30.2.... I 5? W131.“ J. 574.3 1: .I 3 . 7:. (1....) » “vhgu. 3 a. 3.003 {SW/AI '7 l. (o LlBRARY Michigan State This is to certify that the UI'IIVG I'SlIy thesis entitled EVALUATING GEOLOGICALLY-CONTRAINED MODELS WITH PUMPING TESTS IN A HETEROGENEOUS ALLUVIAL AQUIFER, HELIPAD SITE AT LAWRENCE LIVERMORE NATIONAL LABORATORY, CALIFORNIA presented by Robert Stirling Trahan has been accepted towards fulfillment of the requirements for the MS. degree in Environmental Science 6%me \ é Major Professor’s Signature 8 403 cat 20° 3__ W Date MSU is an Affirmative Action/Equal Opportunity Institution PLACE IN RETURN Box to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 6/01 cJClFiC/DatoDuepBS-p. 15 EVALUATING GEOLOGICALLY-CONTRAINED MODELS WITH PUMPING TESTS IN A HETEROGENEOUS ALLUVIAL AQUIFER, HELIPAD SITE AT LAWRENCE LIVERMORE NATIONAL LABOMTORY, CALIFORNIA By Robert Stirling Trahan A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Geology 2003 ABSTRACT EVALUATING GEOLOGICALLY-CONTRAINED MODELS WITH PUMPING TESTS IN A HETEROGENEOUS ALLUVIAL AQUIFER, HELIPAD SITE AT LAWRENCE LIVERMORE NATIONAL LABORATORY, CALIFORNIA By Robert Stirling Trahan Pumping test results and numerical groundwater simulation at the Helipad Site, Lawrence Livermore National Laboratory (LLNL), are used to evaluate four geologically-constrained conceptual models. These include a homogenous-layered model, a homogeneous-layered model, a transition probability geostatistical model, and a stratigraphic transition probability geostatistical model. Each of these models incorporates different aspects of physical heterogeneity observed at this site. At the Helipad Site at LLNL, relatively mature paleosols within the alluvial deposits mark unconforrnities that separate this alluvial aquifer into a series of stratigraphic zones. This provided the stratigraphic framework in which conceptual models were developed. The results of this study show that multiple realizations for the distributions of hydrofacies within stratigraphic and paleosol units (modeled using transition probability geostatistics) better match pumping test results then conceptual modeling approaches that did not incorporate these finer scale heterogeneities. Based on multiple realizations of hydrofacies distributions, characteristics for the distributions and dimensions of gravel and sand channel were identified that could be later used to filter potential poorly performing realization. ACKNOWLEDGMENTS This material is based on work supported by the National Science Foundation under Grant No. EAR-0133666. Additional support was received from Lawrence Livermore National Laboratory under subcontract #B503592. I benefited greatly from discussions with Fred Hoffman, Zafer Demir, Rick Blake, Charles Noyes, Mike Maley, Walt McNab, John Karachewski, and Souheil Ezzedine. Additionally, I appreciate work on this project by Leslie Mikesell, George Bennett, Beth Apple, and Jill Schlanser. Special thanks go toward my adviser, Gary Weissmann, and committee members David Hyndman and M. S. Phanikumar, and especially my loving wife, Kimberly. iii TABLE OF CONTENTS LIST OF TABLES ............................................................................................................ vi LIST OF FIGURES ......................................................................................................... viii Chapter One Introduction and Scope of Work .......................................................................................... 1 Introduction .............................................................................................................. 1 Purpose of Study ...................................................................................................... 7 Thesis Outline .......................................................................................................... 9 Chapter Two Geologic Setting and Stratigraphic Assessment ................................................................ 11 Introduction ............................................................................................................ 1 1 Background Geology ......................................................................................................... 12 Regional Geology Setting for the Livermore Valley ............................................. 12 Geologic evolution and Stratigraphy of Middle California ................................... 14 Helipad Site Geology ........................................................................................................ 20 Core and Facies Descriptions ................................................................................. 20 Comparison of Core to Geophysical Well Logs .................................................... 2] Correlations and Cross Sections ............................................................................ 22 Hydrology in the Livermore Valley ....................................................................... 25 Chapter Three Pumping Test Evaluation in a Heterogeneous Alluvial Aquifer ....................................... 30 Introduction ............................................................................................................ 30 Study Area ............................................................................................................. 33 Site Geology and Stratigraphy ............................................................................... 