volume of units in the Lower Salt __________________________________________ 20 4. Estimated mineral compositions of saline layers in the Lower Salt ____________________________ 35 5. Estimated mineral compositions of mud layers in the Lower Salt _____________________________ 38 6. Mineral abundance in mud units of the Lower Salt, core GS—14, determined for total sample and acid—insoluble fraction of sample ______________________________________________________ 40 7. Chemical analyses of solids in the Lower Salt _________________________________________ 41 8. Comparison of chemical analyses with composition indicated by visual estimates of mineral percentages ______ 42 9. Chemical analyses of brines in the Lower Salt ________________________________________ 44 10. Estimated percentages and total quantities of water soluble components in the upper two and lower five units of the Lower Salt ________________________________________________________ 48 11. Megascopic mineral composition of the Parting Mud ____________________________________ 51 12. Size distribution of acid-insoluble material in four samples of the Parting Mud, core L—12 ______________ 53 13. Partial chemical analyses of samples from core GS—16 in Parting Mud __________________________ 55 14. Mineral composition of salines in the Upper Salt by contour interval ___________________________ 59 15. Chemical analyses of cores from the Upper Salt and Overburden Mud __________________________ 61 A._A.- TABLE 16. 17. 18. 19. 20. 21. 22. CONTENTS Chemical analyses of brines from the Upper Salt and Overburden Mud _______________________ Estimated percentages and total quantities of water-soluble components in the Upper Salt ____________ Partial chemical analyses of core GS—40 from the Overburden Mud __________________________ New “C dates on disseminated organic carbon in mud layers of the Lower Salt and top of the Bottom Mud, core L—31 Depositional rates in Parting Mud _____________________________________________ Comparison of amount of selected components carried by Owens River in 24,000 years with amount now in Owens Lake and Upper Salt of Searles Lake _________________ Analyses of brines pumped to chemical plants, 1938—51 _________________________________ Page 62 66 70 77 99 117 \ 11+Mw‘ SUBSURF ACE STRATIGRAPHY AND GEOCHEMISTRY OF LATE: QUATERNARY EVAPORITES, SEARLES LAKE, CALIFORNIA By GEORGE 1. SMITH ABSTRACT Searles Lake is a dry salt pan, about 100 km2 in area, that lies on the floor of Searles Valley, in the desert of southeast California. Several salt bodies of late Quaternary age lie beneath the surface, mostly composed of sodium and potassium carbonate, sulfate, chloride, and borate minerals. Mud layers separate the salt bodies, which contain interstitial brine that is the source of large quanti- ties of industrial chemicals. The value of annual production from the deposit exceeds $30 million; total production to date exceeds $1 billion. The salts and muds were deposited during Pleistocene and H010- cene times by a series of large lakes (200 m maximum depth, 1,000 km2 maximum area) that fluctuated in size in response to climatic change. Salts were deposited during major dry (interpluvial) epi- sodes, muds during wet (pluvial) episodes that correlate with gla- cial advances in other parts of North America and the world. Data based on cores from the deposit are used in this paper to establish the stratigraphy of the deposit, the chemical and mineral composi- tions of successive units, and the total quantities of components contained by them. These parameters are then used to determine the geochemical evolution of the sedimentary layers. The results provide a refined basis for reconstructing the limnology of Searles Lake and the regional climate during late Quaternary time. Six main stratigraphic units were distinguished and informally named earlier on the basis of their dominant composition: Typical Unit thickness ”C age, uncorrected (in meters) (years B.P.) Overburden Mud _ _ 7 0 to >3,500 Upper Salt _____ 15 >3,500 to 10,500 Parting Mud _ _ _ _ 4 10,500 to 24,000 Lower Salt _____ 12 24,000 to 32,500 Bottom Mud _ _ _ _ 30 32,500 to 130,000 Mixed Layer _ _ _ _ 200+ >130,000 F (The age of 130,000 years for the Mixed Layer is based on extrapo- lated sedimentation rates.) The Lower Salt is subdivided into seven salt units (S—1 to 8—7) and six mud units (M—2 to M—7), the Mixed Layer into six units (A to F). For each salt unit, the areal extent, volume, shape, mineralogy, and chemical composition of the solids and brines have been determined; for each mud unit (which originally extended over much of the basin), the shape and volume within a standard area, and the mineralogy, have been de- termined. The bulk compositions (brines plus salts) of the com- bined Lower Salt units S—1 to 8-5 and units S—6 and 8—7, and the Upper Salt, were determined so that the total quantities and ratios of ions in the initial brines could be reconstructed. The 74 published 1‘C dates on Searles Lake core samples from all but the oldest unit are supplemented by 14 new dates (det rmined by Minze Stuiver) on the Lower Salt.- Most of the age control comes from dates based on disseminated organic carbon; two dates are on wood; dates on carbonate minerals are less reliable. Although the probable disequilibrium between the carbon in the lake and atmo- sphere (because of contamination, slow equilibrium rates, nd oth- er factors) causes disseminated carbon dates to be an es imated 500—2,500 years “too old,” the ages of the major and minor nits are relatively well established. The list above indicates roun ed and uncorrected ages for the contacts of major units. The age of the only salt bed in the Lower Salt which indicates desiccation (8—5) is about 28,000 years. The average uncorrected sedimentation rate in the Parting Mud is 46 yr/cm. Correcting the indicated Parting Mud sedimentation rate on the basis of the greater acid-i ‘soluble content of the Bottom Mud suggests an average rate for it of 33 yr/cm. Using this figure, the age of the base of the Bottom Mud 1s estimated to be near 130, 000 years. The stratigraphy and mineralogy, combined with known and in- ferred dates, provide a basis for reconstructing the climatically controlled history of Searles Lake. The best known part is for the past 150,000 years, for which more than 30 major changes in lake level are reconstructed. Prior to that time, less detail can be deter- mined, but there seems to have been a very long period dominated by dryness. Close similarities in age exist between the major fluctuations in the levels of Searles Lake and the advances and retreats of glaciers in eastern and western North America, fluctuation in atmospheric and sea surface temperatures, and changes in world sea level. The Upper Salt and Overburden Mud are correlated with Holocene and Valderan deposits of the Laurentide ice sheet in easter ‘ North America, the Parting Mud with Twocreekan and Woodfor ian de— posits, the upper part of the Lower Salt (S—5 and abov ) with Farmdalian deposits, the lower part of the Lower Salt and part of the Bottom Mud with Altonian deposits Thin salt layers‘ 1n the Bottom Mud that are estimated to be about 105, 000 years old are provisionally correlated with Sangamon deposits, the 10 m of mud below them with Illinoian deposits, and most of the Mixed Layer with Yarmouth deposits. , The late Quaternary lakes in Searles Valley contained {nough dissolved solids for chemical sediments to form at all time . Rela- tively insoluble minerals precipitated from the large f esh to brackish lakes and salts precipitated from the small highly saline lakes. The geochemical reconstruction of the sedimentatiou pro- cesses allows many of the chemical, physical, and limnological con- ditions to be approximated. Primary aragonite, calcite, and 2 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. possibly dolomite precipitated from even the largest lakes. Some of their precipitation was caused by temperature change and evapo- ration, but laminae of aragonite (and possibly some northupite) were apparently caused by mixing along the chemocline between a fresh surface layer and an underlying saline layer of higher density. Diagenesis after burial produced gaylussite and pirssonite (when Na-carbonate brines reacted with Ca-carbonate minerals), anal- cime, searlesite, K-feldspar, and phillipsite. Primary and secon- dary carbonate minerals now constitute 60 to 85 percent of most mud units; diagenetic silicates, clastic silicates, and organic mate- rial form the balance. The salt layers contain varying mixtures of saline minerals that are zoned both concentrically and vertically. Published logs pro— vide data that allow quantitative estimates of the saline mineral distribution and volume. In the Mixed Layer, halite, trona, and nahcolite are the dominant saline minerals. In the Lower Salt, trona, halite, and burkeite are most abundant; northupite, thenar— dite, hanksite, borax, nahcolite, sulfohalite, and tychite occur in minor to trace quantities. In the Upper Salt, halite, trona, and hanksite are most abundant; borax, burkeite, thenardite, aphthita— lite (glaserite), and sulfohalite occur in that order of decreasing abundance. Trona constitutes most of the edge facies of all units; other min- erals are more abundant in central or intermediate facies. Vertical zonation reflects the original crystallization sequence modified by diagenesis. Units 8—1 to 8—5 in the Lower Salt represent one se- quence that was interrupted by four deep lakes that deposited mud: trona is the dominant mineral in units 8—1, 8—2, and 8—3; trona and burkeite make up most of 8—4; and halite, trona and bur— keite occur in 8—5, which represents desiccation. This sequence is best explained by the phase relations in the NazCOg—NaHCO3— NaQSO4—NaCl-H20 system at 20°C which relates these minerals. Units S—6 and 8—7 of the Lower Salt represent two episodes of in- complete desiccation; trona and halite make up most of both units. Phase relations in the same system at 15°C best explain the com- positions of these units. The Upper Salt represents a single desic- cation episode. Borax and trona occur at the base, and halite and hanksite constitute most of the upper part. The initial crystalliza- tion sequence was trona (and borax), then thenardite and burkeite, halite, and finally aphthitalite; diagenesis resulted in some hank- site forming at the stratigraphic position of the original thenardite and burkeite, some at the stratigraphically higher position of the aphthitalite. Initial crystallization temperatures were low, causing borax to crystallize, then increased during later stages. Phase rela- tions at 20°C in the 5-component system given and in the Na2C03— Na2SO4—NaCl—KCl—H20 system satisfactorily explain the ob- served sequence of primary and diagenetically produced minerals in the Upper Salt. Salts in the Overburden Mud are zoned concen- trically because the small lake that formed them shrank and depos— ited successive assemblages in successively smaller areas. Diagenetic reactions account for a high percentage of the miner- als in the mud layers and some minerals in the salt layers. In the muds, gaylussite and pirssonite formed by reactions between Na- carbonate brines and Ca-carbonate minerals, the species of min- eral determined by the chemical activity of H20 (“H20)- This re- flects differences in pore water salinity in the upper 166 m of sedi— ments and temperatures greater than about 35°C in deeper sediments. Small accumulations of trona nahcolite, northupite, and tychite probably reflect postburial changes in the chemical ac- tivity of CO2 (acoz). Moderate to large quantities of microcrystal- line halite in the muds probably result from post-depositional increases in salinity during compaction. Aragonite, a thermodyna— mically unstable form of CaCO3, altered spontaneously to calcite in sediments older than about 50,000 years. Slow reactions between elastic silicates and pore brines produced a suite of authigenic sili- cate minerals consisting of monoclinic K-feldspar, searlesite, ana- cime, and phillipsite. Pyrite was noted in one sample. Diagenesis of minerals in salt layers includes increases in the sizes of crystals. This is probably a result of cycles that cause slight undersaturation and then supersaturation, the smallest crystals being dissolved during undersaturation cycles and their ingredi- ents subsequently added to the larger ones that survived. Hanksite formed by the diagenetic reaction of metastable assemblages of burkeite, thenardite, and aphthitalite. Observations of other saline lakes now forming crystal layers suggests that the values of (1002 and aH20 at the surface, where crystals first formed, differ from those found a short time later in the accumulating salt layer. This environmental change results in an almost immediate change in saline mineral suites, but that change in species was generally the last; most saline minerals found in Searles Lake appear to be the same species as those formed after that initial change. The composition of the interstitial brines changed some time after initial deposition of both salts and muds, apparently as a re— sult of the downward migration of water from the lake surface which produced some solution of the salts. Movement of ground water in this direction is caused by the high hydrostatic head of the brines in the lake relative to the water in the surrounding valley fill. An estimated 480 X 1012 g of brine has been pumped as the source of 57 X 1012 g of produced salts. This amount represents about 40 percent of the extractable salts in the original brine, but only about 6 percent of the total commercial salt in the deposit. The pumped brines are replaced by waters that dissolve enough salts to reach local chemical equilibrium, but the ratio of components in the re- placement brines is less favorable for commercial operations. The quantity of chemicals that can be extracted depends partly on de- velopments in extraction technology and partly on the level of un- derstanding that can be achieved of the chemical and hydrologic parameters of the deposit that dictate the optimum methods of pumping and recharge. INTRODUCTION The Searles Lake evaporites consist of several lay- ers of flat-lying Quaternary deposits that underlie the dry lake in the middle of Searles Valley, San Bernar- dino and Inyo Counties, Calif. The valley lies near the southwest corner of the Basin and Range province and just north of the Mojave Desert; Los Angeles is about 200 km to the south-southwest, Bakersfield about 160 km to the west (fig. 1). Searles Valley has a drainage area of about 1,600 kmz. The valley floor, which trends south then curves to the southwest, is about 60 km long and at places 15 km wide. Searles Lake itself is a nearly dry playa about 15 km long and 11 km wide that covers approxi- mately 100 kmz; about two-thirds of this area is mud, one-third hard salt. The climate in the valley is hot and arid. In the 36- year period ending in 1973, rainfall averaged about 96.3 mm (3.79 in.) per year, with extremes (calculated for rain years beginning July 1) ranging from 23 to 291 mm (0.92—11.47 in.). During this period, the mean an- nual temperature was 19.1°C (66.3°F); record tem- INTRODUCTION peratures were 47.8°C (118°F) and —12.2°C (10°F) (fig. 2). The strongest winds come in the spring and fall. During most winters, several centimeters of water stand on Searles Lake for a few weeks or months. In the summer and early fall, the surface is normally free of standing water except for isolated brine pools and areas where the chemical plants and towns have dis- charged waste water onto the surface. The surface of Searles Lake supports no vegetation. The surrounding alluvial fans and mountains are c0v- ered by a sparse growth of plants typical of the lower Sonoran zone; dominant types include creosote bush (Larrea tridentata), hop sage (Grayia spinosa), de- 119° 118° O 0 10 10 20 30 4O 50 60 7O KILOMETERS l 3 sert holly (Atriplex hymenelytra), and burro-bush (Franseria dumosa). Pinyon pine grows on th ‘ high- est parts of the Argus Range; Joshua trees on th high- est parts of the Slate Range. Searles Valley is connected by California State Highway 178 from Ridgecrest to the west, and fry the unnumbered continuation of that highway to Pana- mint and Death Valleys to the northeast. Within Searles Valley, numerous dirt roads allow access to many parts of the valley floor. The Trona R ilway, which carries only freight, connects the town ofETTrona with the Southern Pacific Railroad about 55 km to the southwest. In the valley lie the towns of Pioneer Point, EXPLANATION Present playa or lake Pleistocene lake ——> Present river 20 saw 30 40 50 MILES — ----- —> Pleistocene river 36° 0 Bakersfield l l " Searles Lake \ Manly Lake \ (Death Valley) l FIGURE 1.—Location of Searles Lake and other lakes connected with it in Pleistocene time. 4 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. 100 — -- _ Record high — 35 90 _ for month _ — '— 30 80 — — Mean annual temperature — 20 .L 66.3°F (19.1 °C) 1 7O —_ '|' 60 —— —— 15 TEMPERATURE, lN DEGREES FAHRENHEIT 50 —— ‘_ 1° Record Iow/ for month 40 —_ _— 5 _ — 0 30 , | l l l l | | l | l | l J F M A M J J A S O N D 3.5 — Average annual rainfall 3.79 in (9.63 cm) 3.0 —_ _, —— 75 __ (I) Record high 1 E a) f Lil 2.5 .‘ / or month _ E U _ — 60 E _z_ j E 2.0 — — E 2 _ z o — 45 _ _ Z L} 1.5 — ‘ 9 t l- u- < 9 _ — 30 E I 1.0 — ‘ a n. Lu _ .L 15 E 0.5 — - 1 l L ‘L _ FIGURE 2.—Present climate of Searles Valley. Measurements by weather station at Trona, for period 1937—73. Data from climatological records of US. Department of Commerce (Weather Bureau and Environmental Data Service). Trona, Argus, South Trona, and Westend. Trona, population about 2,000, is the largest. DISCOVERY AND ECONOMIC DEVELOPMENT OF SEARLES LAKE The value of the Searles Lake evaporite deposit is now known to be worth billions of dollars, yet the lake was clearly regarded by the earliest transients as a worthless obstacle. The water was undrinkable and TEMPERATURE, IN DEGREES CELSIUS the surface was difficult to traverse. Records of several groups that commented on the nearly dry lake in the valley are available, among them the nearly starved Bennett-Arcane (Manly) party as they left Death Val- ley in 1849. Soon after 1860, prospectors found a rea- son for staying in this part of the desert longer than necessary: gold and silver deposits were discovered in the surrounding mountains. These and subsequently discovered deposits were worked intermittently, and their recorded production of gold, silver, lead, zinc, and copper is valued at nearly $3 million (Norman and Stewart, 1951; Smith and others, 1968, p. 28). The mines were small when compared with Darwin, Pana- mint City, Cerro Gordo, and others nearby, but their discovery did serve to bring prospectors and miners into the valley. Among them were the brothers John and Dennis Searles, for whom the lake was later named. The discovery of borax occurred on February 14, 1873,1 when Dennis Searles and E. M. Skillings first noted the similarity of material from the lake to the samples of the Nevada borate playas (Hanks, 1883). After confirming the borate content of the crusts (sub- sequently found to average about 8 percent), patents for 160 acres were applied for in April 1874 at the US. Land Office at Independence by John Searles and J. D. Creigh on behalf of all four men. Production of borax started in that year at a plant on the west side of the lake, and the company holdings were increased in size, eventually covering 2,000 acres (Dyer, 1950). On January 1, 1878, they incorporated under the name San Bernardino Borax Co. The refining process, which employed about 30 men, operated as follows (De Groot, 1890): The crusts from the lake were collected and dumped into boiling vats of saline solution to dissolve the borax (and other soluble components) and settle out the sediment. The clear liquid was then transferred to wooden crys- tallizing tanks and allowed to settle for 5 to 9 days. This formed an impure grade of product that was car- ried in that state, or after further purification, to the Southern Pacific Railroad at Mojave by twenty-mule- team wagons. By 1889, the Searles’ refinery had pro— duced 10,500 tons of borax (Hanks, 1889). Production continued until 1895, when the Pacific Coast Borax Co. purchased the operation and closed it. 'Some uncertainty exists about the exact dates of these events and the size of the original operation. Hanks (1883, p. 26) is the source of the three dates given here for the discovery of borax, patenting of land, and incorporation of the holding company, and he says that the first patent application was for 160 acres. (His report, however, also de- scribes Searles Lake as lying in T. 30 S., R. 38 E., the site of Koehn Lake.) Teeple (1929, p. 20) and Dyer (1950, p. 39) report 640 acres as the original amount of land under location. Present records of the US. Bureau of Land Management show that patent applications were filed by J. W. Searles and others on October 20, 1874, under the com- pany name of Ohio Borax, and that it included 640 acres in section 20 and 21, T. 25 S., R. 43 E. (SE1/4 sec. 20, E1/2NE1/4 sec. 20, W1/2 sec. 21, W1/2NE1/4 sec. 21). INTRODUCTION Following this, little activity occurred at Searles Lake until about 1905 or 1906, when the soda ash? po- tential of the lake was realized. This started another episode of economic development that ultimately led to its present state; the first 20 years is described in more detail by Teeple (1929, p. 21—25). By 1908, the California Trona Co. had located claims on most of the deposit and had borrowed a considerable amount of money to develop them, but the firm went into re- ceivership the following year. To keep the property in- tact, development and exploration were continued by the receiver. By 1914, the reorganized company, called the American Trona Co., had constructed a small plant and completed the Trona Railway, but the new plant never produced. The operation was idled until 1916, when another new plant was finished and com- mercial production was started at the town of Trona— but the product was potash, not soda ash. Earlier, in 1913, the US. Government had with- drawn the unowned parts of Searles Lake as a potash reserve, realizing that the European supply was about to be cut off. The same realization created a strong incentive to the companies on Searles Lake to start producing potash, known to exist in the brines of the lake at least since 1898 (Gale, 1914, p. 309). Potash production began in 1916 at two plants, one at Trona that had been constructed by the American Trona Co., and one at Borosolvay, constructed in that year as a joint effort of the Pacific Coast Borax Co. and the Solvay Process Co. During the acute potash shortage of World War I, the price increased ten times over the prewar level and both producers flourished. But by the end of 1920, the Borosolvay plant closed for good, and the American Trona Co. continued only with dif- ficulty. Following the period of readjustment after World War I, the American Trona Co. added borax to its pro- duction (in 1919) and began a long period of research on plant design, the details of which are described by Teeple (1921, 1929). In 1926, it merged into the American Potash and Chemical Corp. Production has continued from that time; new processes have been developed and new products added to the list of mate- rials produced. The West End Chemical Co., at the settlement of Westend, was organized in 1920 by F. M. Smith (widely known as “Borax” Smith), shortly after he lost control of the Pacific Coast Borax Co., which he founded (Hellmers, 1938). The original plant was de- signed to produce borax and potash but it was found to be inadequate and was redesigned. In 1927, soda ’Soda ash is the industrial mineral term for sodium carbonate; other terms used in this paper are salt cake, for sodium sulfate, borax, for sodium borate, salt, for sodium chloride; and potash, for potassium oxide and other potassium salts. l 5 ash was first produced, and in 1930, borax was added to the list of products. In 1955, production of sqdium sulfate was begun. The production techniques are de- scribed by Hell’mers (1938) and Ver Planck (1957, p. 480—482). In 1956, the Stauffer Chemical Co. absorbetl the West End Co., and in 1967, the Kerr-McGee Ch mical Corp. absorbed the American Potash & Chjmical Corp. In 1974, the Kerr-McGee Chemical CorpL pur- chased Stauffer’s West End plant. The chemical plants at Trona and West End now operate under one ownership. Descriptions of the development an pre- sent nature of these operations are given by Gale (1938, 1945), Hellmers (1938), Dyer (1950), Hig ower (1951), Ryan (1951), Leonardi (1954), Bixler an Saw- yer (1957), Ver Planck (1957 ), Chilton (1958), Garrett (1960), Garrett and Phillips (1960), Goudge and ,Tom- kins (1960), and Hardt, Moyle, and Dutcher (i972). Both plants extract chemicals from brine p mped from the interstices of the saline layers that underlie the dry lake surface. The plant at Trona produces so- dium carbonate and sulfate, potassium chloride and sulfate, lithium carbonate, sodium borate, phosphoric acid, and bromine. The plant at West End prdduces sodium carbonate, sodium borate, and sodium sulfate. Annual production from'both plants is now valued at about $30 million, and total production since 1926 ex- ceeds $1 billion. 1 Leases held by a subsidiary of the Occidental etro- leum Corp., the Searles Lake Chemical Corp., 11 ar the south edge of the lake were under development ‘in the early 1970’s by two other subsidiary companies, the Garrett Research and Development Co. anld the Hooker Chemical Corp. Anticipated production in- cluded sodium borate, sodium carbonate, and potas- sium sulfate (Phosphorus and Potassium, 1971; Industrial Minerals, 1971) by means of a combiried so- lar evaporation and plant process (Kallerud, (1966). Seven large evaporation ponds were completed in 1971 and pumping of brine for test purposes started in 1972 (California Geology, 1972). The operation is in- active at this time (1978). GEOLOGIC ENVIRONMENT AND HISTORY ‘ OF SEARLES LAKE i Along the west side of the valley, in the southern part of the Argus Range and in the Spangler‘ Hills, most of the rocks are late Mesozoic plutonic bodies cut by numerous dikes. Near the northwest corner of Searles Valley, these rocks are in contact with large areas of Paleozoic limestone and late Cenozoic basalt and pyroclastic rocks. In the Slate Range, alohg the east side of Searles Valley, rocks representing all geo- logic eras crop out: Cenozoic lake beds, gravel, pyro- l l 6 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. clastic rocks, and mafic lava, late Mesozoic plutonic rocks, early Mesozoic metavolcanic rocks, Paleozoic limestone, and Precambrian metamorphic rocks. No— table areas of gypsum-bearing fault gouge crop out along the valley edge. The south edge of Searles Valley is bounded by the Lava Mountains and an unnamed range of low hills that consist mostly of late Tertiary sandstone, pyroclastic rocks, and andesite, and Qua- ternary silt, sand, and gravel. The geology of these ranges is shown on the Geologic Map of California Trona Sheet (Jennings and others, 1962) and on maps by Smith (1964) and Smith and others (1968). Searles Valley is a closed tectonic depression sur- rounded by hills and mountains that project 1,000— 1,500 m above the valley floor. Gravity and seismic data show that the pre—Cenozoic(?) bedrock floor of the basin is as much as 1,000 m below the present sur- face and 500 m below sea level (Mabey, 1956). This depth and the structural characteristics of the sur- rounding ranges indicate that tectonic activity created Searles Valley. Depression of the valley floor and ele- vation of the surrounding ranges probably began in late Cenozoic time (Smith and others, 1968, p. 13, 25). By late Quaternary time, Searles Valley had evolved into its present form. Water that drained into it either formed a lake that deposited muds 0r evaporated to form salines. It is the depositional and geochemical re- cord of these late Quaternary lacustrine events, as in- dicated by the stratigraphy and mineralogy beneath the valley floor, that is described in this report. The late Quaternary history of Searles Lake was first described in detail by Gale (1914): Searles Lake was third in a chain of five lakes that received water from the Owens River during pluvial periods of the late Pleistocene (see fig. 1). This river received most of its water from the eastern slopes of the Sierra Nevada and transported it to Owens Lake. When the lake had filled to a level about 60 m above the present surface, water overflowed through a narrow gorge to Indian Wells Valley, where it formed China Lake. China Lake, in turn, attained a depth of about 12 m above the present playa and overflowed into Searles Valley to form a lake as much as 200 m above the present playa; during the highest stages, Searles Lake co- alesced with China Lake to form a continuous body of water about twice the size of the present-day Lake Ta- hoe. From Searles Lake, water overflowed around the south end of the Slate Range and into Panamint Val— ley, where a lake more than 280 m deep formed. This lake spilled over Wingate Pass into Death Valley. Core holes have been drilled in four of these basins «and their logs published (Smith and Pratt, 1957). The cores show that the following lithologies are probably representative of the fill in the central parts of these basins. Owens Lake basin contains fossiliferous fine- grained sediments throughout most of the 280 m tested; the top few meters of evaporites were deposited after 1913, when the Owens River was di- verted into the Los Angeles Aqueduct. The 220 m of China Lake basin tested consists of silt- and sand— sized clastic sediments plus some calcite and gaylus— site. Searles Lake, cored to a depth of 267 m, contains alternating layers of salines and fine-grained carbon- ate muds; the details of this sequence are discussed in this paper. Panamint Lake basin, tested to a depth of 303 m, contains elastic deposits ranging from clay to gravel, small amounts of gypsum, and thick layers of nearly pure halite. Death Valley was not cored be- cause logs were available from earlier work (Gale, 1913a, p. 16) that reported alternating layers of mud or clay and rock salt in the upper 30 m of the basin fill. PREVIOUS GEOLOGIC STUDIES OF SEARLES LAKE Searles Lake3 first received specific attention from geologists and mineralogists after borates were suc- cessfully extracted from it in 1874. Early accounts of these commercial operations are given by Hanks (1883, p. 26—28; 1889) and DeGroot (1890). Hanks (1889) described the deposit and speculated on its geologic origin; he suggested that the water in Searles Lake was derived from Owens Lake and the Owens River, but the statement is not clear, and apparently the connection he proposed consisted of underground seepage rather than a series of surface streams and lakes as was subsequently established. Gilbert (1875, p. 103) had previously traced the path of waters that spilled from Owens Lake as far south as Indian Wells Valley, but as his search for an outlet was restricted to the south edge of that valley, he missed its narrow spillway to the east leading into Searles Valley. In 1896, Fairbanks (p. 69) noted that Searles Lake had shorelines as much as 500 feet above the present val- ley floor, and reported that Owens Lake had drained southward into Indian Wells Valley during pluvial periods, but he did not discuss the possibility of a con— nection between Indian Wells and Searles Valleys. Bailey (1902) might be cited as the first to propose a surface waterway connection between Owens Lake and Searles Lake. This proposal was part of a grander concept which postulated (Bailey, 1902, p. 10—12, and map facing p. 32) an extremely large Quaternary lake (“Lake Aubury”) that submerged all but the higher "Searles Lake has been known by several other names that are used in older records. The commonest are Slate Range Lake, Alkali Flat, Borax Lake, Borax Marsh, Borax Flat, and Searles Marsh. The name Borax Lake has caused confusion in geologic litera- ture because it is commonly not distinguished from Borax Lake in Lake County, Calif., the site of the earliest borax mining in the State. INTRODUCTION peaks in the southwestern Basin Ranges and Mojave Desert of California and unspecified parts of Nevada' and Arizona to the east. Bailey further postulated that on partial desiccation, this large lake became frag- mented into-more modest-sized lakes, and these too are shown on his map (Bailey, 1902, facing p. 32). One of these lakes includes China Lake and Searles Lake, joined as one, and the text (Bailey, 1902, p. 94) indi- cates this body as the destination of the overflow from Owens Lake, a relation now known to be true. Bailey does not cite the evidence for these smaller lakes—to say nothing of the larger one—and it is not possible to tell whether his correct conclusion regarding the Owens-China-Searles chain was a result of observa— tion or serendipity. The attempt to commercially extract soda ash (so- dium carbonate) from Searles Lake in 1908, and its recognition in 1912 as a domestic supply of potash to replace the European supply being threatened by the events preliminary to World War I stimulated another series of geologic and engineering studies. Reports that include geologic descriptions and interpretations were published by Hamman (1912a, b, c), C. E. Dol- bear (1913), Gale (1913b; 1914), S. H. Dolbear (1914), Free (1914, p. 38—40), and Young (1914, p. 48—53). By the time these investigations were made, Searles Lake seems to have been generally recognized as a member of the chain'of lakes that received water from the Owens River. Hamman (1912a, p. 373) was the first to clearly state that the salines in Searles Lake were de- rived from the desiccation of a long-term overflow from Owens Lake, but both Gale (1913b, p. 886; 1914, p. 251, 252) and Free (1914, p. 39), whose fieldwork in the area was started around 1912, also indicated this interrelation. Of these papers, Gale’s report (1914) was the most complete. It is primarily a discussion of the strati- graphy, chemistry, and mineralogy of the Searles Lake deposit but it also presents the factual data on Owens, China, and Panamint Basins that support the geologic history of Searles Lake described herein. This factual support consists chiefly of a systematic record of the shorelines in the several basins as related to the elevations of their spillways. Except for descriptions of the tufas, the lake deposits exposed around the edge of Searles Valley were not described; the possible correlations between the subsurface deposits in Searles Valley and those of other basins or glaciated areas were discussed only indirectly. At the time of Gale’s investigations (1914), about 65 shallow core holes (mostly 15—25 m deep) and one deep hole had been drilled by private organizations, and his knowledge of the stratigraphy of the deposit was based on the results of their work. The shallower holes penetrated only those layers now known as the Overburden Mud and Upper Salt. Chemical analyses of core samples from these units were included in Gale’s report, but detailed lithologic and minerahogic logs were not made during drilling. The log of the eep hole, called the “old Searles deep well” and drilled near the west edge of the saline body (see fig. 3),Lis so generalized that little can be said of the deeper 5 line and mud layers in the deposit. . Between 1914 and 1952, in the course of commelrcial development, a wealth of data accumulated. Several new minerals were discovered and new analys s of brines and salines became available. Continued core drilling provided more and better information o the parts of the lake fill investigated by H. S. Gale an has revealed the presence of the deeper saline andmud layers (see Dyer, 1950, p. 41; Ryan, 1951, p. 447 and fig. 2). In 1952, as part of a study of borates, a new study of Searles Lake was started by the US. Geologicall Sur- vey. The core and analytical data previously obtained by the companies operating on the lake (the‘ the American Potash & Chemical Corp. and the Wes End Chemical Co.) were made available to the Surve and were used as a basis for a program of core drillin that was carried out between 1953 and 1955. Core logs were later published (Smith and Pratt, 1957; Haines, ‘1957, 1959). i In 1958, Flint and Gale described the subsurface stratigraphy of the Searles Lake evaporites on tl‘le ba- sis of the many cores obtained by the American Pot- ash & Chemical Corp. and the deep core obt ined during the Geological Survey’s program (Smit and Pratt, 1957). The names used in that paper for the ma- jor subsurface units were informal names tha had been in use for some years by company geologis s and engineers working on the deposit, and those names are used in the present report. l Later, after logs of 41 cores from the upper part of the deposit had been published by Haines (1957, 1959), subdivisions of two of the stratigraphic units described by Flint and Gale (1958) seemed justified and were proposed by Smith (1962). A description of the nonclastic mineral components and a summary of their distribution within these stratigraphic layers was later published by Smith and Haines (196 ). The stratigraphic framework established by Flint and Gale (1958) provided the basis for their own linter- pretations of ”C ages of the sediments and for the lat- er study by Stuiver (1964). The mineral associ tions described by Smith and Pratt (1957) and aines (1959) were used by Eugster and Smith (1965‘) in a study of chemical equilibrium relations in the deposit. Samples of the cores obtained by the Survey were 8 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. used in studies by Droste (1961) on clays, Hay and Moiola (1963) on authigenic silicates, and Leopold (1965) on fossil pollen. The stratigraphy and mineral- ogy described in these papers was later used to inter— pret the geologic history of the lake and its climatic implications (Smith, 1968, 1976). The deuterium con— centration in hydrated minerals of this suite allows es- timates of the temperatures of salt crystallization to be made (Smith, Friedman, and Matsuo, 1970). The ground water in Searles Valley was first studied by Thompson (1929, p. 177—182), who noted the un- usually low gradient of the water table near Searles Lake. Moyle (1969) subsequently compiled the much larger volume of well and spring data now available from the valley and confirmed Thompson’s earlier observations. These data led to the conclusions pre— sented in a report by Hardt, Moyle, and Dutcher (1972, fig. 10) that even before the activities of the chemical companies on Searles Lake, the hydrostatic head of the high-density brines in the salt body was greater than in the less saline waters in the surround- ing parts of the valley. This head means that for much or most of the time since Searles Lake desiccated, brine has migrated, though very slowly, from the cen- tral surface of the lake downward and toward its edges, thereby allowing surface waters to move down- ward and mix with the brines. ACKNOWLEDGMENTS This report would not have been possible without the complete cooperation of the companies holding land on the Searles Lake deposit. All gave permission for the Geological Survey to drill core holes on their holdings and to publish the factual results of its find- ings. Data from the past drilling and analytical re- cords of the American Potash and Chemical Corp. and the Stauffer Chemical Co. (and their successor, the Kerr—McGee Chemical Corp.) were made available to me, and written permission was granted to publish those data necessary for this report. Many individuals associated with those companies participated in helpful discussions and aided in com- piling data from Searles Lake. Special thanks are due D. S. Arnold, R. L. Cremer-Bornemann, L. J. Czel, F. J. Dluzak, H. S. Eastman, W. A. Gale, D. E. Garrett, F. C. Hohne, D. A. Holmes, J. F. Phillips, and F. J. Weishaupl, all then on the staff of the American Pot- ash and Chemical Corporation or the Kerr-McGee Chemical Corp., and P. Cortessis, C. F. Cowie, and L. E. Mannion, all then of the Stauffer Chemical Co. Colleagues with the Geological Survey who pro- vided special help in compiling this report include P. F. Irish, R. J. McLaughlin, J. D. O’Sullivan, and S. Walsh, who did many of the compilations and cal- culations; R. D. Allen, R. C. Erd, and Beth Madsen, who helped identify some of the uncommon minerals; and the USGS chemists cited in the tables of analyses. Company chemists were not identified on the records of their analyses. D. V. Haines helped compile a pre- liminary version of this report. This report is in part the result of a cooperative agreement with the State of California, Department of Natural Resources, Division of Mines and Geology. STRATIGRAPHY OF THE EVAPORITE DEPOSITS The term Searles Lake evaporites‘ is used in this re- port as an informal term for the sequence of salines and muds beneath the surface of Searles Lake. The stratigraphy of the Searles Lake evaporites described here, with the emphasis on the mineral composition, extent, and stratigraphic arrangement of the units, comes chiefly from published core logs (Smith and Pratt, 1957; Haines, 1957, 1959). Logs and chemical analyses of the Kerr-McGee Chemical Corp. (then the American Potash & Chemical Corp.) were examined as a check on the extent to which the published data apply to other parts of the deposit. With the permis- sion of the company, some of those data are included in this report. Knowledge of the Searles Lake evaporites is based on samples and logs of cores (fig. 3). The stratigraphy of the upper 120—150 ft (35—45 m)5 of the deposit was initially determined from 30 logs of cores obtained for the Geological Survey (Haines, 1959), and subse- quently applied to about 70 core logs made by the American Potash & Chemical Corp. Samples from several other cores were later obtained from the Kerr- McGee Chemical Corp. for special studies. The strati- graphy of the deeper deposits is determined mostly from the log of the core L-W-D (Smith and Pratt, 1957); six other cores have sampled parts of this inter- ‘Throughout this report the term “evaporites” is used as a nearly all-inclusive word for chemical sediments that precipitated initially from natural bodies of water. It in- cludes minerals such as calcite, aragonite, and dolomite that precipitated from nearly “fresh" waters, and minerals such as borax, trona, and halite that precipitated from “saline” waters. Solutions containing 1 percent dissolved solids fall near the boundary between “fresh" and “saline" waters as the term is used in this report. The terms “salt," “salts," and “salines” refer to mixtures of solid minerals that crystalized from saline waters; mono-mineralic bodies of halite, though called “salt” in many reports, are re- ferred to as “halite”. “In this report, the English system of units (inches, in.; feet, ft; miles, mi; pounds, lbs) has been used in some instances where the data being discussed are directly or indirect- ly derived from core logs, contour maps, or older publications that use these units; this facilitates relating this report to the original sources ofdata. When these data are gener- alized or used in calculations, they are converted and expressed in metric units (milli- meters, mm; centimeters, cm; meters, m; kilometers, km; kilograms, kg). STRATIGRAPHY OF THE EVAPORITE DEPOSITS NAVAL RESERVATION BOUNDARY WW; R. 43 E. 117°20’ R. 44 E. I i 6 5 )b/ 4 fl 2 1 6 5 9/ Pioneer 7/ Point / . _Gs—4o I 63—41 . GS-1' ‘ 7 8 9 .Gs-39 10 11 12 7 8 ‘ ‘7 . ‘31 | . 