35 Pumping Test Data ................................................................................................. 36 Modeling Pumping Tests ....................................................................................... 41 Results and Discussion .......................................................................................... 47 Conclusion ............................................................................................................. 52 Chapter Four Conclusions ........................................................................................................................ 54 Alluvial Fan Heterogeneity .................................................................................... 54 Conceptual Model Development ........................................................................... 54 Evaluation of Conceptual Models .......................................................................... 55 Analytical Solution Evaluation .............................................................................. 56 Future Considerations ............................................................................................ 57 iv Appendix A Core Descriptions ............................................................................................................... 59 Appendix B Geophysical to Well Core Correlations ............................................................................. 97 Appendix C Cross sections ................................................................................................................... 129 Appendix D Analytical Solution .......................................................................................................... 146 Introduction .......................................................................................................... 146 Assumptions for Analytical Solutions ................................................................. 147 Methods ................................................................................................................ 148 Results .................................................................................................................. 149 Conclusions .......................................................................................................... 15 0 Appendix E Isopach maps .................................................................................................................... 162 Appendix F Transition Probability Geostatistics ................................................................................. 177 Introduction .......................................................................................................... 1 77 Transition Probability Model Development ........................................................ 177 Conclusions .......................................................................................................... 182 Appendix G GMS Model Development ............................................................................................... 201 Introduction .......................................................................................................... 201 Model Development ............................................................................................. 201 Assumptions ......................................................................................................... 205 Optimization ........................................................................................................ 207 Calibration ............................................................................................................ 209 Conclusions .......................................................................................................... 209 Appendix H Drawdown Results ........................................................................................................... 213 Appendix I FORTRAN Code ............................................................................................................. 244 Bibliography .................................................................................................................... 259 It}; It's-130.1 List, Table 0.3 Surf Table El Pro;- la‘aic F2 Table [3 Title E4 We F5 Table 3.1 Table D.1 Table D2 Table F.1 Table F.2 Table F .3 Table F.4 Table F .5 Table F .6 LIST OF TABLES Listed above are hydraulic parameters assigned to either layers or hydrofacies used in each of the modeling approaches. Listed values have been calibrated to best match observed drawdown histories ..................... 46 Listed above are accumulative drawdown readings at a specified time after pumping commencement. Drawdown histories for wells 1650, 1651, 1652, 1653, 1654, 1655, 1656, and 1657 are recorded ...................................... 152 Summary table of hydraulic parameters from analytic solutions, result are recorded in meter and day units ............................................................... 157 Proportions of hydrofacies for the TPG simulation and for each stratigraphic zone of the layered simulation. E and F paleosol zones lack the silty sand hydrofacies and are modeled excluding this category .................................................................................................... 184 Embedded transition probability matrices for the transition probability geostatistical simulation. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) ........................................................ 186 Embedded transition probability matrices for the C zone. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) ................................. 187 Embedded transition probability matrices for the D paleosol. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) ............. 188 Embedded transition probability matrices for the D zone. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) ................................. 189 Embedded transition probability matrices for the E paleosol. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) ............. 190 vi ia‘alc F .7 lab}: E8 Table E9 This (3.1 ,' I.) W . In. 6.- Llsi: Table F.7 Table F.8 Table F.9 Table G.1 Table G.2 Embedded transition probability matrices for the E zone. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) ................................. 191 Embedded transition probability matrices for the F paleosol. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) ............. 192 Embedded transition probability matrices for the F zone. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) ................................. Listed above are hydraulic properties assigned to either layers or material (hydrofacies) used in each of the four modeling approaches. Values listed are initial values incorporated into each modeling approach ................... 211 Listed above are hydraulic properties assigned to either layers or material (hydrofacies) used in each of the four modeling approaches. Values listed have been calibrated for to best match drawdown histories observed (for comparison purpose, geostatistical models all used the same values calibrated from realization one) ............................................................... 212 vii 'V II 71:?qu . . figze 1.3 II.“ 9’; 1 s n.‘.\ a. o :,...,, 1 5 F153}; -_. Emu F I332 2.8 Fin. I; '3 ‘ a“ 3] Figure 1.1 Figure 1.2 Figure 2.1 Figure 2.2 Figure 2.3 Figure 2.4 Figure 2.5 Figure 2.6 Figure 2.7 Figure 2.8 Figure 3.1 LIST OF FIGURES Lawrence Livermore National Laboratory (LLNL) is located in west central California approximately 50km east of San Francisco (modified from US Department of Energy, 1998) ........................................................ 3 Block diagram showing distributions of VOC contaminants within hydrostratigraphic units (HSU’s) (Modified from US Department of Energy, 1998) .............................................................................................. 3 Regional geologic map of western-central California. The Livermore Valley is outlined with a dashed line (Modified from USGS, 1966). Image is presented in color ................................................................................... 13 A) Generalized depiction of Late Jurassic to early Cretaceous landscape of western central California (modified from Howard, 1979). B) Generalized depiction of Late Cretaceous landscape of western central California (modified from Howard, 1979) ................................................................. 15 Figure 2.3 Stratigraphic column for the Livermore Valley (modified from Huey, 1948) ............................................................................................... 16 Generalized geologic map of the Lawrence Livermore National Laboratories and vicinity (Modified from Dibblee, 1980) Image presented in color ....................................................................................................... l8 Stratigraphic cross sections at the Helipad Site. Top paleosol surfaces are shown (dashed line). Paleosols are informally named following Weissmann (2001) Image presented in color ............................................. 24 Illustrated in this figure are major streams within the Livermore Valley. Streams are marked by dashed lines; the position of the LLNL property is outlined with a box and labeled ................................................................. 26 Aerial photograph (from 6/8/1940) of the LLNL property area. Indicated are northwest trending river channels of the Las Positas and the Arroyo Seco. Also indicated is the more recent location of the Arroyo Seco stream channel ....................................................................................................... 27 Figure 2.8: Groundwater table beneath the LLNL property, recorded in 1997 (Modified from Harrach, 1998) ........................................................ 29 Lawrence Livermore National Laboratory (LLNL) is located east of San Francisco, within the coast range province of California (modified from US Department of Energy, 1998) .............................................................. 34 viii . a 1‘ 1v“ \ Pint - .- U ._‘_1 (£4 8 '4) ‘4) figure 3.4 -11 u 1 E: (D '4) 'Jo ti“ 9 q tl‘tie 36 Figure 3.7 '17 if) '1- L.“ 2f W, U» 77‘ W. Figure 3.2 Figure 3.3 Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7 Figure 3.8 Figure 3.9 Figure A.1 Shown above is an illustrated conceptual model of the Helipad Study Site. Six paleosol surfaces (colored) have been identified which are used to separate depositional sequences (transparent) (Weissmann and Bennett, 2001; Appendix C). Also indicated is the saturated zone (translucent blue shading) Image presented in color ............................................................. 