9" , ”B" MG-6 ,. , 3:9 (a i 18 16 )6\ /" 4 13 0° 18 17 . T / \/"" K ‘5 ' / .KK \ .GS—37 . 25 x_1. \ 5- MM / c EE \ ; 35°45'— / 55-33 'L-21 \ GS 3 1 I 68—36. .\ - 5 / Argus Uand . ,, _ (as—20 _ _ KM-a 4/19 20 6.3 2' . 21/'I I L U. ' . .N 5542- 7 20 ‘ / . / l/ 22 23 24 68—33 / / "Searles deep\\:vall” / . 65-34. 1 , “ I L-,/ x—2o KM-a 0 65—31. GS;2 ‘ l / 1 GS- 19 34 | / \ ' GS—28. - 30 29 fi" HH L12 2.7 51- 34, x 25 I \ 28 . . X '16 26 o .W, LW, and ‘ \ 1 x- 23 L -w - D \ I 1 1 NN 68— 16 135-14 ’ ' 115-15 68— 13 \ __‘ r‘ \ ‘I \/ .GS—17 23%;? ‘ I 31 32 . 33 34 "’are 39.5 36 32 s - 25. 68-18 L30 0'99\.5—2 W} i \ *% 'css-g _ 68—10 _L31 - .129 4e}_\ 254 I \. / \ 65-8 6—75 9’r6}\ L181 ‘| Westend 1; 01’ \_‘ . 6 \ 5 4 3 95'” 2 5'31 1 I t m RID MOUNT/UN; ' _ GS-5 . l 9 1o 12 § 28 7 o 8 ‘ . T. H ‘ 26 W S. \ _23. 65—4. _| 15 14 .GS-7 18 63—3. 17 ‘ ' 13 ‘ 1 K / ‘ I 22 23 M/‘{ 20 ‘ i 0 1 2 MILES ‘ 1—__—'—l__r_.—'_J 0 1 2 3 KILOMETERS FIGURE 3.—Core holes on Searles Lake discussed in text. 10 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. val: core L—30, described in this report, extends to 160 ft (49 m); core 254, described here with permission of Kerr-McGee Chemical Corp., extends to about 260 ft (80 m); cores KK, SL—34, and 8—2 (Flint and Gale, 1958, figs. 3, 4) extend to about 300 ft (90 m); a core described by Gale (1914, p 289—290) extends to 628 ft (191 m). Flint and Gale (1958) described the major stratigra- phic units of the deposit and applied the following in- formal names: Overburden Mud, Upper Salt, Parting Mud, Lower Salt, Bottom Mud, and Mixed Layer. Smith (1962) subdivided the Lower Salt into 13 units and the Mixed Layer into 6 units (fig. 4). The stratigraphic subdivisions of the Mixed Layer are thick units distinguished by changes in evaporite mineral content that indicate significant differences in the chemical nature of the lake in which they were deposited. The Bottom Mud, Lower Salt, Parting Mud, and Upper Salt are relatively homogeneous units defined on the basis of lithology. Contacts be- tween them are sharp and considered to be about the same age throughout the deposit. The Overburden Mud, which is more heterogeneous, consists of inter- bedded mud and saline layers. In the central part of the deposit, its basal contact is gradational and con- sists of a zone having an upward increase in the per- centage of Clastic-rich mud layers and a decrease in saline layers; around the edges, the contact is at the base of a zone of solid mud that includes near-shore equivalents of horizons in the Upper Salt as well as in the Overburden Mud. In studying cores, it is generally easy to separate the mud units from the saline units. The muds6 are mostly dark green to brown, soft, and appear nonporous. They consist chiefly of chemical precipitates made up of Ca, Na, and Mg combined with C0,. The major minerals are fine-grained aragonite and dolomite and fine- or coarse-grained gaylussite and pirssonite (the names and chemical compositions of the nonclastic minerals in the Searles Lake evaporites are given in table 1). A few mud layers have large percentages of clay-sized halite, or small percentages of fine- or coarse-grained borax or northupite. Galeite, schairer- ite, and tychite occur in traces. Authigenic silicates ”The term “muds” is used throughout this report for the layers having the physical properties listed here. By most conventional sediment terminologies, these muds would be designated as marls rich in organic material. The term “mud“ is preferred, however. because it conveys the concept of moist plasticity, one of the striking properties of these materials, better than the conventional sediment term, and because it is an established part of the local terminology for both the sediment itself and the stratigraphic units composed of it. such as K-feldspar, analcime, phillipsite, and searle- site are locally abundant. Clastic silt and clay, and partially decomposed organic material, are always present but subordinate. Almost all mineral identifi- cations of fine-grained components in the muds are by X-ray diffraction techniques. The salines are mostly white to dark gray, hard, and porous. They consist chiefly of precipitates made up of Na, K, and Mg combined with C0,, HC03, 80,, C1, or B,O,. The major minerals are coarse-grained halite, trona, hanksite, burkeite, borax, nahcolite, mirabilite, thenardite, northupite, and aphthitalite (glaserite). Small quantities of sulfohalite, teepleite, and tincal- conite occur locally. Mineral identifications of saline minerals are by visual inspection (of large crystals) and X-ray diffraction methods. The identification and correlation of mud and sa— line units in cores is based upon multiple criteria. Thickness is the most reliable single parameter, but mineralogy, crystal size and habit, bedding character, and the number and positions of thin layers of muds in salines (or, rarely, salines in muds) provide sup- porting evidence. The Upper Salt, Parting Mud, and units within the Lower Salt generally maintain similar thickness over an extent of a kilometer or so, and these thicknesses form a pattern that can be matched with confidence in nearby cores. The mud beds in these se- TABLE 1.—N0nclastic minerals in the Searles Lake evaporites Mineral Composition Adularia ______________ KAlSi30, Analcime ______________ NaAlSiZO6 . H20 Aphthitalite (glaserite) ______ K,Na(SO,)2 Aragonite _____________ CaCO3 Borax ________________ Na2B407 - 10H20 Burkeite ______________ 2NaZSO, - Na2C03 Calcite _______________ CaCO3 Dolomite ______________ CaMg(C03)2 Galeite _______________ Na,SO, - Na(F,Cl) Gaylussite _____________ CaCOa - Na2CO3 - 5H20 Halite _______________ NaCl Hanksite ______________ 9Na2SO, - 2Na2C03 - KCl Mirabilite _____________ NaZSO, ~ 10H20 Nahcolite _____________ NaHCO3 Northupite ____________ Na2C03 ~ MgCOa ~ NaCl Phillipsite ______________ KCa(Al,Si,O,6) - 6HZO Pirssonite _____________ CaCO3 ' Na2C03 . 2H20 Schairerite _____________ NaZSO4 . Na(F,Cl) Searlesite _____________ NaBSiZO6 . H2O Sulfohalite _____________ 2NaZSO4 ‘ NaCl ~ NaF Teepleite ______________ Na,B.,O4 ' 2NaCl ~ 4H20 Thenardite ____________ NaZSO, Tincalconite ____________ NazB,O7 - SHZO Trona _______________ Na2C03 - NaHCO3 - 2H2O Tychite _______________ 2Na2CO3 - 2MgCO3 - NaZSO, fix STRATIGRAPHY OF THE EVAPORITE DEPOSITS DEPTH, UNIT IN METERS Overburden Mud Upper Salt — Parting Mud ...~...~.---u m: Lower Salt Bottom Mud Mixed Layer O 15 20 25 100 150 200 250 267 LITHOLOGY Interbedded halite and brown mud in central facies grading edgeward to brown mt'd Halite, trona, hanksite, and borax grading downward and edgeward to trona Green mud containing gaylussite and pirssonite Interbedded green muds containing gaylussite, pirssonite, borax, and northupite, and salines consisting of trona, halite, burkeite, and borax Green mud containing gaylussite Trona and nahcolite, some interbedded brown mud containing gaylussite Trona, nahcolite, and halite, some interbedded brown mud containing gaylussite Halite and trona, some interbedded brown mud containing pirssonite Brown mud containing pirssonite, some interbedded trona, halite, and nahcolite Green to brown muds containing pirssonite, some interbedded halite Green mud containing pirssonite FIGURE 4.—Summary of stratigraphic units in Searles Lake evaporite sequence. 11 12 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. quences are more constant in thickness than the saline beds, and for this reason provide more reliable mark- ers; toward the edges of the deposit, the saline beds pinch out, and the overlying and underlying mud beds are not separable. Detailed correlations have not been attempted between sections of the Bottom Mud and Mixed Layer. Very few cores penetrate complete sec- tions of the Bottom Mud; only one has been examined carefully during this study. A schematic cross section of the fill in Searles basin is shown in figure 5. The bedrock profile is based on Mabey (1956). The top 45 m of the fill shown in this diagram is based on the data of Haines (1959), Smith (1962), Smith and Haines (1964), and this report. The section between 45 m and 267 m is based on the core log L-W-D from the central part of the basin (Smith and Pratt, 1957) and the “old Searles deep well” from near the west edge (Gale, 1914, p. 275, 289—290); these two logs are so different that correlations are not pos- sible, although the older log reports saline minerals from enough horizons to indicate that some saline lay- ers extend at least that far west. The cross sections of the upper part of the Searles Lake evaporites, plate 1, are based on core holes de- scribed by Haines (1959) and on subsurface data adapted from the isopach maps presented in the fol- lowing sections (figs. 8—20). Most contacts are drawn at the boundaries between mud and saline material; the base of the Overburden Mud is drawn at the lowermost thick lens of mud in the part of the core considered to represent the Overburden Mud. The mud layers that can be correlated with the greatest confidence are those near the center of the deposit where the saline layers that separate them are the most distinct and thickest. Near the edges, some saline layers are missing and successive mud layers are combined. Since most of the saline layers have about the same areal extent and position, both the saline units and the mud layers they separate can be recog- nized over about the same area. W E 4000 Argus Range Slate Range — 1500 2000 Searles Lake 1:- 100° \ |:—\ /— 500 SEA SEA LEVEL ’ ‘ LEVEL 2000 *‘ 50° 4000 NO VERTICAL EXAGGERATION —1000 1500 (I) II ’— 3 — 1500 E LIJ LL E E 2 2‘ 4000 —\ E O a e > — 1000 ‘1 fl 3000 5 Lu Upper Salt and _l Overburden Mud L” Parting Mud Lower Salt 2000 ”Old Searles Bottom deep well” '— 500 1000 SEA _ LEVEL _ LEE/2L 1000 VERTICAL EXAGGERATION x 5 L / _ — 50° 2000 ‘ PM); \ / FIGURE 5.—Cross-section of Searles Valley showing bedrock profile (dashed where position uncertain; after Mabey, 1956, fig. 6) and stratigraphy of upper part (top 45 m) of Cenozoic fill (from data of Haines, 1959; Smith, 1962; Smith and Haines 1964; and this report). Profile is east-west and through central part of lake, along section line one mile north of boundary between T. 25 S. and T. 26 S. (see fig. 3). Two deep core holes are projected to this section: “old Searles deep well” (Gale, 1914, p. 289, 290) is 2 km north of this section line, L—W—D (Smith and Pratt, 1957, p. 25—51) 0.8 km north of it. It ‘1 MIXED LAYER 13 MIXED LAYER All the interbedded salines and lake muds below the Bottom Mud were included in a single stratigraphic unit by Flint and Gale (1958, p. 694) and designated the Mixed Layer. The explored part of this unit, 219 ft—875 ft (668—2667 m) in core L-W—D, was previ- ously inferred by Smith (1962) to be of Illinoian and Sangamon age, but the present paper tentatively re- vises this correlation, making the Mixed Layer of Yar- mouth age and possibly Kansan. Saline layers have an aggregate thickness estimated to be near 280 ft (85 m). This thickness is slightly less than half the entire se- quence, but as most of the salines are in the upper 60 percent of it, that part is dominated by them. The sa- lines are mostly trona, nahcolite, and halite that form individual beds several meters thick; small to trace amounts of burkeite, northupite, sulfohalite, thenar- dite, and tychite are also found. The muds consist of a variable percentage of megascopic crystals of gaylus- site or pirssonite, embedded in a silt— to clay-sized ma- trix of gaylussite, pirssonite, dolomite, calcite, northupite, K-feldspar, analcime, searlesite, clastic silicates, and organic material. The Mixed Layer has been divided into six strati- graphic units (Smith, 1962) whose boundaries sepa— rate deposits characterized by different suites of evaporite minerals that indicate significant changes in the chemical nature of the depositing lake. Relative to the overlying units of the Searles Lake evaporites, these Mixed Layer units are thicker and the contacts between them less sharp because the long periods of distinctive chemical sedimentation that they record changed gradually over a long period of time. AREAL EXTENT AND THICKNESS The interbedded sequence of salines and lake muds that constitute the Mixed Layer grade laterally into alluvial deposits that crop out on the valley sides, but the transition zone between them is not exposed and has not been observed in cores; the areal extent of the lacustrine facies therefore can only be estimated. The three deep cores described by Flint and Gale (1958) and the L-W-D‘core (Smith and Pratt, 1957) are all in the central part of the basin, and it is clear that all four cores are composed entirely of lake deposits. The log of the “‘old Searles deep well” (Gale, 1914), near the west edge of the present lake (fig. 5), notes halite, thenardite, trona, and northupite at several horizons within the 628 ft (191 m) of fill tested; although saline layers are apparently few and mud layers predomi- nate, the lower part of that log describes nothing that could be interpreted as the toe of an alluvial fan. The lacustrine deposits of the Mixed Layer appear to be at least as extensive as the present playa lake (100 kmz), but outcrops of subaerial gravels considered to be contemporaneous with the Mixed Layer limit the pos- sible extent of most lacustrine deposits to less than 400 kmz. The total thickness of the Mixed Layer has not been determined. Core L- W- D includes 655 ft (200 ) of section assigned to this unit. Three other core holes in the central part ofthe deposit (Flint and Gale, 1958) penetrate the top 65, 100, and 135 ft (20, 30, and 41 m) of this layer. The “old Searles deep well” penetrated- the lateral equivalent of the unit, and, because f its position on the flanks of the bedrock basin (f1; 5), may have included the lateral equivalent of deep r ho- rizons than did L-W—D in the center of the basin. The deepest samples in the old well consist mostly of clay containing halite and calcite; these lithologies indicate that the base of the lake deposits were not reached. MINERAL COMPOSITION AND LITHOLOGY i The entire known thickness of the Mixed Layer consists of deposits formed in a moderately to highly saline lake. These have been divided into six zones on the basis of evaporite mineralogy, each zone repre- senting a period when the lake waters, on the average, had a characteristic salinity or composition. The ones of highest salinity are represented by saline 1 ers. Decreases in salinity correspond to increases in the percentage of material logged as mud, but the mud layers themselves are actually composed mostly of evaporite minerals that indicate something of the lake’s salinity and composition. 1 The estimated volume percentages of the t’ain components in the Mixed Layer of core L- -D (Smith and Pratt, 1957) are plotted quantitatively (pl. 2A) such that the horizontal sum of all percentages to- tals 100. The core that was lost (52 percent) is graphi- cally interpolated by diagonal lines that connect the lithologies plotted for recovered core above and b low. The elastic minerals in the mud (col. 1) and th: few beds of sand (col. 2) are mostly quartz, feldspar, and clay, although biotite and amphibole are common. The clastic quartz and feldspar have not been studied in detail, but X-ray patterns of bulk core samples in- dicate that they are commonly present in about equal quantities. Hay and Moiola (1963, p. 315—320) note that the grains are generally subround to round and pitted or frosted. Cementing materials include calcite, halite, and searlesite. The heavy minerals in the clas- tic fraction were studied by Gan (1961, table 3). In the Mixed Layer, he found a small but generally persis- tent percentage of amphibole, opaque mineral, mica, 14 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. epidote, apatite, sphene, pyroxene, zircon, and topaz fragments; in a few samples he found rutile, cassiter- ite, olivine, anatase, corundum, tourmaline, and spinal fragments. The clay minerals in the Mixed Layer were studied by Droste (1961) who shows (fig. 3) that the illite:- montmorillonitezchlorite or kaolinite ratios average about 6:3:1. Estimates by Hay and Moiola (1963, table 1) confirm this ratio. Droste (1961) used the pattern of stratigraphic variation in these ratios to correlate the deposits in Searles Lake with those in the other basins that once contained lakes in the same chain. The field evidence now shows that much of the clastic sediment in each basin was derived locally, the observed vari- ations are likelier to be the result of local factors such as the position of the shoreline of the lake, the pattern of streams relative to the site of the core hole, the cir— culation pattern of strong currents within the lake, or the nature of local rock-weathering processes. Of the fine-grained authigenic silica minerals found in the Mixed Layer, and the searlesite(?) reported at 248 ft (75.6 m) by Smith and Pratt (1957 ), plotted in column 4, K-feldspar is by far the most common; it is estimated by Hay and Moiola (1963) to make up 10— 20 percent of many layers and as much as 50 percent of some. Analcime may be slightly more common that K-feldspar in units A and B, but the reverse is clearly true in units C, D, E, and F. Searlesite was reported by Smith and Pratt (1957) from the middle of unit A at 248 ft (75.6 m) and by Hay and Moiola (1963) from units E and F. These three authigenic silicates occur in the overlying Bottom Mud, but only analcime and phillipsite are found in the younger units above it. This stratigraphic distribution suggests either that less time is needed for the formation of analcime and phillipsite than for searlesite and K-feldspar, or that analcime and phillipsite are here precursors to K-feld- spar and possibly searlesite as is found elsewhere (Sheppard and Gude, 1968, p. 35, 36). Eighteen of the X-ray determinations of the fine— grained carbonates (col. 5) were made by Hay and Moiola (1963; R.L. Hay, written commun., 1964) on samples selected for study of authigenic silicates; the rest were made during the present study. Although the samples studied by X-ray are widely spaced, the following observations are probably applicable to the unstudied segments of this unit: northupite, dolomite, pirssonite, and calcite are the only carbonates detect- ed in this fraction; dolomite is more common than cal- cite; and northupite is most common in zones containing saline layers. Gaylussite and aragonite,7 ”The light-colored laminae in this unit reported by Smith and Haines (1964) to be aragonite are now known to consist of other carbonate minerals. though common in overlying units, were not found in ‘ the fine-grained fraction of the Mixed Layer. The percentages of coarse-grained evaporite miner- als are plotted in columns 6 to 10 of plate 2A. Gaylus- site and pirssonite are relatively insoluble Na-Ca carbonates, and their percentages are shown in col- umns 6 and 7. The percentages of more soluble eva- porite minerals—the salines—are plotted in columns 8, 9, and 10. About 45 beds of trona and nahcolite are shown, and these total about 45 ft (14 m) thick. Nearly the same number of beds of halite are shown, but they total about 95 ft (29 m). As about half the core was not recovered, actual thickness of salt beds are likely to be about twice these figures, meaning that a total of about 280 ft (86 m) of salts occur in the Mixed Layer. In terms of the overall composition of the 655-foot (200 m) section of the Mixed Layer discussed here, a little less than 15 percent is trona plus nahcolite and 30 percent is halite. Chemical analyses of the water-soluble fraction of seven samples of the Mixed Layer are given in table 2. Six of these samples are of salines. These analyses confirm that trona, nahcolite, and halite make up most of the saline portions of the core as reported in the log of L-W-D; they also show that saline minerals containing K and 3,07 are nearly absent,” and sulfate minerals are subordinate to rare. The seventh sample is composed of silt, and the analysis of its water-solu- ble fraction indicates the presence of small quantities of carbonates, chlorides, and sulfates. Generalized de- scriptions of the individual stratigraphic units follow: Unit A.—Saline layers composed of trona and nah- colite predominate in unit A. Halite is missing (except near the basal contact), and this characteristic distin- guishes unit A from unit B. The mud layers are mostly dark yellowish brown and contain megascopic gaylus- site and smaller quantities of pirssonite. Locally there are detectable quantities of megascopic northupite and tychite and microscopic analcime, searlesite(?), and dolomite. A thin bed of basaltic or andesitic vol- canic ash lies near the top. Unit B.—Saline layers again predominate in unit B; they consist of both trona and halite with smaller quanitites of nahcolite. Some of the mud layers are dark yellowish brown, others dark olive green. They contain megascopic gaylussite, traces of megascopic pirssonite, sulfohalite, and northupite, and local con- centrations of microscopic crystals of analcime, dolo- mite, K-feldspar, and northupite. sCore drilling was done with brine pumped from the top of hole W; table 16 shows the uppermost sample of brine to contain 0.70 percent 8.0, and 1.39 percent K. Some of the components in this brine must have adhered to the cores that were later analyzed, and some or all of the detected B.O7 and K could have come from this source. BOTTOM MUD TABLE 2.—Chemical analyses of core samples from the Mixed Layer l [Analyses, by Henry Kramer and Sol Berman, are of material dissolved in boiling water. Samples from core L-W-D described by Smith and Pratt (1957)] Depth Weight percent l (ft) Unit Litholugy Total Na K Cl SO. COJ HCO, 8,01 water soluble 464.0—465.3 _ _ _ _ C _____ Trona with halite, 32.2 0.09 11.9 3.4 18.6 20.1 0.08 86.4 some burkeite. ‘ 580.0—583.0 _ _ _ _ C _____ Halite, trona, 38.2 .17 47.4 8.9 2.4 1.0 .12 98.2 and thenardite. 594.0—597.6 _ _ _ _ C _____ Trona and mud. 30.0 .30 5.3 2.0 21.3 23.2 .23 82.3 6403—6500 _ _ _ _ D ______ Halite, some trona 38.8 .02 56.9 .5 .9 1.4 .02 98.5 and pirssonite l mud. 707.0—710.0 _ _ _ _ E _____ Halite and mud. 33.2 .06 47.9 .4 2.0 .08 .06 83.7 7 22.8—730.0 _ _ __ _ E _____ Halite and 38.3 .04 56.5 .3 1.0 .08 .02 96b pirssonite mud. 8650—8680 _ _ _ _ F _____ Silt cemented by 4.1 .19 1.05 1.2 3.7 .08 .21 10.5 pirssonite and searleSIte. Unit C.—Unit C consists mostly of halite beds, but contains some relatively thin beds of trona. The pre- ponderance of halite over other saline minerals and the virtual lack of gaylussite distinguish this unit from units A and B above it. Traces of nahcolite, burkeite, sulfohalite, and thenardite occur locally. The subordi- nate mud layers are mostly yellowish to orange brown; they contain megascopic and microscopic crystals of pirssonite and traces of gaylussite, northupite, K-feld- spar, analcime, and dolomite. Unit D.—In unit D, pirssonite-bearing yellowish— or greenish-brown mud layers are more common than sa- line layers, and this distinguishes these deposits from those of unit C above. Most mud layers contain only pirssonite; some contain a little northupite, calcite, and dolomite. Gaylussite is absent from the muds of this and all deeper units. The saline layers consist of about equal percentages of halite and trona, and traces of nahcolite, sulfohalite, and tychite. A 1-cm bed of devitrified glass or andesitic tuff lies near the middle of this unit. Unit E.—About two-thirds of unit E consists of yel- lowish-green mud containing megascopic crystals of pirssonite and locally northupite and sulfohalite. The saline layers are composed of halite that locally con- tains traces of included northupite and sulfohalite; trona and nahcolite are absent from the saline layers, and this distinguishes this unit from unit D above it. A thin bed of devitrified andesitic(?) volcanic ash that has been partly altered to K-feldspar, analcime, and searlesite, occurs near the upper contact. Microscopic crystals of analcime, K-feldspar, searlesite, northu— pite, dolomite, and calcite make up small to major percentages of some beds in the unit. Unit F.—Unit F consists chiefly of mud containing pirssonite. Megascopic crystals of northupite and mi- croscopic crystals of K-feldspar and dolomite are com- mon to abundant, crystals of analcime, searlesit‘e, and calcite less common. A few thin beds of fibrousltrona and cubic halite are found, but the rarity of such beds distinguishes unit F from other units in the Mixed Layer. 1 Brine samples from the Mixed Layer were not col- lected for analysis during the drilling of core hole L-W-D because they were contaminated during cor- ing by the surface brines forced down into these deep layers. However, the general composition of the rines that permeate the saline bodies of the Mixed Layer can be inferred from the mineral phases present. The saline layers are composed predominantly of minerals made up of Na, C0,, HC03, and Cl. Minerals contain- ing Mg and SO, are rare. Saline minerals contai ing B and K have not been noted although the autliigenic silicates searlesite and K-feldspar do contain these components. As the brines in contact with these min- erals appear to be in equilibrium, limits can be placed on their chemical compositions by use of phase dia- grams (Smith and Haines, 1964, fig. 14) or relative chemical activity diagrams (Eugster and Smith, 1965). Estimates based on these data indicate, in qualitative terms, that the brines in the Mixed Layer hawe high percentages of Na, low to high percentages of Cl, H003, and 003, (with the ratio of HCO3/CO3 mostly low), and very low percentages of K, 80,, and B. The chemical activity of H20 is higher in unit A, and1 possi- bly in unit B, than in older units; the activity pf CO2 varies but may be higher in units B, C, and the‘ upper part of D than in other units. \ BOTTOM MUD l The Bottom Mud is a unit deposited by a sdries of perennial lakes that occupied Searles Valley th ough- out most of early Wisconsin, Sangamon, and I1 inoian l 16 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. times. It consists largely of dark-green to black mud that contains megascopic crystals of gaylussite. The muds consist of fine-grained carbonates and other evaporite minerals, authigenic silicates, clastic sili- cates, and partially decomposed organic debris. Thin lenticular beds of nahcolite and mirabilite are found at several horizons, and a few thin layers of trona, thenardite, and borax have been noted near the top. These represent low stands of the lake but not desic- cation. AREAL EXTENT AND THICKNESS The Bottom Mud has been penetrated by only a few test holes. Core hole L-W and its downward extension L-W-D (Smith and Pratt, 1957, p. 25—51) is in the east-central part of the deposit. Core holes S—2, KK, and SL-34, which extend to depths of about 300 ft (90 m), penetrated the entire unit (Flint and Gale, 1958, figs. 2, 3, and 4); their core logs are not published, but two were examined during this study. The “old Searles deep well,” drilled near the west edge of the deposit (Gale, 1914, p. 289), penetrated the Bottom Mud, but its upper and lower limits cannot be identi- fied in the log. Core hole 254, drilled in the southeast part of the lake by the Kerr-McGee Chemical Corp., is described here with their written permission. In most parts of the deposit, the thickness of the Bottom Mud is about 30 m. Flint and Gale (1958, figs. 3 and 4) show Bottom Mud thicknesses in cores SL— 34, 8—2, and KK ranging from about 65 to 90 ft (20—27 m), but application of the criteria used by Smith (1962) to the original logs of SL—34 and 8—2 indicates Bottom Mud thicknesses in these cores of about 94 and 100 ft (28 and 30 m) respectively. In core L-W-D, the Bottom Mud is 101 ft (31 m) thick; in core 254, 97 ft (29 m) thick. Near the edge of the lake and outside of it, the areal variation in thickness cannot be mea- sured; unpublished mapping shows that the layer was at one time continuous with sections of coarser lake sediments now preserved as remnants around the sides of the valley. MINERAL COMPOSITION AND LITHOLOGY The Bottom Mud consists mostly of mud that con- tains megascopic gaylussite crystals. About a quarter of the unit has faint or widely spaced laminae or thin beds, but most of it is massive. The mud is clay- to silt- sized material that is moist and plastic. It is com- posed, in major to minor percentages, of micro- crystalline gaylussite, dolomite, aragonite, calcite, an- alcime, searlesite, authigenic K-feldspar, halite, and elastic minerals plus a few percent organic material. Thin lenticular beds of nahcolite and mirabilite are found at several horizons, and small amounts of trona, burkeite, and borax occur locally. The mirabilite and borax dehydrated to thenardite and tincalconite prior to logging and X-ray identification. Plate 28 shows graphically the distribution of saline layers in three cores of the Bottom Mud, and the min- eral composition of the mud fraction. Core 254 repre- sents samples collected from the entire thickness of Bottom Mud, core L—30 and core L-W—D samples from the upper third and lower two thirds. The strati- graphic intervals represented by the two partial cores probably overlap slightly, and the beds of nahcolite at about 152 ft (46.3 m) in both are probably correlative and equivalent to the one at 155 ft (47.2 m) in core 254. The bed of nahcolite at 134 ft (40.8 m) in core 254 may be equivalent to one or both of the beds of mira- bilite9 reported in core L-W. The other nahcolite beds are not found in the correlative cores, either because not present or not recovered during coring. The top of the Bottom Mud is at 123.4, 116.8, and 120.7 ft (37.61, 35.60, and 36.79 m) in the cores plotted (p1. ZB). The base of the Bottom Mud in core 254 is at 226.8 ft (69.13 m) and in L-W-D at 219.3 ft (66.84 m). The top of the unit is placed at the base of the salts that make up unit 8—1 of the Lower Salt, and the base at the top of the uppermost saline layer in unit A of the Mixed Layer, a closely spaced series of nahcolite beds. The Bottom Mud, as represented by the cores plot- ted, consists mostly of mud containing as much as 70 percent gaylussite crystals. The crystals are subhedral to anhedral, and generally cut across bedding, demon- strating that they grew after burial (see Smith and Haines, 1964, fig. 15; Eugster and Smith, 1965, pl. 1). Crystal sizes range from a fraction of a millimeter to 20 mm, and inclusions of mud similar to the host ma- terial occur in most crystals. The largest crystals com- monly are found in the top few meters. Beds and lenses of highly soluble saline minerals are critical in reconstructing the history of the lake inas- much as they represent periods of low lake levels. A thin bed of borax occurs about 3 ft (1 m) below the top of the Bottom Mud in the core L—30, and thin discon- tinuous beds and pods of northupite, borax, trona, nahcolite, and thenardite are found in this zone in cores GS—8, 10, 11, 15, 16, 17, 18, 19, and 27 (Haines, 1957, 1959). Beds of mirabilite were reported 10—20 ft (36 m) below the top contact in core L—W (Smith and Pratt, 1957, p. 30) and traces of mirabilite and borax were found in core 254 although none were detected in this zone in core L—30. In cores shown in figure 9, nah- ‘Samples of the mirabilite reported in the log of core L-W were not available for con- firmation by X-ray diffraction. However, chemical analyses and descriptions in other logs of the rapid dehydration character of the minerals that constitute saline layers in the Bottom Mud show that mirabilite is probably present in some parts of the unit. BOTTOM MUD 17 colite forms prominent beds 35—45 ft (10—14 m) below the top, and traces of mirabilite and borax were found in core 254. Core 254 penetrated several beds of nah- colite at depths 60-70 ft (18—21 m) below the top, and core L-W-D contained nodules of trona and borax at this depth. Two thin beds of nahcolite occur near the base of the unit in core L-W-D, 1—4 ft (0.3—1.2 m) above several thicker beds of nahcolite assigned to the Mixed Layer. The mineralogy of the fine-grained fraction of the segments logged as mud in cores 254, L—30, and L—W-D was studied by X-ray diffraction and found to consist chiefly of carbonate minerals, halite, and auth- igenic silicates. Clastic minerals are generally subordi- nate. Figure 9 shows the mineral composition of the samples studied. Determinations on 10 of the X-rayed samples from core. L-W-D are by Hay and Moiola (1963, table 1, R. L. Hay, written commun., 1964); the remaining determinations from L-W-D and all the de- terminations from cores 254 and L—30 are products of this study. Although the abundances of mineral com— ponents are only relative, their presence or near-ab- sence is generally certain, and the indicated changes in relative concentrations for any component in a giv- en core are probably reliable. Gaylussite (G, on pl. 2B) is a component in about 90 percent of the samples. Except for the top few meters, its abundance tends to be less in the upper third of the unit, the zone containing the largest amount of calcite (C). Dolomite (D) is a component in half to two-thirds of the unit, but it may be slightly less abundant in the zones containing salines. Aragonite (A) is found in samples from the upper 20 ft (6 m), but it has not been detected below those depths in this unit or in underly- ing deposits. Analcime (An) is concentrated in two zones, one about 40 ft (12 m) thick near the middle of the unit and one about 10 ft (3 m) thick at the base. A concentration of another zeolite, phillipsite, is reported by Hay and Moiola (1963, table 1) from a tuff bed 8 ft (2.4 m) below the top of the Bottom Mud in core GS—2. Searlesite (81) is detected in significant amounts in a 10-ft (3 In) zone a little below the middle of the unit. Monoclinic K-feldspar (K) that is pre- sumed to be authigenic is present in detectable con- centrations in three zones that lie above the main saline beds, and its abundance is roughly proportional to the percentage of Clastic minerals (Cm); possibly some of the monoclinic feldspar is clastic rather than authigenic, but its diffraction patterns resemble those described by Hay and Moiola (1963, fig. 3) and by Sheppard and Gude (1968, fig. 2; 1969, fig. 2) from authigenic material. The halite (H) concentration may partially reflect the amount of NaCl—saturated interstitial brines in the core sample which dried be- fore X-raying rather than crystalline material in the original core, but crystalline material can be shown to exist in the Parting Mud and may be present in these samples also; its concentration may increase toward the top of the unit. The small amounts of other saline materials (Sx) may also come from the dried brines, but where shown present in minor amounts, the min- erals were observed as crystals. The concentration of Clastic minerals (Cm) seems highest in the zones de- posited immediately after the main saline layers were deposited, a time when the lakes were smaller and the shores nearer the center of the basin. Traces of microscopic crystals of pyrite were found in greenish-gray silts from a depth of 152.3 ft (46.4 m) in core L-W-D. None were found in four other sam- ples of similar-appearing mud from cores L—30 and L-W-D, which were studied in comparable detail. CHEMICAL COMPOSITION OF THE BOTTOM MUD Complete chemical analyses of samples from the Bottom Mud are not available. Determinations of the acid-insoluble percentage of 41 samples from the Bot- tom Mud (fig. 6) range from 11 to 65; the average of all determinations is 31. The acid-insoluble fraction in- cludes clastic and authigenic silicates (plus the much smaller organic fraction); the fraction that dissolved in the acid or water consists of Na, Ca, and Mg carbon- ates plus any salts. As each portion contains several components, the variations cannot be attributed to in- creases or decreases in any one. Comparing these data with the X-ray data plotted for this core (pl. 28) shows that the zones containing the highest percent- ages of acid-insoluble material are those containing the highest percentages of authigenic silicates, chiefly analcime (138—176 ft, 219—227 ft) and K-feldspar (176—196 ft). Clastic mineral percentages tend to be high in these zones, which presumably originally in- cluded much higher concentrations of elastic silicates that provided the Si and Al now contained in the authigenic minerals. In the parts of the core that do not contain authigenic silicates in abundance, the per- centages of acid-insoluble material are mostly be- tween 10 and 30 and average about 20. The mineral components plotted in figure 9 provide a more detailed estimate of the chemical composi- tion of the unit. Megascopic crystals of gaylussite (Na2COa ' CaCO3 - 5H20) were estimated visually to constitute about 10 percent of core L—30 and 55 per- cent of L-W-D. The finely crystalline gaylussite in the mud determined from the X-ray data appears to be more abundant and consistently present than any other component. Finely crystalline dolomite (CaCO3 - 18 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. MgCO,) appears to be next most abundant; halite (NaCl) is third. Less persistent and abundant quanti- ties of analcime (NaAlSiZO, - H20), authigenic K-feld- spar (KAlSi,O,), and calcite (CaCO,) are indicated. Relatively minor amounts of searlesite (NaBSi20, - H20), aragonite (CaCO,), trona (NaHCO, - Na,CO, - 2H20), burkeite (2Na,SO4 - Na,CO,), and borax (Na,B,O7 - 10H20) are present. Thin beds of nahcolite (NaHCO,) and mirabilite (Na,SO, - 10H20) make up 5-10 percent of the unit. This balance of mineral components shows that the authigenic and evaporite minerals in the mud portions of the Bottom Mud are predominantly made up of Ca, Na, Mg, and C0,, with smaller quantities of Cl, and still smaller quantities of Si, K, B, and HCO,. The salts are dominat 100 93 22 M-G ______ > 100 1 1 6 27 M—5 ______ > 100 28 7 M—4 ______ > 100 59 14 M—3 ______ > 100 57 13 M—2 ______ > 100 74 17 Total ______ — 427 100 The contouring of these data is a subjective process, and in some areas the actual distribution of thick- nesses may be quite different from that shown. This is most likely to be true near the edges, where control points are few. Furthermore, compaction of material or core losses during drilling makes some reported thicknesses uncertain over a range of several tenths of a foot. Many parts of the deposit have closely spaced control points that are fairly consistent; it seems prob- able that the thicknesses represented by the contours in these areas are essentially correct. The volumes of saline units were calculated from these isopach maps. Planimeter measurements were first made of the areas bounded by successive con- tours. The area between each of the crudely concen- tric contours was then calculated, the volume of the vertical zone beneath each area was computed, and the volumes of all zones were added together. This cal- culation, expressed as a formula, is as follows: V = h1(Ao — A1) + h2(A1 _A2) + he (A2 — A3) where V is total volume, A0, A], A2, and A3 are areas bounded by the zero, first, second, and third contours, and h,, hz, and h3 are the midpoints between the zero and first, first and second, and second and third contours (so that maps with a 1-ft contour interval thus have h1 = 0.5, h2 = 1.5,h3 = 2.5, etc.). The volumes obtained in this way, given in table 3, are a few percent larger than if calculated from formulas for the prismoid or frustrum of a cone”, but this tech— nique is preferred because the volume of each vertical zone is subsequently used to estimate mineral zona- tion and bulk mineral composition of the unit. MUD UNITS The six mud units that separate the salt layers in the Lower Salt extend beyond the edges of those lay- ers. Their lateral equivalents crop out in some parts of the valley, mostly a kilometer or more outside of the limits of the salts. Within the cored area, though, the individual mud layers can be identified in virtually every core, and their variations in thickness plotted as isopach maps (figs. 15—20). The contours are terminated at an arbitrary boundary which approxi- mates the limits of the salt layers. Volume measure- ments have been made in the same manner as the measurements on the salt layers; the results are given in table 3. ”For example, the volume of unit 8-7 by the technique used is calculated to be 62.3)(106 ' m“. Applying the formula for the prismoid in the form: A.,+Al 2 1 1 v: ?c [A.