37 Well and cross section locations within the Helipad Site study area ......... 39 Observed drawdown history for 1650 (left) and 1250 (right) series wells. Three different clusters of drawdown are observed due to the physical heterogeneity of the alluvial system .......................................................... 39 A) Well locations and interpreted distribution for gravel and sand hydrofacies within the E3 channel. Pumping wells are indicated with black circles. B) Analytical solution to observed drawdown histories at the Helipad Site showing a contour map of hydraulic conductivity, values are measured in m/s ......................................................................................... 40 Homogeneous-layered model. B) Homogeneous-layered model incorporating paleosol layers. C) Transition probability geostatistical simulation. D) Stratigraphic transition probability geostatistical simulation (stratigraphic zones are simulated separately and later merged). Image presented in color ....................................................................................... 42 Drawdown results for modeling approaches are plotted for each observation well. Positions of plots are similar to there locations at the Helipad Site (a representation of well locations are illustrated in the center of this figure). For each graph, drawdown is in meters (y-axis) and time is in days (x-axis). Pumping wells are 1551 and 1552. Observed drawdown for well 1651 is not included due to poor well construction ...................... 48 Cross section and map views of drawdown propagation at time of 0. 1 5 day A) Simulated drawdown for the homogeneous-layered conceptual model. B) Simulated drawdown for the homogeneous-layered model incorporating paleosol layers. C) Simulated drawdown for the transition probability geostatistics conceptual model. D) Simulated drawdown for the stratigraphic transition probability geostatistics conceptual model. Image presented in color ............................................................................ 50 Contour map of measured drawdown at a time of 0.15 day. Drawdown values for 1650 series observation wells are indicated .............................. 51 Gravel Facies o Clast supported. 0 Massive. ix Figure A.2 Figure A.3 Figure A.4 Figure A.5a Figure A.5b Figure A.5c Figure A.6a Figure A.6b Figure A.6c Figure A.7a Figure A.7b Figure A.7c Figure A.8a Figure A.8b 0 Typically contains thick, reddish, clay coats on grains ....................... 60 Sand Facies 0 Typically fine to medium sand, with some coarse to very coarse sand and/or gravel, commonly silty. o Grains are typically subrounded. 0 Well to moderately sorted with some clay coats on grains .................. 60 Silty Sand Facies 0 Massive. 0 Common root traces. 0 Typically brown (lOYR) with mottling around root traces. 0 Light pedogenic alteration (thin to no clay coats, minor MnO, no visible pedogenic structures) ............................................................... 61 Paleosol Facies -Evidence for pedogenic alteration: Thick clay coats on ped faces. Preservation of soil structures (blocky to prismatic). Reddish color (5 to 7.5 YR typical). Common Mn- and F e-oxides. Common carbonate in upper paleosols (above 50 ft) and Lower Livermore. Formation; Carbonate is present in lower paleosols (below 50 ft), however are less common. 0 Occasional root traces .......................................................................... 61 Core description of facies identified in well 220 ....................................... 62 Core description of facies identified in well 220 continued ...................... 63 Core description of facies identified in well 220 continued ...................... 64 Core description of facies identified in well 653 ....................................... 65 Core description of facies identified in well 653 continued ...................... 66 Core description of facies identified in well 653 continued ...................... 67 Core description of facies identified in well 906 ....................................... 68 Core description of facies identified in well 906 continued ...................... 69 Core description of facies identified in well 906 continued... .................. 70 Core description of facies identified in well 1205 ..................................... 71 Core description of facies identified in well 1205 continued... ................ 72 ligre A93 Figure A.% Figs: ADC Figure .A.9d Figre A. l 03 fig: A. 1 0b Figure Al la Figu'e Al lb Fig: A. 1 la Figure A.le Figure A. 13c Figure A. 1 3a fire A.le Fire A.13c Figure A. 13c Figure A. 143 5m Altb C Ii.» , “51": .‘Uk Fine A153 FEE-lit? AISb ET“ .lsue AISC In. I ride ‘.\'16a COIL . COR Cort Core Core Core Core . Core 1 C ore Core Core ore Core C(Ifc Core Core Core Core Figure A.9a Figure A.9b Figure A.9c Figure A.9d Figure A.10a Figure A.10b Figure A.11a Figure A.11b Figure A.12a Figure A.12b Figure A. 12c Figure A.13a Figure A.13b Figure A.13c Figure A.13c Figure A.14a Figure A.14b Figure A.14c Figure A.15a Figure A.15b Figure A.15c Figure A.16a Figure A. 1 6b Core description of facies identified in well 1206. .. ................................. 73 Core description of facies identified in well 1206 continued... ................ 74 Core description of facies identified in well 1206 continued... ................ 75 Core description of facies identified in well 1206 continued. ................. 76 Core description of facies identified in well 1207 ..................................... 77 Core description of facies identified in well 1207continued ..................... 78 Core description of facies identified in well 1208 ..................................... 79 Core description of facies identified in well 1208 ..................................... 80 Core description of facies identified in well 1223 ..................................... 81 Core description of facies identified in well 1223 continued... ................ 82 Core description of facies identified in well 1223 continued... ................ 83 Core description of facies identified in well 1303. .. ................................. 84 Core description of facies identified in well 1303 continued. ................. 85 Core description of facies identified in well 1303 continued... ................ 86 Core description of facies identified in well 1303 continued... ................ 87 Core description of facies identified in well 1306. . . ................................. 88 Core description of facies identified in well 1306 continued... ................ 89 Core description of facies identified in well 1306 continued... ................ 90 Core description of facies identified in well 1401... ................................. 91 Core description of facies identified in well 1401 continued... ................ 92 Core description of facies identified in well 1401 continued... ................ 93 Core description of facies identified in well 1416... ................................. 94 Core description of facies identified in well 1416 continued. ................. 95 xi Figure A.16c C or. Figre 81 Figure 8.3 Figure 8.3 Frgre BA Figure 3.5 Figure 8.6 Figure 8.8 Figure 89 Figure B. 10 Figure 8.1 1 5 "'e 8.1: (I‘D 5:36 8.13 We 8.14 Figure 315 We 3.16 ERIE 8.17 Fig”? B. 13 Remit 8.19 PIERRE 8:0 K1512“ 8‘] F1, ‘59-"? 1 8‘2 Get Gen. Figure A.16c Core description of facies identified in well 1416 continued... ................ 96 Figure B.1 Figure B.2 Figure B.3 Figure B.4 Figure B.5 Figure B.6 Figure B.7 Figure B.8 Figure B.9 Figure B.10 Figure B.11 Figure B.12 Figure 3.13 Figure 8.14 Figure B.15 Figure 3.16 Figure B.17 Figure 8.18 Figure 3.19 Figure B.20 Figure B.21 Figure 3.22 Geophysical well logs and core descriptions for well 220 ........................ 98 Geophysical well logs and core descriptions for well 653 ........................ 99 Geophysical well logs and core descriptions for well 906 ...................... 100 Geophysical well logs and core descriptions for well 1205 .................... 101 Geophysical well logs and core descriptions for well 1206 .................... 102 Geophysical well logs and core descriptions for well 1207 .................... 103 Geophysical well logs and core descriptions for well 1208 .................... 104 Geophysical well logs and core descriptions for well 1223 .................... 105 Geophysical well logs and core descriptions for well 1250 .................... 106 Geophysical well logs and core descriptions for well 1251 .................... 107 Geophysical well logs and core descriptions for well 1252 .................... 108 Geophysical well logs and core descriptions for well 1253 .................... 109 Geophysical well logs and core descriptions for well 1254 .................... 110 Geophysical well logs and core descriptions for well 1255 .................... 111 Geophysical well logs and core descriptions for well 1303 .................... 112 Geophysical well logs and core descriptions for well 1306 .................... 113 Geophysical well logs and core descriptions for well 1307 .................... 114 Geophysical well logs and core descriptions for well 1401 .................... 115 Geophysical well logs and core descriptions for well 1416 .................... 116 Geophysical well logs and core descriptions for well 1550 .................... 117 Geophysical well logs and core descriptions for well 1551 .................... 118 Geophysical well logs and core descriptions for well 1552 .................... 119 xii Figrre 8.33 Figure 8.34 Figs: 8.35 Figre 8.36 Figrre 8.37 First 838 Figrre 8.39 Figre 8.30 Flg‘rre 8.31 Figure C .l Fierce: 51mm Fm CA West ' area... “€81 1 area... LOCari CharaL 3013.1] Ma... Figure B.23 Figure B.24 Figure 8.25 Figure B.26 Figure B.27 Figure B.28 Figure B.29 Figure B.30 Figure B.31 Figure C] Figure C.2 Figure C.3 Figure C.4 Figure C.5 Figure C.6 Figure C.7 Figure C.8 Figure C.9 Geophysical well logs and core descriptions for well 1553 .................... 120 Geophysical well logs and core descriptions for well 1650 .................... 121 Geophysical well logs and core descriptions for well 1651 .................... 122 Geophysical well logs and core descriptions for well 1652 .................... 123 Geophysical well logs and core descriptions for well 1653 .................... 124 Geophysical well logs and core descriptions for well 1654 .................... 125 Geophysical well logs and core descriptions for well 1655 .................... 126 Geophysical well logs and core descriptions for well 1656 .................... 127 Geophysical well logs and core descriptions for well 1657 .................... 128 Location of wells at the Helipad Site where wells with or without core are indicated ................................................................................................... 