+4< +A.1+ 6 c A,+A, 2 [A. + a +A.] . . . where C = contour, interval, and A0, A., A1, etc. = total areas bounded by each contour, the volume of 8—7 is calculated to be 60.3)(106 m“. Applying the formula for the frustrum of a cone: v = 1/3C[A,+ A, (A,A.)‘/2] + ‘/aC[A.+A , +(A.A,)V2] where symbols are as above, the volume of 8—7 is calculated to be 60.0)(10‘ m“. 21 LOWER SALT Tabumuusv )1 R. 43 E. 117°20' R. 44 E. ‘ l 6 5 4 3 2 1 6 5 Pioneer Point . _o.9 l l 7 9 | / .0.” l 18 16 - T. l 25 S. g 4. 35°45' — 0'7 .‘ Argus 2 0 . 2.6 _2.2 _ . 3// 12's 3.3\ 30 2 Q ‘ o W 1.5 / 31 32 I1.4 NAVAL‘RESERVATI‘ON BOUNDARY 10 use Maumw EXPLANATION 4 Isopach Hachured in areas of thinning. Contour interval 1 foot 27 I Core-hole location Thickness of unit S — 1, in feet 0 0 22 ‘l 2 23 24 2 MILES 3 KILOMETERS FIGURE 8.——Isopach map of unit S—1, Searles Lake. 20 22 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. f R. 43 e. 117°20' R" 44 E' ' 1 6 5 4 3 2 1 6 5 , Pioneer Point . l 7 8 I 18 17 , T. 1 25 s. 3 35°45’ — ‘1 Argus ‘ 19 20 ‘ >‘ ' Ir. 1 <1 D | Z 1 D 0 la: 30 29 z e 1 E- < > fl H m ‘m . ad 31 h] 1< > E 1 Westend \ .08 ' 6 \ 5 1 0'1. 8 l EXPLANATION ’ 0 _ - Isopach '0-3+ Hachured in areas of thinning. Contour interval 1 foot \ N 1 ' 15 14 00-5 18 .2-7 13 0.7' ‘ Core-hole location ' Thickness of unit S _ 2, in feet | 1 p4)” I 22 23 24 19 20 | O 1 2 MILES 1—_“—|_‘__lfi—~—I_J O 1 2 3 KILOMETERS FIGURE 9.-—Isopach map of unit S—2, Searles Lake. $4 ‘uv’ f " LOWER SALT 23 f R. 43 E. 117°2o' R.44 E. ‘ I 6 5 4 3 2 1 6 5 , Pioneer Point . l 7 7 8 1 18 18 17 l. L Argus 1 19 2 00'9 o . ‘9 20 l in ' a: l < - Q 1 Z . a O 29 1 m 30 Z .13 ‘ 2 [-1 |< > a: 511 m ‘3‘: . 0.17 . 31 32 u 1 5 31 32 0 2‘ / 1‘0 2‘0 / . l i 111 L 12 ‘13 '15 '1 3 \1.0 K _1.4 \ iz 1.0 ' 5 .1.4 .03 1 .018 . 5 ‘ 0.7 A1,2 1.0 1~0 9 10 EXPLANATION T 11 -— 4 —-—- 26' Isopach s. 0'4 Hachured in areas of thinning. Contour interval 1 foot .2] 15 14 Core-hole location Thickness of unit S -3, in feet | 22 23 24 19 20 ' O 1 2 MILES H—l—_L_I—T—J O 1 2 3 KILOMETERS FIGURE 10.—Isopach map of unit S-3, Searles Lake. 24 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. f R. 43 E. 117°20’ R. 44 E. 4 3 2 Pioneer Point 0‘) 01 L0 U1 0 A 01- ._J__.__._.._______.___.__..__ N (.0 NAVAL RESERVATION BOUNDARY m am MOUNTAIN EXPLANATION T 4 26 . 0.1 Isopach S Hachured in areas of thinning. \ Contour interval [foot 15 14 .24 .2] 13 Core-hole location Thickness of unit S - 4. in feet I 0 22 23 24 / 19 20 O 1 2 MILES }—-—I——'—1‘—‘1"J 0 ‘l 2 3 KILOMETERS FIGURE 11.———Isopach map of unit S—4, Searles Lake. LOWER SALT )./ WW; R. 43 E. 117°20’ R_ 44 E, ' < 6 5 4 3 2 1 6 5 / Pioneer Point . 0/\ I .12 l 7 I 18 . i Argus ‘ 19 ‘ | 30 i 31 l NAVAL RESERVATION BOUNDARY m up MGWVYAM 9 EXPLANATION T 4 26' Isopach 5' Hachured in areas of thinning. Contour interval 2 feet .2.7 15 Core-hole location Thickness of unit S — 5, in feet 22 0 1 2 MILES F—r—L-W—TJ o 1 2 3 Kl LOMETERS FIGURE 12.—Isopach map of unit S—5, Searles Lake. 26 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. TOW f R. 43 E. 117°20' R. 44 E. l 6 5 4 3 2 1 6 5 Pioneer Point .0.4 2.9 316. . O N \a _\ N \1 00 o 4.6 18 / 35°45’ — —1 a \ a H 0 fl 5» 3R (0 0w 0;“? '5 _ . a _ .- C U! I; I /. “/3 1) u 3 _L_._____._..__________.__ Argus 9 19 2 .315 353 21 . ‘9 20 0132 >.. ' i =3“ 5'8: l E | Z . D O "3- 30 29 I m 30 29 I . \' 5.7. 2'9. .g 15 . _ 4.9 \ > / '41 g 49 ‘0 ° '2 9 '5: 31 1 33 34 3 36 31 ' 3 - ,4 .4 s 4 3. 4.1. . 02 5 l < 315 > V < v. =3] - - \ 3 7 I Z L/ 3.9 '33 ! Westend \5/ 3.7. e \\\W 2 1 .55 411. s 5 l m m uowvmw 0 EXPLANATION 11 4 s, 2.4 Isopach _4.9 m :§ 2.0 ., Hachured in areas of thinning. ' Contour interval 2 feet 211. 13 Core-hole location Thickness of unitS — 6, in feet y— 1. 22 23 24 19 20 O 1 2 MILES i—h—l—f—a 0 1 2 3 KILOMETERS FIGURE 13.—Isopach map of unit S—6, Searles Lake. LOWER SALT f R. 43 E. 117°20' R, 44 E_ ' 1 6 5 4 3 2 1 6 5 Pioneer Point 1 A I 0.3 1 . 0.7 . 0.1 ' 7 0". 1o 11 2 7 8 I 1.4 116 0.6 0.7.7 I ' ’0-7 ' 'o.7 ' ' I 18 17 . i \ o —1 19 20 I I 9.5 30.08 29 i 1 32 , .3114 ‘0'1 _2.3 0.6 O . 26% 5 m ~50 "GHANA/N; 014 1'5 1.5 1 5 0~3 g o o o ' 12 2'7 o EXPLANATION T ‘0 11 ' 1.9“ ° 7 8 , _ 4 _ 1 Isopach 5' 0‘4 1,1 ; Hachured in areas of thinning. ' Contour interval 2 feet .217 15 14 .015 18 1.1 17 ' . 13 Core-hole locatlon ‘ Thickness of unit S — 7, in feet ' I I 22 23 24 19 20 I o 1 2 MILES o 1 2 3 KILOMETERS FIGURE 14.——Isopach map of unit S—7, Searles Lake. 27 NAVAL RESERVATION BOUNDARY 28 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. mmmvm )1 R. 43 E. 117°20' R. 44 E. l 6 5 7W 4 2 1 6 5 5/ Pioneer 7/ Point >\ 7 I Q / _, 3 / 2 3. 2.3-. 8 . _l___.____..___..._______ NAVAL RESERVATION BOUNDARY 70 M0 MOUNTAIN; TO EXPLANATION 26' 4 S. Isopach Hachured in areas of thinning. Contour interval 1 foot . . 1‘8 15 l Core-hole location . Thickness of unit M-2, in feet I I 22 23 24 19 20 ‘ O 1 2 MILES 1—‘~“T“‘—'—I—7‘J 0 1 2 3 KILOMETERS FIGURE 15,—Isopach map of unit M—2, Searles Lake. f R. 43 E. LOWER SALT 117°2o’ R.44 E. 4 Pioneer Point ] 6 5 3 5/ p 4 ”\3/ 2.9.7. 7 ONZ/ 1.1 U K .1141 18 16 15 T. 25 w s. :03 _ \‘f, 35°45._ K 0.9 09. 00.4 Argus 4-6 > 09 .0.9 ~ 19 2 o 21 . 0.7 0.7 1,0 22 / .0.5 _0.5 017‘ .0.6 O \ 0.9.1 , 1.1 27 30 Q3 29 Lo 0’5 0‘7" g) .0] l 0.7 , " 1 / '5’), 0.8 _p.7 / \/ f LC . 2— .0.9 /_ ‘ A . 31 32 .05 33 0.6. 3 .0] k \ e1.2 1 5 .0.6 'o.9 Am 1 '1 Westend \ 6 _\ 2\ 3 0o] 2 .03 .112 1 .1.1 '—\ e 7\\ .1-2 .1-5 \ \ m 9 12 EXPLANATION T 4 26 I S. sopach Hachured in areas of thinning. Contour interval 1 foot .2.7 1.2_ . ’ 13 Core-hole locatlon Thickness of unit M- 3, in feet 22 23 24 19 20 ‘ 0 1 2 MILES |—_l—|—I_T’J O 1 2 3 KILOMETERS FIGURE 16.—Isopach map of unit M—3, Searles Lake. 29 NAVAL RESERVATION BOUNDARY 3O SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. WW} R. 43 E. 117°20’ R. 44 E, ' 1 6 5 4 3 2 1 6 5 Pioneer Point . 7 I A ‘ 1 2.2 . 2‘. 2.0 . .25 o 8 - 1.8 1.7 6 16 15 14 17 T. 25 1.7 s. _1.4 1.4 1.5: 35°45’— ' :14 . _ 0.5? 2 15.1.9 1.6 1.5 ' o 21 o h 20 20 N 22 .18 >1 '35 14 1.6 1 1.5 : : Q " Z 5 .3 6 O 9 a: 30 9 32 29 ‘z ”m '9 a \ E S /— 04 ‘1'!) " 5 31 3 3 - 2 A 7 <1 3 :13 >2 \ 2\\ 0.7 3\ 0.3 EXPLANATION 9 2.8 1' ' 12 T. 4—— 26 / [sopach 5' 1‘7 Hachured in areas of thinning. \< Contour interval 1 foot \11 2.7 15 14 .1.1 ’ ~ 13 Core-hole locatlon Thickness of unit M—4, in feet I 22 23 24 19 20 I 0 1 2 MILES F_—|_—lfi_‘-—l_l 0 1 2 3 KILOMETERS FIGURE 17.—Isopach map of unit M—4, Searles Lake. LOWER SALT 31 f R. 43 E. 117°20’ R. 44 E. x I I ‘ . 6 5 4 3 2 1 6 5 IA , Pioneer Point \ . k I A? I 1 A . 0‘4 .05 ' 1 7 .05 10 11 8 l g . R '0] I ' °o.4 '0.3 1 n 18 16 15 14 17 . T. | I 25 0.7. S. 0 5s 0.5 ' 0,6= _ -0.5 0.5? ' t 35°45’ — 0-8 _ Argus 0 1 ' 19 20 o ' .0-4 20 ’ 23 K / 00-3 ' C: K ‘ . _0,3 _0.6 at - 1. - - z . . D . '16 C1J >¢ o 0.3. 7 9, an 30 0.2. . .02 0.3. .0.7 0.3. 25 003+ 30 0'4 ”1. 29 z 1 28 0.2 26 . 9; . O " F , 1< \ =0.3? _o_4 0.3+ -02 0.6 -03 0.3 g m 0.3 l a ’ a: _ 31 34 35 36 31 32 - 0,5. .o.5+ 0.5. . 0.7 33 j > , V < \ _o.6.> _o.5 =0.4? -o.4 lz Westend \\ 6 \\ 3 0013 2 .0.7 1 V _o17 _o.1 .02 _ \_/ 3.0 o 5 9 10 O ' 12 0 EXPLANATION T 1 1‘ 26' \ 4 S. 1_2 Isopach > \ Hachured in areas of thinning. \’ Contour interval 1 foot ‘4‘ 15 14"3 \ .05 .27 13 Core—hole location \/ Thickness of unit M -5, in feet l 22 23 24 19 20 ’ O 1 2 MILES 0 1 2 3 KILOMETERS FIGURE 18.—Isopach map of unit M—5, Searles Lake. 32 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. f R. 43 E. 117°20' R. 44 E. l 4 3 2 1 6 5 ,1 Pioneer Point /A 6.034» 1. l J F v I. 1 18 1 Se. 1 35°; — :1 Argus 19 NAVAL RESERVATION BOUNDARY _2.2 _28 _2.1 \ _117 _1.9 / 2.1_ 2.0 ' - / \ 5 \ ’1. o' 5 4 01.6+ 2 ‘1? y .25 .2] 'fi \—/ _214 /'_\ .215 / m an MDUNTAW \ 3.4 \J’3 31‘ 2.1 3.3 37 33 33 45 9 10 G o o o' 12 o ' o' 7 ' 0 8 EXPLANATION T 3 1 4 26 b1 Isopach 5' 3'0 / .45 Hachured in areas of thinning. /—\_/ Contour interval 1 foot / 2.7 15 14 .5.7 18 4,4. 17 ' . 13 1 Core-hole location /6 \/ 6 . Thickness of unit M—6, in feet l f I \ ___—— | 22 23 24 19 20 I o 1 2 MILES 0 'I 2 3 KILOMETERS FIGURE 19.——Isopach map of unit M—6, Searles Lake. 25 35°45' LOWER SALT 33 mummy / R. 43 E. 117°20' R. 44 E. I ‘ 6 5 4 3 2 1 6 5 / Pioneer Point /\ . , m % 7 l 18 . ‘4 NAVAL-RESERVATION BOUNDARY Westend\‘\ 6 7:: no MDUNTAW EXPLANATION Isopach 4—— Hachured in areas of thinning. Con tour interval 1 foot . 2.7 Core-hole location Thickness of unit M - 7, in feet 9 10 T. 26 5- 3.0 15 \ I 22 23 24 19 20 ’ o 1 2 MILES P——'——;T——-TJ o 1 2 3 KILOMETERS FIGURE 20.—Isopach map of unit M—7, Searles Lake. 34 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. MINERAL COMPOSITION AND LITHOLOGY The Lower Salt is composed of alternating saline layers and mud layers. The salines are mostly hard, white or light gray, porous, and composed of megasco- pic crystals. The muds are generally soft, black to dark greenish gray, impervious, and composed mostly of microscopic or submicroscopic components that form a matrix for megascopic crystals of gaylussite, pirsson- ite, borax, and northupite. Thin discontinuous beds of mud occur locally in the saline units, and lenses and small clusters of saline minerals occur in the mud units, but they do not make up significant percentages and generally the horizons containing them cannot be traced throughout the deposit. SALINE UNITS The main mineral components of the saline units in the Lower Salt are, in approximate order of decreasing abundance, trona, halite, burkeite, northupite, borax, and thenardite. Small amounts of hanksite, nahcolite, sulfohalite, tychite, and tincalconite are found mixed with the other salines, and gaylussite and pirssonite are found in some of the thin interbeds of mud. The volume percentages of megascopic minerals in the saline units given in table 4 are based on visual estimates of mineral percentages (Haines, 1957, 1959). As described later, these estimates have systematic but moderately sized errors; on the average, they over- estimate trona by about 14 percent and underestimate halite by 8 percent, burkeite and hanksite by 5 per- cent, borax by 2 percent, and gaylussite and pirssonite by 1 to 2 percent. The volume percentage estimates of the mineral content are made as follows: The isopach map of each stratigraphic unit is considered to represent a solid body that has a base that is flat and a top that is di- vided into “steps” of different heights along the con- tours. It is treated as if the steps are a series of nested and crudely concentric rings that have rectangular cross sections. The inner and outer edges of each irreg- ular ring are vertical and defined by successive iso- pach contours, and the top and base are flat with the height of the top surface being midway between the thicknesses represented by the bounding contours. The area of each of these concentric bodies is calcu- lated from planimeter measurements as described in the previous section, and the volume computed on the basis of their height. The relative volume percentages of the saline minerals in each of these concentric bo- dies are then calculated. First, the volume percentages of all minerals in that particular stratigraphic unit in each core are computed. These data are then grouped according to the isopach contours the core falls be- tween and averaged. Where there are no core holes be! tween a given pair of contours, a reasonable value is interpolated or extrapolated (shown in table 4 in par- entheses). The average composition of each of the rings making up the isopach map body is then weighted according to the percentage of the unit’s to- tal volume it accounted for, and then all are added. The data in table 4 clearly show two types of compo— sitional variation, lateral and vertical. Lateral vari- ation of the minerals trona, halite, and burkeite is marked; borax, thenardite, and nahcolite also appear to change systematically. The minerals halite and burkeite are clearly concentrated in the thicker cen- tral parts of the units, and trona, nahcolite, and then- ardite are concentrated near the thinner edges. This distribution, a function of the relative solubilities of these minerals, is discussed in a later section. Vertical variation in composition is evident from the data in this table. The most marked vertical changes are as follows: the lower three units (8—1, 8—2, and 8—3) consist almost entirely of trona; the next unit (8—4) is mostly trona and burkeite; the next (8—5), the thickest, consists of trona and halite with some bur- keite; the upper two (S—6 and 8—7) contain large per- centages of both trona and halite but are relatively low in burkeite. Several other vertical changes are evident. Northupite is most abundant in 8—2, 8—3, and 8—4. Borax becomes gradually more abundant from 8—1 to 8—5 and is notably abundant in S~7. Nahcolite is restricted to 8—1 and 8—7. More complete descrip- tions of the saline units, based on the above data and on the mineralogical and textural details presented by Haines (1959) and Smith and Haines (1964), follow. S—1.—Unit S—l consists chiefly of bladed and fine- grained trona with subordinate amounts of the fibrous form. In the central and thicker parts of the body, few other minerals are associated with this bed except for some massive northupite in the interstices of trona blades in GS—ll and 12. Toward the thinner marginal areas, there are local pockets of borax crystals (in GS— 18 and 21) and nahcolite (in GS—9, 27, and 41). Tincal- conite is reported from this layer (Pabst and Sawyer, 1948; see also Smith and Haines, 1964). Visible halite is absent. Mud as impurities in the trona and as thin beds is common, especially near the lower and upper contacts; in the central areas, mud rarely forms as much as 10 percent of the unit, but nearer the edges it commonly forms more than 20 percent. LOWER SALT TABLE 4.——Estimated mineral compositions of saline layers in the Lower Salt [Numbers in parentheses either interpolated or extrapolated. t = trace] 35 Composition (in volume percent) indicated between contour lines Weighted Stratigraphic Mineral total unit 0—2 2-4 4—6 6—8 8—10 10—12 12—14 14+ percent S—7 __________ Halite ________ — — 13 35 38 (40) (40) — — — — 18 Trona ________ 84 80 58 61 (60) (60) — — — — 73 Borax ________ 3.5 3.7 — — — — — — — — — — — — 2.2 Thenardite _____ 1.7 — — — — — - — — - — — — — — .6 Sulfohalite _____ — — — — .1 .2 — — — — — — — — .05 Pirssonite ______ .1 .8 .1 .2 — — - — — — — — ‘ .3 Burkeite _______ —— —— 5.7 —— —— —— —— —— 1.1 Nahcolite ______ t — — — - — — — — — — — — — — t Mud _________ 11 2.7 .8 .4 —— —— —— -— 4.6 Number of cores between contours _ _ _ _ 21 7 2 1 0 0 — — — — Percentage of unit volume lying between indicated contours __________________ 35 27 18 15 3 2 — — — — S—6 __________ Halite ________ — — 5.4 30 36 (40) (40) — — — — 21 Trona ________ 92 93 68 64 (59) (59) — — — — 77 Borax ________ —— .8 —— —— —— —— —— —— .3 Burkeite _______ — — — 1.2 — — — — — — - — — — .6 Sulfohalite _____ - - t — — — — — — — — — — — — t Mud _________ 7.5 .6 .3 — — (1.0) (1.0) — — — — .7 Number of cores between contours _ _ _ _ 2 18 8 1 0 0 — — — — Percentage of unit volume lying between indicated contours __________________ 5 34 49 10 1 1 — — — — S—5 __________ Halite ________ 9.5 17 40 66 43 61 49 (50) 51 Trona ________ 79.9 71 50 31 33 31 35 (35) 37 Hanksite _______ — — — — - - — — — — t t — — t Borax ________ — — — — .1 — — 2.2 — — .8 (1.0) .5 Burkeite _______ — — — — 8.0 1.5 20 5.3 12 (13) 8.1 Thenardite _____ 1.5 — — — — — — — — — — — — — — .03 Sulfohalite _____ .2 — — t t — — t — — — — t Northupite _____ .3 0.4 t t t — — — — — — .03 Tychite _______ —— —— t t —— -— —— -— t Pirssonite ______ — — — — .3 - - .1 .3 .8 — — .3 Mud _________ 8.6 12 2.0 2.0 1.1 1.8 3.1 (1.0) 2.6 Number of cores between contours _ _ _ _ 7 4 3 3 5 7 2 0 Percentage of unit volume lying between indicated contours __________________ 2 6 8 13 16 38 14 3 Contour lines 0—1 1—2 2-3 3—4 4—5 5+ S—4 __________ Trona ________ 49 82 45 72 (60) — — — — — — 64 Hanksite _______ —— 1.7 —— —— —— —— —— —— 8 Borax ________ 5.3 1.2 .2 11 (1.0) —- - — —— 1.7 Burkeite _______ 32 12 52 6.2 (35) — — — — — — 29 Northupite _____ 3.5 .6 .3 2.5 (1.0) — — — — — — 1.0 Pirssonite ______ — — .2 — — — — — — — — — - — — .1 Halite ________ .3 —— —— —— —— —— —— —— .03 Sulfohalite _____ .2 — — — —— — — — .03 Mud _________ 9.6 2.2 2.5 7.5 (3.0) — — — — — — 3.5 Number of cores between contours _ _ _ _ 15 9 5 1 0 — — — — — — Percentage of unit volume lying between indicated contours __________________ 14 46 36 3 1 — — — — - — 36 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. TABLE 4.—Estimated mineral compositions of saline layers in the Lower Salt—Continued Composition (in volume percent) indicated between contour lines Weighted Stratigraphic Mineral total unit 0—2 2—4 4—6 6—8 8—10 10—12 12-14 14+ percent s—3 __________ Trona ________ 88 86 (86) — — — — — — — — — — 87 Hanksite _______ — — 2.0 (2.0) — — — - — — — — — — 1.5 Borax ________ 3.7 1.1 (1.0) — — — — — — — — — — 1.8 Northupite _____ .8 1.5 (2.0) — — — - — — — — — — 1.3 Gaylussite ______ .3 — — - - — — — — — — — ~ — — .08 Burkeite _______ — .6 (2.0) — —- — — — .5 Mud _________ 7.1 8.8 (7.0) — — -— —— -— — — 8.2 Number of cores between contours _ _ _ _ 19 14 0 — — — — — — — — - — Percentage of unit volume lying between indicated contours __________________ 27 65 8 - — — — — — — — — — S—2 __________ Trona ________ 98 93 89 91 — — — — — — — — 92 Borax ________ —— .1 2.2 —— —— —— —— —— .9 Burkeite _______ .5 .7 1.3 — — — — — — — — — — .9 Thenardite _____ — — — — 1.3 — — — — — — — — — ~ .5 Northupite _____ .8 .6 1.4 .3 — — — — — — — — .9 Mud _________ .9 5.2 4.5 8.8 - — — — — — — — 4.6 Number of cores between contours _ _ _ _ 11 11 6 3 — — — — — — — — Percentage of unit volume lying between indicated contours __________________ 12 43 40 5 - — — — — — — — S—l __________ Trona ________ 71 92 97 98 85 (100) — - — — 92 Borax ________ .3 1.2 —— —— —— —— —- .3 Northupite _____ .4 — — 3 — — — — — — — — — — .1 Tychite _______ t —— —— —— —— —— —— —— t Nahcolite ______ 4.5 — — — — — - — — — — - — — — .6 Pirssonite ______ — — .4 .4 — — — — — — — — — — .2 Mud _________ 24 .6 2.5 1.9 14 — — — — — — 7.1 Number of cores between contours _ _ _ _ 16 4 5 3 1 0 - — — — Percentage of unit volume lying between indicated contours __________________ 14 24 23 29 8 2 — — — — S—2.—Unit S—2 is in many respects similar to S—1; they are similar in area and volume and both consist largely of trona in the fine—grained and bladed forms, but the fibrous form is subordinate. Locally S—2 has pockets containing major percentages of burkeite (in the central facies, GS—15 and 16), borax (in the east edge facies, GS—2 and 27), and thenardite (in GS—6). In about a third of the cores, small quantities of fine— grained northupite form beds or interstitial fillings. Visible halite is absent. Impurities of mud are less common than in S—1 and seem to be concentrated in the central areas. S—3.—Like the underlying two saline beds, unit S—3 consists primarily of trona, although it is slightly smaller in volume and covers less area. Local pockets contain uncommonly high percentages of burkeite (in GS—15), hanksite (in GS—10), and borax (in GS—15, 19, 22, and 27). Northupite occurs in about a quarter of the cores. Visible halite is absent. Thin beds and im- purities of mud make up several percent of the unit in most cores; their distribution does not seem to be re- lated to areal position. S—4.~—In bulk composition, unit S—4 is largely trona plus major amounts of burkeite; their relative per- centages do not seem related to the thickness of the unit. Halite is present but in very small quantities (in GS—10 and 18). Hanksite (in GS—2 and 27) is present but in smaller quantities than in S—3. Borax (in GS—2, 8, 15, 17, 20, and 26) and northupite (in GS—23, 24, 26, 27, 39, and 41) are commonly found in percentages similar to those of the two underlying units. Mud lay- ers are generally subordinate, especially in the central areas, and most of them contain neither pirssonite nor gaylussite. S—5.—Unit S—5 has the largest volume of saline lay- ers in the Lower Salt, although its areal extent is about the same as the other units. Halite is the chief component, but trona approaches the same concen- tration. Burkeite is subordinate to halite. Borax forms about half a percent. Minor amounts of hanksite, thenardite, sulfohalite, and northupite occur sporadi- cally. In many areas, burkeite and halite show a ten- dency to be more abundant in the lower three- quarters of the bed, trona to be more abundant in the LOWER SALT 37 upper one-quarter, but the layering within the unit is not consistent. With respect to thickness, trona per- centages show a clear tendency to diminish toward the thick parts of the saline body, whereas halite percent- ages complement this trend. Both burkeite and borax are clearly concentrated near the central part of the deposit. A 2- 3-cm mud layer is generally present about half a meter above the base of the unit, and a small percentage of mud, much of which contains pirs- sonite, occurs throughout. S—6.—Trona is clearly the predominant mineral in 8-6. The relative percentage of halite is less than half that of the underlying saline unit. The percentages of borax and burkeite are small. In the thicker central facies, the lower two-thirds of the layer contains high percentages of halite that is mixed with trona, and the upper third of this layer is mostly trona; the thinner edge facies are almost exclusively trona. A little borax is found near the edges (GS—6, 19, and 20), burkeite near the center (GS—15). Very small quantities of sul- fohalite and mud occur in this unit. S—7.—The preponderant mineral in unit 8—7 is trona. As in 8—6, halite is concentrated in the lower part of the central thicker facies, whereas trona forms the upper part; trona becomes very common toward the edges. Borax is found locally in the west, central, and eastern parts of the body. A little nahcolite (GS— 40), burkeite (GS-15), thenardite (GS—3), sulfohalite (GS—16), and aphthitalite (GS—15) are noted. Mud beds, some of which contain pirssonite, are more abundant in this unit than in S—6 and become more predominent toward the edges. MUD UNITS The six mud units in the Lower Salt are dark organic-rich marls in which megascopic carbonate minerals are embedded. Quantitative estimates of the megascopic mineralogy of these mud layers (within the area sampled by cores) given in table 5, have been made by the same techniques used for the saline lay- ers. These data clearly show that the percentages of these minerals vary according to stratigraphic posi- tion. In units M—2 — M—5, megascopic crystals of gay- lussite are uniformly distributed and form percentages diminishing from about 18 to 2; small quantities of pirssonite occur in M—4 and M—5. In M—6 and M—7, gaylussite forms 30—40 percent of cores from the edge facies but is subordinate to pirssonite in the central facies. Crystals of borax locally form pockets in most of these layers, but are most abundant in M—4, M—5, and M—6. Northupite forms a small amount of all units but is most concentrated in M—3. Schairerite, tychite, and sulfohalite form fractions of a percent of some units. In addition, some layers have lenses or small pods of megascopic saline minerals such as ha- lite, trona, nahcolite, burkeite, thenardite, or hanksite that were probably formed by postdepositional cry- stallization of migrating brine. The microscopic size fraction of these muds consists predominantly of smaller crystals of most of these same minerals plus aragonite, analcime, clastic sili- cates, and partly decomposed organic material. An X- ray study by R. C. Erd (written commun., 1958) of eight samples of muds from the Lower Salt in GS—14 is summarized in table 6. Gaylussite, pirssonite, northupite, and halite are the major evaporite compo- nents of the microscopic fraction. These data show that fine-grained gaylussite and pirssonite coexist; northupite is a prominant component of some units; dolomite is not detected; and halite is present in only minor amounts. Studies of the fine-grained fraction of mud samples from GS—2 by Hay and Moiola (1963, ta- ble 1) report small quantities of analcime in units M—6 and M—7. The small percentages of saline minerals found in many of these samples may come from the evaporation of brine that was in the pores or entered the core during drilling. The mineralogy of these mud layers differs from that of the underlying Bottom Mud and the overlying Parting Mud, which contain larger percentages of ara- gonite, dolomite, and halite, smaller percentages of elastic minerals, and almost no northupite. Summaries of the megascopic lithology of the mud layers follow. M—2.——Unit M—2 is the lowest mud layer within the Lower Salt sequence. It most commonly consists of mud that contains 15—20 percent gaylussite crystals. Fine laminar bedding is well-developed in most cores. A few parts of the bed contain disseminated crystals of trona, but halite is not reported. Other minerals found are nodules and thin beds of northupite (in about a third of the cores), disseminated crystals of borax (in GS—8 and 21), and small pockets of crystal- line galeite (in GS—17,22, and 41) (misidentified in original logs as schairerite; see Smith and Haines, 1964, p. P32), tychite (in GS—l, 3, and 24), thenardite (in GS—6 and 10), hanksite (in GS—26), and sulfohalite (in GS—ll). The northupite is concentrated around the edges of the deposit; thenardite may be also. Unit M-2 is generally 1—3 ft (0.3—1 m) thick and is thinnest near the south end of the deposit (fig. 15); the only marked thickening is to the southeast and southwest, and this thickening is probably a reflection of the clas- tic contribution coming from the large drainages of these areas. M—3.—Unit M—3 is similar to M-2, consisting chiefly of euhedral gaylussite in mud, although the average percentage of gaylussite is slightly lower.Lam- 38 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. TABLE 5.—Estimated mineral compositions of mud layers in the Lower Salt [Compositions, in volume percent, estimated within arbitrary boundaries shown in figs. 15—20] Composition (in volume percent) between indicated contour lines Weighted Stratigraphic Mineral total unit 0-1 1-2 23 3-4 4-5 5-6 6-7 percent M—7 ____________ Mud ___________ (62) 62 63 67 62 60 60 63 Gaylussite _________ (18) 21 19 33 38 40 40 28 Pirssonite ________ (20) 16 15 — — — — — — — — 8.0 Northupite ________ — — — — t — — — — — — — — t Borax ___________ —— —— .6 —— —— —— —— .2 Trona ___________ —— .1 2.6 .1 —— —— —— 1.1 Sulfohalite ________ - — — — t — — — — — — - — t Number of cores between contours __________ 0 7 20 5 1 0 0 Percentage of unit volume, within arbitrary boundary shown on map, lying between indicated contours __________________ 1 11 41 18 24 4 2 M—6 ____________ Mud ___________ (56) 59 62 65 69 66 (65) 65 - Gaylussite _________ (2.0) 3.5 16 9.8 26 33 (34.0) 20 Pirssonite ________ (39) 34 21 23 4.2 — — — — 13 Northupite ________ — — — — .1 t .2 .4 — — .1 Borax ___________ (1.0) 1.0 .8 1.9 .6 .9 (1.0) 1.1 Halite ___________ (1.0) 1.0 —— -— —— —— —— .1 Trona ___________ (1.0) 1.1 —— —— .2 —— —— .1 Number of cores between countours _________ 0 5 9 8 9 3 0 Percentage of unit volume, within arbitrary boundary shown on map, lying between indicated contours __________________ 1 6 21 23 25 18 6 M-5 ____________ Mud ___________ 86 95 (98) (98) — — — — — — 92 Gaylussite _________ 2.6 2.5 (2.0) (2.0) — — — — — — 2.4 Pirssonite ________ 1.5 — — — — — — — — — — — — .7 Northupite ________ 4.2 — — — — — — — — — — — — 1.9 Borax ___________ 3.3 —— —— —— —— —— —— 1.5 Trona ___________ 2.3 2.5 — — — — — — — — — — 1.7 Tychite __________ .1 — — - — — — — — — — — — 0.5 Sulfohalite ________ .1 — — — — — — — — — — — — .05 Number of cores between contours __________ 25 2 0 O — — — — — — Percentage of unit volume, within arbitrary boundary shown on map, lying between indicated contours __________________ 46 26 18 10 — — — — — — M—4 ____________ Mud ___________ 85 85 84 83 — — — — — — 84 Gaylussite _________ 11 11 11 13 — — — — — — 11 Pirssonite ________ — — .1 .2 — — — — — — — — .1 Northupite ________ 3.0 1.1 1.9 1.6 — — — — — — 1.5 Borax ___________ .2 .9 1.4 1.2 —— —— —— 1.1 Trona ___________ .5 2.0 1.9 — — — — — — — — 1.7 Burkeite __________ — — t — — 1.3 — — — — — — .1 Tychite __________ —— t —— —— —— —— —— t Hanksite _________ — — .5 — - — — — — — — — — .3 Number of cores between contours __________ 3 19 7 3 - — — — — — Percentage of unit volume, within arbitrary boundary shown on map, lying between indicated contours __________________ 2 53 34 11 — — — — — — inar bedding is well-developed in most cores, and, massive white nodules or thin beds in more than half after partial drying, the beds separate into paper-thin the cores, and averages about 2 percent of the unit. A layers that have a marked flexibility. Laminae near few crystals of trona are found in many cores; crystals the base are commonly contorted. Northupite forms of halite are not found. Tychite crystals have been LOWER SALT 39 ‘ TABLE 5.—Estimated mineral compositions of mud layers in the Lower Salt—Continued Composition (in volume percent) Weighted Stratigraphic Mineral total unit 0-1 1-2 2-3 3-4 4-5 5-6 6-7 percent M—3 ____________ Mud ___________ 84 83 84 82 83 83 83 83 Gaylussite _________ 13 14 9.2 11 17 15 15 13 Northupite ________ 1.5 1.2 2.9 5.9 — — 2.0 2.0 2.2 Borax ___________ — — t — — — — — — — — - — t Trona ___________ 2.1 2.3 2.4 1.4 — — — — — — 1.5 Tychite __________ .2 —— 1.1 —— —— —— —— .2 Number of cores between contours __________ 16 6 5 2 1 0 0 Percentage of unit volume, within arbitrary boundary shown on map, lying between indicated contours __________________ 11 30 18 15 14 8 4 M—2 ____________ Mud ___________ 82 82 81 76 69.2 — - — — 79 Gaylussite _________ 17 17 18 21 13.6 — — — — 18 Northupite ________ — — .1 .4 2.2 t — — — — .6 Trona ___________ — — .4 .4 .6 14 — — — — 1.6 Tychite __________ 1.3 — — t .2 — — — — — — .1 Galeite ___________ - - - — .1 — — - — — — — — .05 Thenardite ________ — — 7 — — — — 3.6 — — — — .4 Borax ___________ — — 2 .1 — — — — — — — — 1 Sulfohalite ________ — — — — — — — — — — — — t Hanksite _________ — — — — t — — — — — — ~ — t Number of cores between contours __________ 3 11 12 5 1 — — — — Percentage of unit volume, within arbitrary boundary shown on map, lying between indicated contours __________________ 3 16 54 18 9 — - — — found only in the southeast part of the deposit (in GS— 3 and 4). A trace of borax is reported from GS—27. Current marks were noted in GS-17 . Generally, this unit has a thickness of a foot or less (fig. 16), but it thickens rapidly toward the edges of the contoured area. M—4.—Like the underlying two mud units, M—4 consists of dark mud containing gaylussite, but the gaylussite percentage has fallen to about 10 percent. Small euhedral crystals of pirssonite occur locally (in GS—2O and 41). Faint laminar bedding is noted in most cores. Northupite, as white massive nodules or thin beds, forms about 1.5 percent, the maximum con- centration in any mud layer in the Lower Salt; its dis- tribution is probably random but it may be more concentrated in the thin central facies. Euhedral, sub- hedral, anhedral, and massive borax crystals lie in an irregular east-west belt through the center of the de- posit. Crystals of trona are embedded in the muds of this unit in about a third of the cores. A little tychite (GS—3), hanksite (GS—26), and burkeite (GS—18 and 22) are noted. The thickness of this unit is commonly between 1 and 2 ft (0.3—0.6 m), but the areal pattern of the variations in thickness is erratic (fig. 17); the unit thickens toward the west and southwest edges but maintains a nearly uniform thickness to the other edges of the drilled area. M —5.—Unit M—5 is normally the thinnest mud unit in the Lower Salt. Faint to indistinct laminar bedding is noted in most cores. Gaylussite crystals, generally much smaller than in other units, are reported from only about a third of the cores studied by Haines (1959) and are estimated here to constitute only 2—3 percent of the unit. Pirssonite crystals are reported from less than half as many cores. Northupite was found in three cores (GS—20, 21, and 26). A little trona (in GS—5, 15, 23, and 39), tychite (in GS—26), and sul- fohalite (in GS—26) are reported, but halite is not found. Borax reaches a higher concentration in this mud unit than in any other, most of it being in the elongate prismatic form (see Smith and Haines, 1964, p. P10—P12 and fig. 5); much of the borax is concen- trated near the base of the unit and in the central part of the deposit. Ripplemarks were noted in the mud at GS—16. The thickness of this unit is generally less than 1 ft (0.3 m) (fig. 18); except for an anomalous area in the northwest-central part, greater thicknesses are found only very near the west and south edges of the evaporite body. 40 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. TABLE 6.—Mineral abundance in mud units of the Lower Salt, core GS—14, determined for total sample and acid-soluble fraction of sample [Semiquantitative determination from X-ray diffraction charts; M = major, I = intermediate to minor, t = trace, ? = determination uncertain, — = not detected. Determinations by R. C. Erd, R. J. McLaughlin, G. 1. Smith] Total Sample Acid-soluble fraction Acid Depth (ft) Unit Halite Gaylussite Pirssonite Northupite Calcite Aragonite insoluble Quartz Feldspar Mica Clay Amphibole mlnera S 86.1 ______ M—7 M‘ I I I — — I — M I — M 93.0 ______ M-6 I — M — ~ t" I t M I — I 94.4 ______ M~6 I — — — t — M I M I — I 108.0 ______ M—5 t M — — - — I t? M I t — 110.0 ______ M—4 M — — — v — I — I M — — 110.6 ______ M—4 I M — M ~ — I — M M — — 112.6 ______ M~3 I M — — — — t — M M — ~ 114.3 ______ M—2 I ~ — - — A M — M I — — 115.3 ______ M—2 I M — — — ~ I t? M I — — ‘Sample also contains intermediate amounts of trona. M—6.—About two-thirds of unit M—6 consists of mud, one—third of gaylussite or pirssonite. Laminar bedding is almost nonexistent. In 11 cores, gaylussite exists alone; in 13, pirssonite exists alone; and in 5, both exist in the same unit. The cores that contain megascopic gaylussite in this unit are mostly from the edge facies, whereas the pirssonite-bearing cores are from the more central facies, and the cores containing both come from a transitional zone between them. These zones, though distinctly concentric, do not fol— low closely the isopachous patterns of the unit; they do, however, approximately follow the present-day depth contours on the top of the mud layer (which are similar to those on the base of the Parting Mud shown in fig. 25). Disseminated crystals of euhedral borax are found in about half of the cores described by Haines (1959), most of which are from the central and eastern parts of the deposit. Northupite is found in this layer in six cores, all from areas near the edges. The total volume of unit M—6 (within the boundary used for measurements) is about 1.2 X 108m3; this volume makes it the largest of any mud unit in the Lower Salt. Its pattern of thickness variation, shown in figure 19, is notable because it varies more symmetrically around a central thin area than the other mud units, probably because the surface on which it was depos- ited was nearly flat (see p. 97), whereas the other mud units were deposited on irregular surfaces. Typical thicknesses are mostly between 1 and 3 ft (0.3—0.9 m), but some are nearly 6 ft (almost 2 m). M—7.—In most respects, unit M—7 is similar to M—6. Indistinct laminar bedding is found in this unit in about half the cores. Mud, gaylussite, and pirssonite form most of the layer. The percentage of gaylussite is slightly higher than in M—6 and the areal distribution slightly wider, the average percentage of pirssonite correspondingly lower and the distribution more re- stricted. The pattern of mineral zonation is broadly the same. Borax is found sporadically along the east and west edges of the deposit. Northupite is almost nonexistent. A little trona (in GS—7, 14, and 17), sulfo- halite (in GS—8), and nahcolite (in GS—40) are noted. The volume of this unit (table 3) is a little less than unit M—6 but still greater than any of the lower four mud units of the Lower Salt. The thickness of the unit is commonly between 1 and 3 ft (0.3~O.9 m) (fig. 20), somewhat less than M—6, and it shows only a slight thickening toward the west and south. CHEMICAL COMPOSITION CHEMICAL ANALYSES OF THE SOLIDS Chemical analyses were made of 34 samples from four cores of the Lower Salt (table 7). The 11 samples from GS—16 are representative of the area near the central part of the deposit, the 6 samples from GS—21 of the parts near the edge, and the 9 samples from GS— 11 and the 8 from GS—12 of facies between the edges and center. At the time these samples were taken for chemical analyses, the stratigraphic subdivisions of the Lower Salt had not been established. The inter- vals chosen for sampling and analysis thus bear no re- lation to them. The samples were taken from large-diameter cores (Haines, 1959, p. 147—148, pls. 9 and 10) by sawing a uniform wedge from one side of the core. This entire sample, which represented approximately 5 percent of the volume of the core, was then crushed and split to get a representative sample for analysis. Table 7 lists both chemical and X-ray analyses. The X-ray data are not quantitative, but when combined with the chemical analyses, allow one to verify and semiquantify the stratigraphic trends in mineral com- position inferred from the visual estimates and sum- marized in table 4. The dominant trends verified are the tendencies for the lower saline units (S—l, S—2, and S~3) to contain larger amounts of CO2 in the form of trona; for the middle units (S—4 and S—5) to contain larger amounts of SO3 in the form of burkeite and Cl in 41 2.231225: :2 22032222222322 9283 2222.225 2:23.592 3222822222» .20 moo—20.2.5950. LOWER SALT 2m + .322 22.2222 .3232” .229 2.922 2222.22 2.2 2.22 2.2. 2.2. o2 22.22 2.2. 9.22 22 2.9 22.2 + 2m + 22.2 2.99 2.2.9 ......... 2 .222 .322 ..32:8_8:22 £22225 2.222 222.22 9.22 2.22 22.2 2.2 2.2 2.2.2 8.2 2.222 22.2. 2.22 22.2 + 22 2.2.9 22.2222 IIIIIIII x 2m + 2.52 + 322222 .322 £22222 .229 9.222 82.22 2.22 2.92 2.2 2.. 2.2 2.92 22. 2.22 22.2 9.2. 2.m + 22 + mm 22.2222 '23 ........ w .32.. .229 9.82 2222.2 22. 9.92 22.2 2.2. 22.2 2.22 .2... 2.222. 22. 222.. mm 2.9.2 2.2.2 ......... .2 .322 2822222 .3223? .229 2.922 222.22 2.22 2.92 2.2 22.. 2.2 2.9.2 222. 22.22 2.2 2.22 mm + 2222 + 8 2.22 .229. ......... m 3233.22 .322 .229 22.22 89.2 92. 2.22 2.2 22. 2.2 2.22 22. 2.22.2 2.22 22. cm 2.3 no.2... ........ 