130 South to north trending cross-section through the western central helipad area ........................................................................................................... 131 South to north trending cross-section through the central helipad area ........................................................................................................... 132 South to north trending cross-section through the eastern central helipad area ........................................................................................................... 133 West to east trending cross-section through the northern central helipad area ........................................................................................................... 134 West to east trending cross-section through the central helipad area ........................................................................................................... 135 West to east trending cross-section through the southern central helipad area ........................................................................................................... 136 Location of all wells within the Helipad Site study area used for site characterization ........................................................................................ 1 3 7 South to north trending cross-section through the helipad study area ........................................................................................................... 138 xiii Figrre C10 Figure C.ll Fizure C.l3 Figs: C .13 FigrreClA HEEDFO 5:333 [)1 1 \ch C urn Cur. Cur. C ur. Cur. C ur. a C un ; C un e 1m: hfihk used 1 Come. Sh0“3 COUhl Weft]. 1654.. Figure C.10 Figure C.11 Figure C.12 Figure C.13 Figure C.14 Figure C.15 Figure C.16 Figure D.1 Figure D.2 Figure D.3 Figure D.4 Figure D.5 Figure D.6 Figure D.7 Figure D.8 Figure D.9 Figure D.1O Figure D.11 South to north trending cross-section through the helipad study area ........................................................................................................... 139 South to north trending cross-section through the helipad study area ........................................................................................................... 140 South to north trending cross-section through the helipad study area ........................................................................................................... 141 West to east trending cross-section through the helipad study area ........ 142 West to east trending cross-section through the helipad study area ........ 143 West to east trending cross-section through the helipad study area ........ 144 West to east trending cross-section through the helipad study area ........ 145 Curve match and analytical solution for well 1650 ................................. 153 Curve match and analytical solution for well 1651 ................................. 153 Curve match and analytical solution for well 1652 ................................. 154 Curve match and analytical solution for well 1653 ................................. 154 Curve match and analytical solution for well 1654 ................................. 155 Curve match and analytical solution for well 1655 ................................. 155 Curve match and analytical solution for well 1656 ................................. 156 Curve match and analytical solution for well 1657 ................................. 156 Location of observation wells at the central helipad study area. Drawdown histories at wells 1650, 1651, 1652, 1653, 1654, 1655, 1656, and 1657 are used for an analytical evaluation of pumping test results ........................ 158 Contour map of transmissivities. The distribution of transmissivities shows a channel passing through wells 1651 and 1654 ........................... 159 Contour map of storage coefficient values. The distribution of storage coefiicients shows a channel passing through wells 1651 and 1654 ........................................ 160 xiv fgmeDl3 CC hguan.l EgnEl hnmES Figzeii4 Fgure E5 Figure E6 Est-re E7 Figure 58 Fists 13.9 Figure E10 finnEJ] Here E12 EQREJ3 5. rig-TC E14 C0 n: U! r C0: Iou C01 ”F? Cor mid Con I0“: Con ”PW Com mid, Con, L IOVVQ COm. ”Chm COHh l(“Vet Figure D.12 Figure E] Figure E.2 Figure E.3 Figure E.4 Figure E.5 Figure E.6 Figure E.7 Figure E.8 Figure E.9 Figure E.10 Figure E.11 Figure E.12 Figure E.13 Figure E.14 Contour map of hydraulic conductivity values. The distribution of hydraulic conductivity shows a channel trending north and passing through well 1654 .................................................................................... 161 Contour isopach map of channel deposits (i. e. sand and gravel) for the Aa unit ........................................................................................................... 163 Contour isopach map of channel deposits (i.e. sand and gravel) within the upper portion of the Ab unit ..................................................................... 164 Contour isopach map of channel deposits (i.e. sand and gravel) within the lower portion of the Ab unit ..................................................................... 165 Contour isopach map of channel deposits (i.e. sand and gravel) within the B unit ........................................................................................................ 166 Contour isopach map of channel deposits (i.e. sand and gravel) within the upper portion of the C unit ....................................................................... 