2222-0 .822 22.2222 .8222» .229 2.92 2222.2 2.2 2.92 9.2 22. 2.2. 2.22 on. 2.22 2.2 2.2 22 + 2m + 22 22.222.22.922 llllllll 22 .3222: $232325: _.3_:825=22 .3222“ .229 2.222 622.2 22.22 22.22 9.2 2.... 2.2 2.92 92. 2.22 2.2 2.2. 22.2 + 8 + 2.2.2 22.922.22.222 ........ o .322 .222 .2222: .3222?» .322322 2.92 292.2 2.2. 9.2 2.2. 2.2 2.22 9.2: 2.... 2.9.2 2.2 2.2. 2.2.2 + 2m 2.2222222 ........ 2 22:82:52 .222 6232222 .3232 2.22 32.3 222. 2.2. 2.2.2 2.2 2.222 2.22 22. 2.22. 2.2. 22.. .252 + mm 2222.22.32 ‘‘‘‘‘‘‘ 22 2322828222 .2222 .222 .822: 2.22 2922.8 22. 2.2 2.222 9.2 22.22 22.22 22. v.92. 222. 22. am 22.82.2222 ........ .2 2322282822 .222 .322 .322222 2.92 82.2.2 22.222 2.222 22.22 .22 2.2 2.92 2.2 22.22 2.2 22.9 mm + 222.2 222222.92 ........ v2 .222 .3222: 2.22 2.22.22 8. 2.2. 22.2.2. 222. 22.2 2.2 9.2. 9.22. 22. 222. mm 2.92 .222 ........ 2. 22222222 .322 .229 2.82 222.22 22.2 9.22 2.2. 22. o2 9.22 2.2 2.22 229. 2.2 mm + 92.2 22.222 [2.222 ........ 2 2322222222 22222.22 .222 .3222: 22.92 229.3 2.2 22.2 2.22 2.2 2.2 2.22 222. 2.22 2.2 2.2 92.2 + 9m 2.22 .2222 ....... 22 .822 .229 22.22 $22.22 22. 2.22 2.92 9.2. 22.2 2.222 22. 9.222. 222. O2. 9w 2.8 [9.222 ‘‘‘‘‘‘‘‘ 0 2.3222558: 9m + .222 .3232» ..32=828:2 .229 2.92 89.2 22.22 22.222 2.2 2.222 2.2 0.22 8. 2.22 2.2 2.2. 2522 2222222 9.222 -222 llllllll 22222292 .322 2322222 .2222?» .229 2.22 82.2 2.2. 2.222 2.2 22. 22.2 2.22 9.2. 2.22 9.2 22.2. 22 + 222 222222.222 ........ 2 22.2 + 2 + .322 2822222 .3222?” .229 2.92 2.9.2 9.2 2.222 2.2 222.. 2.2 22.22 222.. 22.22 2.2 2.22 22.2 + 8 + 2.2.2 222222.22 ........ v2 222 + 3222222 .222 .3212 .3222 2.222 28.2.2 2.2 22.9 2.222 2.2 2.2.2 2.22 22. 22.22. mm. 2.2 2.2 + 9.2 + mm 22.22.3222 ........ 2. 223222 .222 .322: 2.2222 292.22 22. 22.2. 2.92 2.2 2.2 2.9 222. 2.22. 2.22 222. mm 0.222.222 ........ 2 .222 222222 .2222 2.222 32.22 2.2 2.2 2.22 2.2 2.22 22.92 .22. 22.22 o2 2.9 mm + 22 2.22.222 ........ 2.2 2.322222 .2222 .3222 2.202 222.222 2.2 2.2 2.22 22. 2.2 2.22 22. 2.92. 2.2.. 2.2. 8 22.222 22.2.2 ........ 0 23222222 .2222 .322 .229 9:92 82.22 2.2 2:22 2.22 2.2. 2.222 2.222 29. 2.222 22. 2.2 mm + 922 + 92 22.2.22 2.22 ........ .2 .322 .222 232282829 9.22 222.2 22. 2.2.2 222. 2.22 22. 22.222 92. 2.222 92. 92. 9m 2.22 2.222 IIIIIIII 2.222222 2322022 :32222222 .3222 .229 22.82 2222.2 on. 2.92 9.2 222. 9.2 2.9.2 2.2. 2.22 222.. 22. 252.2 .82 + 2m 222222.222 ........ 2 2322022 .322 .32 2222 .3223?” .229 2.22 222.2 2.22 22.92 22.2 22. 2.2 22.22 92. 2.22 22.2 9.2. 2m + 22.2 + 2m 222224.222 yyyyyyy 2.2 .322 222222 .3232» .229 22.222 83 2.22 22.222 22.2. 2.2 22.2 22.22 2.2 2.22 22.2 22.22 mm + 2.2.2 12279.82 ........ 2 2222222 6223222 .222 .3222 2.22 222.92 22.2 2.2 2.22 22. 2.22 2.22 2.2.. 2.22. 222.. 2.2 «m + 22 + mm 9.8222222 ........ 2 323222 .222 .3222: 2.2222 82.2222 8. 9.2 2.222. 22.. 2.22 22.2 92. 9.8 .222. 8. mm 22.252.22.82 ........ 2. 93222222 28222 .222 .8222 2.2222 82.22 2.2. 2.9 2.92 229. 2.9. 2.2.2 222. 22.22. 22.2 2.2 mm + 222.2 2.822.222 ........ 2 2.322222 £22232 222.2222 .222 222222222 9.82 2222.2 2.2 2.222 2.22 22. 9.22 9.22 2.2 2.22 2.2 2.22 22.2 + 8 2.22 22.22 ........ : 62:282.: 6222.222 2222222 .222 .8222 22.8 292.5 2.9. 22.2 22.222 22.2 2.2 2.222 22.2 22.222 2.2 2.2 mm + 92 22.222 .2922 llllllll .22 2.322222 .2222 .3222: 22.22 222.9 2.2 9.2 2.222. 222.22 22.2 2.222 22.... 2.2.92. 2.22.22 2.2 9m 2.922 [2.22 »»»»»»»» .222me 28222222 222282. 20 222.5 322.82% :2 220 2a 222 .22 0.: 20 6.2 Mem .8 0.2 02.2 0222 95 22.22.2222 22 62 222.22 .3...X .3 gas—$3 £22052 55m 22322222 .3025 322.2: 23 :2 5an numb—02 Eo< 03222222235 Efifim A .0 «Eu 92:49:21: A. .m .3 3222232 “2335:: 133 «e 222322222 >5.X 22052252928 MES“. “20:35 E «o 2325 min—223:2 .3 6222:2233. 233 02: A5 ”Coo: a: 9.2.6 .5Im+2 :2 $29.22 .20 2:22.222 ”—22:25 .3 2222223229 262292 2.23 2.2—£30225 3222 A3 ”mu—22222222 20:3 .3 8:83.522: zuhdn .3 20m :3 2.2323 3:22.25 5 2392222 .20 2:22.222 anion .3 “5.2229222 2:22:28 20 222 ”5:212 E $299222 «o 2255.222 ”22:202— ..3 6222222322 22522302 :2 9222229222232 395 202mg: .09 .0222 .032 .0220 A: ”22:22:2203 2523222252 38:52.5 .3 .3 was 63532223 .0 .: ..2w2_ooEuw .m .4 .3 22.222252 2222:2228 223m. .8325 2: 2.2 2.22.223 \o mmmbasxls 2225.8 42 the form of halite (as well as CO2 in the form of trona); and for the upper units (S—6 and 8—7) to contain Cl and CO2 in the forms of halite and trona. The analyt- ical data also confirm that the edge facies (exempli- fied by GS—21) are low in C1- and SOs-bearing minerals relative to central (GS—16) and intermediate (GS—11 and GS—12) facies. It is difficult to assess the vertical variations in CaO, MgO, K20, and B203 be- cause the statigraphic composition of the analyzed in- tervals is so variable. Horizontal variations are more marked; the central and intermediate facies appear to have about three times as much B203 as the edge fa- cies, about half as much CaO and MgO, and about the same K20. The acid-insoluble material averages about 4 percent; because about 45 percent of the Lower Salt consists of material logged as mud, more than 90 per- cent of most mud layers is soluble in acid. The chemical analyses provide a way to check the reliability of the visually estimated mineral composi— tions. Table 8 compares the visual estimates of min— eral volume percentages (converted to weight percent— ages of major element oxides) with the chemical anal- yses of the same intervals. In compiling table 8, the chemical analyses of cores from both the Lower Salt and Upper Salt, given in tables 7 and 15, were com- bined into a single list and the analysis percentages ' compared with the modal percentages calculated from the estimated mineral composition of the same inter- vals cited in the published logs (Haines, 1959). Differ- ences were tabulated, with a positive sign used for differences where the chemical analyses gave higher values, and a negative sign used for those where the SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. chemical analyses gave lower values. To test for sys- tematic errors in visual estimates, these positive and negative values were added algebraically, and the sum divided by the number of analyses. Results of these computations are shown in table 8 both for the four individual cores and for all 51 analyses for Lower and Upper Salt combined. Also listed are mean deviations and standard deviations. 7” Although the analyses provide a more accurate measure of the composition of the cores, not all differ- ences are attributable to errors in the visual estimates. The chemical analyses are of a relatively small wedge cut from one side, whereas the visual estimates were based on the surface area of the entire core. Because of crystal sizes and the lateral variability in composition found in most cores, the large surface is likely to pro- vide a different, and possibly better, sample of the layer than the small wedge. The chemical analyses were of samples that included the fractured material logged as “probably cuttings, not core,” whereas the modal estimates omit these segments. And the analy- ses include the components that were in the brine en- trapped in the pores and were not actually present as salts. Nevertheless, it is clear that the visual estimates have small but consistent errors which should be taken into account when evaluating the bulk mineral compositions of the saline units given in table 4. The 5.0 percent negative error in CO2 and negative error in H20 probably means that trona was overestimated in the cores by about 14 percent (assuming that some of the analyzed CO2 came from underestimated or unde- TABLE 8—Comparison of chemical analyses with composition indicated by visual estimates of mineral percentages [Positive values indicate components for which chemical analyses were higher, negative values indicate chemical analyses lower. All values in weight percent. Data for chemical analyses from tables 7 and 15, visual estimates of mineral percentages from Haines (1959)] Oxygen CaO MgO Na20 K20 CO2 SOa BZOa Cl H20 equivalent Sum2 of Cl Average error‘ GS—ll ___________ +0.4 +0.7 —0.8 +0.6 —4.6 +0.8 +0.5 +5.2 —1.6 (—1.7) —0.5 GS—12 ___________ 0.0 +0.8 +0.3 +0.5 —5.7 +3.3 +0.6 +4.7 -3.2 (-1.1) -—0.2 GS—16 ___________ +0.6 +0.7 —1.8 +0.9 —3.8 +1.6 +1.1 +2.2 —1.1 (—0.5) —0.1 GS—21 ___________ +1.1 +1.3 —2.2 +0.9 —6.5 +3.8 +0.6 +5.1 -3.0 (—1.2) —0.1 All samples __________ +0.5 +0.8 —1.1 +0.7 —5.0 +2.2 +0.7 +4.8 —2.1 (—1.1) —0.4 Mean deviation3 ________ 1.2 0.8 2.4 0.8 5.9 4.0 0.9 6.4 3.0 Standard deviation‘ _____ 2.0 1.2 3.1 1.0 6.9 5.5 1.7 8.2 3.9 Ed si n retained N E where d = difference between chemical analysis value and visual estimates value, N = number of analyzed samples; 14 in GS—ll, 12 in GS—12, 16 in GSv16, and 9 in GSV21, total of 51. 2Algebraic Sums are negative because sums of chemical analyses (tables 8 and 15) are mostly less than 100. 1 EM N. ( Ed“ N /d/= same but sign ignored. ) “2 LOWER SALT 43 tected gaylussite or pirssonite as the positive error in CaO suggests). The 4.8 percent positive error in C1 suggests that halite was underestimated by an average of 8 percent; the 2.2 percent positive error in SO3 may mean that burkeite or hanksite was underestimated by about 5 percent; and the 0.7 percent error in B203 means that borax was underestimated by 2 percent. Positive errors in CaO may mean that megascopic gaylussite or pirssonite were underestimated by 1—2 percent, but the CaO may also represent small amounts of microcrystalline minerals in the mud. Similarly, the 0.8 percent MgO may represent about 5 percent megascopic northupite or about 4 percent mi- croscopic dolomite. The 0.7 percent error in K20 theo- retically suggests an error of nearly 25 percent hanksite, but probably indicates about 2 percent aphthitalite or contamination by the entrapped brine. The errors in visual estimates indicated by these data are of the type easily made while logging core. Fine-grained trona, burkeite, and halite in a core may be very similar in appearances, and when in doubt, the core logger generally chooses the mineral that most commonly has this habit—trona. Borax, prior to de- hydration of the surface to white powdery tincalcon- ite, is easily misidentified as one of the other glassy conchoidal-fracturing minerals such as hanksite or gaylussite. Small amounts of northupite (especially the fine-grained variety), thenardite, aphthitalite, and other saline minerals are easily overlooked. All fine- grained components in the muds—mostly aragonite, calcite, dolomite, or halite are not identifiable visually. CHEMICAL ANALYSES OF THE BRINES The current economic value of Searles Lake lies in its brine content. The major commercial operations on the deposit pump the brines from the interstices of the Upper and Lower Salts and process them to ex- tract chemicals. The brine varies in composition from place to place and from depth to depth, and an under- standing of these variations is essential for the opti- mum utilization of the deposit. Table 9 gives 68 brine analyses from the Lower Salt portion of 11 core holes, 10 in the central and interme- diate parts of the deposit, 1 (GS—1) in the outer parts. All samples were taken by lowering a hose to the de- sired level, pumping brine from that level until an equilibrium was established, then taking a sample. Much of the brine in samples collected in this way un- doubtedly came from levels well above and below the bottom of the collection hose. The contribution from each salt horizon exposed in the uncased drill holes depended on its permeability and its proximity to the point of collection. The samples thus approximate moving averages of the brines that existed at succes- sive levels, but these averages overrepresent, to an un- known degree, the brines in the more permeable zones. The analyses from core L—31, however, repre— sent definite stratigraphic units because a packer was used during their collection. The compositions of the brines given in table 9 are plotted in figure 21. It shows their compositions pro- jected to one face of the tetrahedrons that represent 5- component systems. Figure 21A shows projected boundaries of the mineral stability fields in the Na2C03-NaHC03—Na2SO4-NaCl-H20 system at 20°C; figure 2lB, projected boundaries in the NaZCOs- Na2SO4-NaCl-KCl-H20 system at 20°C. Many of the points in figure 21A are alined along zones parallel to the burkeite-thenardite and burkeite-halite bound- aries. They are probably exactly on these boundaries; the burkeite field expands with increasing tempera- tures, and temperatures in the Lower Salt are mostly 20°—24°C which would cause the burkeite boundary to move toward the alined points. The points in figure 218 show more scatter although most are within the boundaries of the hanksite field; the small amount of hanksite in the Lower Salt means that the removal of much of the K-bearing brines from the unit would re- sult in the phase disappearing. The remaining points lie a short distance away in the adjoining fields. The specific gravities of brines from the central areas of Searles Lake lie mostly between 1.29 and 1.31 and tend to increase with depth. Those from the edges are as low as 1.25. Except for samples from sites near the edges, the total percentages of dissolved solids, calculated by summation of the percentages, mostly lie between 33 and 35 percent. Values of pH given for brines from four holes (table 9) range from 9.1 to 9.9; those from GS—4 (which is nearest the edge and higher in sulfate and lower in carbonate) are the lowest. These values may be slightly in error because they were measured several days after collection, and it is likely that some CO2 was lost, thereby changing the pH value. The brine analysis given in table 9 shows some ver- tical variation in the weight percentages of the dis- solved components. The Na percentages of brines in the upper seven analyses in the table are calculated by equivalent difference; the percentages in the lower four are based on analysis. The Na percentages in most core holes increase downward. The percentage of K increases upward in 9 of the 11 wells analyzed and increases downward in two (GS—1 and GS—10). Seven of the eight analyses wherein K increases upward come from near the center of the deposit, but neither 44 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. TABLE 9.——Chemical analyses of brines in the Lower Salt [Analyses of core holes HH, MM, U, W, X, S—28 and S—31, by chemists of American Potash & Chemical Corp., published with permission of company. Analyses of GS—1, GS—4, and GS—10 by Henry Kramer and Sol Berman, US. Geological Survey, of L—31 by Shirley L. Rettig, US. Geological Survey. Compositions in weight percent except where indicated as parts per million (ppm)] Core hole (depths in feet Brine to top and base sample Total of Lower Salt) depth Specific dissolved solids Li PO. F Br S As Si I Date of Sampling (feet) gravity (by summation) pH Na K (ppm) COa SO. C1 B.O, (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) H11 95 1.300 34.75 2.40 _ _ _ _ (90.7—129.8) 100 1.310 34.98 1.69 _ _ _ _ November 1949 105 1.310 34.99 1.74 _ _ _ _ ‘110 1.311 35.02 1.73 _ _ _ _ 115 1.314 35.12 1.61 ____ 120 1.317 35.33 1.56 _ _ _ - 125 1.320 35.40 1.54 _ _ _ _ MM 70 1.297 34.37 2.08 _ _ _ _ (66.5—98.3) 75 1.299 34.46 1.96 _ _ _ _ November 1949 '80 1.300 34.52 1.96 _ _ _ _ 85 1.301 34.57 1.95 _ _ 1 _ 90 1.301 34.64 1.96 _ A _ _ 95 1.301 34.60 1.96 , _ _ _ U 80 1.292 33.95 2.03 _ _ _ _ (78.4—114.6) 85 1.294 34.22 2.01 _ _ _ _ December 1949 ‘90 1.294 34.11 1.99 _ _ _ , 95 1.299 34.45 1.85 _ _ _ _ 100 1.300 34.42 1.82 _ _ _ _ 105 1.300 34.47 1.80 _ _ , _ 110 1.300 34.47 1.77 _ - _ _ W 85 1.301 34.84 2.10 _ _ _ _ 850—1201) 90 1.304 35.99 1.92 _ _ _ _ arch 1950 95 1.304 34.85 1.89 _ _ ‘100 1.304 34.91 1.89 _ _ 105 1.303 34.86 1.89 _ _ 110 1.303 34.76 1.92 _ A 115 1.302 34.57 1.89 _ _ _ _ 120 1.302 34.59 1.90 _ 7 _ _ X 90 1.304 35.00 2.13 _ _ _ _ 89.2—127.2) 95 1.304 35.04 2.13 _ _ _ _ arch 1950 100 1.305 35.00 2.14 _ _ _ _ I105 1.305 35.01 2.14 _ _ _ _ 110 1.305 35.05 2.14 _ _ a _ 115 1.306 34.97 2.10 - 120 1.306 34.94 2.10 _ 125 1.306 34.91 2.05 _ S—28 80 1.289 33.50 1.46 , _ _ _ (77.4—102.1) 85 1.290 33.50 1.43 A _ _ _ September 1950 ‘90 1.293 33.94 1.39 A _ _ _ 95 1.295 34.11 1.35 _ _ _ _ 100 1.295 33.97 1.34 _ _ _ _ S—31 85 1.297 34.27 1.60 _ _ (84.0-117.1) 90 1.297 34.23 1.55 11 December 1950 '95 1.298 34.29 1.52 11 100 1.303 34.50 1.39 32 105 1.305 34.48 1.35 _ _ _ 110 1.305 34.50 1.33 32 115 1.305 34.62 1.35 _ _ _ _ GS—l 68 1.253 29.8 . . 0.95 8.8 (65.0—88.6) 70 1.255 30.1 9.3 10.3 .96 8.8 July 1954 75 1.253 30.5 9.3 10.7 1.06 12 ‘80 1.265 31.3 9.3 10.6 1.36 21 85 1.264 31.6 9.3 10.8 1.36 22 86 1.263 31.7 9.3 10.7 1.42 20 64.6 1 275 31.8 9.15 11.2 0.93 11 . . . . (64.6—89.5) 70 .274 31.9 9.19 11.1 1.01 11 3.00 . . . _____ September 1954 75 1.274 31.8 9.12 11.1 .94 11 2.81 5.90 10.39 .61 _ _ _ _ _ ‘80 1.276 31.5 9.18 10.8 .90 11 3.11 6.30 9.82 .57 ____________________________ 85 1.274 32.0 9.15 11.1 .92 11 3.11 6.36 9.91 .57 ____________________________ 86 1.272 31.3 9.10 11.0 .58 7 3.91 8.20 7.17 .45 ____________________________ GS—10 85 1.280 33.0 9.32 10.7 1.70 23 2.58 4.53 12.34 1.10 ____________________________ (83.1-114.5) 90 1.281 33.3 9.41 10.9 1.85 23 2.70 4.49 12.24 1.16 _______________________ January 1955 ‘95 1.288 34.6 9.37 11.8 1.95 51 2.66 4.73 12.24 1.19 ____________ 100 1.288 33.7 9.39 11.0 2.00 51 2.58 4.70 12.20 1.22 _____ 105 1.288 33.6 9.42 11.0 1.95 52 2.56 4.71 12.22 1.20 _____ 110 1.286 34.0 9.38 11.0 2.02 50 2.69 4.71 12.22 1.38 ____________ L—312 102 1.299 33.0 9.75 11.1 2.39 _ _ _ _ 3.23 3.78 11.70 1.23 536 (915—1291) 127 1.302 32.2 9.90 11.1 1.77 _ v A _ 4.32 4.08 9.82 1.45 357 November 1964 ‘Uppermost brine sample used to calculate composition of brine in units S] + S2 + S3 + S4 + S5. 2Sample from depth of 102 feet represents brine from unit S7; sample from 127 feet represents brine from $1, S2, and S3. the areal position in the deposit nor the total thick- ness of the Lower Salt bears a consistent relation to the K content. Ryan (1951, p. 449) reports 1.5 percent K as typical of brine pumped from the central part of Lower Salt; Dyer (1950, p. 41) reports 1.4 percent K as typical. In most wells in the central part of the deposit, total carbonate in the Lower Salt brines, expressed in the analyses as CO3 percentage, increases downward; to- ward the edges, the vertical variation is small, reflect- ing the more nearly monomineralic composition of this zone. The percentages of SO4 are generally be- tween 3 and 6. The percentages of SO4 in the brines increase toward the top in four core holes, increase to- LOWER SALT 45 ward the bottom in five, and are nearly constant in two.11 Brines in core holes GS—4 have abnormally high percentages of SO, although the associated salines are not high in sulfate minerals. This high SO, content in the brine may result from the relatively large quanti- ties of sulfate that were contained by the waters drain— ing from the gypsum-bearing lake deposits near the south end of the Slate Range, and the gypsiferous gouge of the Sand Canyon thrust fault along the nearby edge of the Slate Range (Smith and others, 1968, p. 14, 21). The percentages of Cl mostly fall be- tween 10.5 and 12 except near the edges, where the brines have a lower total salinity. The Cl content of the brines generally increases upward. The B,O7 per- centage normally increases downward; most percent- ages lie between 1 and 2 except near some parts of the edge, where they are substantially less. Analyses of minor elements are available for some sets of samples. The brine wells for which PO, analy— ses are available come from the central part of the de- posit. Although there is appreciable variation in P0, content of brines from different wells, the percentages in brine from a single well are relatively uniform. Ryan (1951, p. 449) reports an average of 590 ppm PO, and W. A. Gale (written commun., 1952) reports 535 ppm in brines from the central part of the Lower Salt. The analyses for F and Br are all of brines from the central parts of the deposit. The amount of F shows no clear trend; Br tends to increase toward the base of the layer. Ryan reports 30 ppm Li and 540 ppm Br, and Gale estimates 30 ppm Li and 580 ppm Br in the brines pumped by the American Potash and Chemical Corp. The amounts of S may increase downward, al- though the total range of variation is small. The values are mostly between 1,000 and 1,500 ppm. Ryan and Gale report average quantities of about 1,500 and 1,800 ppm S, respectively. Data on the As, Si, and I content are given in table 9 for core hole L—31. Esti- mates by Ryan and Gale of the I content of brines pumped from the central part of the Lower Salt are the 20 ppm and 25 ppm. The Sr content of a brine sample from the top part of the Lower Salt in GS—26 is 1.5 ppm (J. D. Hem, analyst, written commun., 1960). The W content of brine in the Lower Salt is reported by Ryan and W. A. Gale to be about 32 ppm. Samples from GS—1, GS—4, GS—10, and L~31 were analyzed for Ca and Mg, but the concentrations were below the limit of detection. ”Analyses showing higher percentages of SO. in the upper part of the Lower Salt may be of brines that were largely or entirely drawn from units S—4 and 8—5, as those units are more porous and were not blocked off during sampling of most cores. Analysis of a brine from core L481, which was collected when that layer was blocked off from other layers, show lower SO, values for brine from 577 than from units S—l through S43 in that core; although not sampled, units S—4 and 8—5 are almost certainly higher in SO, than 8—1, 872, and 8—3. The vertical variations in brine composition follow a broad pattern but do not bear a close relation to the mineral composition of the surrounding solids. The inconsistency of relations result in part from post-de- positional changes in the brines caused by the follow- ing processes: (1) fresh waters from the surface and the ground water of surrounding valley sediments have encroached on the salt bodies, probably affecting each salt layer somewhat differently; (2) the brines were allowed to mix when the mud layers that sepa- rated the Lower Salt salt layers were perforated by drill holes; (3) additional mixing occurred when these brines were pumped for processing by the chemical companies; and (4) the brine sampling procedure pro- vides a sample for analysis that is still more mixed be- cause it draws brines into the sample from porous zones above and below the sampled depth in the un- cased hole. Of the brines represented by analyses in table 9, those from S~31 are probably least affected by post— depositional changes as that core hole is well away from the edges of the body, and is in the southern part of the deposit, which, at the time of sampling and analysis, was not so extensively drilled nor as heavily pumped as other areas represented by analysis. In 8—31, the depths sampled for brine lie near or within the following salt units: 85 ft (25.9 m) ____________ S—7 90 ft __________________ S—6 95 ft __________________ 8—5 100 ft __________________ 8—5 105 ft __________________ 8—4 110 ft __________________ 8—3 115 ft (35.1 m) _____________ 8—1 The brines in 8—31 show their greatest change above and below the sample from 95 ft. This change is as ex- pected because, as noted in earlier papers (Smith and Haines, 1964, p. 47—50, 54; Smith, 1968, fig. 4), the only dry-lake stage inferred during deposition of the Lower Salt is represented by the top of unit 8—5. The two samples representing S—7 and 8—6 have greater percentages of K, 80,, Cl and Br. The five samples re- presenting S—5, 8—4, 8—3, 8—2, and 8—1 have greater percentages of Na, 00,, 3,0,, Li, S, and total dissolved solids. Of these lower five samples, the lowest three (representing 8—1, 8—2, S—3, and 8—4) have the great- est concentrations of these components, and the two above them (representing 8—5) have intermediate concentrations. 46 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. Na2504 EXPLANATION ® . S—1toS—5 S-GandS—7 Average brine Composition from table 10 A c 4 Mirabilite i ) Na2C03 NaCl Na2$04 EXPLANATION ® 0 HANKS'TE S—1toS—5 S—6andS—7 (mineral composition) Average brine Composition from table 10 B BURKEITE APHTHITALIT X (mineral (mineral A l _ 6 com osition) composition) na ys:s D From table 22 Thenardite Burkeite Aphthitalite Natron Sylvite Total C03 as Total K Nazcoa as KCI PARTING MUD 47 ESTIMATED BULK COMPOSITION OF THE UNIT The estimates of mineral composition of the salt layers (table 4), the measure of the probable error in those visual estimates (table 8), and analyses of the brines (table 9) provide a basis for calculating the rela- tive percentage and total quantity of the major water- soluble components in the Lower Salt. Because the top of S—5 represents a depositional break, the per- centage and total quantities are calculated for the lay- ers above and below this level. Table 10 tabulates the steps followed in the calculation. Combining these data and weighting them according to the volumes they represent yields these totals for the Lower Salt: Total quantity Relative amounts (grams X 1012) (percent, water-free) Na _____ 234 35.1 K ______ 9 1.4 Mg _____ 3 0.4 CO, _ _ _ _ 99 14.9 HCOa _ _ _ 83 12.4 SO, _ _ _ _ 56 8.4 Cl _____ 171 25.7 B,O, _ _ _ _ 11 1.7 H20 _____ 312 __ In a much more extensive study of the boron in Searles Lake, D. V. Haines, (unpub., 1956) estimated the total amount of B,O7 in the Lower Salt to be 17.1 X 1012 g. That amount, based on chemical analyses of brines from 155 wells and cores from 86 core holes, in- cludes the B,O7 that was in borax in the interbedded mud layers of the Lower Salt and is predictably larger. Calculations based on tables 3, 4, and 5 show that about 63 percent of the solid borax observed in the Lower Salt is in salt layers and 37 percent in mud lay- ers. As about 8.6 X 1012 of the 11 X 1012 g of 3,07 listed in table 10 comes from solid crystals in the salt layers, the amount of additional B407 in borax crystals em- bedded in M—2 to M-7 is estimated to be near 5.0 X 1012 g. The resulting total of 16 X 1012 g B,O7 in the salt < FIGURE 21—Composition of brines given in table 9 plotted on dia- grams that indicate phase relations in two 5-component sys- tems. A, Shows phase boundaries in, and projected from, Na2C03—NaZSO,-NaCl—NaHC03—H20system.B,Thosebound- aries in, and projected from, Na,COS—Na2804—KCl—NaCl—H2O system. Brines are plotted on the basis of their compositions as projected to plane of diagram; some points represent more than one analysis. See figures 35 and 39 and associated text for explanation of phase boundaries and method of plotting. and mud layers in the Lower Salt can be compared with Haines’ 17.1 X 1012 g to give a measure of the probable accuracy of the amounts given here. PARTING MUD The Parting Mud, of late Wisconsin age, is a layer generally 12—14 ft (3.7—4.3m) thick that rests on the top of the Lower Salt. It is composed chiefly of mud that contains megascopic crystals of gaylussite, pirs- sonite, and a little borax. Gaylussite is found in almost every core hole, commonly forming 5—15 percent of the unit. Pirssonite is less abundant, commonly 2—5 percent of the unit. In about half the core holes, a layer of megascopic crystals of borax is present near the top or bottom of the unit. The mud matrix con- sists of a dark-green to black mixture of microscopic crystals of halite, dolomite, clastic silicates, authigenic silicates, organic debris, and entrapped brine. In the upper one- to two-thirds of the unit, thin laminae of white aragonite are numerous. _ The Parting Mud has for many years been recog- nized as a stratigraphic unit by members of the chemi- cal companies operating on the deposit. This unit separates the Lower Salt and the Upper Salt, which differ chemically, and the Parting Mud helps main- tain those differences so that they can be utilized in the commercial operations. AREAL EXTENT AND VOLUME The outer limits of the Parting Mud are well beyond the limits of the area sampled by cores; mapping in progress shows that its lateral equivalents once ex- tended over an area of about 1,000 km2. Scattered remnants of it crop out around the edges of the basin, and it forms a persistent and unbroken layer in sub- surface sections. The Parting Mud is easily identified in almost all cores from Searles Lake. Its top is most commonly 15— 25 m below the present lake surface, its thickness gen- erally 3—4 m. Variations in the depth to the base and top of the unit are shown by contour maps (figs. 22 and 26); variations in its thickness are shown by an isopach map (fig. 23). The isopach contours show that the unit has a large area of relatively uniform thick- ness in the center and a zone of rapid thickening around the edges. Toward the southwest, a zone of thickening is shown, despite the lack of core evidence, because the outcropping equivalent of the Parting Mud is abnormally thick southwest of the lake. 48 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. TABLE 10.—-Estimated percentages and total quantities of water soluble components in the upper two and lower five units of the Lower Salt Units 8—1 + S—2 + 8—3 + S—4 + 8—5 Units 8—6 + 8—7 Na K Mg COa HCO, SO. Cl B,0, H20 Na K Mg CO, HCO3 SO, Cl 13.01 H20 Chemical composition of solids inferred from visual estimate‘ --—— 33.7 trace trace 17.1 15.7 50 18.8 02 9.5 32.2 0 0 20.7 21.0 0.7 12.3 0.3 128 Average error of visual estimatesx —-— -0.8 +0.6 +0.5 -3.4 “-3.4 +2.6 +4.8 +0.8 —‘1.6 -0.8 +0.6 +0.5 —3.4 3-3.4 +2.6 +4.8 +0.8 ‘1.6 Chemical composition of solids, wei ht percent adjusted {or probable error ————————————— 32.9 0.6 0.5 13.7 12.3 7.6 23.6 1.0 7.9 31.4 0.6 05 17,3 17.6 3.3 17.1 1.1 11.2 Chemical composition of brine" ————— 11.5 1.7 0 3.3 50 4.6 11.3 1.3 66.3 11.3 1.7 0 3.1 60 4.8 11.3 1.1 66.7 Chemical composition of combined solids and brines’ ——————————— 24.3 1.0 0.3 9.5 7.4 6.4 18.7 1.1 31.3 23.4 1.0 0.3 11.6 10.6 3.9 14.8 1.1 33.4 Total quantity of component in included salt layers, grams X 10" ' —————————————— 161 6 2 6 49 43 124 73 3 1 36 34 13 47 105 'Data in table 4, converted to weight percent. ’Bssed on comparison of visual estimates and chemical analyses of cores (converted to ions), data from table 8, average of all samples; table 8 also lists error in Ca of +0.4, reducing totals in this and underlying columns to 99.6. 'Error in CO, listed in table 8 divided equally between COJ amd HCOS, ‘Reduced by amount of H in HCO,. sArithmetic aver ‘Percentage of H O, is low, assumed to be 0. 1In weight percent, assumed porosity, 40 percent. 'Assumed specific gravity of salt plus brine-filled pores, 1.80; volumes from table 3. The volume of the unit, calculated within the arbi- trary boundary by the methods described previously, is about 480x 106 m3, about 10 percent more than the combined volumes of the mud layers M-2 to M—7 in the Lower Salt. MINERAL COMPOSITION AND LITHOLOGY The chief megascopic components of the Parting Mud are mud, gaylussite, pirssonite, and borax, in de- creasing order of abundance. Thin beds and isolated crystals of trona, halite, and northupite are found 10- cally. Prominent laminar beds of microscopic argonite crystals characterize the upper one- to two-thirds of the layer. The weighted average composition of the megascopic minerals, based on the visual estimates of mineral percentages and calculated by the method de- scribed in the section on the Lower Salt, is given in table 11. The data in table 11 indicate a tendency for the per- centage of megascopic gaylussite to increase laterally toward the thinner (more central) areas and the per- centages of pirssonite to remain nearly constant. Bo- rax has its highest percentages in the edge and central facies, its lowest percentages in the intermediate zone. Trona and northupite, which occur as thin beds and isolated pods of crystals, are mostly in the intermedi- ate zones. Megascopic crystals of halite occur sporadi- cally in the central zones. In general, though, areal variations in the megascopic mineralogy of this unit are not great because the area sampled by cores repre- sents only the most central facies of the entire unit which originally covered much of the floor of Searles Valley. Although there is local variation within and be- tween cores, megascopic gaylussite and pirssonite e of brine analyses given in table 9; averages exclude nalyses from core hole L—31. commonly form a higher percentage in the top and bottom meter of the Parting Mud, lower percentages in the middle. In more than half of the pirssonite- bearing cores, this mineral is concentrated near the upper or lower contact, possibly indicating the pene- tration of more saline waters from the enclosing saline layers which would tend to alter gaylussite to pirsson- ite (Eugster and Smith, 1965, p. 478—483). Borax is concentrated in the top few centimeters of the unit, but some occurs 20—60 cm up from the base. Arago- nite, mostly in the form of white laminar beds com- posed of microscopic crystals, is common in the upper 55—60 percent of the section except for the uppermost 30—50 cm of the unit. Variations in the color, abundance of gaylussite and pirssonite, the spacing and color (mineralogy) of la- minae, and the chemical composition of the organic components were used by Mankiewicz (1975, p. 7—8) to subdivide the Parting Mud into five parts. In his core B, which came from the northwest part of the lake, the Parting Mud had a thickness of 5.4 m, and changes in lithology were noted at levels 0.30, 0.75, 1.65, and 2.70 m below the top of the unit. The top unit (0.30 m thick) is characterized by laminae and abundant large gaylussite and pirssonite crystals; the next (0.45 m) contains fewer crystals but is finely laminated; the third (0.90 m) is finely laminated but contains more widely spaced yellowish-white (dolo- mite?) laminae and also contains two rhyolitic tuff beds; the fourth (1.05 m) contains more abundant and closely spaced dolomite(?) layers and some gaylussite (or borax?) vugs, and it is lighter colored, and the fifth (2.70 m) which is light gray and faintly laminated, is characterized by an absence of yellowish and white la- minae and the presence of numerous reddish-orange PARTING MUD 49 ,1 R43 E. 117°20’ R.44 E. l 4 3 / Pioneer Point I /% / ////q «a / 1 Argus (g 19 20 21 / / “° 3 3 32 35 36 1 32 V W / 1 k 1 NAVAL RESERVATION BOUNDARY fl 6‘\ 5 \g/W V L/ 12 7 EXPLANATION 9 \\ 1 T. 80 26 5. Structure contour Drawn on base of Parting Mud. \ Contours show depth below pre- \// sent surface, in feet. Interval 15 14 18 I 0 feet 13 I \ 1 . 22 23 1 24 19 20 I 0 1 2 MlLES 0 ‘l 2 3 KILOMETERS ) FIGURE 22.——Contour map on base of Parting Mud, Searles Lake. 50 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. W; R. 43 E. 117°20’ R. 44 E. Pioneer 21.2 4 3 2 1 6 5 ‘ | l 25 35°45’ V____J_ NAVAL RESERVATION BOUNDARY m my uouwm ¢ 9 10 EXPLANATION T 14 26' S. 131 Isopach ' Hachured in areas of thinning. \ Contour interval 2 feet /\ .127 15 14 .17‘4 Core-hole location NK Thickness of unit, in feet I \ \ 22 23 24 19 20 l 0 1 2 MILES F—fi—%——H 0 1 2 3 Kl LOMETE RS FIGURE 23.—Isopach map of Parting Mud, Searles Lake. PARTING MUD 51 TABLE 11.—Megascopic mineral composition of the Parting Mud [Based on visual estimates of Haines (1959), compositions in volume percent; estimated for intervals without cores shown in parentheses; calculated for contours and within boundary shown in figure 23; t = trace] Between contours Mineral Welghted 24+ 24—22 22—20 20—18 18—16 16—14 14—12 12—10 10—8 8—6 total percent Mud _______________ (90) 79 91 89 90 88 86 88 76) 68 88 Gaylussxte _____________ $6.0; 14 5.6 5.9 7.5 9.0 9.5 9.8 E20) 25 8.7 Pirssonite ____________ 3.0 3.2 3.6 3.4 2.6 3 3 3.3 1.7 (3.0) 4.2 3.0 Northupite ____________ -- —— —— —— —- —— t ,- —— ~— t Borax _______________ (1.0) 3.2 — — 1.1 1 .4 6 (1.0) 3.1 .5 Halite ______________ —— —— —— -— —— —— .1 ~— —— —— .03 Trona _______________ —— —— —— .3 .1 .3 —— —— —— 2 Number of cores between contours _______ 0 1 2 2 4 6 9 7 0 1 — — Percentage of volume between contours 1 5 8 11 14 14 32 14 1 t — — laminae. These units can be identified in most logs of the Parting Mud cores described by Haines (1959), and a sixth unit, composed of about 0.5 m of faintly laminated green mud, is found above the other five in cores from many parts of the lake. Hand-specimen and thin-section studies of these muds indicate that most of the megascopic gaylussite and pirssonite crystals grew after burial. Large crys- tals generally cut directly across bedding planes. Where such crystals cut aragonite laminae, the la- minae are not bent (Smith and Haines, 1964, fig. 15; Eugster and Smith, 1965, pl. 1); this shows that most of the mud’s compaction occurred prior to crystal growth, and that the growth was by volume-for-vol- ume replacement. Thin-section studies of the textural characteristics of muds are useful in the laminated segments of the Parting Mud. Sections show that the aragonite la- minae are thin bands produced by variations in the purity of calcium carbonate. Where laminae pairs are closely spaced, thickest beds may be either pure or im- pure. Where laminae pairs are thicker, the dark im- pure beds are thicker and the light beds of nearly pure calcium carbonate are thinner, rarely exceeding 1 mm. The top and bottom contacts of the laminae are usually sharp, even on a microscopic scale. Laminae defined by slight differences in carbonate percentage are commonly lenticular, and most laminae cannot be traced more than a few millimeters before pinching out. Beds of nearly pure carbonate tend to be continu- ous at least over the width of the core, and distinctive sequences of them could probably be recognized in cores from an area of several square kilometers. The thickness of calcium carbonate laminae ranges from several hundredths of a millimeter to about a millimeter. Several series of measurements give aver- age thicknesses for pairs of beds (one light bed and one associated dark bed) ranging from an average of 3.5 mm per pair in zones that represent sections of widely spaced types of laminar beds to an average of 0.75 mm per pair for zones that typify sections of fine laminar bedding. The finest laminae are in beds that average 0.35 mm per pair. If these laminae are inter- preted as annual cycles, sedimentation rates repre- sented by these three sets of measurements would be about 3, 13, and 29 yrs/cm, respectively. Sedimenta- tion rates implied for these muds by ”C dating average 38 yrs/cm. Presumably the discrepancy results from the lack of distinct episodes of rapid calcium carbon— ate deposition during some years. The mineralogy of the fine-grained muds of this unit cannot be determined satisfactorily with a petro- graphic microscope. Study with immersion oils per- mits identification of some crystals of carbonates, halite, clay, and silicates, but most of these minerals are coated with submicroscopic crystals of carbonates making consistent identifications impossible. The abundance of minerals in Parting Mud samples from L—12 and KM—3, as determined by X-ray dif- fractometer studies, is schematically represented in figure 24. Comparable data for the analyzed core GS— 16 are given in table 13. The amounts of gaylussite and pirssonite are variable, but they tend to be more con- centrated near the top and base. In core L—12, micro- scopic gaylussite and pirssonite crystals occur in beds also containing microscopic crystals of the same spe- cies. Dolomite is found in major or intermediate amounts at the top of two cores and at the base of all three, but it is also found in detectable amounts in other parts of the core. Aragonite is most abundant in the middle and upper parts; it is mixed with calcite in many samples. Halite, as microscopic and submicros- copic crystals, is a major constituent throughout most of the section in all three cores. The origin of this solu- ble salt is problematical, but the lower abundances near the top and base of the Parting Mud (at its con- tacts with salt bodies containing sodium chloride-rich brines) is the opposite of what should be found if the 52 DEPTH, IN FEET DEPTH, IN FEET SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. 31.9 24.96 84 — E 86 — u, 2 l- LIJ a 2 E 88 — . I l— l LIJ D 90 — 92 — — 93.3 28.44 75.0 22.86 76 — 78 — (I) 0: LLI 80 ‘ l- LIJ 2 Z 82 — — 25 :5 '- D. Lu 0 84 — 86 — 37.5 26.67 CE Segments of core recovered G gaylussite P pirssonite D dolomite A aragonite C calcite H halite CORE L—12 G P D A C H Sx Cm %% Tr, Bx -Tr \l/\l/ \l/ EXPLANATION Estima E ted abundance from X— ray charts Trace to doubtful component Minor to trace component Intermediate to minor component Major component Abbreviations Sx salts in minor amounts ~ Tr, trona; Bx, borax; Np, northupite clastic minerals — mostlv quartz, feldspar, mica, and clay halite formed after burial from brines that migrated into the mud layer. The smaller amounts of trona and borax occurring sporadically in core L—12 may come from the desiccation of entrapped brines. The quanti- ty of northupite at the base of the unit in L—12 is nota- ble and is similar to the quantities noted by Mankiewicz (1975, p. 87) in the top centimeter of the Parting Mud. Clastic silicates form minor to interme- diate amounts of the muds in the middle and upper middle parts of the section. Authigenic silicates are not so abundant in the Parting Mud as in underlying layers. Hay and Moiola (1963, table 1) report traces of analcime from the low- er part of the Parting Mud in GS—2, and a thin zone of tuff altered to phillipsite from 1—2 ft (0.3—0.6 m) below the top of the Parting Mud in GS—14 and 17. The acid-insoluble fraction of these muds is largely composed of elastic fragments plus organic material that is amorphous to X-rays. The mineralogy of this fraction of a sample from the interval between 86.7 and 87.3 ft (26.43 and 26.61 m) in core hole L—12 was studied by R. C. Erd (written commun., 1957). The clay minerals were found to be mostly mont- morillonite with a little hydrous mica. About 10—15 percent of the sample was composed of the following, in the approximate order of abundance: biotite (subangular to subrounded grains, dark green), chlo- rite, feldspar (fresh subangular grains of albite-oligo- clase and some fresh subangular microcline), quartz (subangular), hornblende (angular, blue green), glass shards (n below 1.51), zircon (traces, subround, pris— matic, colorless), sphene (traces, subround), epidote (traces, subround), magnetite (traces). Four Parting Mud samples from core L—12 were studied by Paul D. Blackmon, of the Geological Sur- vey, for grain-size distribution of the elastic minerals; the results are shown in table 12. Coarse sand and gravel grains are absent. Except for the uppermost sample, medium sand grains are nearly absent. Fine sand and very fine sand consistently form about 15 percent of the total, but most of the clastic material is silt or clay sized. The quantities of sand seem greater than would be transported by lake waters to this location. Core L—12 is in Parting Mud sediments that were mostly deposit- ‘FIGURE 24.——Mineral abundance in samples from two cores of Parting Mud. Samples from core L-12 represent only silt— and clay-sized components; samples from KM—3 include both mega- scopic and microscopic components. Determinations for L—12 by R. C. Erd; for KM—3 by R. J. McLaughlin and G. 1. Smith. PARTING MUD 53 TABLE 12.—Size distribution of acid-insoluble material in four samples of the Parting Mud, core L—12 [Original 350 g samples digested in cold diluted (4:1) HCl; washed in cold water by decanting; dried on a steam bath. Each residue was then treated as follows: the sample was quartered to an appropriate size and stirred vigorously in distilled water for 15 minutes utilizing a milkshake-type stirrer. Sodium metaphospbate was added as a dispersing agent. After stirring, each sample was further dispersed in an ultrasonic separator for approximately V2 hour. Sandsized particles were then removed by wet seiving, dried, and seived using screen sizes of Wentworth’s scale. Percentages of silt and clay were determined by standard pipette analyses. Analyses by P. D. Blackmon. All values are in weight percent] Size distribution (sizes in mm) Interval sampled Percent Coarse sand Medium Fine Very fine (ft below ground acid to gravel sand sand sand Silt Clay surface) insoluble (>05) (025—05) (0.10—0.25) (00625—010) (0002—00625) (<0.(X)2) 82.2—82.6 — — — — 20.7 0.0 4.2 10.5 8.2 50.2 27.0 86.7—87.3 — — — — 28.9 .0 .5 4.8 11.2 61.4 22.0 89.4—89.9 — — — — 23.9 .0 .8 4.7 10.3 55.0 29.2 91.1-91.7 - — — —— 22.9 .0 .0 2.9 5.4 66.5 25.1 ed in 100 In or more of water, 5 km from the nearest shore, and several times that distance from the inlet for most of the water that flowed into the basin during high stands of the lake. It seems unlikely that fine and very fine sand grains could have been transported by current to that area. Transport by wind seems likelier, and the common occurrence of frosted sand grains (Hay and Moiola, 1963, p. 320, 322) supports this premise. If all the fine and very fine sand is wind- blown, that depositional process accounts for 3.8 per- cent of the Parting Mud volume in L—12. The Parting Mud in this core is about 3.44 m thick and accumulat— ed during a period of about 13,500 years; this indicates an average depositional rate for sand that was air- borne for a distance of several kilometers—to the mid- dle of the basin—of about 1 cm per 1,000 years. As finer material was undoubtedly also introduced dur- ing wind storms, aeolian deposition was actually more rapid. If all the acid-insoluble material was transport- ed by wind, aeolian deposits accumulated at the rate of about 6 cm per 1,000 years. The types of clay included in the clastic fraction of the Parting Mud were studied by Droste (1961, fig. 5) and reported to be predominantly illite, with much smaller percentages of montmorillonite; chlorite or kaolinite were detected only in samples from the edge facies where chlorite was introduced from the adja- cent mountain ranges (especially by the Slate Range on the east, where large areas of chlorite-bearing rocks and fault gouge are exposed). R. C. Erd (written com- mun., 1957) and Hay and Moiola (1963, table 1), how- ever, report montmorillonite as the dominant clay mineral and illite as subordinate. Organic material in samples from the lower middle part of the Parting Mud in GS—12 was studied by Val- lentyne (1957), who reported several types of caro- tenes and xanthophylls. Samples for depths of 1.46— 1.65 m below the top of the Parting Mud in the same core were studied by R. C. Erd (written commun., 1957). In the acid insoluble residue, he noted abun- dant fragments of chitinous material, apparently de- rived from the brine shrimp Artemia, closely associated with the montmorillonitic (and possibly the aragonitic) mud was an orange-yellow compound that was clearly organic and thought probably a carot- enoid. Green stains on some mineral grains suggested algae as a source. Mankiewicz (1957), in a more recent study, extract- ed and identified by chromatographic techniques a large number of hydrocarbons, terpenes, and sterols in four cores of the Parting Mud. Green algae, blue-green algae, and vascular plants were the dominant sources of organic components. Their abundances varied in response to changes in stream inflow, lake salinity, and the degree of anaerobic conditions that affect preservation. Zones containing concentrations of vas- cular plant debris, as indicated primarily by increases in the amounts of straight chain saturated hydrocar- bons (n-alkanes) that have odd-carbon numbers in the 023—036 range, are interpreted to represent periods of strong inflow which introduced plant remains from other basins. Zones containing high percentages of al- gal remains, as indicated by concentrations of hydro- carbons that have odd-carbon numbers in the 015—0“, range, are interpreted as resulting from times of re- duced inflow and high algal production within the lake. The extent of diagenetic reaction of original or- ganic compounds to related compounds is used as a measure of the development of anaerobic conditions on the lake floor. The Parting Mud contains enough brine in its inter- stices to give the material a moderate plasticity. The percentage of brine in this mud was estimated by re- cording the weight of water lost when samples were heated to about 60°C12 for several hours. Amounts of water range from about 10—42 weight percent (fig. 25). '1 Gaylussite (and presumably pirssonite) is dehydrated to Na,Ca(C03)2 at tempera— tures approaching 100°C. At 60°C, no dehydration was noted in test samples. 54 75 Top of Parting Mud | I | I — ~23 oo o | I ESTIMATED DEPTH, IN FEET ESTIMATED DEPTH, IN METERS (—— Base of Parting Mud I I I I 0 1O 20 30 40 50 FORE WATER, IN WEIGHT PERCENT FIGURE 25.—Weight percent of pore water in samples of Parting Mud. Samples from core KM—3; values obtained by measuring weight loss from fresh moist sample during 17 hours of heating at 60°—65°C. All this evaporated water, though, was in the form of brine that probably had a density of about 1.25 con- sidering the amounts of halite that exist in the sample. For this reason, 10 weight percent water would indi- cate 12.5 weight percent of entrapped brine, and 42 percent water 52.5 percent brine. The highest per- centages are about 1 meter below the top of the unit; distinctly lower percentages are at the top and in the basal zones. The greatest changes occur between units of massive and laminated lithologies; the top three and lower three samples were massive, and the middle six samples were distinctly laminated. These litholo- gic changes coincide with changes in the amounts of aragonite and calcite in the solids and clearly reflect shifts in the character of deposition. It is likely that the higher percentages of pore water in the middle six samples are products of these changes. CHEMICAL COMPOSITION Nine samples from core GS—16 were partially ana- lyzed to determine their major element composition. Table 13 gives those analyses and lists X-ray diffrac- tion data for the minerals in the same samples. The analyses list the evaporite constituents as “water- and acid-soluble components”, and the elastic constitu- SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. ents as “acid-insoluble components.” In making the major element analyses, difficulties were discovered that introduced small errors into the results, and insufficient amounts of the samples re- mained to complete the analyses or repeat the deter- minations. For example, one portion of all the samples was heated to 500°C before treatment with acid in an attempt to quantify the percentage of dolomite in the sample. Under correct conditions, dolomite'decom— poses at this temperature to CaCO,, MgO, and C0,, but the results showed that this reaction was incom- plete, possibly because of interference from other components in the mixture. When heated, however, some of the elastic silicate minerals apparently fused with the NaZCO3 or Na.,,B,O7 in these samples and their components subsequently dissolved in acid. This ap- pears to be the reason that about 65 percent of the total A1203 and 90 percent of the total Fe203 were found in the acid-soluble fraction. Inasmuch as SiO2 was not determined on this fraction and A1203 was de- termined by difference, all SiO2 in this fraction was reported as A1203 making the total alumina in the acid-insoluble fraction too high. The acid-soluble CaO, MgO, NaZO, and K20 percentages were deter- mined from this heated fraction. Although the report- ed amounts of these components seem consistent with the observed evaporite suites, they may be slightly high if they include contributions from the clastic minerals. For the same reason, reported amounts of SiO2 and other elements in the elastic fraction may be low because they represent the concentrations that re- mained in the acid-insoluble residue after the treat- ment described above. Total H20 could not be determined accurately. The samples were first ana- lyzed by heating to about 800°C and the weight loss determined, but the loss was found to include some C02, organic H, and possibly organic C, as well as H20 and OH; subsequent analyses on the three samples for H2O+ and H20— are given in a footnote to table 13. The analyses of table 13 show the approximate ma- jor-element content of each fraction. On the average, about 20 percent of each sample consists of elastic ma- terial, nearly 80 percent is evaporite, and about 1—2 percent is organic. Combining this information with the X-ray diffraction data verifies that the water- and acid—soluble evaporites are clearly dominated by gay— lussite, pirssonite, dolomite, calcite, aragonite, and halite. The acid-insoluble clastic minerals are composed mostly of SiO,, although the X-ray identifications show chlorite, feldspar, mica, and clay to be more abundant than quartz in most samples. The combined percentages of acid-insoluble components (exclusive of organic C) show that the greatest amounts of elastic PARTING MUD 55 TABLE 13.—Partial chemical analyses of samples from core GS—16 in Parting Mud [Major-element analyses by M. J. Cremer, L. B. Schlocker, H. N. Elsheimer, and F. 0. Simon; minor-element analyses by Harry Bastron. Analytical techniques: (1) Except for CO2 analysis, water» and acid»soluble components were separated from insoluble components by igniting sample at 500°C for 48 hours, immersing sample in dilute (2N) HCl (ca. 80°C) until C02 evolution ceased, then filtering. (2) The filtrate included some components that presumably would have remained in the acid-insoluble fraction if the sample had not been heated sufficiently to cause limited Na,COa fusion. Of these, the R20, group was precipitated as hydroxide and total Fe determined as Fe203 by potentiometric titration methods, TiO2 and MnO determined by colorimetric methods, and A120a by difference. The AL203 and Fe203 percentages were added to the percentages of these components determined on the insoluble fraction. (3) CaO and MgO were determined in this filtrate by gravimetric methods, and NaZO and K0 by flame photometer methods. (4) S03 and 132.0J were determined on separate samples; they were immersed in boiling water for about 1 hour, SO, determined on the filtrate by X-ray fluorescence and 3203 by volumetric methods. (5) Cl was determined on separate samples by two methods: five samples were immersed in water at 80°C for several hours, filtered, and Cl determined gravimetrically in the filtrate; four samples (values followed by a “7" because considered less reliable) were fused in LiBO2 and total Cl determined by a colorimetric method. (6) Total carbonate determined on dried but unheated sample by adding HJPO. and trapping the evolved C02 in a train. (7) The acid insoluble portion represented by the heated residue after filterin , washing, and drying was fused with LiBOz, dissolved in HNOM and from this solution, the A1203; total Fe and FeZOa, CaO, and MgO determined by atomic absorption metho s, the Na.,0 and K20 by flame photometric methods, and the SiO2 by difference. (8) Organic carbon determined by igniting dry sample at 1,000°C, determining evolved CO2 in a train, and subtracting the C02 present as carbonate from this quantity. (9) Minor elements determined by emission spectrographic techniques on the acid insoluble portion; the results are semiquantitative and are assigned to geometric brackets reported as their midpoints (such as 1, 0.7, 0.5, 0.3, etc.) with their one standard deviation precision estimated at plus or minus one bracket; n.d., not detected. (10) X-ray diffraction analyses of total untreated samples by R. J. McLaughlin and G. 1. Smith] Major elements (percent of total sample) Water- and acid-soluble components Acid-insoluble components Sample Depth in CaO MgO Nap K20 Total SO3 B20J Cl SiO,2 A120a Total Fe CaO MgO N820 K20 Or anic No. core (ft)‘ Carbonate as e2 3 car on, as CO2 as PM 1“ 70.6—71.5 13.3 8.6 14.4 2.9 20.1 1.9 0.8 9.9 13.2 2.8 1.4 0.1 0.04 0 2 0.2 1.7 PM 2 71.5—72.5 14.7 5.0 19.0 2.9 18.9 2.2 0.9 1.9 12.5 2.3 1.1 0.1 0.04 0 02 0.02 1.6 PM 3 72.5—73.3 12.2 5.3 19.8 3.0 13.2 (‘) 1.1 18.8 15.4 3.1 1.4 .2 .5 3 .3 1.6 PM 4 73.3—74.2 6.2 4.7 20.7 4.1 8.0 ‘) 1.5 15.9? 19.8 4.5 1.8 .3 .1 5 .5 1.8 PM 5 74.2—75.1 10.7 3.9 17.6 4.3 10.8 ‘) 6 19.6? 19.9 4.3 2.0 .3 .09 7 .5 1.6 PM 6 75.1—77.0 19.6 4.3 14.6 3.0 17.6 1.3 9 12.0? 15.1 2.2 1.3 .2 .06 3 .4 (‘) PM 83 78.0—79.0 17.1 4.8 13.6 3.4 16.4 1.2 6 11.5 17.3 3.6 1.6 .2 .05 3 .2 2.2 PM 9: 79.0—80.0 12.9 5.3 15.8 2.9 18.4 1.5 8 7.3 15.2 3.8 1.6 .03 .01 02 .02 1.3 PM 10 80.0—81.1 11.2 6.8 17.2 2.9 18.2 1.5 8 12.0? 14.0 3.1 1.5 .1 .04 3 .4 0.7 Minor elements (parts per million of aCid-insoluble portion) SaIrInple B Ba Cr Cu Ga Mn Mo Nb Sc Sr Ti Yb V Y Zr 0. PM 1 300 500 15 1.5 5 50 3 7 3 200 2000 0.7 15 7 150 PM 2 300 500 20 2 5 150 7 7 3 200 2000 .7 15 7 200 PM 3 300 500 15 2 7 70 5 7 3 200 2000 .7 20 7 150 PM 4 500 500 15 1.5 7 70 5 7 5 200 3000 1 50 10 150 PM 5 500 500 15 1.5 7 7O 5 7 5 200 3000 1 50 10 150 PM 6 300 500 15 1.5 7 70 5 n.d. 5 200 2000 0.7 30 7 150 PM 8 100 300 15 0.7 5 70 n.d 5 3 200 2 n.d. 15 n.d. 70 PM 9 100 50 n.d. n.d 30 n.d 5 n.d. 10 700 n.d. 7 n.d. 100 PM 10 500 500 15 0.7 5 70 n.d 5 3 200 2000 0.7 20 7 100 Minerals identified by X-ray in untreated sample (in apprornmate order of decreasrng abundance) PM 1 __________ Dolomite, halite mica chlorite, clay, calcite('{). . ‘ PM 2 __________ Pirssonite, dolomite, halite, mica, clay, chlorite, ara onite, feldspar, calcrte. PM 3 __________ Halite, aragonite, mica, chlorite, clay, dolomite, fel spar quartz, cac1te. PM 4 __________ Dolomite halite, mica, clay, chlorite, felds ar, quartz, calcrte. _ PM 5 __________ Halite, aragonite, calcite, mica, chlorite, c a , feldspar, dolomite, quartz(?). PM 6 __________ Aragonite, dolomite, mica, chlorite, clay, ha ite, feldspar quartz, calcite. PM 8 __________ Ara onite, halite, dolomite, calcite, mica, chlorite, clay, feldspar, quartz. PM 9 __________ Ga ussite, dolomite, mica, clay, chlorite, halite, calc1te felds ar(.). PM 10 _________ Do omite, gaylussite, mica, clay, chlorite, halite, amphibole(? . ‘Parting Mud in (ES-16 lies between 70.6 and 82.2 ft (21.52 and 25.05 m). 2Percentage possibly low because of loss during original heating and acid treatment. “Analyses for HZO‘ (weight lost on heating at 100°C) as follows: PM 1 = 6.7, PM 8 = 3.1, PM 9 = 11.1. Analyses for 1‘120+ (wei ht percent of H lost during heating from 100° to 800°C, expressed as H20) as follows: PM 1 = 1.7, PM 8 = 2.8, PM 9 = 2; percentages may be high because H may have been erived from both H20 and organic compounds. ‘Not analyzed because of insufficient sample. minerals are in a zone just above the middle of the sin center. Parting Mud. A high concentration of clastics in these Analyses for U in 29 samples of the Parting Mud by horizons is consistent with the data on sediment sizes Mankiewicz (1975, table 2—1) shows an average value in table 12. Slightly increased concentrations of Ga, of 12 ppm with a range from 1 to 42 ppm. The higher Sc, Ti, Yb, V, and Y in this zone may also reflect this values occur near and below the middle of the Parting distribution. Unpublished field observations of the Mud, a zone he interprets on the basis of the organic lake sediments exposed in the surrounding valley content to have been deposited during a period of show that there was a lake recession during this part maximum lake size and inflow. of its history that would have allowed more and The analyses of MgO give values that average 5.4 coarser elastic material to be transported into the ba- percent but are as high as 8.6 percent; if all the MgO 56 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. was contained in dolomite, the dolomite percentage would average 25 percent and be as high as 40 percent. The analyses for carbonate, expressed as C0,, average 15.7 percent. If all the CO2 was in dolomite, the aver— age dolomite percentage would be 33; if it was in cal- cite or aragonite, their combined average percentage would be 35; if in pirssonite, its average percentage 43; and if in gaylussite, 53. The actual percentages of each are much less because most samples contain two or more of these minerals, but these values provide up- per limits. The average SOa content is 1.6 percent, the B203 content 0.9 percent. These values imply a thenar- dite content in the analyzed samples of 2.8 percent and a borax content of 2.5 percent. Using the average brine values in table 17, combined with an estimated interstitial brine content in the Parting Mud (fig. 25) of about 25 percent, it appears that desiccation of the entrapped brine could account for all but 1.1 percent of the thenardite and 1.9 percent of the borax. Al- though borax is commonly observed, thenardite and other sulfate minerals have not been observed in the Parting Mud; some or all of the reported SO3 may have been produced during analysis by oxidation of pyrite, a mineral that has been observed in trace amounts in the Bottom Mud. The Cl content averages 13.1 percent and is as high as 19.6 percent; the implied average halite content is 22 percent with a maximum of 32 percent. As the Cl content of the average brine in the Upper Salt (table 17) is 12.2 percent, the Cl con- tent in the Parting Mud clearly reflects the presence of crystalline halite in the mud. The CaO content is of special significance in recon- structing the nature of the depositing lake. If the aver— age value of 13.1 percent CaO is not inflated greatly by contribution from the elastic material, it virtually re- quires the lake to have been stratified with a fresh layer at the surface that introduced additional cal- cium each year; details of this reasoning are given in the section on the processes of sedimentation of the mud layers. The average organic C content of 1.6 percent indi- cates an average content of organic compounds that probably exceeds 2 percent. The Parting Mud in core GS—16 is 12.2 ft (273 cm) thick and has an approxi- mate density of 2.2. If the average organic content is 2 percent, each square centimeter of the Parting Mud represents a column of sediment that contains 16.4 g of organic matter. As the unit was deposited in about 13,500 years, the minimum rate of organic production and accumulation was about 1.3 X 10'3 g/cmZ/yr. Ac- tual rates were higher because some of the organic ma- terial has decomposed and been lost as CO2 and CH,. These data show that the Parting Mud is a marl dominated by carbonate minerals but containing ap- preciable quantities of halite and subordinate amounts of clastic components and other evaporites. The organic percentage is low but contributes dispro- portionately to the physical properties of the mud. UPPER SALT The Upper Salt is the largest of the commercially worked saline layers in the Searles Lake deposit. It ex- tends over an area of about 110 km2 and has typical thicknesses near its center of about 15 m. This unit was the first to be exploited as a source of commercial brine; from it have come the bulk of the industrial chemicals produced from Searles Lake. The salt body is lens shaped; its upper contact with the Overburden Mud is gradational but generally concave downward, and its lower contact is concave upward as it rests in the symmetrically shaped basin formed by the top of the Parting Mud (fig. 26). Most of the Upper Salt layer is composed of salines. The most abundant minerals are halite, trona, and hanksite. Minerals occurring in smaller quantities are borax, burkeite, pirssonite, thenardite, aphthitalite, sulfohalite, and mud. Compared with the Lower Salt, the solids of this layer contain more halite and hanksite, about the same amount of borax, and less trona, burkeite, then- ardite, and mud. The brine is higher in K and Cl and lower in total CO3 and B,O,. AREAL EXTENT AND VOLUME The areal limits of the Upper Salt, are about the same as those given for the saline layers within the Lower Salt (114 kmz). The variations in thickness of the Upper Salt are shown by the isopach map (fig. 27). In making this map, deciding on the position of the contact between the Upper Salt and the Overburden Mud was found to be a very subjective matter. In general, the contact was placed at the top of the uppermost zone in which the salt beds dominate. According to the isopach map, the configuration of the Upper Salt is symmetrical; it has a zone of rapid thickening near the edges and a large area of more gradual thickening near the center. This distribution of thickness is similar to that of 8—5, the largest saline layer in the Lower Salt, but it appears more regular than the other saline layers within that unit. This reg- ularity is partly due to the large contour interval used for the map of the Upper Salt which minimizes the small variations introduced artificially by differences in the measuring techniques used during coring and logging. UPPER SALT O1 \1 ”mummy ,4 R. 43 E. 117°20' R. 44 E. l 6 5 @7/ 1 6 5 Pioneer Point m w ._I__2___..__._._.__2_2_._ r —NAVAL-RESERVATION BdUEERY EXPLANATION :\ 0 11 12 50 Structure contour Drawn on base of Upper Salt. Contours Show depth below pre- \\\ m “gum/Nam; \ 90 \ \z 1 sent surface, in feet. Interval ‘7 \ \ 1 0 feet 1 3 \k 22 23 24 o 1 2 MILES }—_T——|—T——I_J o 1 2 3 KI LOMETERS FIGURE 26.—Contour map on base of Upper Salt, Searles Lake. 58 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. ,4 R. 43 E. 117°20’ 12.44 E, ' 1 6 5 4 3 2 1 6 5 Pioneer Point . ‘ .1 1.3 I | 16.7. 14.9. 7 8 .11V’\.\ 12 7 8 11.1 39.2 :48\\ 1 o . / O o . Trona / 27.1? 18 % 16 15 . fl 7 T. 0 2 5 ‘3) 5‘ 464' 44 0' 53.1 c 45.2 c e . 35°45' — 37 0 V7 71 4‘ 46 3 9 1.90 19 I 2 ° . .39. 24 '35 1 20 l 2 . 23 ¢o . _ 58.1 50 \ o ' 50.6: ' ' I ' 69.5\6.0\ , Argus 19 2 59,9 3O . \ .1? “38 30 29 24.0 5 56.9 56.2 \ o ‘ o 4 4.72 _ _ . ‘ _ v . \ - \_/_\ God 241 32 1 34.3. I 31 32 - . 33 34.5 C i v D 4 50m 36 41.2 3 2 C . . 46.0 46.0 3.1 NAVAL RESERVATION BOUNDARY 30.1 20.1_ 29.3_ 48.0 m 55.0 40 .\.\\§ 4 \ 45.0 - ' 39.5= 48.7 \é’olj 3 0 3 45.0, 249.0. ‘30 35]. 42.4. ‘30 n; An MDUN mm )0 0 1O EXPLANATION T 26 70 S- 12.0 Isopach Hachured in areas of thinning. Contour intervalIOfeet 15 .20.1 Core-hole location Thickness of unit, in feet | 22 23 24 19 20 I O 1 2 MILES 1-——T——'—1'—_I—l 0 1 2 3 KILOMETERS FIGURE 27.—Isopach map of Upper Salt, Searles Lake, Calif. UPPER SALT 59 Using the isopach map of the Upper Salt (fig. 27), the volume is calculated to be about 1,050 X 106 mg, or about 1 kma, about twice the total volume of the saline layers in the Lower Salt. More than half the total vol- ume of the Upper Salt lies within those parts of the unit that are more than 12 m (40 ft) thick. MINERAL COMPOSITION AND LITHOLOGY Trona and halite are the major mineral components of the Upper Salt, together forming more than three- fourths of the total volume. Hanksite is the third most abundant mineral, borax the forth. Smaller amounts of mud, burkeite, pirssonite, thenardite, aphthitalite, and sulfohalite occur. Quantitative estimates of the mineral composition of the Upper Salt were made in the same manner as for the saline layers in the Lower Salt. The results of these calculations, given in table 14, illustrate an areal mineral zonation that can be correlated with the unit’s thickness. Some of the mineral percentages in individ- ual cores vary markedly. The averaged values indicate the following trends. Trona reaches its maximum per- centage in the thinner parts of the layer and dimin- ishes toward the thicker parts.13 Halite follows the op- posite trend. Hanksite, the predominant sulfate min- eral in the Upper Salt, is slightly more abundant in the intermediate thickness zones. The highest per— centages of borax are normally in the thicker zones. Aphthitalite is concentrated in the thicker parts of the deposits. Thin beds of mud occur sporadically throughout the Upper Salt. Pirssonite is in the mud layers. Other areal concentrations noted are unrelated to mmmin the 0—10 contour area, 08—25, is not representative of that facies; it consists of an unusually large amount of borax that is associated with halite, whereas most cores obtained by the chemical companies from that facies mostly con- tain only trona. thickness. Hanksite is most abundant in the western, central and eastern parts of the unit, and is deficient to the north and south. Burkeite and thenardite are higher than average along the south and southeast edges of the deposit, apparently because of the source of nearby sulfate in the Slate Range (Smith and oth- ers, 1968, p.14). Borax is more common in the western, central, and eastern parts of the Upper Salt body. The vertical distribution of minerals varies some- what from one area to the next. A few centimeters of borax occurs at the base of the Upper Salt (or in the top of the Parting Mud) in about half of the GS series of cores, most are from the central part of the deposit. Almost all cores of the Upper Salt have a 1- to 3—m layer of trona at the base. The basal few centimeters of this bed contains major amounts of northupite as a fine-grained interstitial material. This trona layer maintains its thickness toward the edges of the de- posit and constitutes the entire Upper Salt in most of the outermost facies. Above the bed of trona, except in the outermost facies, halite is the predominant com- ponent. Its percentage increases in an irregular man- ner toward the top of the Upper Salt. In some parts of the deposit, the halite-rich zone extends to the top of the Upper Salt; in others, in particular in the central facies, it is overlain by a layer 2—5 m thick composed largely of hanksite and borax. The stratigraphic position of hanksite concentra- tions in the Upper Salt is variable but is generally re- lated to areal position. In the central part of the deposit, the largest concentrations of hanksite are generally at the top of the Upper Salt, as in GS—15 and GS—16. In the edge facies of the deposit, hanksite is more commonly concentrated just above the basal lay- er of trona, and may be nearly pure or mixed with ha- lite and trona, a distribution pattern found in GS—10, GS—12, and GS—17. Other cores have distributions of TABLE 14.—Mineral composition of salines in the Upper Salt by contour interval [Based on visual estimates of Haines (1959), compositions in volume percent; calculated for contour shown in figure 27; t = trace] . Between isopach contours Weighted Mineral total 0—10 10—20 20—30 3040 40450 50—60 60+ percent Halite _____________________ 61 33 46 48 43 46 43 Trona _____________________ — — 55 33 27 33 39 34 Hanksite ___________________ — — 6.1 19 18 19 10 17 Borax _____________________ 39 8 .0 .8 4.6 1.4 2.8 3.1 Burkeite ____________________ — — .4 .8 — — — — .8 — — .3 Pirssonite __________________ — — 1.4 .1 .1 .3 1 — — .2 Thenardite __________________ — — .8 3 — — — — - — — — 1 Aphthitalite __________________ — — — — t — — t .5 .03 Sulfohalite __________________ — — t — — t t .1 .01 Mud ______________________ .3 2.7 2.0 1.0 2.2 2.6 1.3 2.0 Number of cores between contours _ _ _ _ 1 7 4 4 4 1 — — Percentage of total volume between contours _____________ 2 7 20 31 20 6 — — 60 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. hanksite that are combinations of these two patterns; hanksite may be at both stratigraphic horizons (for example, GS—2, GS—19) or be distributed throughout the section (for example, GS—12, GS—27). Borax characteristically occurs at two stratigraphic positions in the Upper Salt. The lower and the thinner of the two is at the base of the section, and crystals may extend down into the top few centimeters of the Parting Mud. It is most commonly associated with fine-grained trona and northupite and in some places with a little hanksite. The second characteristic posi- tion is a meter or so below the top of the Upper Salt, at the base of the hanksite concentration if any occurs at that level. The Upper Salt is a porous aggregate of interlocking crystals. Published estimates of the total voids range from 20 to 60 percent, estimates between 40—47 are most common. In this paper, a porosity of 40 percent has been assumed. Visual observations suggest that about half of the total pore space contributes to the useful permeability of the crystal mass. The permeability varies from one layer to another. The muds or mud-rich zones appear almost imperme- able. Burkeite, most commonly in the form of hard massive vuggy material, probably has a very low per— meability except where the vugs are numerous. Trona, especially zones composed of fine-grained aggregates, likewise would have a lower permeability (see Haines, 1959, pl. 10); where the coarse-grained bladed or fi- brous habit is predominant, the permeability is high. Beds of halite probably have low permeability where they are tightly intergrown but may have a high per— meability where the crystals are loosely packed or vuggy (see Smith and Haines, 1964, fig. 8). Hanksite zones have, in many instances, the highest permeabil- ity because of the abundant large crystal-faced cav- ities that characterize the thick beds of this mineral (see Haines, 1959, pl. 9). Much of the bedding in the Upper Salt is indistinct because the salts have been recrystallized since burial. Some bedding, however, is clearly defined. The best examples of this are in the fine-grained trona beds near the base of the Upper Salt and in scattered zones containing interbedded layers of salines and mud. The bedding in trona is defined either by fluctuations in the percentage of mud impurities or by differences in the transparency of the fine-grained trona (see Haines, 1959, pl.10; and Smith and Haines, 1964, figs. 6 and 7 in which the opaque beds appear white; the more transparent beds appear light gray). Most such layers range in thickness from a millimeter to a few centimeters, but there is wide variation. Calculations based on the solubility of trona and the maximum probable evaporation rate show that layers to a half meter thick could represent annual deposits if no new water was added, but most annual layers would be thinner because there generally was some inflow and rainfall on the lake surface. Beds defined by layers of mud are clearly the result of inflowing muddy water during temporary wet periods. Most of these are found in the upper part of the Upper Salt, and their lack of lateral continuity suggest that they were formed in locally flooded areas rather than in a large lake capable of distributing the sediments evenly throughout the basin. CHEMICAL COMPOSITION CHEMICAL ANALYSES OF THE SOLIDS The composition of the solids in the Upper Salt is represented by the analyses of four cores, GS—ll, GS— 12, GS—16, and GS—Zl given in table 15. The sampled profiles include both the Upper Salt and Overburden Mud; the estimated positions of the contacts are indi- cated by footnotes in table 15. The five samples from GS—16 are representative of the central part of the de- posit, the three samples from GS—Zl of the edges, and the five from GS—ll and four from GS—12 of facies be- tween the edges and center. The techniques used in sampling and conventions for tabulating the data are the same as the Lower Salt. The chemical analyses are expressed as weight per- cent of the hypothetical oxides; X-ray analyses are of the bulk sample. The X-ray data support the areal and stratigraphic trends in mineral composition in- ferred from the visual estimates (table 14). The most notable tendencies are for the lowest sample to con- tain large amounts of CO2 in the form of trona; for the middle and upper samples to contain larger amounts of C1 in the form of halite; for the upper samples (that include the Overburden Mud) to contain more acid- insoluble material; and for the base and the middle of the section to contain higher percentages of B407 in the form of borax. The vertical and horizontal distri- bution of K20 in the form of hanksite (and aphthita- lite?) is erratic. CHEMICAL COMPOSITION OF THE BRINES Most of the production of industrial chemicals from Searles Lake has come from brines from the Upper Salt. Commercial operations using brines from the in- terstices of the Upper Salt have been in existence on a major scale since the mid—1920’s, and some production has come from earlier operations. The brine composi- tion varies from place to place and with depth, partly as a result of its original distribution and partly as a result of the circulation and dilution caused by the pumping. UPPER SALT 61 TABLE 15.—Chemical analyses of cores from the Upper Salt and Overburden Mud [Chemical analyses by L. B. Schlocker, H. C. Whitebread, and W. W. Brannock. Analytical techniques: (1) CaO, MgO, Nazo, K,O, and 13,0, were determined in solutions prepared by boiling portions of samples in 1+9HC1; (2) C] was determined in solutions prepared by boiling portions of samples in distilled water; (3) S0, was determined by X-ray fluorescence of whole samples; (4) acid insoluble was residue obtained by boiling portions of samples in 1+9HC1, dried at 110°C; (5) H20 was determined by measuring weight of H expelled during combustion. X-ray analyses of total untreated sample by R. J. McLaughlin and G. 1. Smith] Depth in core Acid (Oxygen Minerals identified by X-ray Sample No. (ft) CaO MgO Na20 K,O C02 SO3 B10, Cl H20 insoluble equivalent Sum (in approximate order of decreasing abundance) of Cl) GS—ll—A ——————— 0 — ‘17.1 1.4 .49 43.8 1.1 3.2 6.8 .23 41.2 2.5 7.6 (9.30) 99.0 Halite, hanksite, pirssonite? B ——————— 17.1 — 30.7 1.5 .35 45.6 1.1 5.2 9.9 .48 38.8 2.9 2.7 (8.76) 99.8 Halite, hanksite, trona. C ——————— 30.7 — 55.6 .55 .16 46.3 1.2 7.8 9.6 1.2 36.1 4.4 .84 (8.15) 100.0 Halite, hanksite, trona, tincalconite?z D ——————— 55.6 — 68.6 .09 .04 49.0 .47 7.6 8.3 .08 40.6 3.3 .22 (9.16) 100.5 Halite, trona, hanksite. E ——————— 68.6 — 72.73 .30 .30 42.7 .65 23.5 13.6 2.2 6.9 11.6 .55 (1.56) 100.7 Trona, burkeite, halite, tincalconitef. GS-lZ—A ——————— 26.9— 33.3‘ 5.4 1.1 40.4 .59 9.2 .71 .15 37.9 5.1 5.4 (8.56) 97.4 Halite, pirssonite. B ——————— 33.3 — 60.7 .63 .16 47.8 .35 8.2 9.6 .10 38.4 3.4 .51 (8.67) 100.5 Halite, trona. C ——————— 60.7— 70.8 .10 .03 45.2 2.0 9.3 23.9 .10 18.8 3.6 .08 (4.24) 98.9 Hanksite, halite, trona. D ——————— 70.8 — 76.3 .73 .64 37.8 1.1 24.7 5.0 7.3 7.4 16.4 .35 (1.67) 99.8 Trona, tincalconite”, halite. GS—16—A ——————— 0 — 13.85 1.4 .32 43.4 1.0 3.2 9.4 .12 38.3 2.0 7.6 (8.65) 98.1 Halite, hanksite. B ——————— 13.8— 25.3 1.5 .16 45.8 1.4 4.6 20.2 .07 30.3 1.4 1.2 (6.84) 99.8 Halite, hanksite. C ——————— 25.3 — 45.8 .50 .20 45.2 1.4 8.5 11.0 2.0 31.1 5.0 .44 (7.02) 98.3 Halite, trona, hanksite, tincalconite’. D ——————— 45.8— 64.0 .13 .06 44.5 4.6 6.7 10.8 .44 37.2 3.6 .22 (8.40) 99.8 Halite, trona, hanksite. E ——————— 64.0— 70.6 .19 .22 40.3 .78 31.4 2.7 1.6 7.9 16.8 .14 (1.78) 100.2 Trona, halite, tincalconitez. GS—2l-A ——————— 5.4 — 10.9‘ 1.4 .59 37.6 1.2 13.8 25.8 2.4 2.2 8.0 3.5 (0.50) 96.0 Hanksite, trona, tincalconitez, halite? B ——————— 10.9— 30.6 .11 .44 40.7 1 7 14.8 24.0 .85 9.4 6.8 2.0 (2.12) 98.7 Hanksite, trons, halite. C ——————— 30.6- 35.5 .15 .54 40.7 1 5 28.3 4.2 .57 10.2 15.6 .51 (2.30) 100.0 Trona, halite, northupite? ‘ Contact of Overburden Mud at 16.3 ft. ’ Occurrences of tincalconite represent borax before dehydration in atmosphere. 3 Includes 0.2 ft of Parting Mud. The 104 brines analyzed (table 16) are from the Up- per Salt portion of 10 core holes, 5 (HH, U, W, X and S—31) in the central part of the deposit, 1 (GS—4) in the outer part, and 4 (MM, S-28, GS—l, and GS—lO) in intermediate facies. The samples were collected in the same manner as those from the Lower Salt. The compositions of the brines listed in table 16 were projected to one face of the tetrahedrons used to represent the two pertinent 5-component systems (fig. 28). Projected boundaries of the mineral stability fields in the Na,CO,-NaHCO,-Na,SO,-NaCl-H,O sys- tem at 20°C are shown in figure 28A; projected bound- aries in the Na,CO,-Na,SO,-NaCl-KCl-H,O system at 20°C in figure 28B. The points in figure 28A are closely grouped and confirm the equilibrium assem- blage in the Upper Salt of trona, halite, and thenar- dite without burkeite. The points in figure 28B are mostly within the boundaries of hanksite field; few are on the hanksite-aphthitalite boundary or in the aphthitalite field, confirming the equilibrium pres- ence of that mineral in the assemblage at the time the analyses were made. The specific gravities of brines from the central areas are mostly near 1.26 at the top and 1.30 at the base. Those from the edges of the deposit are as low as 1.25. Except for GS—l and GS—4, which lie nearer the edges than the others, the total percentages of dis- solved solids are between 33 and 35 percent. Values of pH from three holes range from 9.1 to 9.4. The percentage of Na fluctuates over a small range but not systematically. The average is near 11 percent. The K content of the brine increases downward in a1- ‘ Contact of Overburden Mud at 31.3 ft. 5 Contact of Overburden Mud at 13.8 ft. 5 Contact of Overburden Mud at 6.3 ft. most every well. Most of this increase occurs in the top 5-10 m, but further increases generally occur below this zone. Near the top of the sampled profiles, in the Overburden Mud, the percentage of K is generally be- tween 1 and 2 percent; at the base, it is generally about 2.5 percent in the central areas and under 2 percent nearer the edges. Teeple (1929, p. 18), Gale (1938, p. 869), and Ryan (1951, p. 449) list average percentages of K in the brine pumped from the Upper Salt as 2.53, 2.46, and 2.63, respectively. The analyses list total carbonate expressed as 00,, and in brines having these pH values, almost all the carbonate must actually be in this form. In most core holes, the CO3 percentage shows a marked increase downward. The percentages of SO, are generally be- tween 4 and 5. The brines in six core holes have in- creasing percentages of SO, toward the bottom, in three increase toward the top, and in one SO4 is nearly constant. The percentages of Cl mostly fall between 12 and 13. The CI content of the brines generally in- creases upward. Most B407 percentages, which lie near 1, increase downward. Analyses of minor elements in Upper Salt brines are available for some sets of samples. The phosphate analyses arefrom brine wells in the intermediate and central parts of the deposit. Although areal position seems to be unrelated to the values, there is consider- able variation in PO4 content of brines within and be— tween different wells. Ryan (1951, p. 449) reports an average of 940 ppm P04 and W. A. Gale (written com- mun., 1952) reports 920 ppm P04 in brines from the central part of the Upper Salt. As is reported only for 62 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. TABLE 16.