167 Contour isopach map of channel deposits (i.e. sand and gravel) within the lower portion of the C unit ....................................................................... 168 Contour isopach map of channel deposits (i.e. sand and gravel) within the upper portion of the D unit ....................................................................... 169 Contour isopach map of channel deposits (i.e. sand and gravel) within the middle portion of the D unit ..................................................................... 170 Contour isopach map of channel deposits (i.e. sand and gravel) within the lower portion of the D unit ....................................................................... 171 Contour isopach map of channel deposits (i.e. sand and gravel) within the upper portion of the E unit ....................................................................... 172 Contour isopach map of channel deposits (i.e. sand and gravel) within the middle portion of the E unit ..................................................................... 173 Contour isopach map of channel deposits (i.e. sand and gravel) within the lower portion of the E unit ....................................................................... 174 Contour isopach map of channel deposits (i.e. sand and gravel) within the upper portion of the F unit ....................................................................... 175 Contour isopach map of channel deposits (i.e. sand and gravel) within the lower portion of the F unit ....................................................................... 176 XV Fzgrre Fl VC - mt Fir-gar F3 FL- CJIN the 8:22.183 Sc;- pro. h} d CL‘AK Figure E4 Full app: to c3 FAélzre F5 Full app: to CI” Figure F6 Full ERIE F7 FUIF 1 Figure F .8 F. {in ., edCFIW .~ dra\\( Figure F.l Figure F2 Figure F3 Figure F.4 Figure F.5 Figure F6 Figure F.7 Figure F8 Figure H.1 Figure H.2 Vertical Markov chain model fit to measured facies data for the simulation modeling approach ................................................................................... 185 Full realization following methods for a TPG simulation approach. Four categories are used (gravel, sand, silty sand, and paleosol) to characterize the geology at the Helipad Study Site, LLNL LLNL .............................. 194 Separately modeled zones (C through F zones) are merged together to produce one final whole realization for the spatial distribution of hydrofacies at the Helipad Site. Image is presented in color ......................................................................................................... 195 Full realization (1 through 5) following methods for a layered simulation approach. Four categories are used (gravel, sand, silty sand, and paleosol) to characterize the geology at the Helipad Study Site, LLNL ................. 196 Full realization (6 through 10) following methods for a layered simulation approach. Four categories are used (gravel, sand, silty sand, and paleosol) to characterize the geology at the Helipad Study Site, LLNL ................. 197 Full realization (11 through 15) following methods for a layered simulation approach. F our categories are used (gravel, sand, silty sand, and paleosol) to characterize the geology at the Helipad Study Site, LLNL ....................................................................................................... 198 Full realization (16 through 20) following methods for a layered simulation approach. Four categories are used (gravel, sand, silty sand, and paleosol) to characterize the geology at the Helipad Study Site, LLNL ....................................................................................................... 199 Full realization (21 through 25) following methods for a layered simulation approach. Four categories are used (gravel, sand, silty sand, and paleosol) to characterize the geology at the Helipad Study Site, LLNL ....................................................................................................... 200 Observed drawdown histories for 1650 series wells and 1250 series wells are depicted for the Helipad Site (1650 series wells are screened within HSU 3, 1250 wells are Screen across HSU 4). As a result of pumping (1650 series wells) there is little to no observed drawdown in1250 series wells ......................................................................................................... 215 Simulated drawdown histories for 1650 series and 1250 series wells within the layered homogeneous conceptual model. Simulated results show minor drawdown in1250 series wells ................................................................. 216 xvi . Q “*1"; H 1 .LN‘ " s F igre 11.43 Figre HAb Figure H.521 Fisk's H.5b SF S r in; the Cs. TV I l m0t Figure H.3 Figure H.4a Figure H.4b Figure H.5a Figure H.5b Figure H.6a Figure H.6b Figure H.7a Figure H.7b Figure H.8a Figure H.8b Simulated drawdown histories for 1650 series and 1250 series wells within the layered homogeneous conceptual model with paleosol layers. Simulated results show minor drawdown in1250 series wells ................ 