—Chemical analyses of brines from Upper Salt and Overburden Mud [Analyses of cores HH, MM, U W, X, S—28 and S—31 by chemists of American Potash & Chempial Corp., published with permission of company. Analyses of GS—1, GS-4, and GS— 10 by Henry Kramer and 801 Berman, US. Geological Survey. Composition in weight percent except where indicated as parts per million (PDm)] Total Sample Total dissolved carbonate depth Specific solids as PO. As Li Br S F (ft) gravity (by ‘iM-l pH Na K CO, SO. Cl 8.0, ppm ppm ppm ppm ppm ppm Core HH,‘ June 1948 5 1.276 32.33 _____ 10.81 1.95 1.74 4.02 13.02 0.72 _____ — — _____ 700 ..... 13 10 1.278 32.37 ' _____ 10.80 1.98 1.83 4.05 12.88 .76 _______________ 700 _____ 10 15 1.279 35.60 _____ 10.89 1.98 1.89 4.10 14.33 .77 _______________ 700 _____ 13 20 1.280 32.66 _____ 10.89 1.99 1.92 4.18 12.82 .79 _______________ 700 _____ 13 25 1.287 33.27 _____ 11.03 2.03 1.98 4.87 12.50 .79 _______________ 700 _____ 13 30 1.301 34.36 _____ 11.07 2.49 2.45 5.09 12.10 1.07 _______________ 900 _____ 13 35 1.301 34.25 _____ 11.04 2.48 2.45 5.03 12.09 1.07 _______________ 900 _____ 13 40 1.301 34.34 _____ 11.06 2.48 2.46 5.10 12.08 1.07 _______________ 900 _____ 13 45 1.301 34.31 _____ 11.06 2.48 2.45 5.08 12.08 1.07 _ _ _ _ _ - -- _______ 900 _____ 16 50 1.301 34.31 _____ 11.05 2.48 2.45 5.10 12.07 1.07 _______________ 900 _____ 15 55 1.301 34.34 _____ 11.06 2.49 2.45 5.10 12.09 1.06 _______________ 900 ..... 15 60 1.301 34.37 _____ 11.05 2.47 2.45 5.08 12.06 1.06 _______________ 900 _____ 16 65 1.301 34.42 _____ 11.05 2.49 2.45 5.08 12.09 1.06 _______________ 900 _____ 1 70 1.301 34.46 _____ 11.06 2.50 2.46 5.08 12.09 1.07 _______________ 900 _____ 15 75 1.301 34.38 _____ 11.03 2.49 2.46 5.05 12.08 1.07 _______________ 900 _____ 15 Core MM,1 October 1948 5 1.2844 32.39 _____ 11.34 1.06 2.20 4.55 12.06 0.82 540 __________ 650 __________ 10 1.286 32.60 _____ 11.39 1.09 2.21 4.56 12.16 .82 540 __________ 670 .......... 15 1.287 32.57 _____ 11.38 1.06 2.21 4.55 12.13 .82 660 __________ 650 __________ 20 1.288 32.69 _____ 11.42 1.06 2.22 4.52 12.20 .81 700 __________ 670 __________ 25 1.292 34.26 _____ 11.42 1.30 2.50 4.64 11.90 1.03 700 .......... 750 __________ 30 1.295 33.40 _____ 11.49 1.30 2.55 4.60 11.96 1.04 700 __________ 740 __________ 35 1.295 33.37 _____ 11.47 1.27 2.54 4.66 11.91 1.04 740 __________ 740 __________ 40 1.292 33.39 _____ 11.46 1.31 2.52 4.68 11.88 1.03 800 __________ 740 __________ 45 1.297 33.90 _____ 11.52 1.47 2.64 4.79 11.90 1.04 860 __________ 720 __________ 50 1.298 34.01 _____ 11.42 1.65 2.72 4.69 11.86 1.07 940 __________ 790 __________ Core U,‘ December 1949 5 1.278 32.68 _____ 11.30 1.31 2.23 4.00 12.62 0.75 740 __________ 630 __________ 10 1.279 32.60 _____ 11.29 1.28 2.25 4.00 12.55 .76 740 __________ 630 __________ 15 1.286 33.15 _____ 11.40 1.33 2.34 4.41 12.38 .78 800 __________ 670 __________ 20 1.293 34.40 _____ 11.08 1.73 2.03 4.88 12.16 .96 900 __________ 680 __________ 25 1.297 34.10 _____ 11.38 1.73 2.52 4.95 11.91 1.04 900 __________ 750 __________ 30 1.297 34.10 _____ 11.37 1.73 2.52 4.96 11.91 1.04 900 __________ 750 __________ 35 1.297 34.11 _____ 11.35 1.76 2.52 4.96 11.91 1.04 900 __________ 760 __________ 40 1.297 34.11 ..... 11.35 1.77 2.52 4.96 11.91 1.04 880 __________ 760 __________ 45 1.297 34.11 _____ 11.34 1.78 2.52 4.96 11.91 1.04 880 __________ 760 __________ 50 1.297 34.42 ..... 11.12 2.17 2.51 4.96 11.91 1.07 880 __________ 760 __________ 55 1.299 34.39 _____ 10.99 2.37 2.51 4.96 11.91 1.09 880 __________ 770 __________ 60 1.301 34.76 _____ 11.05 2.46 2.73 4.82 11.89 1.21 920 __________ 870 __________ 65 1.303 34.75 _____ 11.06 2.45 2.73 4.82 11.88 1.21 920 __________ 880 .......... Core W,1 February 1935 5 1.256 32.62 _____ 10.71 1.39 1.66 2.64 13.53 0.700 588 70 __________ 40 _____ 10 1.277 34.90 _____ 11.04 1.65 2.18 3.69 12.76 .950 752 81 __________ 140 ..... 15 1.284 35.65 _____ 11.04 1.83 2.38 4.00 12.52 .988 830 92 __________ 180 _____ 20 1.291 36.09 _____ 11.14 1.88 2.41 4.25 12.32 1.067 858 86 __________ 220 _____ 25 1.292 36.25 _____ 11.12 1.96 2.55 4.24 12.26 1.061 868 32 __________ 270 _____ 30 1.293 36.49 ..... 11.14 1.99 2.59 4.30 12.27 1.082 896 32 __________ 280 ~~~~~ 35 1.293 36.34 _____ 11.05 2.06 2.57 4.29 12.17 1.097 900 32 __________ 280 _ .— _ _ _ 40 1.294 36.72 _____ 11.02 2.28 2.56 4.48 12.21 1.101 856 54 __________ 270 _____ 45 1.296 36.82 ..... 10.93 2.42 2.58 4.52 12.18 1.094 872 65 ,,,,,,,,,, 270 _____ 50 1.298 37.02 _____ 10.92 2.51 2.62 4.52 12.19 1.099 920 59 __________ 250 _____ 55 1.298 37.19 _____ 10.90 2.62 2.66 4.46 12.19 1.129 950 92 __________ 200 _____ 60 1.298 37.33 _____ 10.91 2.63 2.72 4.48 12.15 1.139 988 86 __________ 180 _____ 65 1.298 37.50 _____ 10.98 2.60 2.74 4.54 12.17 1.143 996 92 __________ 180 _____ 70 1.298 37.42 _____ 10.98 2.60 2.71 4.51 12.19 1.140 986 92 .......... 190 _____ Footnotes on page 64. UPPER SALT 63 TABLE Iii—Chemical analyses of brines from Upper Salt and Overburden Mud—Continued Total Sample Total dissolved carbonate depth Specific solids as P0. As Li Br S F (ft) gravity (by summation) pH Na K C0I SO. Cl B‘O, ppm ppm ppm ppm ppm ppm Core X, ‘ March 1950 5 1.264 30.37 _____ 11.16 1.05 1.96 3.36 13.05 0.57 660 __________ 510 100 _____ 10 1.265 31.59 _____ 11.17 1.04 1.96 3.37 13.05 .57 660 __________ 520 200 _____ 15 1.265 31.60 _____ 11.18 1.04 1.96 3.37 13.05 .57 660 __________ 520 200 _____ 20 1.283 32.97 _____ 11.48 1.13 2.04 5.14 12.16 .60 620 __________ 520 300 _____ 25 1.294 33.88 _____ 11.35 1.66 2.36 5.00 12.06 .98 720 __________ 610 100 _____ 30 1.299 34.39 _____ 11.25 2.05 2.47 5.10 12.02 1.03 720 __________ 650 100 _____ 35 1.300 34.49 _____ 11.24 2.11 2.47 5.18 12.02 1.01 700 - _ ._ _ _ _ _ _‘_ _ 650 100 _____ 40 1.301 34.60 _____ 11.17 2.23 2.41 5.30 12.02 1.01 700 __________ 630 100 _____ 45 1.301 34.54 _____ 11.17 2.23 2.41 5.26 12.02 1.01 680 __________ 610 100 _____ 50 1.301 34.59 _____ 11.18 2.23 2.41 5.28 12.02 1.01 700 __________ 630 100 _____ 55 1.301 34.62 _____ 11.19 2.23 2.41 5.33 11.99 1.01 700 __________ 630 100 _____ 60 1.301 34.62 _____ 11.19 2.23 2.41 5.33 11.99 1.01 700 __________ 630 100 _____ 65 1.300 34.60 _____ 11.17 2.23 2.41 5.33 11.99 1.01 700 __________ 630 100 _____ 70 1.301 34.62 _____ 11.18 2.24 2.47 5.23 11.99 1.04 700 __________ 720 100 _____ 72 1.302 34.50 _____ 11.13 2.24 2.46 5.10 11.99 1.06 760 __________ 750 200 _____ Core S—28,‘ August 1950 30 1.290 33.57 _____ 11.34 1.70 2.71 4.22 12.19 0.98 560 __________ 630 500 _____ 35 1.291 33.53 _____ 11.35 1.72 2.73 4.22 12.18 1.00 560 __________ 640 500 _____ 40 1.297 34.40 _____ 11.32 2.15 2.99 4.16 12.09 1.21 640 __________ 740 600 _____ 45 1.301 34.78 _____ 11.26 2.38 3.14 4.03 12.06 1.35 720 __________ 760 800 _____ 50 1.301 34.67 _____ 11.20 2.42 3.14 4.00 12.00 1.36 700 __________ 770 800 _____ 55 1.301 34.81 _____ 11.24 ' 2.42 3.16 4.05 12.01 1.37 720 __________ 770 800 _____ 60 1.302 34.81 _____ 11.21 2.44 3.15 4.02 12.02 1.39 780 __________ 840 700 _____ 64‘ 1.302 34.85 _____ 11.23 2.44 3.15 4.05 12.02 1.38 780 __________ 840 700 _____ Core 8—31,1 December 1950 25 1.285 32.99 _____ 11.49 1.30 2.51 4.23 12.33 0.76 500 __________ 570 400 _____ 30 1.285 33.02 _____ 11.51 1.28 2.55 4.25 12.30 '.76 500 __________ 590 400 _____ 35 1.291 34.62 _____ 11.44 2.06 2.97 4.31 12.12 1.20 720 __________ 740 500 _____ 40 1.306 35.20 ..... 11.16 2.62 3.09 4.21 12.11 1.33 980 __________ 930 500 _____ 45 1.306 31.13 _____ 11.11 2.66 3.08 4.21 12.06 1.34 960 __________ 940 500 _____ 50 1.307 35.26 _____ 11.13 2.68 3.11 422 12.09 1.36 960 __________ 930 500 _____ 55 1.307 35.27 _____ 11.13 2.70 3.14 4.21 12.06 1.37 940 __________ 930 500 _____ 60 1.307 35.19 _____ 11.10 2.71 3.12 4.20 12.06 1.37 900 __________ 940 500 ..... 65 1.307 35.17 _____ 11.10 2.70 3.12 4.22 12.04 1.36 900 __________ 930 500 _____ 70 1.307 31.25 _____ 11.16 2.67 3.12 4.23 12.09 1.36 880 __________ 930 500 _____ brines from core hole W in the central segment of the lake; values range from 32 to 92 ppm with no clear ver- tical trend. Li was analyzed in brines from intermedi— ate and edge facies but no areal or vertical trend is evident; Ryan (1951, p. 449) reports 70 ppm in pumped brines. The analyses for Br are all brines from the intermediate and central parts of the depos- it; Br tends to increase in concentration toward the base of the layer. Ryan reports 810 ppm Br, and Gale estimates 860 ppm Br in brines pumped for commer- cial use. The amounts of S are variable and tend to increase downward. Ryan and Gale report average amounts of S as 330 and 390 ppm, respectively. Amounts of F in core hole HH are mostly between 10 and 15 ppm; Ryan and Gale estimate 20 and 15 ppm F, respectively. Ryan and Gale estimate the I content of brines pumped from the central part of the Upper Salt to be 30 and 29 ppm. They and Carpenter and Garrett (1959), report W in average amounts near 55 ppm. Hem (written commun., 1960) reports Sr in two sam- ples from 60 and 70 ft (18.3 and 21.3 m) in GS—14 as 4.5 and 3.2 ppm, and in samples from 25 and 30 ft (7.6 and 9.1 m) in GS—26 as 3.7 and 2.2 ppm. The brines from GS—1, GS—4, and GS—lO were analyzed for Ca and Mg but their quantities were found to be below limitsiof detection. Vertical variations in brine composition bear a spo- radic relation to the mineral composition of the salts although there is a marked increase in the total dis- solved solids below the zone immediately below the Overburden Mud. Percentages of CO. and B407 in- crease downward in accord with the downward in- crease of trona and borax, percentages of Cl increase upward as does the percentage of halite. Percentages 64 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. TABLE 16 . ~Chemical analyses of brines from Upper Salt and Overburden Mud -—Continued Total Sample Total dissolved carbonate depth Specific solids as P0. As Li Br S F (ft) gravity (by summation) pH Na K COa SO. Cl B.07 ppm ppm ppm ppm ppm ppm Core GS—l,‘ july 1954 30 1.264 32.4 9.2 10.9 1.52 2.54 3.91 12.64 0.84 __________ 28 _______________ 35 1.277 33.5 9.2 11.3 1.80 2.56 4.62 12.30 .966 __________ 31 _______________ 40 1.281 34.2 9.3 11.3 2.10 2.71 4.68 12.36 1.08 __________ 35 _______________ 45 1.276 34.4 9.3 11.3 2.02 3.04 4.64 12.33 1.07 __________ 35 _______________ 45.8 1.277 33.8 9.3 10.9 2.10 2.72 4.67 12.35 1.07 __________ 35 _______________ Core GS—4,1 September 1954 24.8 1.274 32.0 9.12 11.0 1.07 2.45 4.56 12.14 0.78 __________ 42 _______________ 30.0 1.278 32.3 9.18 10.9 1.19 2.62 4.62 12.08 .85 __________ 43 _______________ 35.0 1.277 34.0 9.23 12.8 1.20 2.50 4.72 11.99 .83 __________ 42 _______________ 40.0 1.282 33.0 9.28 11.3 1.22 3.03 4.73 11.81 .86 __________ 43 _______________ 45.0 1.284 33.0 9.31 10.9 1.78 2.81 4.42 12.01 1.11 __________ 72 _______________ 48.9 1.287 32.9 9.32 10.7 1.79 2.97 4.38 11.92 1.10 __________ 69 _______________ Core GS-lO,‘ January 1955 36.3 1.285 33.6 9.30 10.9 1.98 2.37 5.09 12.32 0.94 __________ 58 _______________ 40 1.283 33.6 9.30 10.9 1.98 2.33 5.12 12.30 .94 __________ 58 _______________ 45 1.285 33.5 9.30 10.7 1.93 2.32 5.11 12.28 .95 __________ 58 _______________ 50 1.284 33.4 9.26 10.9 1.95 2.24 5.10 12.30 .95 __________ 57 _______________ 55 1.285 33.6 9.24 10.9 1.93 2.39 5.12 12.26 .95 __________ 57 _______________ 60 1.282 33.7 9.25 10.9 1.95 2.40 5.19 12.28 .95 __________ 58 _______________ 65 1.289 33.6 9.38 10.8 2.05 2.48 4.84 12.32 1.10 __________ 59 _______________ 70 1.280 33.4 9.31 11.3 1.75 2.48 4.54 12.26 1.10 __________ 22 _______________ 'Depth to top and base of Upper Salt; HH: 13—78.? feet. MM: 6—534 feet. U: 12(?)-65.0 feet W: 22-71.3 feet. X: 19—76.7 feet. of SO4 do not show any clear trend, and percentages of K increase downward. Both patterns are in contrast with the general tendency for hanksite, the principal mineral in these cores containing these components, to be more abundant near the top of the Upper Salt. The effect of pumping by chemical companies cannot be inferred from data in table 16 because the samples come from widely spaced sites and represent a wide range in dates of collection. ESTIMATED BULK COMPOSITION OF THE UNIT The estimates of mineral composition of the Upper Salt (table 14), the measure of the probable error in the visual estimates (table 8), and analyses of the brines (table 16) provide a basis for calculating the to- tal quantity of the major water—soluble components in the Upper Salt. The total is about 1,330 X 1012 g of salts. Table 17 gives the quantity of each component and tabulates the steps followed in the calculation. A more detailed study of the boron in Searles Lake (D. V. Haines, unpub. report, 1956) estimates a total of 23 S—28: 28.6—64.7 feet S—31: 23.0—72.0 feet. GS—I: 28.9—46.0 feet GS—4: 24.8—48.9 feet. GS—IO: 32.4—70.2 feet X 1012 g of B0 in the Upper Salt, whereas table 17 indi- cates 28 X 1012 g; this discrepancy indicates the ap- proximate accuracy of these estimates. The relative percentages of components, calculated water-free, are: Na _____________________ 34.8 K ______________________ 2.1 Mg _______________________ 4 CO3 _____________________ 7.5 HCO3 ___________________ 4.9 SO, ____________________ 14.5 C1 _____________________ 33.7 B40, _____________________ 2.1 FIGURE 28.-—Compositions of analyzed brines from the Upper Salt > (table 16), plotted on diagrams that indicate phase relations in two 5-component systems. A Phases in NaZCOK—NastrNaCl —NaHC03—H20 system, B, Phases in Na2C03—Na2S04—KC1— NaCl—HZO system. Brines plotted on the basis of their compo- sitions projected to plane of diagram; some points represent more than one analysis. See figures 35 and 39 and associated text for explanation of phase boundaries and method of plotting. UPPER SALT 65 NaZSO4 EXPLANATION Average brine Composition from table 17 Mirabilite Nazco3 NaCl NaZSO4 EXPLANATION Average brine Composition from table 17 Aphthitalite (mineral X composition) EKHankslta (mineral composition) B Burkeite (mineral composition) Analysis 1 and 4 From table 22 Thenardite Bu rkeite Aphthitalite N atron Sylvite Total 003 as Total K Nazcos as KCI 66 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. TABLE 17.—Estimated percentages and total quantities of water-soluble components in the Upper Salt Na K Mg co. Hco. so. C1 13,0, H70 Chemical composition of solids inferred from visual estimate‘ _______________ 34.2 0.5 0 10 5 9.1 11.4 26.8 1.0 6.5 Average error of visual estimates2 ______ —0.8 +0.6 +0.5 —3 4 8—3.4 +2.6 +4.8 +0.8 ‘-1.6 Chemical com osition of solids, in weight per- cent, ad'uste for probable error ______ 33.4 1.1 0.5 7.1 5.7 14.0 31.6 1.8 4.9 Chemic composition of brine5 _______ 11.2 2.0 0 2.5 "’0 4.6 12.2 1.0 66.5 Chemical composition of combined solids and bl‘ines'l _____________________ 24.5 1.5 0.3 5 3 3.4 10.2 23.8 1 5 29.5 Total quantity of component in Upper Salt“, grams X 1012 __________________ 462 28 5 100 64 192 448 28 556 ‘ Data in table 14 converted to weight percent. ’ Based on comparison of visual estimates and chemcial analyses of cores (converted to ions), data from table 8, average of all samples. Table 8 also lists error in Ca of +0.4, reducing totals in this and underlying columns to 99.6 5 Error in CO, listed in table 8 divided equally between CO, and HCOT OVERBURDEN MUD The Overburden Mud, commonly about 7 m thick in the center of the deposit, is composed of a succes- sion of discontinuous beds of mud and salts. Most of the salt is halite, some as partially dissolved and rounded crystals; locally beds of hanksite, trona, bo- rax, and thenardite are found. Toward the edges of the lake, the Overburden Mud is characteristically a black or dark-gray pirssonite-bearing mud that grades up- ward into greenish or olive-colored clay and silt; at the surface, when dry, it is a light—tan or pinkish—tan silt. In many parts of the Searles Lake deposit, the subsur— face contact between the Overburden Mud and the upper part of the Upper Salt is gradational and not easily identified. Stratigraphic criteria used to place the contact in cores are described in the section on the Upper Salt. Outside the area covered by the Upper Salt, the base of the Overburden Mud is in contact with a sand layer that is the lateral equivalent of the Upper Salt. In most areas, the outcrop of this contact is known or interpreted to be a kilometer or more out- side the salt body limits. Along the southwest part of the body, southeast of Westend, this contact is much closer to the inferred edge of the Upper Salt body be- cause the chemical and Clastic sediments carried into the basin from upstream formed a delta in this area, steepening the slopes upon which both salt and mud layers were deposited. AREAL EXTENT AND VOLUME The Overburden Mud covers the entire surface of Searles Lake within the area underlain by the Upper Salt (fig. 27), and in most areas extends a kilometer or more beyond this limit. The average thickness of the Overburden Mud in the 88 cores in which it could be measured is 7 m (23 ft). It is only 2—5 m thick in an area of about 10 km2 that lies within a kilometer or two ‘ Reduced by amount of H in HC03. "‘ Arithmetic average of brine analyses listed in table 16, in weight percent. ‘ Percentage of HCOJ is low, assumed to be 0. 7 In weight percent, assumed porosity 40 percent. ° Assumed specific gravity of salt plus brine filled pores, 1.80; volume from text. of the northwest edge of the lake, but elsewhere, it is mostly 6—9 m thick. Thickness greater than 9 m were noted only in cores that are near the north, south, and southwest edges of the Upper Salt unit; the unit thins to zero about a kilometer outside this edge. These thickness data were plotted on an isopach map and the volume of the unit calculated. This infor- mation was used in making the very general calcula- tions of the mineral composition of the Overburden Mud. The map is not included in this report because the position of the basal contact of the Overburden Mud, owing to its gradational nature, is chosen by means of a set of arbitrary criteria. As a result, the trends in thickness indicated by subsurface data are relatively unsystematic, and the changes in thickness near the edge of the deposit are too abrupt to allow meaningful extrapolation in areas of no data. The to— tal volume of the unit, calculated from the isopach map and within the arbitrary boundary described pre- viously, is 380 X 106 m3. MINERAL COMPOSITION AND LITHOLOGY The composition of the Overburden Mud is dis— tinctly different from the other mud layers in this de- posit. Clastic fragments are commonly larger and make up higher percentages of this layer. Sand is abundant in parts of the Overburden Mud, whereas clastic fragments larger than silt size make up less than 20 percent of the Parting Mud and older mud units. In the outer parts of the deposit, the Overbur- den Mud is composed almost entirely of mud and this grades into interbedded mud and salt layers in the center. In contrast, the Parting Mud and Bottom Mud are nearly constant in composition between their edge and central facies. The megascopic composition of the Overburden Mud was estimated by the methods used for older units, although the results are not as reliable for rea- OVERBURDEN MUD 67 sons described below. The calculated mean composi- tion (in volume percent) is as follows: Mud (elastic and other fine-grained minerals) _ _ _ _ 83 Halite ___________________________ 13 Pirssonite _________________________ 2 Hanksite _________________________ 1 Trona ____________________________ 1 Gaylussite ______________________ Trace Borax _________________________ Trace Thenardite ______________________ Trace Sulfohalite _______________________ Trace Near the edge of the deposit, evaporite components are generally reported only from the bottom meter, but this is probably not an accurate record of their dis— tribution. Most test holes in those areas did not take cores in the upper part of the Overburden Mud, and the composition of that segment was inferred to be pure mud. Pirssonite and halite are the minerals likely to have been underestimated as a result of this prac- tice. In the central segment of the deposit, evaporite minerals are about evenly distributed vertically throughout the unit. X-ray diffraction studies of samples from this unit (tables 15 and 18) confirm the megascopic mineral identifications and show that halite, pirssonite, and hanksite are the most common evaporite minerals, and that little dolomite and no aragonite are present. The elastic minerals are mostly quartz, feldspar, mica, and clay. The clay fraction of surface samples, studied by Droste (1961, p. 1,717—1,719), consists mostly of montmorillonite and illite; there is a marked increase in chlorite in the east part of the deposit which re- ceives sediment from the several chlorite-rich zones of fault gouge in the Slate Range. In the central part of the deposit (Haines, 1959, fig. 6), the surface of the Overburden Mud is generally pure halite. Much of this material is firm enough to support a car or truck, even when covered with water. The cubic crystals of halite are several millimeters to a few centimeters across and are intergrown in a ran- dom orientation. As a result of microorganisms that live in the near-surface brine, some of the salt has a light pinkish color. In a few areas where the surface has been modified by the evaporation of waters re— turned from the chemical plants and towns to the lake, the surface crust includes more thenardite and a number of introduced impurities. The surface crusts commonly form a pattern of po- lygonal cracks and ridges (fig. 29). The center of each polygon is saucer shaped, and the edges turn up to form ridges a few millimeters to more than half a me- ter high. These polygons appear to grow from the cen- ter outward, and edges are thrust outward over the adjacent polygon. The zone between polygons is a cha- otic zone of slabs broken off when they met crusts growing from the opposing side. FIGURE 29.—The Polygonal cracks and ridges on surface of north- ern part of Searles Lake (sec. 11, T. 25 S., R. 43 E.). Metal clipboard is 20 cm long. A remarkably linear zone of saline crusts and ef- florescences known as the “trona reef” occurs along the northeast edge of the lake surface (Gale, 1914, p. 275, 294—296). The efflorescences range from a surface powder to crusts several centimeters thick. Saline minerals constitute the bulk of this material; the per- centages of minerals vary from place to place. Gale (1914, p. 294) reports an analysis of a composite sam- ple of crusts that suggests their content to be about 54 percent trona,” 20 percent halite, 5 percent thenar- dite, nearly 3 percent borax, a few percent a K-bearing mineral, and the remainder a material that is insolu- ble in water. X-ray diffraction analysis of a sample of crust from the northern part of the zone (NWIA sec. 12) shows it to be composed of halite, a smaller amount of trona, and minor quantities of burkeite and gaylussite(?). Gale (1914, p. 295) attributed the trona reef to evapora- tion of saline ground water that rose to the surface at the edge of the playa as a result of the impermeability of the sediments that characterize the playa floor. This does not explain the absence of similar zones around other parts of the playa where ground water is encroaching. Seismic re- fraction data (Mabey, 1956, p. 846), however, show a ve- locity discontinuity beneath the trona reef that could be a fault having upward displacement on the east side. It seems likely that this fault extends to the surface, where it acts as a linear barrier to eastward—moving brine.15 The “ The published analysis reports components that total 89.20, not 100 as indicated. However, the percentage of H20 that would accompany 54.04 percent NaZCOJ' NaHCO, as trona and 1.52 percent Na.,B,07 as borax brings the total to 100.77. ‘5 As noted by Hardt, Moyle, and Dutcher (1972, fig. 10), as the hydrostatic head of the brines in the center of the valley is greater than in the surrounding areas, brine migrates from the lake into the surrounding ground water reservoir areas. 68 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. TABLE 18.——Partial chemical analyses of core GS—40 from the Overburden Mud [Analysts and analytical techniques as listed in table 13] Major elements (percent of total sample) Water-and acid-soluble components Acid-insoluble components Total carbo- Total Organic Depth nate Fe as carbon in core CaO MgO Na,0 K,0 as CO2 SO3 B20l C1 ‘SiO2 AlZO3 FeZO, Cao MgO N820 K20 in C 1.0 _ _ 4.2 1.2 5.0 4.0 1.2 0.22 0.19 1.5 61.3 11.7 0.9 1.9 0.5 2.9 2.7 0.04 230.3 _ _ 9.2 6.8 14.5 2.6 15.3 1.8 .64 7.9 26.9 3.0 .3 .5 .1 .8 .6 .5 Minor Elements Depth in core B Ba Cr Cu Ga Mn Nb Sc Sr Ti Yb V Y Zr 1.0 g _ i - 200 700 15 0.7 15 150 5 10 700 3000 1.5 50 10 150 30.3 _ _ _ _ 100 500 15 0.7 10 150 7 7 300 3000 1.5 30 10 150 Depth in core Minerals identified by X-ray of untreated sample (in approximate order of decreasing abundance) 10 __________________________________ Feldspar, quartz, mica, amphibole, clay, halite 30.3 Dolomite, pirssonite, gaylussite, halite, mica, clay, calcite ‘Percentage possibly low because of loss during original heating and acid treatment. brine is thus forced to the surface and, on evaporation, produces an elongate strip of efflorescences. This ex- planation accounts for the notable linearity of the fea- ture, and for the absence of similar features along other parts of the lake periphery. CHEMICAL COMPOSITION Analyses of the Overburden Mud segment of four cores are included in the analyses of Upper Salt given in table 15, analyses of individual specimens logged as pure mud in core GS—40 in table 18. Two of the sam- ples given in table 15 (GS—ll—A and GS—16—A) are al- most entirely Overburden Mud, but samples from GS—12 and GS—21 include only a small part of the unit. The GS—11 and GS—16 cores, from the central segment of the deposit, contain ony 7.6 percent acid- insoluble material. Most of the remaining portions of these cores are halite (as shown by X-ray data com- bined with the high Na and Cl percentages) with some hanksite (as shown by X-ray data combined with the high S03 and low CO2 percentages). Some pirssonite is present. The sample from near the top of the unit in core hole GS—40 (table 18) contains 81.8 percent acid— insoluble material, the rest being mostly halite with small amounts of other saline minerals forming the balance.16 The mud sample from near the basal con- tact of core GS—40 contains only 32.4 percent acid-in- soluble material and larger amounts of dolomite, pirssonite, gaylussite, and halite. '5 The percentages of acid-soluble K20 appear too high. The discussion of analytical problems encountered in making analyses of samples from the Parting Mud (p. 54) applies to the two analyses in table 18. 2Base of Overburden Mud in GS~40 at 31.9 feet. RADIOCARBON AGES OF STRATIGRAPHIC UNITS By MINZE STUIVER17 and GEORGE 1. SMITH INTRODUCTION Radiocarbon ages, together with the stratigraphy and mineralogy, provide a basis for reconstructing the climatically controlled history of Searles Lake. The large number of published “C dates that have been de- termined on core samples from Searles Lake (fig. 30) are taken from publications by Flint and Gale (1958), Rubin and Berthold (1961), Ives, Levin, Robinson and Rubin (1964), and Stuiver (1964). Of the 74 dates shown in figure 30, 40 are based on carbon from inor- ganic carbonate minerals, and 2 are on wood frag- ments. Fourteen previously unpublished dates on dissemi- nated organic carbon from the Lower Salt and the top of the Bottom Mud are given in table 19.‘8 These sam- ples were collected in November 1964 from core L—31 in order to obtain a detailed chronology of the wet-dry episodes in the Lower Salt deposits. The dated samples plotted in figure 30 are grouped according to the stratigraphic unit from which they came. Stratigraphic assignments were based on the re- ported sample depths and logs of the sampled cores. The relative positions of samples from the Parting ‘7 Quaternary Research Center, University of Washington, Seattle, Wash. 98195. ”‘ Support of the “C work was through N. S. F. grant GS—36762 to M. Stuiver. RADIOCARBON AGES OF STRATIGRAPHIC UNITS AGE IN YEARS AGE, IN YEARS BEFORE PRESENT UNIT BEFORE PRESENT LITHOLOGY Rug'fi'ulggggffigflgsgm, UNIT (STU'VER'1964) AND IVES AND OTHERS, 1964) (6,890 1 140) 6,630 1 390 Overburden Mud 3,520 1- 190 (wood) (9,700 1' 180) \ 11,800 1 1,000 . : 12,390 1400 } Overburden Mud (11,100 1 180) 11,010 1 150 i’Ii/ (10,600 1100) (10,460 1 170) \ 10,270 1 450 ”pp“ 33" 11.510 1150 11,400 1 600 (9,720 i 200) (9,900 i 500) Upper Salt (9,850 1 180) k (9,840 1 80) (8,550 1 250) , 10,680 1 90 10,900 1 90 10 230 1 80 ' 10,494 1 560 183?: f 320 10,700 1 130 (10,630l80) ”1'81“ 14°) ’ ' 15,089 1 1,000 (11'4702100) (12420+160) (11,0101110) 12'7301210 10,590 1110 : 18'000 1 730 (10,440 1 90) ‘ ’ 11,800 1 130 if/ 16'62° 3 32° Parting Mud Parting Mud < . (16,890 1 210) (13,100 1 230) , 13 710 + 270 - 22,600 11,400 ’ ' ' 21,200 1 2,170 (17,710 1 280) / 23,923 1 1,800 18'80" 1 24° (23,610 1 1,750) (19,380 i 250) . . 24,690 11,070 19,970 1 280 22,800 1 1,400 (21,380 1' 340) 22,400 1 1,000 22,300 1 280 (20,650 1 210) 24,630 1 460 (22,450 1 380) " L 23:7” i 32° (22,620 1 200)‘ (25,600 1 230) 26,700 1: 2,000 (wood) Lower Salt 24,600 1 400 29,500 1 2,000 27,500 1 800 (23.000 1 1,400) L s It (25,800 1 500)' (3.331233) ower a 1 (24,400 1 400)' , 28,000 1 600 i’!' (31,100 1 400)' 32,000 11,000 \m (32,300 1 900) Bottom Mud 32,700 1 800 Bottom Mud FIGURE 30.—Summary of published l‘C dates on subsurface samples from Searles Lake. Dates in parentheses are on carbon in inorganic carbonates and are considered relatively unreliable; dates followed by asterisks (*) were sus- pected by Stuiver (1964) of being contaminated; remaining dates, except for two on wood, are on organic carbon disseminated in lake mud. 70 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. TABLE 19.—New ”C dates on disseminated organic carbon in mud layers of Lower Salt and top of the Bottom Mud, core L—31 Sample Depth Years B.P. Y—2230‘ ______ Top 5 cm of M—7 23,750 i 300 Y—2231 ______ Bottom 5 cm of M—7 26,350 i 350 Y—2232 ______ Top 5 cm of M—6 24,760 i 300 Y—2233 ______ Bottom 5 cm of M—6 28,880 i 500 Y—2234 ______ 5 cm thick layer near base 8-5 27,550 i 400 Y-2235 ______ Top 5 cm of M—5 28,380 i 350 Y—2236 ______ Bottom 5 cm of M—5 29,040 i 350 Y—2237 ______ Top 5 cm of M—4 28,620 i 350 Y—2238 ______ Bottom 5 cm of M-4 30,160 i 400 Y—2239 ______ Top 5 cm of M-3 30,510 i 400 Y-2240 ______ Bottom 5 cm of M—3 30,270 i 500 Y—2241 ______ Top 5 cm of M—2 30,280 i 300 Y—2242 ______ Bottom 5 cm of M—2 32,620 i 500 Y—2243 ______ Top 5 cm of Bottom Mud 32,800 i 600 Mud are approximate; dated samples came from many cores, and the level within the unit was esti- mated and plotted on the basis of its proportionate position between the base and top of the Parting Mud in that particular core hole. RELIABILITY OF SAMPLED MATERIALS Dates on wood are considered to be the most reli- able. Dates on disseminated organic carbon are con- sidered less reliable because most are probably somewhat too “old.” Dates on carbonate minerals may be either a little too “old” or too “young” and are therefore the least useful. The reasons leading to this ranking of reliabilities are given below. The dates on wood are considered best because they are on material that is relatively unsusceptible to re- placement by younger carbon, and because they pre- sumably derived carbon from atmospheric CO2 rather than from the lake. The only predictable discrepan- cies between 1“C dates and true dates are produced by the variation in the atmospheric 14C through geologic time which is known for the past 7,400 years from comparison of ”C dates with tree-ring dates, and the minor discrepancies that result from fractionation of 1“C by some plants and trees during their growth. The 1“C date of 3,520 i 190 on wood corresponds to a tree- ring age of 3,7 20—3,800 years (Ralph and others, 1973, p. 11; Stuiver and Suess, 1966, p. 539). The 1“C date on wood of 26,000 : 2,000 years is too old for comparison with tree-ring chronologies. Carbon isotope fractiona— tion during plant growth can be calculated by measur- ing 13C/‘ZC ratios. When 13C measurements are lacking, age errors of a few hundred years may be introduced. . Dates derived from disseminated organic carbon form the main basis for the time scale used to inter- pret and correlate the stratigraphic section in Searles Lake. These dates had to be used because wood frag- ments are rare in the sediments, and a few hundred grams of almost any subsurface sample of mud pro- vides enough disseminated organic carbon for dating. Most mud samples contain several percent carbon in the acid-insoluble fraction that commonly makes up 15—25 percent of the total sample (Stuiver, 1964, table 4; this report, table 13). A sample adequate for dating may therefore be obtained from normal-sized cores by a horizontal slice 1—2 cm thick that probably repre- sents less than 100 years of sedimentation. The or- ganic compounds in these muds do not seem to have changed their composition greatly since deposition (Vallentyne, 1957) although some change is indicated by the secondary organic components noted by Man- kiewicz (1975). Broecker and Kaufman (1965, p. 554) suggest that dates on such material from Searles Lake are likely to be about 2,200 years too “old.” An error in this direc- tion is anticipated because the original 1“C/"C ratios in most lacustrine organisms reflects the ratios in the lake waters, and those ratios were probably not ,the same as in the contemporaneous atmosphere because of the slow rate at which atmospheric CO2 exchanges with water. The suggested magnitude of error is based on a calculation of the amount of CO2 in solution and the probable exchange rate that indicates that the ra- tio of 1“C in the CO2 dissolved in the original lake was only 77 percent that of the atmosphere, giving an ap- parent age of 2,200 years. The lake area of 250 km2 assumed by them is about correct when the lake stood at a level near 1,750 ft (530 m), and this is reasonable for the lakes that existed during the time the Lower Salt was being deposited (Smith, 1968, fig. 4). At the time the Parting Mud was deposited, however, the average elevation of the lake surface was near 2,000 ft (610 m) and its area about 400 kmz. Using their for- mula and revised values for the average lake area dur- ing its expanded stages of 400 km2 and for the number of moles of Na,CO3 of 2.7X1012 (table 17), the percent- age of 1“C in the lake relative to that of the atmosphere becomes 83, indicating an original apparent age of 1,500 years. Broecker and Kaufman (1965) found that modern organic material from Mono and Pyramid Lakes gives apparent ages of 1,800 and 800 years, re— spectively. Some confirmation of this magnitude and direction of error predicted by Broecker and Kaufman (1965) results from comparison of two dates from unit M—3 of the Lower Salt (fig. 30). The date of 29,500 : 2,000 on disseminated organic carbon is 2,800 years older than the date on wood from the same horizon, but the large experimental uncertainty on both samples must be considered in drawing conclusions from this compari- son. A similar comparison can be made with the date on wood from the Overburden Mud where one date on disseminated organic carbon from a slightly greater RADIOCARBON AGES OF STRATIGRAPHIC UNITS 71 depth is about 3,000 years older than the dated wood. An age difference twice as large results if other dates on disseminated organic carbon from nearby horizons are compared with the wood date. It is difficult to assess the extent of disequilibrium between atmospheric and lacustrine CO2 during var- ious stages in the history of Searles Lake. One reason is that some of the lakes appear to have been density stratified (Smith and Haines, 1964, p.52; Smith, 1966, p. 174—176). In such lakes, most organisms live in the less saline and less dense surface layer where photo— synthetic activity is greatest. The volume of such a surface layer is only a fraction of the total volume of the lake, and the proportion of total carbonate in it even less because of its lower salinity. The CO2 in the surface layer of a stratified lake, must therefore be more nearly at equilibrium with the atmosphere than is the near—surface concentration of CO2 in an unstra- tified lake having the same total salinity (Deevey and Stuiver, 1964, p. 6). During the times represented by organic-rich mud layers in Searles Lake, both strati- fied and unstratified lakes probably existed, but it is not possible to reconstruct the physical and organic regime of the lake at all stages represented by samples. Additional sources of error come from the fact that during most stages of deposition, older lake deposits were undergoing erosion. They contributed both de- trital organic carbon and older CO2 as their carbonate was dissolved. Pre-Quaternary carbonate rocks in Searles Valley are virtually restricted to the north quarter of the Slate Range. Broecker and Walton (1959, p. 24) consider the effects of this process to be negligible in lacustrine environments. Oana and Dee- vey (1960, p. 265) found evidence based on 13C ratios in Searles Lake muds that the process was potentially a major source of error, but with a larger number of samples, Stuiver (1964, p. 389—390) was unable to de- tect the trend on which their conclusion was based. The many pairs of dates on coexisting carbonate and organic carbon have differences that range from insig- nificant to 6,500 years. Such differences may be pro- duced in part by variations in the ratio of contaminating carbon introduced in the two forms; dates on both carbonate and organic carbon would re- flect the CO2 dissolved from older carbonates, but only the dates on organic carbon would reflect the de- trital older carbon. These diverse sources of error lead to diverse possi- ble combinations. For example, the quantity of car- bonate dissolved from older lake sediments and transported as dissolved CO2 into the existing lake was partly a function of the area of sediments preserved and exposed around the lake; the CO2 derived from this source affected both organic carbon and inorganic carbonate in the new lake sediments. The quantity of detrital carbon was influenced by the area of expo- sure, but it affected only the organic carbon fraction of the new sediments. The rate of new organic carbon production in the lake changed as the concentrations of nutrients and other salts varied, and this variation in productivity affected the ratio of new to reworked organic carbon in the sediments. Moreover, precipita- tion of CaCO3 was the result of two mechanisms, an— nual evaporation and mixing with the underlying saline water; the relative importance of the two mechanisms determined how much carbon came from the CO2 in the surface layer of fresh water (and was thus nearly modern) and how much came from the CO3 in the underlying layer (which was less equili- brated and therefore not so modern). As dates on carbonate minerals involve uncertain- ties as to the original 1“C ratio of the CO2 in solution and other factors, neither the direction nor the magni- tude of displacement of 1“C ages on carbonate minerals can be evaluated. Like the organisms that produced the organic carbon in muds, they reflect the 1“C con- tent of the carbonate and CO2 dissolved in the water, and this ratio was contemporaneous only to the extent that it was equilibrated with the atmosphere. Many of the dated carbonate minerals (such as gaylussite and pirssonite) were formed from moderately saline brines and others (such as trona) came from highly saline brines. Brines represent periods when the lake area was smallest and the rate at which CO2 was equili- brated with the atmosphere at its lowest. Most car- bonate minerals formed, therefore, at times when equilibration with the atmosphere was less complete than during periods when the lake contained rela- tively fresh water and the deposits consisted of organic muds. Additional change can occur after burial if either younger or older carbon is introduced during recrys- tallization of the carbonate minerals. Older carbon from deeper horizons may be introduced by carbonate brines that migrate upward during compaction. Car- bon from older zones may have come from upward moving CO2 produced by microorganisms that contin- ued to assimilate organic matter after burial (although it is uncertain how long CO2 can avoid being converted to methane in the reducing environment provided by these muds). At some time during the history of Searles Lake, when the hydrostatic brine level became like that of the present, carbon from younger horizons began to be introduced into deeper levels because of the downward movement of brines (Hardt and others, 1972, fig. 10). The hydrogen-deuterium ratios of the interstitial brines and hydrated salts in the mud units 72 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. suggest, according to unpublished data, that most of the gaylussite crystallized from brines that had mi- grated from above; dates on those crystals are there- fore expected to be too young. Many trona layers are bedded and fine grained, and their crystals appear to have stopped recrystallizing soon after the layer was deposited. If the brines in which they grew or recrystallized were near equilib- rium with the atmosphere, they are correspondingly reliable carbonate minerals for dating purposes. Whether the brines had reached the required state of near-equilibrium is difficult to determine by an inde- pendent method. All gaylussite and pirssonite crystals in the mud layers were crystallized after the muds in that horizon were deposited and largely compacted (Smith and Haines, 1964, fig. 15; Eugster and Smith, 1965, pl. 1), and all dates on those minerals are suspect. Stratigra- phic unit M—3, the unit that provided wood and dis- seminated organic carbon for dating, contains gaylussite crystals that yielded a 1“C date (fig. 30) of 23,000 : 1,400 years. This is 3,700 years younger than the wood. If the date on wood is correct, the gaylussite contains about 2.1 percent more modern 1“C than the wood and may represent contamination to that ex- tent. A liklier mechanism is that the gaylussite formed by the reaction between aragonite or calcite and NazCoa-rich brines from above: caco,+ 2Na+ + co ,=+5H,0 = CaC03- Na,co,- 5H,o If the original 1“C content of the aragonite was the same as the wood, 3.6 percent modern, and the gaylus- site is 5.7 percent modern, then by the above equation, the CO,= added from brine must have had a 1“C con- tent of 7.8 percent modern, or an apparent age of 20,500 years. Most other pairs of carbonate-organic carbon dates do not differ by this much, and about a third differ in the opposite direction. Of the 22 paired dates on car- bon and carbonate listed by Flint and Gale (1958, ta- ble 2) and Stuiver (1964, tables 1, 2, and 3), 14 of the carbonate ages are younger than the organic carbon (average difference = 1,710 years), and 8 are older (average difference = 520 years). It seems, therefore, that carbon exchange mecha- nisms that operate during recrystallization of carbon- ate minerals are not predictable in terms of either direction or amount. Even when dates on carbonate and organic carbon are determined in pairs, their rela- tive ages cannot be used to indicate whether “old” or “young” carbon was introduced during carbonate re- crystallization because the amount of “old” carbon in the organic sample cannot be independently determined. An opportunity to make a detailed study of paired dates is provided by the 10 pairs listed by Stuiver (1964, table 2) obtained from samples of three cores from the Parting Mud. The three secondary carbonate samples from the top 10 percent of the unit are 200— 1,410 years “older” than the organic carbon in the sur- rounding muds, and the seven samples that lie in the lower 90 percent of the unit are 590—3,980 years “younger”. The relation between dates appears too consistent in direction to be accidental. Stuiver (1964, p. 384—387) suggested that this rela- tion could be explained by the downward diffusion of 1‘C from the top part of the unit, as this isotope had a higher concentration in the brine in that part of the unit and substantial diffusion could have taken place during the 10,000-year period that followed its deposi- tion. Those calculations however, required the top of the Parting Mud to be sealed off by the Upper Salt against post-depositional diffusion of any “C from above, which we now consider unlikely. It seems liklier that the porous carbonate salts at the base of the Up- per Salt provide a poor seal, and that the brines in the Upper Salt constitute an available source of carbonate brine having a greater concentration of 1“C available for diffusion. If this is so, the mechanism is removed by which the carbon as CO3 near the top could have become depleted in 1“C by downward diffusion while the underlying deposits became enriched and thereby appear younger; downward diffusion of 1“C from the Upper Salt would instead have maintained or in- creased the amount of ”C in the carbonate of the top part so that its apparent age remained the same or be- came younger. It is unlikely that this relation between dates can be entirely explained by the diffusion of high carbonate brines into the Parting Mud. Postdepositional move- ment of such brines was probably responsible for the recrystallization of aragonite and calcite to form the gaylussite or pirssonite crystals that were dated. This mechanism would have required the migration of car- bonate-bearing brines in opposing directions; the car- bonate minerals in the top 10 percent of the Parting Mud would have had to incorporate CO3 from older brine that migrated upward from underlying horizons, the carbonate in the lower 90 percent to incorporate CO3 from brine that migrated downward from overly— ing horizons. An alternative explanation stems from the possi- bilities that (1) during deposition, all the original 1“C ages were too old and that successive 1“C ages on both carbonates and organic carbon differed from the cor- rect age and from each other by varying amounts, (2) that during diagenesis, all apparent dates on carbon- ate minerals decreased by a relatively uniform RADIOCARBON AGES OF STRATIGRAPHIC UNITS amount as they incorporated C03 from downward-mi- grating brines. The upper 10 percent of the Parting Mud appears to represent a period when the lake was strongly stratified. In those thin and relatively fresh surface layers, organic carbon was probably generated with apparent 1“C ages that were nearly modern, whereas the submerged and unequilibrated Cos-rich brines of the lower zone (which supplied the carbon for most of the crystallizing CaCO,) was producing ap- parent 1‘C ages that were substantially old. If those apparent ages were too old for the subsequent recrys- tallization of carbonate minerals to reverse, the pre- sent deposits would have organic carbon dates “younger” than the carbonates—the relation we now see. The lower 90 percent of the Parting Mud, on the other hand, appears to represent a time when the lake was unstratified or less frequently stratified so that most of the organic carbon and carbonate was produced in the same layer and probably slightly old. When more modern carbon was later introduced dur- ing diagenesis of the carbonate minerals, it caused the apparent ”C age of the minerals to appear less than the organic carbon. A change in the structure of the lake according to this pattern is reasonable in View of its history during Parting Mud time as deduced from surface mapping (Smith, 1968, fig. 4). The last brief episode of lake ex- pansion (and mud deposition) required introduction of a large amount of new water each year into a basin that had contained smaller and more saline lakes for several thousand years. The new and relatively fresh water introduced by streams was probably nearly in equilibrium with the CO2 of the atmosphere (Broecker and Walton, 1959, p. 32). Carbon in the underlying body of saline water probably would have been “older” , because incompletely equilibrated and possibly be- cause large areas of older lake deposits were exposed to erosion so that their carbon and carbonate were added to the lake. Earlier stands of the lake during Parting Mud time were mostly lower and changed less rapidly; new water had more time to mix and form an unstratified lake that was less completely equilibrated and thus depositing both organic and inorganic car- bon that was slightly old. Considering the possible sources of error in most 1“C dates from Searles Lake, the agreement between dates obtained from different materials and determined in different laboratories is surprisingly good. What the discussion showsis that the dates must be considered individually. The nature of the material, the lake his— tory at the time the sample material was deposited, the subsequent history of the layer, and the size of the experimental uncertainty all must be considered. The discussion here shows that 1“C dates on wood 73 samples are best although wood samples from the Searles Lake deposits are rare. They can be used with— out a correction and with confidence, especially if the experimental uncertainty is doubled, increasing the liklihood from two—out-of—three- to nineteen-out-of— twenty that the correct date lies in this range. Dates on disseminated organic carbon and carbonate are much more abundant, but several factors other than age influence their 1‘C content. Most factors tend to make the disseminated organic carbon samples ap- pear old relative to their true 1“C age. Present-day ana- logs and calculations of reasonable models show that this error may range in size from several hundred to a few thousand years. The age assignments that follow are mostly based on disseminated organic carbon; the “corrected” 1“C ages assume that the reported dates are 500—2,500 years too old. Carbonate mineral sam- ples may have an error of comparable size but either young or old, and these samples provide only a supple- mentary basis for estimating the ages of units in Searles Lake. With these considerations in mind, the contacts of the mud units are interpreted to have un- corrected and corrected 1“C ages as follows. PROBABLE TRUE AGES OF STRATIGRAPHIC UNITS In the Overburden Mud, only the one date on wood is regarded as reliable. The wood was recovered from a depth of 2.4 m, about one-third of the depth to the base of this unit, and its age was 3,520tl90 years (Stuiver, 1964, p.381). The unit was deposited in a much smaller lake than most other mud units (Smith, 1968, fig. 4), and large amounts of older lake beds con- taining carbon and carbonate minerals were exposed to erosion. Possibly it was contamination from these widespread deposits that accounts for the extreme discrepancy between the date on wood from this unit and the dates on both detrital carbon and carbonate from nearby horizons. The 1“C age of the basal contact of the Overburden Mud is more than 3,500 i- 400 years and less than the age of the Parting Mud. The seven 1“C dates from organic carbon derived from the top of the Parting Mud are within a few hun- dred years of each other (fig. 30). Their average is about 10,500 i 165 years. An earlier estimate of the ”C age of this contact was 10,200 years (Stuiver, 1964, p. 382). Quite likely all the dates are too “old” because of the unequilibrated older CO2 that existed in the lake water. The corrected 1“C age of this contact is therefore likely to be between 8,000 and 10,000 years B.P. The minimum correction leading to the older date is favored because Searles Lake was apparently stratified and the ”C that became photosynthetically fixed in the upper layer was therefore more nearly contemporaneous. 74 The base of the Parting Mud is represented by several dates on organic carbon (fig. 30). The two nearest the base average about 23,000:900 years. The seven dates on the lowest 0.5 In average 23,300 : 1,175 years. Stuiver (1964, p. 382) estimated the 1“C age of the basal contact to be about 24,200 years old, the rounded average of the two most reliable dates from the base of the unit. An age of 24,000 years is used here, but the discussion shows that ages spanning more than 1,000 years can be derived. Again, there is reason to suspect that most of the organic carbon was too “old” when deposited the corrected 1"C age of the base of this unit probably lies between 21,500 and 23,500 years. Published dates on units within the Lower Salt in- clude six dates on disseminated organic carbon and one on wood (fig. 30). In addition, 13 dates are now available on disseminated organic carbon from core L—31. (table 19). Where possible, samples were col- lected from the top and bottom sections of the mud layers in the Lower Salt. The ages increase downward from 23,750 i 300 at the top of M—7 to 32,800 :600 at the top of the Bottom Mud, a range of 9,050 years. A minor reversal is represented by the age for the top of M—6 (24,760: 300), although it is consistent with the age of 24,600:400 years previously determined on a sample from the upper part of M—6 (Stuiver, 1964 and fig. 30). Plotting this series of 1“C ages against depth (fig. 31) points up the relatively long periods repre— sented by the muds and the short intervals represented by the salts. For instance, units M—7, M— 6, and M—2 together account for approximately 6,500 years of the 9,000 year long Lower Salt episode. The top of unit S—5 represents the only period of desicca- tion in this unit. It is bracketed by ages of 27,550 i 400 and 28,880:500 years; its ”C age is thus near 28,000 years B.P. The lake history during Lower Salt time was com- plex and changed rapidly (Smith, 1968, fig. 4). It is thus difficult to evaluate the probable degree of CO, equilibration between the atmosphere and the lake waters that contained the majority of organisms or the amount of older carbon and carbonate that was washed into the lake from the surrounding slopes. The difference between dates from unit M—3 on dissemi- nated carbon (29,500 1 2,000) and wood (26,000:2,000) suggests a 2,800-year lag in the ex- change of CO, in the lake with the atmosphere, but the large experimental uncertainty in both dates makes it unwise to use the size of this difference rigorously. The top of M—6 may represent a time when the CO, in the lake was close to equilibrium with atmospheric C0,, and its age, 1,500 years younger than the pro- jected age of this horizon in figure 31, may be the ones SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. 14c AGE, lN YEARS PARTING 24,000 26,000 28,000 30,000 32,000 34,000 MUD WW l l | l l l ‘—* POI i-O—I\ I-lO-i l I I I M - 6 \\\ |——~—’ " ' \ H—i *— .1 s C: S — 5 u! E o .1 -:;'.: \ 3.5.2": \ BOTTOM 2 MUD ' ’VVV} FIGURE 31.—Relation between depth and new HC ages of mud layers in Lower Salt. that are too old by 500—2,5OO years. The corrected 1“C age of the period of major dessication (8—5) is there- fore interpreted to be between 25,500 and 27,500 years, and the age of other mud units to be corre- spondingly less than reported. The two 1“C dates on organic carbon from the top part of the Bottom Mud give ages of 32,700 i 800 years (fig. 30) and 32,8001600 years (table 19). These ages are interpreted to mean that the uncorrected and rounded age of the contact is about 32,500 i 700 years, and that the deposition of the Bottom Mud actually ceased some time between 30,000 and 32,000 years ago. The base of the Bottom Mud is too old for radio- carbon dating. The period of time required for its de- position is inferred by extrapolation of sedimentation rates determined from the Parting Mud (table 20) and from the top 3 m of the Bottom Mud. This period is then added to the age of its top contact, which is ap- proximately 32,500 years. This calculation is not greatly affected by uncertainties about the degree of CO, equilibration in a lake because the sedimentation RATES OF DEPOSITION 75 rates are based on the differences between dates on disseminated organic carbon which are subject to comparable errors. As discussed in the next section, it probably should be corrrected on the basis of the per- centage of acid-insoluble material, which is higher in the Bottom Mud than in the Parting Mud. An unpublished study by Goddard (1970) compares dates on salts obtained by 230Th/mU techniques with dates obtained by 1“C techniques.19 Twenty samples from two cores of the Lower Salt and two samples from the Upper Salt were dated by both methods and found to be in general agreement. Better agreement among dates on the Lower Salt was found when the 1‘C age was reduced by 1,900 years to account for the lag in isotopic equilibrium of lake waters with the atmo- sphere. He concludes that the absolute ages of salt units in the Lower Salt are as follows: Unit Age (x103 years) 8—7 _____________ 23.0 + 1.0 8—6 _____________ 23.9 + 1.0 8—5 _____________ 26.1 i 1.0 8—4 _____________ 26.3 i 1.0 8—3 _____________ 28.5 i 1.0 8—2 _____________ 29.8 i 1.0 8—1 _____________ 31.3 i 1.0 Samples from depths of 7.6 and 21.5 In in the Upper Salt gave 230Th/mU dates of 13.8:1.2X103 and 10.0 : 0.2X103 years, respectively. Using the average sedimentation rates for mud (ta- ble 20) without any correction, the base of the Bottom Mud is calculated to have an age of 143,000i6,000 years. (In calculating these numbers, the age of the top contact of the Bottom Mud was rounded to 33,000 years.) (In calculating these numbers, the age of the top contract of the Bottom Mud was rounded to 33,000 years.) The faster and slower sedimentation rates given for cores 129 and X—20 in table 20 indicate ages of 132,000:12,000 and 155,000:9,000 years. Correcting the sedimentation rates on the assumption that the sedimentation rate of acid-insoluble compo- nents increased so that they accounted for 30 percent of the sample (while the carbonate sedimentation rate remained constant), the base of the Bottom Mud is calculated to have an age of 129,000 years, with the corrected faster and slower sedimentation rates indi- cating ages of 119,000 and 139,000 years. Correcting the sedimentation rate on the assumption that the carbonate sedimentation rate decreased enough for the percentage of acid-soluble material to increase to 30 percent indicates the base of the Bottom Mud to be about 195,000 years old. In this paper, an age of 130,000 years, a rounded value based on the corrected Wded in : Peng, T.-H., Goddard, J. G., and Broecker, W. S., 1978, A direct comparison of “C and 23"Th ages at Searles Lake, Calif: Quaternary Research, v. 9, no. 3, p. 319—329. average sedimentaion rate, is used for the base of the Bottom Mud, but the above discussion indicates the level of uncertainty. RATES OF DEPOSITION A knowledge of approximate rates at which salts and muds were deposited in Searles Lake helps inter- pret the depositional history in the basin. The deposi- tional rates of salts and muds can differ by about three orders of magnitude. Estimates of the rates for each follow. Salts crystallize from a body of brine at a maximum rate that is determined by the annual evaporation. Some salts (such as natron and mirabilite) crystallize at an accelerated rate during winter because their solubility is markedly reduced by low temperatures, whereas others (such as halite) are virtually un- affected by temperature change. The net accumula- tion in a year, though, reflects the amount of water lost by evaporation in the preceding year reduced by the amount of new water that was added to the lake during the year. The maximum rate of saline deposi- tion is thus determined by the annual evaporation when no new water is added during the year. Unless the lake begins to expand, the minimum rate is deter- mined by the amount of new dissolved material that is introduced during the year because the water that brought it in must be evaporated and the dissolved salt crystallized out. Evaporation rates of brines can be derived in two ways: (1) by calculations based on evaporation rates from standard evaporation measurement pans, or (2) by comparison of measured evaporation rates from natural brine bodies. The results, derived below, sug- gest maximum rates of accumulation of porous salt layers to be between 25 and 40 cm per year. These rates mean that all the salt layers in the Bottom Mud, and all but one of the layers in the Lower Salt could be the result of less than 5 years of crystallization. Unit 8—5 of the Lower Salt might represent only a quarter of a century. The Upper Salt might represent half a century. Rates based on evaporation pan data must first be multiplied by a factor approximated in a desert envi- ronment by 0.6 to adjust for the differences in the heat distribution, surface character, circulation, and size of large natural bodies of water relative to small shallow test pans (Blaney, 1955, 1957). The result of this cal- culation must then be multiplied by a factor that nor- mally lies somewhere between 0.6 and 0.8, depending on the species and concentrations of ions in the solu— tion,20 as well as the evaporation rate, humidity, and temperature of the brine (Harbeck, 1955). Using rea- 2“ This factor is applicable to most solutions whose anions are dominated by Na and 76 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. sonable values for these factors and applying them to the saline lakes that existed in Searles Valley during the periods represented by salt layers suggests maxi- mum depositional rates about 15—25 cm of solid salts per year, or 25—40 cm of porous salts. A series of measurements on evaporating brine from Owens Lake allows a second means of estimating maximum salt crystallization rates. That lake, 100 km northwest of Searles Lake, is 600 In higher, so there- fore has a slightly less arid climate. Owens Lake is nor- mally almost dry, but becomes flooded whenever run- off in Owens Valley greatly exceeds the capacity of the Owens Valley aqueduct, which carries water to Los Angeles. In the winter of 1937—38, Owens Lake flooded, and between mid-April 1939 and mid-April 1940, evaporation lowered the level of Owens Lake by 46.9 inches (119 cm) (Dub, 1947, table 3). During the same period, 6.69 inches (17 cm) of rain fell on the lake, meaning that total evaporation from its surface was 53.6 inches (136 cm). The brine had a Na2C03 + Na2B407 + NazSO, + NaCl content of 154,000 ppm at the beginning of this period and a content of 310,000 ppm at the end. Saturation with sodium carbonate probably occurred during July 1939 when a salinity of about 220,000 ppm was reached. Approximately 29 inches (73 cm) of net evaporation occurred during the 9-month period that followed. A brine body, 73 cm deep and containing 220,000 ppm solids (about 245,000 mg/L), would precipitate 9 cm of solids (as- suming a specific gravity of 2.1) or about 15 cm of po- rous salts containing 40 percent brine. Extrapolating linearly from 9 months to a full year indicates nearly 19 cm of porous salts as an annual accumulation rate, but the actual rate might be nearer 25 cm because the three unrepresented months are May, June, and July, when about 40 percent of the annual evaporation takes place (Lee, 1912, table 50). Work by Friedman, Smith, and Hardcastle (1976) on Owens Lake following another period of flooding in the spring of 1969 permits another estimate. A maxi- mum water depth of about 2.4 m was reached in Au— gust 1969, and this water had virtually all evaporated by September 1971; the net evaporation rate was therefore about 120 cm per year. Sodium carbonate salts began to crystallize in mid-August 1970 from 168 cm of brine. Since the salinity of the brine at that time was near 250,000 mg/L, about 20 cm of nonporous salts (density = 2.1) should have accumulated upon desiccation, which took place a year later. This would be increased to nearly 33 cm if its porosity was 40 per- K. Solutions containing much Mg have their evaporation rates much more strongly af- fected by increased salinities because of the very large hydration energy of that ion. Turk (1970) studied the evaporation rate of MgCl,-rich brines from the Bonneville Salt Flats and found that the most concentrated solutions reduced the evaporation rate to as little as 9.5 percent of the freshwater rate. Brines dominated by CaCl2 would exhibit intermediate effects. cent.21 Unless the saline lake begins to expand, the mini- mum rate of salt deposition is controlled by the amount of dissolved material introduced into the ba- sin each year. If one assumes that the Owens River was the source of all solids that entered Searles Valley during a given year and that the quantity was the same as at present, an estimate can be made of the rate salts would have to be crystallized annually to maintain a steady state. In 1908, at a station in the lower Owens River, the water contained an average of 339 ppm solids and had an annual flow of 218,000 acre feet (269x106 m3) (Gale, 1914, p. 263). This flow car- ried, therefore, about 9X1010 g of solids in solution, which, if crystallized in Searles Valley over the area of the Upper Salt (110 kmz), would amount to an annual layer of porous salts about 0.06 cm thick (17 yrs/cm). This rate is about three orders of magnitude less than the maximum estimated rate for salts and is compara- ble to the estimate of depositional rate for muds. Bradley and Eugster (1969, p. B35) used a deposi- tional rate of 0.2 cm/yr (5 yrs/cm) in their calculations of saline depositional rates of trona beds in the Green River Formation. This rate was taken from Fahey’s (1962, table 17) estimate based on the percentage of acid insoluble material in the saline beds. Alternating light and dark beds of trona at the base of the Upper Salt and elsewhere have thicknesses ranging from about 1 to 30 cm (Haines, 1959, pl. 10; Smith and Haines, 1964, p. P22 and figs. 6, 7). These were formerly considered to be annual layers and thus were a means of estimating the rate of saline accumu- lation. However, the salines deposited in Owens Lake in 1970 and 1971 have many similar layers that appar- ently reflect weather cycles of a few days or weeks. Similar beds in the Searles Lake saline layers may also represent much less than a year. Mud layers are deposited at much slower rates. Us- ing thicknesses of the material compacted to its pre- sent form, the rate near the middle of the basin is probably near 0.025 cm/yr or 40 yrs/cm. Table 20 gives all of the 1“C dates from the Parting Mud that used organic carbon and represent a sequence of two or more from the same core. Rates implied by many of the 17 pair are meaningless because the differences between sample ages have large experimental uncer- tainties, or the intervals in the core were too close to allow the differences in ”C dates to be significantThe differences between the uppermost and lowermost samples in cores X—20, X—23, L—U—l, and 129 are con- sidered the most meaningful. They average 381-2 yrs/cm and range from 34:4 to 42:3 yrs/cm. The 2‘ Actually, by January 1971, a substantially greater amount of salt had accumulated at the Owens Lake sample site. The discrepancy is apparently caused by wind, which drove crystals growing on the surface of other parts of the lake to the sample site where they sank to the bottom. RATES OF DEPOSITION TABLE 20.—Depositional rates in Parting Mud [Based on |‘C dates in Flint and Gale (1958, table 2) and Stuiver (1964, table 2). Only dates on organic carbon are used, and only cores with two or more such dates are listed] 77 Midpoint Depositional rates, of sample “C Difference between Difference between Depositional rates between uppermost to lowermost (ft below top of age sample depths sample ages between intervals sample in core Core Parting Mud) (years) (feet) (years) yrs /ft yrs / cm yrs /ft yrs / cm X46 ————— 13532 33,288: 14:88 4.40 200 :r 1,980 45 1.5 X-20 .35 10700 E ’130 _____ ’ 1.28 100 i 184 78 2.6 $1253 $9338}: fig 1.57 930 : 247 592 19 3'40 13’7“): 270 .20 980: 342 4,900 161 1,260:97 42:3 690 16,620: 320 3.50 2,910: 419 831 27 7.28 19’970 : 280 .38 3,350 : 425 8,816 289 11:45 24,690:1070 4.17 4,720: 1,107 1,132 37 x—231 .16 210235 E ’105 ““ ’ .61 355: 152 582 19 51; $4338: $118 4.68 8,210 : 264 1,754 58 9'05 22’300 : 280 3.60 3,500 i‘ 369 972 32 1,255 i 41 41 : 1 11:63 24,630: 460 2.58 2,330: 538 903 30 L—U—13 .02 10 680 : 90 — — — ’ .06 220 i 127 3,666 120 ‘1’: 183383 28 .07 -670 : 120 ___________ I 1,065 : 27 35 : 1 12‘25 235710; 320 12.10 13,480: 330 1,114 37 129 ______ .50 10 494 : 560 ’ 3.00 4,595 : 1,146 1,532 50 3'38 1218331511938 3.50 2,911 : 1,238 832 27 1 1,049 r 122 34 : 4 12’00 222,562 1 1 285 5.00 4,562 : 1,478 912 30 ' ’ * ’ Avg. 1,157:72 38:2 'Parting Mud assumed to be 12 feet thick. 2Average of two dates. 3Parting Mud in L—U—l is 12.3 feet thick. date of 17,100 : 700 from a level 2.75 m below the top of the Parting Mud in core B (Mankiewicz, 1975, p. 10) suggests that the depositional rate of mud near the edge of the deposit (fig. 3) was near 24 yrs/cm (assum- ing the top of the Parting Mud in that area to be 10,500 yrs old). The 1“C dates on organic carbon in two samples from the Bottom Mud (fig. 30) indicate a depositional rate of about 46 : 6 yrs/cm. The upper date of 32,700 : 800 represents the top 0.1 ft (.03 m) of the unit; the lower date of 46,350 : 1,500 represents material at about 9.9 ft (3.0 m) depth. There is a strong lithologic similarity between the dated sediments from the Parting Mud and the sedi- ments in the Bottom Mud. Most of the Bottom Mud is too old for 1“C dating, but extrapolation by use of these sedimentation rates provides an approximation. The 30 m of sediments in the Bottom Mud contains 3—6 percent bedded salts (mostly nahcolite and mirabi- lite), probably deposited in very brief periods of time as a result of chilling the lake waters when they were moderately saline. Excluding these salts leaves the equivalent of about 29 m of mud. The average sedimentation rate calculated from dates in the Parting Mud (38:2 yrs/cm) suggests that the Bottom Mud represents a period of deposition about 110,000 : 6,000 years long. The fastest (34 : 4 yrs/cm) and slowest (42 : 3 yrs/cm) rates indicate depositional periods about 99,000: 12,000 and 122,000 :9,000 years long. The slower sedimentation rate indicated by the two dates at the top of the Bottom Mud (46 : 6 yrs/cm) suggests a period of deposition about 133,000 : 17,000 years long. The variable percentage of acid-insoluble material in the Bottom Mud (fig. 6) shows that the relative im- portance of various sedimentation processes during Bottom Mud time was not constant. The increases in the percentage of insoluble material may have re- sulted from an increase in the rate at which water- or air-suspended clastic material was introduced, a very large increase in the rate of organic productivity, or a decrease in the rate of nonclastic precipitation while other rates remained constant. The average of the acid-insoluble percentages plotted in figure 6 is about 30, the average of the percentages given in table 13 for the Parting Mud about 20. Starting with the balance of nonclastic, elastic, and organic sediments found in Parting Mud, an increase of 70 percent in the depositional rate of acid-insoluble clastic sediments, for example, would increase the to- tal sedimentation rate 14 percent (assuming weight and volume percents are interchangeable) and pro- duce a sediment containing 30 percent acid-insoluble material. An increase of 14 percent above the average sedimentation rate determined for the Parting Mud would change the average sedimentation rate for the Bottom Mud to about 33 yrs/cm. A 500 percent in- crease in elastic material would double the sedimenta- tion rate and produce a sediment containing the observed maximum of 60 percent acid-insoluble mate- rial. Its sedimentation rate would be 19 yrs/cm. An average sedimentation rate of 33 yrs/cm for the 29 m 78 of mud in the Bottom Mud indicates a depositional interval that was about 96,000 years long. The increase in the amount of organic material re- quired to change the percentage of acid-insoluble ma- terial from 20 to 30 is too large to be a likely mechanism. The average sample of Parting Mud con- tains about 2 percent organic material, and this would have to increase 3,500 percent to produce the required 70 percent increase in acid-insoluble material. The observed variation in the percentage of acid- insoluble material in the Bottom Mud might also be caused by a decrease in the sedimentation rate of the nonclastics (mostly carbonates). Again, starting with the balance of sedimentation processes found in the Parting Mud, a reduction of about 40 percent in the depositional rate of carbonate would produce a sedi- ment containing about 30 percent acid-insoluble ma- terial. Its sedimentation rate would be about 68 percent that of the Parting Mud, or about 56 yrs/cm. This rate suggests a depositional interval that was more than 160,000 years long. A final uncertainty in estimating the duration of the period represented by the Bottom Mud by using rates derived from the Parting Mud comes from lack of data on their relative densities and thus the relative extent of compaction. Core samples obtained from fresh- water Lake Biwa, Japan, have mud densities at depths of 22—26 m (equivalent to the typical depth of the Parting Mud, although about twice as old) that aver- age about 1.45 g/cm3, whereas samples from depths of 38 m and 68 m (equivalent to depths of the top and base of the Bottom Mud) average about 1.50 g/cm3 and 1.58 g/cm", respectively (Yamamoto and others 1974, fig. 1). These densities imply a compaction of the deeper samples that increases with depth from 3.5 to 9.0 percent above the value found for the shallower samples. The slope of their age/depth curve (fig. 3) in- dicates an apparent accumulation rate (after compac- tion) for the shallower zone of 25 yrs/cm, and rates for the deeper zone that average 28 yrs/cm, an increase of about 12 percent. A similar curve for sediments in Clear Lake, Calif. (Sims and Rymer, 1975, fig. 4) indi- cates apparent accumulation rates for the same depth zones of about 12.5 yrs/cm and 14 yrs/cm, a 12 percent difference. These data indicate that because of the greater compaction of the Bottom Mud sediments, a correction factor of 3—12 percent is probably required when estimating the period represented by its deposi- tion. Such corrections would increase the age of the base of the Bottom Mud by 5,000—20,000 years and would partially or totally offset the corrections made on the basis of the increased clastic sedimentation rate; it would also explain some of the difference be- tween the 38 yrs/cm average rate determined for the SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. Parting Mud and the 46 yrs/cm rate determined for the top 3 m of the Bottom Mud. GEOCHEMISTRY OF SEDIMENTATION All the late Quaternary lakes that existed in Searles Valley contained an appreciable percentage of dis- solved solids (fig. 32). Even at its highest levels, salini- ties may have exceeded 1.5 percent and pH values probably exceeded 9. Chemical sediments deposited under these conditions consisted mostly of aragonite, calcite, or dolomite. These minerals probably con- tained most of the calcium, much of the magnesium, and some of the carbonate that was dissolved in those lakes. When the lakes shrank as a result of evapora- tion, salinities increased and some primary gaylussite may have crystallized. Eventually, saline layers com- posed of trona, halite, and related minerals formed, and they contain most of the other components origin- ally dissolved in the lake. Only the quantities of those components required for phase equilibrium and the most highly soluble ions remained in solution. The stratigraphic sequence of mud and salt layers beneath the surface of Searles Lake reflects the gross changes in chemical sedimentation and thus provides a first approximation of the chemical history of the VOLUME, IN CUBIC METERS x109 0 1O 20 30 4O 50 60 70 80 90100110 2300 l l | | | 1 I l | l m LIJ E < —————— —Spi|lway level—— —— —>» 200 E g: m 2200 — 1- LL LL 2 _1 “J n: E in; 2100 — — 150 :3 ~ 2: — 1- 5m _, 2000 — _, z _ > < ~ 100 m g '- 0 w 1900 — > u, <( m U, ”J u: E < 1800 — " l .1 ‘ 50 Eu: ”‘ 1700 — 1;; p "" m resent surface _ . , 0 E < My 0 5 1o 15 20 25 30 35 40 m' SALINITY,WEIGHT PERCENT 1400 I 1 1 I I 1 1 | I | 800 l 1000 900 1 1 O 200 300 400 100 5 600 700 00 00 AREA, IN SQUARE KILOMETERS FIGURE 32.—Relati0n between elevation, area, and volume of the Pleistocine lake surface and approximate salinities of its wa- ters. Based on USGS 15-minute topographic maps, measure- ments of area enclosed by each contour taken by planimeter, volumes between pairs of contours calculated using the pris- moidal formula (V = 1/3 C [A1 + A2 + (Al A2)1/2 ]) where C is the contour interval, and A1 and A2 are areas enclosed by succes- sive contours. Solid lines indicate volume and areas above base of Upper Salt (elev. 1,560 ft, 475 m); dotted portions indicate quantities above base of Lower Salt (elev. 1,500 ft, 457 m). Present surface is about 1,616 ft (493 m) above sea level. Salin- ity of waters that desiccated to form Upper Salt plotted on a curve based on the assumption that the total quantity of salts (given in table 17) was dissolved in a homogeneous lake having the indicated volume. GEOCHEMISTRY OF SEDIMENTATION 79 lake. Mineral variations within the mud and salt lay- ers mostly indicate variations in the chemistry of the lake waters. When the mineral, chemical, and other lithologic criteria are combined, the lake history and many of the processes of chemical sedimentation can be reconstructed from the chemical sediments now found. In an earlier paper, Smith and Haines (1964) approximated these processes on the basis of mineral assemblages and crystal habits. The data presented here allow a more detailed reconstruction because they include both the relative percentages and the to- tal volumes of minerals that constitute these assem- blages. Mud layers are discussed first because their evapor- ite composition represents the first components to precipitate from the lake waters. Salt layers represent the components that survived earlier precipitation and were eventually crystallized. Some components in the original lake never crystallized; most of these re- mained in the brines that filled the interstices of the salts (prior to extensive pumping by the chemical companies that extract chemicals from them) al- though some were incorporated into the waters of later lakes and lost by overflow. The chemistry of these brines is implicitly considered during discussion of the crystallization of the salts. MUD LAYERS The mud layers beneath the surface of Searles Lake represent deposits formed in relatively fresh lakes that covered much of the floor of Searles Valley. At their maximum size, these lakes covered about 1,000 kmz; geologic mapping of the valley floor around the present dry lake shows that carbonate-rich muds were deposited in almost all parts of those expanded lakes. The muds below the present dry lake surface have been affected by diagenesis, but their lateral equiv- alents exposed around the edge of the valley com- monly have similar lithologies, yet do not contain the minerals attributed to diagenesis and thus serve as samples of the original deposits. Their original simi- larity is not surprising; the chemical compositions of lake waters tend to be uniform over large areas, and primary chemical sediments deposited from them are comparably uniform. Many of the deposits that crop out are light green, fine-grained, and locally laminated, and most of the samples from the deposits contain high percentages of aragonite, calcite, and (or) dolomite. Halite is com- monly present in fresh samples, but not in quantities like those found in subsurface deposits. Analcime, searlesite, K-feldspar, and phillipsite—minerals thought to be produced by authigenesis in the subsur- face deposits—have not been found in outcrops. Gay- lussite and pirssonite are rarely found in outcrops, and there is no textural evidence in the exposed sedi- ments that large quantities of megascopic crystals like those in subsurface muds ever existed; both minerals are soluble in water, and subaerial leaching normally produces some {form of textural expression. Of the nonclastic minerals in the subsurface mud layers, only aragonite, calcite, dolomite, and northu- pite are considered primary. Aragonite is most com- monly an abundant mineral in the Parting Mud and upper part of the Bottom Mud (pl. 23); the mineral is thermodynamically metastable, and the oldest occur- rence is in the Bottom Mud, in sediments estimated to be about 50,000 years old.“ Possible traces were found in the Lower Salt (table 6); but it is not found in the Overburden Mud (table 18) and Mixed Layer (pl. 2A). Calcite is locally abundant in the upper part of the Bottom Mud but is subordinate in the Parting Mud and nearly absent in the Overburden Mud, Lower Salt, and Mixed Layer. It is the only form of calcium carbonate found in sediments older than 50,000 years, and some of the calcite in older sediments may have been produced by diagenetic alteration of primary aragonite. Dolomite is abundant in the Bottom Mud, in some parts of the Overburden Mud, Parting Mud, and Mixed Layer, but it is absent from mud layers in the Lower Salt. Northupite is fairly common in most mud layers of the Lower Salt and apparently is the Mg-bearing mineral in those layers that contain no dolomite. The mineral is found in the top of the Bot- tom Mud and occurs sporadically throughout the Mixed Layer, but it is absent from the Overburden Mud and Parting Mud. The amount of Ca that reached the center of the ba- sin virtually required transportation to that area by relatively fresh water (Smith, 1966, p. 173—174). This conclusion results from evidence like the following. The average CaO content of the Parting Mud in core GS—16 is 13.1 percent (table 13). The unit in this core is 12.2 ft (3.72 m) thick and has an average density near 2.0. Each square centimeter of the surface of the Parting Mud in this part of the deposit, therefore, re- presents a column of mud that weighs 744 g and con— tains 97 g of CaO. As the Parting Mud was deposited in about 13,500 years, about 7.2X10'a g of CaO was de- posited each year over each square centimeter of this part of the lake floor. And as most or all of this was originally deposited as calcite or aragonite, the least- soluble Ca-bearing minerals in the mud, about 12.9X10'3 g/cmZ/yr of CaCO,was precipitated. Evapo- ration from the lake surface determines the minimum amount of CaCO3 precipitated each year, and if annu- 2‘ Reported occurrences of aragonite in the Mixed Layer (Smith and Haines, 1964, p. P25) were not verified by X-ray diffraction. 8O SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. a1 evaporation was 200 cm/yr, each cubic centimeter of evaporated water would have had to contain about 64 ppm CaCO,, or 26 ppm Ca. This is a reasonable quantity to expect from a fresh- water lake in this climate. Pure water in equilibrium with CO2 in the atmosphere (about 0.033 volume per- cent CO,) at a temperature of 16°C is saturated with CaCO3 at this concentration (Hutchinson, 1957, table 84). With increasing alkalinity, however, smaller con- centrations represent equilibrium amounts. Calculat- ing equilibrium quantities of Ca in alkaline brines is difficult, but analyses of comparable natural waters suggest limits. Jones (1965, table 7) cites numerous analyses of stream, spring, and lake waters from Deep Springs Valley, Calif, that have a composition and pH similar to the waters that probably existed in Searles Lake at various stages. Almost all are in contact with CaCO3 or more soluble carbonate minerals. Only a few waters that contain more than 26 ppm Ca have a total dissolved solid content of more than about 1,000 ppm; the few that do have pH values less than 9 and small to unmeasurable amounts of CO3 relative to H003. The waters that desiccated to form the salts now in Searles Lake are very unlikely to have had these properties once concentration began. As noted in the section in diagenesis, mi- crocrystalline halite is found in large quantities in many of the mud layers and is considered the product of diagenesis from highly saline pore waters incorpo— rated in the sediment at the time of deposition. This proposed mechanism conflicts with the evidence for fresh lake waters which was derived from the CaO content of the muds. This enigma can be resolved if one postulates that (1) cool fresh Ca-bearing water flowed into the basin during part of the year and spread as a layer over highly saline waters that occu- pied the basin during the remaining seasons, and (2) that the calcium in the fresh waters precipitated as aragonite, calcite, or dolomite as a result of both evap- oration and mixing in the zone along the interface be- tween the pre-existing saline water and the new fresh water. Salinity stratification seems unavoidable whenever a highly saline lake receives a small supply of new water on a markedly seasonal basis, and it can persist for very long periods with large or uniform in- flow volumes because salinity stratification is very stable. The presence of Ca-carbonates in mud layers in the middle of the basin, therefore, is interpreted as an indicator that fresh water flowed into a lake that was chemically stratified. Aragonite is commonly formed when CaCO3 is pre- cipitated rapidly from solutions with high pH and sa- linity (Zeller and Wray, 1956; Ingerson, 1962, p. 827— 829). Jones (1965, p. A45) reported aragonite “varves” in the older strata of Deep Springs Lake and attribu- ted the crystallization of aragonite to the results of seasonal changes in the lake combined with inflow from fresh springs, possibly as a stratified surface lay- er. The darker layers were composed of dolomite, cal- cite, and elastic minerals attributed to deposition during summer periods of high evaporation. Arago- nite, along with gypsum and calcite, was also reported by Neev and Emery (1967, p. 82—94) from laminated sediments from the Dead Sea. The white laminae there are composed mostly of gypsum and aragonite, and they form during summer “whitenings” which oc- cur at irregular intervals of several years; necessary conditions are the combined result of several seasons of evaporation followed by a summer warming of the surface waters. The dark laminae are composed of gypsum, calcite, and other components deposited more evenly throughout the intervening periods. In the Na-carbonate-rich waters that constituted Searles Lake even during its high stands, aragonite was most likely to have been formed when Ca-bearing fresh waters flowed into the basin as a surface layer. Where the aragonite is in the form of distinct laminae, the inflow is interpreted to have been seasonal and to have formed a thin layer so that most of its calcium was precipitated during a small part of the year by a combination of warming, evaporation, and mixing. Where aragonite is disseminated, or calcite is the dominant mineral, the introduction of calcium and crystallization of CaCO3 is interpreted to have been a process that occurred uniformly throughout the year because the surface layer was thick enough to prolong mixing and minimize the importance of seasonal warming and evaporation. Support for the suggestion that aragonite laminae represent times when the lake was stratified comes from the study by Mankiewicz (1975, p. 115—117). La- minae are most abundant in the Parting Mud above the level dated by him as 17,000 i 700 years B.P. That segment also contains a higher percentage of macer- ated organics and chlorophyll pigments. Mankiewicz interprets this as being partly a matter of preservation and suggests as its cause that a more stable chemical stratification resulted from an increase in the salinity contrast between the upper and lower water layers, and that this reduced the amount of oxygen that could reach the bottom waters and sediments. However, he also attributes the higher percentages of long-chain hydrocarbons (derived from higher plants), aragonite, and uranium in most parts of this zone to vigorous epi- sodes of inflow which brought plant debris, Ca, and U from upstream. The inflow of large volumes of waters into basins containing more saline and unoxygenated waters virtually requires a density stratification. I: GEOCHEMISTRY OF SEDIMENTATION 81 The dolomite in Searles Lake muds is considered most likely to be primary, or nearly so, although a dia- genetic origin at a much later time cannot be dis- proved. The primary origin seems probable because (1) the mineral occurs chiefly in microcrystalline form; (2) it is found in outcropping lake sediments where other minerals known to be of diagenetic origin are missing; (3) it is concentrated in certain stratigra- phic zones as if reflecting episodes when the chemistry of the lake was favorable, (4) dolomite is abundant in samples that apparently had very small percentages of minerals that would have provided Mg during diagen- esis (the Mg content of interstitial brines is also very low), and (5) dolomite is absent in zones that contain minerals that could react diagenetically to form dolo- mite (for example, when the Mg-bearing minerals northupite and tychite occur with calcite, gaylussite, and pirssonite; see Eugster and Smith (1965, p. 497— 504)). Reasoning like that applied here to explain the CaO content of the muds can also be applied to their MgO content. The average percentage of acid-soluble MgO shown in table 13 is 5.4, this percentage implies a de- positional rate of 3.0 X 10‘3 g/cmz/yr from waters con- taining at least 15 ppm MgO. Since this quantity can be contained in alkaline brines having total salities as high as 300,000 ppm (Jones, 1965, table 7), transport of MgO into Searles Valley by highly saline waters was possible. Solutions containing 15 ppm MgO, however, would also have to contain 21 ppm CaO (15 ppm Ca) to form primary dolomite, and dolomite commonly co- exists with other Ca-bearing minerals meaning that the amount introduced by solutions was substantially higher. If twice as much CaO was introduced, Jones’ data (1965, table 7) suggest that waters having a total salinity of more than a few percent would be inad— equate. The conditions under which dolomite forms are im— perfectly known. In the Overburden Mud and Parting Mud, the mineral commonly occurs in the largest quantities near contacts with salines; in the Bottom Mud and Mixed Layer, though, no correlation is evi— dent. Observations by other workers (Alderman and Skinner, 1957, p. 566; Graf and others, 1961, p. 221; Jones, 1961, p. 201; Jones, 1965, p. 44—45; Peterson and others, 1963; Clayton, Jones, and Berner, 1968; Clayton, Kninner, Berner, and Rubinson, 1968; Barnes and O’Neil, 1971, p. 702—705; Barnes and oth- ers, 1973, table 1) present valid reasons for consider- ing dolomite as both primary precipitates and diagenetic products. The studies by Peterson, Bien, and Berner (1963) and by Clayton, Jones, and Berner (1968), though, indicated dolomite to be forming in, or just below the surface of, the muds of Deep Spring Lake, California. That lake represents an environ- ment that is chemically very similar to Searles Lake (Jones, 1965), and the muds that contain dolomite seem to be modern analogs of the muds in Searles Lake that were deposited during times of major inflow and large lakes. In Deep Spring Lake, dolomite is abundant in the area flooded seasonally as the lake rises and falls, and along the side where lake waters mix with perennial, nearly fresh springs (Jones, 1965, p. A44; Clayton, Jones, and Berner, 1968, p. 417). These relations make it additionally plausible to infer that much of the dolomite in the Searles Lake deposit was formed in an environment in which Mg- and Ca- bearing waters mixed with lake waters that had a high pH and total salinity. The presence of northupite as a primary mineral in the mud layers of Searles Lake has been interpreted as a result of introducing Mg-bearing waters into solu- tions having relatively high concentrations of sodium carbonate and chloride (Smith and Haines, 1964, p. P51). The mineral has been formed synthetically (Wilson and Ch’iu, 1934, table 4), but the required percentages of carbonate and chloride, relative to magnesium, are not typical of those created by evapo- ration of normal waters. A likely explanation, there- fore, for the occurrence of primary nOrthupite in Searles Lake muds is that there was mixing along the interface between inflowing low density Mg-bearing waters and pre-existing high density saline waters. The mud horizons that contain northupite may indi- cate times at which the pre-existing saline waters were much more concentrated than those that existed in the basin at the times aragonite, calcite, and dolomite formed by mixing along the interface of a chemically stratified lake. Thickness variations in the mud units also reflect processes that occurred during deposition. Areal vari- ations were caused by areal differences in the volumes of elastic and chemically precipitated minerals. More than three-quarters of the mud in most units is com- posed of chemically precipated (acid-soluble) miner- als in both subsurface samples and in samples exposed around the edges of Searles Valley. The variations in mud-layer thickness shown on the isopach maps are attributed largely to areal variations in the geochemis- try of sedimentation. Chemical sedimentation pat- terns are influenced by proximity of shorelines, water depth, current patterns, wind directions, and the vol- ume, chemistry, proximity, and entrance point of in- flowing water. To some extent, of course, clastic sedimentation patterns also contributed to the thick ness variations; they are chiefly influenced by the shape of the preexisting surface of deposition, the to- tal volume and size distribution of clastic material in- 82 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. troduced from different directions, the proximity of the shoreline, the depth of water, and the wave energy produced by wind. Both chemical and elastic sedimentation are influ- enced by the proximity of the shoreline. The basal and top zones of the mud layers (and possibly some zones in the middle) were deposited immediately after and just prior to saline deposition; therefore, they are de- posits formed in lakes that had shorelines only slightly beyond the edges of the salt beds (approximated by the arbitrary boundary line on the isopach maps). The middle zones of the mud layers, however, were mostly deposited in larger lakes in which the areas plotted in the isopach maps represent only the centermost por- tions. In general, thin mud layers are interpreted to have a higher proportion of material deposited in shallow lakes with nearby shorelines, and thick layers are interpreted to be composed mostly of deep lake deposits. Isopach maps of the six mud units in the Lower Salt, figures 15—20, seem to show changes in the areal distribution of material, although the lack of many data points near the edges makes contouring and in- terpretation of these areas very subjective. Unit M—2, the lowermost mud, has evenly spaced isopach con- tours indicating that the unit thins gradually toward its area of minimum thickness in the central and southern part. Units M—3 and M—5, in contrast, thin abruptly near the edges and have large central areas that are fairly uniform. Unit M—4 is somewhat similar to M—2 except for an anomalous area in the west-cen- tral part of the contoured area. Units M—6 and M—7, the upper two mud units in the Lower Salt, again thin uniformly from their edges toward a small central area. In all mud units, those that have the greatest rel- ative volumes (table 3), and therefore probably repre- sent the greatest periods of time, are those that most clearly thin gradually toward the center of the basin. This configuration may be the natural form of deep lake sediments. The layers with smaller relative vol- umes may have their distribution of thicknesses domi- nated by the shallower stages at the beginning and end of deposition. The distribution of sediment in the Parting Mud re- vealed by the isopach map (fig. 23) shows that the unit tends to have zones of greater thickness near the north, southeast, and south edges, and to be more uni- form throughout the middle (although a small thinner area is present near the very center). The thick zone shown by unsubstantiated contours along the south- west edge is based on the assumption that chemical and elastic sedimentation from the southwest—the area receiving most of the inflow—was more rapid than elsewhere. The thick zones near some of the oth- er edges are possibly products of inflowing waters from those directions or of periods where the deposit- ing lakes were relatively small. However, the history of lake fluctuations during the time this unit was depos- ited was complex (Smith, 1968, fig. 4), and too many episodes are superimposed to make meaningful inter- pretations. The pattern of thickness variation in the Overbur- den Mud was found to be so uncertain that the iso- pach map used to approximate its volume is not included in this report. Part of the difficulty is in con— structing an isopach map from the gradational lower contact of the unit, and part comes from the erosion of this and older units from surrounding areas (deter- mined by geologic mapping of the exposed units) and redeposition of the eroded sediments on the original top surface of the Overburden Mud. SALINE LAYERS The saline layers in the Searles Lake evaporites are composed of relatively soluble salts that crystallized from saline solutions. During the deposition of layers of this type, the minerals that originally crystallize are mostly determined by the compositions and tempera- tures of the solutions at the surface of the saline lake where they form. Most minerals form on the surface of the lake and float until the crystals founder and sink to the bottom. Although some dissolve at the surface if conditions change before they sink, others do sink and become part of a mineralogically heterogeneous layer that is accumulating on the bottom. Conditions at the surface are changeable, and rapid shifts in tempera- ture, salinity, and CO2 content occur on an hourly, daily, monthly, and seasonal basis. As a result, several different minerals form at the surface from similar so- lutions. On the bottom, all are exposed to the more uniform environment pr0vided by the accumulating layers of salts on the floor of the lake, and the changes that take place tend to make the mineral assemblage more uniform. Many of these changes occur within the first few hours or days after deposition. In this section, the present mineral assemblages are used to reconstruct the geochemical environment that existed during their deposition. Many of the following reconstructions, though, apply principally to condi- tions within the accumulating salt layer, not to the lake that first precipitated them. This is in some ways fortunate because the temperatures in the accumulat- ing salt layer have more meaning; depending on the depth of lake water, they approximate monthly or yearly averages of the lake temperatures and thus can be interpreted more easily as a measure of the climate that prevailed during salt deposition. However, some conclusions regarding the chemistry of the lake waters ‘4 GEOCHEMISTRY OF SEDIMENTATION 83 can also be made. The conclusions regarding which temperatures are recorded by saline mineral assemblages are based largely on unpublished observations of salt crystalli- zation processes that occurred in Owens Lake, Calif. during 1970 and 1971. These observations showed (1) that most saline minerals originally crystallized at the water surface and that their geochemical significance was limited to this zone; (2) that some of the crystals that formed at the surface subsequently dissolved whereas others sank to bottom within hours; and (3) that within days to months, most of the carbonate (and sulfate?) minerals that sank to the bottom were altered to another species without any textural or ob- vious chemical evidence of the change. Had crystals not been collected as they formed or within days thereafter, it would have been virtually impossible to reconstruct the mineralogy of the very first crystals and thus the chemistry of the lake waters in which they formed. Most of the minerals that recrystallized in Owens Lake without leaving obvious evidence were those sensitive to P002. Both nahcolite and natron were pri- mary crystalline phases, and both converted to fine- grained and primary-appearing trona within weeks or months after being deposited on the lake bottom. As explained in more detail by Milton and Eugster (1959), Eugster and Smith (1965), Bradley and Eug— ster (1969), and later in this section, the changes in PCO2 indicated by these reactions are opposites. Alter- ing nahcolite to trona requires a lowered FCC, to allow loss of one-third mole of CO2 for each mole of reacting nahcolite; altering natron to trona requires an in- creased P002 to allow the addition of one-third mole of C02 for each mole of reacting natron. It appeared that the final species of mineral in Owens Lake was deter- mined by the PC02 in the environment provided by the accumulating layer rather than by the solutions responsible for the primary crystallization. The P001 in that environment was partly determined by the ex- tent of organic decomposition within the salt layer and in the salt and mud layers immediately below, partly by the buffering effect of other saline minerals, and partly by the amount of CO2 that was able to mi- grate between the saline layer pore waters, the overly- ing waters in the lake, and the atmosphere. The only sulfate mineral observed in the salines ac- cumulating in Owens Lake was burkeite, but several lines of evidence suggest that mirabilite was originally deposited during the winter. When accompanied by sodium carbonate minerals, the stability fields of mir- abilite and burkeite are influenced by 0002, (11.120, and temperature (Eugster and Smith, 1965, p. 489—497), and it appears that postdepositional reaction of sul- fate minerals also occurs in response to the different conditions found in the accumulating sediments. PHASE RELATIONS APPLICABLE TO SEARLES LAKE SALTS The Searles Lake saline layers are composed chiefly of minerals that represent phases in the sodium bicar- bonate, carbonate, sulfate, chloride system (fig. 33). Trona, burkeite, mirabilite, thenardite, and halite are now present in the Searles Lake deposit; natron is not. In that system, at 20°C, the field of trona occupies a wedge-shaped volume above all the burkeite and part of the natron, mirabilite, thenardite, and halite fields. The top of the trona field is shown shaded, and the field of nahcolite occupies all the area above both it and the underlying fields that extend outside the edges of the trona field. At temperatures above 20°C, the trona, burkeite, and thenardite fields expand at the expense of the halite, mirabilite, nahcolite, and natron fields. At temperatures below 20°C, the re- NaHCOa N32804 NazCOa FIGURE 33.——Three dimensional view of phase system NaHCOa- Na2C03-NaZSOrNaCl-H20 at 20°C. Solid phases indicated as follows: B, burkeite (2Na2504-Na2C03); H, halite (NaCl); M, mirabilite (Na2SO.-10H20); Na, natron (Na2C03~10H20); Nh, nahcolite (NaHCOa); Th, thenardite (NaZSO,); Tr, trona (Na2C03-NaHC03-2HZO). Shaded area represents top of trona field which occupies a wedge-shaped volume between overly- ing nahcolite field and parts of underlying natron, mirabilite, thenardite, halite, and burkeite fields. Boundaries of stability fields plotted in terms of weight percent of solid components in equilibrium solutions. Arrows show slopes of boundaries to- ward base of tetrahedron. Compositions of double salts bur- keite (b) and trona (tr) are shown on edges of tetrahedron. Data from Teeple (1929); diagram from Smith and Haines (1964, fig. 16). 84 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. verse happens. The thenardite field does not exist be- low about 17°C; the burkeite field does not exist below 14.4°C. The data in this report show that of the re- lated minerals in figure 33, only trona, halite, bur- keite, and thenardite coexist in appreciable quanti- ties; nahcolite is sparsely represented in a few layers, and mirabilite is found only in the Bottom Mud. The changing composition of the brine during depo- sition of the salt layers can be theoretically recon- structed from phase diagrams representing the Na2 CO,-NaHCO,-NaZSOrNaCl-HZO system. When data are not available for the 5-component system, the 4- component systems Na2C03-NaHC03-NaCl-H20 and Na,CO,-Na,SO,-NaCl-H20 are used. Although the laboratory determination of points in a 5-component system is not experimentally difficult, plotting the re- sults in a quantitative manner requires at least four dimensions—which is difficult. However, by limiting boundaries to those in which the H20 component is saturated, specifying saturation of one or more solid components, and projecting the field boundaries of the saturated solid components onto a two dimen- sional triangular diagram, all five components can be related.230ther components that exist in natural solu- tions can generally be ignored with only minor uncer- tainties introduced. The changing ratios of these components in the lake brines as crystallization and rapid diagenesis pro- ceeded are indicated by the crystallization path. This path represents a greatly simplified version of the chemical events that occurred in the lake because nei- ther the seasonal changes nor the alteration of the ini- tial crystallization products is indicated. The path does show, however, the compositional changes that ultimately occurred when equilibrium was reached as a result of postdepositional interaction between the lake brines and the minerals in the accumulating salt layer. When interaction ceased, the bulk composition of the buried salt layer became largely fixed, and fur- ther changes involved only phases allowed by that fixed composition. 13 A small uncertainty about the precise relations between stability field boundaries and the compositions of the brines comes from the lack of a satisfactory method for plotting more than one cation percentage in this system. Two major cations (Na and K) were present in the crystallizing solutions in mole ratios of about 10:1, and the diagram dimensions relate hypothetical salts (Na,CO,, NaCl, etc.) that contain only one. Consid- ering the average brine compositions in table 17 as an example, the anion equivalent sum of 00,, S0,, and Cl is 0.5284 per 100 g of solution, whereas the cation equivalent of Na is only 0.4870, meaning that about 80 percent of the 0.0512 equivalents of K are required to electrically balance the three anions in solution. In this paper, relative per- centages in these 4- and 5-component systems have been calculated by converting total C0,, 80,, and C1 to NaQCOJ, NaZSO“ and NaCl. This overstates the amount of one or more of the hypothetical salts in the reacting solutions by maximum of a few percent. In the example using the average brine given in table 17, if all the K is calculated as KCl and removed from the system, and the amount of NaCl is calculated from the amount of Cl that remains, the point representing the brine composition in the NazCOZ-NaZSOr NaCl-H,O diagram is shifted by 3 percent directly away from the NaCl corner of the diagram. The general direction and extent of the crystalliza- tion path in a given phase diagram is deduced from the stratigraphic order of the primary minerals“ that crystallized; its exact path is dictated by the geometry of the stability fields and their boundaries. By com- bining the stratigraphy of the observed mineral as- semblages with the requirements of the phase relations, the composition of primary salt assemblage and the composition of the interstitial brine can be ap- proximately reconstructed. This places limits on the physical conditions under which crystallization took place, and defines both the solids and brines that con- stituted the starting point for subsequent diagenetic change. SALINES IN THE MIXED LAYER The primary salts in the Mixed Layer apparently represent chemically simple waters. The saline miner- als in the Mixed Layers are mostly trona, nahcolite, or halite, meaning that the saline lake waters were domi- nated by Na, 00,, H003, and Cl, and that the brines at the time crystallization began were very close to the NaHC03-Na2C03-NaCl face of the tetrahedron shown in figure 33. The small percentages of burkeite, north- upite, sulfohalite, thenardite, and tychite (pl. 2A) may indicate brief changes in the chemical composition of the lake waters, but they do not occur in a pattern that seems significant. Changes in the compositions of the brines that oc- cupied Searles Lake at the times the Mixed Layer was deposited can be inferred from the saline minerals (pl. 2A). A few thin beds of trona and a little halite are in unit F, but unit E is the oldest unit that contains thick beds of salt. Units E, D, and C are similar in that they contain much more halite than trona and no nahco- lite. In the NaHCO3-NaZCOS-NaCl-H20 system at 20°C (fig. 33), the halite field is small; it becomes even smaller above 25° and below 15°C. Many saline beds in units C, D, and E, however, are composed solely of halite. These show that crystallization started from brine represented by a point very close to the NaCl corner, and even though crystallization of halite forced the composition of the remaining brine toward the NazCOa-NaHCO3 edge of the phase diagram, it never reached the boundary of the adjoining field. Other saline beds in these units, however, contain both trona and halite. These beds show that the initial brines were represented by points in the trona or ha- “ For the sake of simplicity in this paper, the term “primary” is used for minerals that appear to have ultimately become the stable phase in the accumulating salt layer even though they may have actually been products of rapid diagenesis from other species. The term “secondary” is reserved for minerals that are thought to have formed many years after deposition and as a result of diagenetic processes not directly related to the geochemistry of the depositing lake. GEOCHEMISTRY OF SEDIMENTATION 85 lite field but near the phase boundary, so that crystal- lization of the first mineral moved the composition of the remaining brines into the other field before cry- stallization ceased. It is evident, therefore, that the brines that produced salines in units C, D, and E of the Mixed Layer had initial compositions that were near or inside the boundary of the halite field in the Na2C03-NaHCOa-NaCl-HZO system. Temperature changes and small compositional variations allowed crystallization of other minerals at times, but the bulk composition of the brines was at all times dominated by NaCl. The composition of salines in units A and B in the Mixed Layer indicate that there was a major change in the composition of water flowing into Searles Valley relative to the waters that desiccated to produce sa- lines in units C, D, and E. In unit B, trona and nahco- lite are much more abundant than halite, and in unit A, they are virtually the only saline minerals found. The saline composition of the unit shows that the brines that produce salts in unit B had shifted in com- position toward the Na2C03-NaHCO3 edge of the dia- gram shown in figure 33, although extensive crys- stallization of carbonate minerals still shifted the composition of the remaining brine into the halite field. The brines that produced salts in unit A were either so dominated by Na carbonates that no degree of desiccation could shift the brine composition to the halite field, the brines of unit A were desiccated less extensively than those that produced unit B, or those that crystallized salts did so at lower temperatures. Evidence of a different type of change in brine com- position during deposition of the Mixed Layer comes from the Br content of halite (Holser, 1970, p. 309— 315). It abruptly rises and falls in the middle of unit E; more gradually increases to a high level in the lower part of unit C and then decreases. Holser (1970, p. 311) concludes that the gradual rise in Br in the base of unit C represents a gradual net increase in the ex— tent of evaporation, and that the decrease represents a relatively large inflow of new water. The presence of about 13 ft (4 m) of clay and silt in the zone of change (Smith and Pratt, 1957, p. A39) supports this conclu- sion and might indicate more than 10,000 years of in- flow that produced a large lake. The new water in the lake apparently had a high Cl/Br ratio as indicated by their ratio when deposition of salts resumed. The stratigraphic distribution of nahcolite and trona is a function of the chemical activities of C02 ((1002) and water (aH,O)- The relation between these controls and minerals has been discussed by Milton and Eugster (1959), Eugster and Smith (1965), Eug- ster (1966), and Bradley and Eugster (1969). Applica- tion of these principles to salts in the Mixed Layer (Eugster and Smith, 1965, fig. 22) demonstrates a gradual but erratic upward increase in a002, and an abrupt increase in aH,O in units B and A. Both changes are consistent with the other evidence of a major change in the chemistry and character of depo- sition in late Mixed Layer time. The existence of nahcolite in units A and B indi- cates values of 0C0, above those provided by equilibri- um with the atmosphere. They may be interpreted as a measure of the extent of bacterial production of CO2 at the time deposition was taking place. Observations like those of Siever, Garrels, Kanwisher, and Berner (1961), Siever and Garrels (1962, p. 54), and Jones (1965, p. A48) show that CO2 from bacterial decompo- sition can increase the aco2 in uncompacted sedi- ments to a level much above that needed to change any sodium carbonate mineral to nahcolite. Evidence is not available as to whether greater production of CO2 during deposition of the upper part of the Mixed Layer might have resulted from more intense bacte- rial activity or from a larger amount of organic materi- al available for decomposition. The amount of organic material in the mud layers in units A and B was prob- ably high. This quantity was initially controlled by the chemistry of the lake waters that determined the or- ganic productivity of the lake and thus the organic material that can become available for decomposition. The study of Owens Lake (Friedman, Smith, and Hardcastle, 1976) showed that an alkaline lake con- centrated to levels both below and above those neces- sary to form Na-carbonate salts can have a very high productivity; at salinities ranging from 136,200 mg/L to 387,500 mg/L, the phytoplankton population limit— ed downward visibility to 5 or 10 cm. SALINES IN THE BOTTOM MUD The several layers of salines in the Bottom Mud (pl. 23) are mostly nahcolite or mirabilite; only the zone of salts about a meter below the top of the unit contains other minerals. These salt beds commonly contain more mud than the others do, and the lack of clear correlation between saline layers in the three cores plotted on plate 2B suggest that the beds are lenticu- lar. These characteristics indicate that these layers were probably the result of winter cooling of the lake during periods when its salinity had increased to the proper levels. Deposition in this manner accounts for the mono- mineralic composition of the beds, the high mud con- tent in some, and the lenticularity of the layers. The original NazCOa-minerals were probably natron or nahcolite and the Na,SO,-mineral was mirabilite. The solubilities of these minerals are markedly sensitive to temperature, and when cooled, solutions containing 86 SUBSURFACE STRATIGRAPHY AND GEOCHEMISTRY, QUATERNARY EVAPORITES, SEARLES LAKE, CALIF. mixtures of dissolved salts crystallize these minerals exclusive of the others. Except for borax, which was initially a relatively minor ingredient,25 no other salts in the applicable systems are similarly affected by chilling. In pure Na-carbonate and Na-sulfate solu- tions, natron, nahcolite, and mirabilite crystallize at temperatures near 0°C from brines having total salini- ties between 3 and 7 percent (Makarov and Bliden, 1938, tables 1 and 4; Freeth, 1923, tables 1 and 17 ); at 15°C, these salts crystallize from solutions having sa- linities between 8 and 15 percent. Addition of NaCl raises the salinity but lowers the percentage of these components necessary for their crystallization (Ma- karov and Bliden, 1938, table 8). The tendency for the salts that are preserved to be mixed with mud and form lenticular beds is inter- preted to be a result of a mechanism that allows salts to survive the warmer seasons that follow crystalliza- tion. Initially, each salt layer probably covered the floor of the lake, but currents in the relatively shallow lake probably mixed the salts in some parts of the lake with mud, a mixture that would have been relatively impervious. During the following warmer seasons, this property would have protected the salts in these areas from being infiltrated by lake waters and dissolved. These saline layers, therefore, are considered indic- ative of crystallization during winter from an interme- diate—sized lake. Salinity of the lake water was probably in the range 5—15 percent, the exact value depending on the species and amounts of the other components in solution which depress the saturation point. The data in figure 32 show that lakes having these salinities could have had areas of 400—600 km2 and depths of 70—120 m. The discontinuous layer about 1 m below the top of the Bottom Mud that contains borax, northupite, trona, nahcolite, and thenardite may also be a result of winter crystallization inasmuch as these minerals are reasonable products of winter crystallization if modi- fied by diagenesis. If the layer was initially composed of salts crystallized during winter, the most likely pri— mary minerals were borax, natron or nahcolite, and mirabilite. The diagenesis that produced the present suite could have been caused solely by a post- depositional decrease in aH,O- Saline brines having a lower aH,O exist in the overlying Lower Salt, and downward migration of brines from that layer would have mixed with the original interstitial brines and lowered the “H20 This, with or without a change in ac02, could have produced the present mineral suite. Remembering that the meter of intervening mud con- tains dolomite and gaylussite, and using aco,-GH,0 2” When the brines that formed the Upper Salt were concentrated to 5 percent, Na,B.O., constituted about 0.15 percent. diagrams of Eugster and Smith (1965, fig. 19), a de- crease in “H20 of brine would have caused minerals of stability field 15 (mirabilite, natron, gaylussite, halite, dolomite, aphthitalite) to alter to those of field 6 (thenardite, nahcolite, gaylussite, halite, dolomite, aphthitalite) or field 18 (thenardite, northupite, trona, gaylussite, halite or dolomite, and hanksite), depending on the “CO; Note that gaylussite, northu- pite, and nahcolite do not coexist in any field of those acoz—aHZO diagrams and that they do not coexist in the salt layers being described; gaylussite exists in the muds of all cores, but nahcolite exists in this zone only in cores GS—15 and GS—16, and northupite exists only in cores GS—18 and GS—27. SALINES IN THE LOWER SALT, UPPER SALT, AND OVERBURDEN MUD The starting compositions of the solutions that pro- duced the Lower Salt and Upper Salt in Searles Lake are now best approximated by the present bulk chemical composition of each salt unit or group of re— lated units. Their compositions are summarized in ta- bles 10 and 17. These calculated compositions are weighted averages of analyses and include the compo- nents in both the present salts and the residual brines that fill their interstices. There is the possibility that small quantities of soluble components remained in the surface brines at the time salt crystallization ceased and these components were incorporated into younger saline units or removed from the basin by subsequent overflow. The only evidence of this hap- pening to an appreciable extent is in the deposition of unit S—7. In the following discussion, the salt layers of the Lower Salt are divided into two groups which repre— sent sequences of crystallization that are interrelated and can be treated almost as if they were continuous. The first is composed of units S—l to S—5, and the sec- ond of units S—6 and S—7. This grouping is emphasized when viewed as a histogram (fig. 34). Crystallization of the Upper Salt is considered as a single continuous event. The crystallization paths followed by the brines during these three episodes are plotted on the phase diagrams shown in figures 35, 36, and 38. The salines in the Overburden Mud were formed by a crystalliza- tion sequence similar to the one that formed the Up- per Salt. LOWER SALT The phase diagrams shown in figure 35 allow the crystallization of units S—l through S—5 to be recon— structed. Field boundaries in the 5-component system with all points and boundaries saturated in both H20 and NaHCO3 are shown in figure 35A. To portray this GEOCHEMISTRY OF SEDIMENTATION 87 MINERAL COMPOSITION, IN VOLUME PERCENT O 2 a u : e 2 '12 a E E m '8 o '7' m £ 0 x 1: § 2% ii ‘55 E a % E E l- m I I l- < Z a: Z [—7 [—1 I“! l_l l_| O 1000100010001000505050505 UpperSalt D34 |<1 [:43 [117 10.1 It |:|3.1 5—7 :74 H [313 |]0.6 It D22 S—6 :77 I1 U21 |0-3 .. 33 s—5 [:37 n8 I:|51 It It |0.5 It 33': s—4 E64 [129 It |<1 [|1.7 |]1.o 3 s—3 1:87|<1 |2 [11.8 |]1.3 5-2 1:192|<1 |0.5 No.9 [10.9 s—1|:]92 no.6 |0.3 lo.1 Nazso. Burkelte (mineral composition) A . .. Mirabmte Outer limits of trona field Thenardite \ Burkeite i s—1+s-2+s—3 1‘ 17-3 Natron b-1 S-4 b—2 8-5 Hallte \ Nazco3 32 NaCI NaHCO; B Trona (mineral Nahcolite composltion, projected) s-1+S-2+S-3 Natron \ 82 percent Na2C03 ["2] Burkeite 72 percent NaCl 18 percent NaCl 3‘5 28 percent N32$O4
— _ _ \\ Burkeite ‘ \ [7-1 1 M b-3 Natron \ ~