217 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 218 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 218 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 219 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 219 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 220 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map View is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 220 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 221 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 221 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 222 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th xvii Figre H93 Figure 8% Figure H. 1 03 Em H.108 Figure 11.1 Ia Figure H.111, . Em H13a F31 '99 ‘t‘w leb €133 H138 Sim the‘ TCSU Mari pro't mod $1131 Figure H.9a Figure H.9b Figure H.10a Figure H.10b Figure H.11a Figure H.11b Figure H.12a Figure H.12b Figure H.13a Figure H.13b model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 222 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 223 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 223 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 224 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24m model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 224 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 225 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 225 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 226 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 226 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 227 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th xviii 1 Fit 'eHl-la n E Figure H.148 . Figure H.153 p—fi ig‘re H.156 . Figure H.163 Sin: Figxe H.161) Mir; prof. 11100:;V Sim: FigreHna Simu the It FESUII FM l‘bns H.17b Kidp From 511311; 11:» . .UA’EI’H83 Slmu; the [n reSuli. F183;: Figure H.14a Figure H.14b Figure H.15a Figure H. 1 5b Figure H.16a Figure H.16b Figure H.17a Figure H.17b Figure H.18a Figure H.18b model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 227 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 228 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 228 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 229 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 229 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 230 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is fi'om the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 230 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 231 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 231 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 232 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24m xix Figure H.191» .\1.. Figure H.303 Sir. Figure H.301) M3- Figure H.313 Sin; Figure H.31b 813" 1 F1 1! We H.333 Figure H21 SIT. 93 51? the" [CJ Figure H.19a Figure H.19b Figure H.20a Figure H.20b Figure H.21a Figure H.21b Figure H.22a Figure H.22b Figure H.23a Figure H.23b model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 232 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 233 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24'h model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 233 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 234 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 234 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 235 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 235 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 236 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24‘h model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 236 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 237 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th , Figure H.-~la Figure H.341) Fieure H.353 Ewe H.373 Figure H.271, 583611.333 Figure 1.1ng | Sitar ‘- Sim. the trI resuj Ma , Prob... mom. Strazf: - _-_ — Slmt; I reSui: Map \ pitibd. Figure H.24a Figure H.24b Figure H.25a Figure H.25b Figure H.26a Figure H.26b Figure H.27a Figure H.27b Figure H.283 Figure H.28b 1 model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 237 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 238 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 238 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 239 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 2431 model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 239 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 240 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 240 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 241 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 241 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 242 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th xxi Figire H.393 Figre H.39b . . Figure H.29a Figure H.29b model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 242 Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells .................................. 243 Map view of the spatial distribution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) ..................................................................................... 243 xxii Chapter One Introduction and Scope of Work Introduction In an alluvial aquifer, the structural architecture of sediments is commonly very complex. For example, the architecture of alluvial fans may consist of multiple sequences separated by unconformable surfaces (Weissmann et al. 20023). Within individual sequences, channel sediments, which consist of highly permeable sand and gravel bodies, typically form the main aquifer. These channels while elongate are not necessarily laterally extensive or layered. Intermediate- and fine-grained materials, such as levee, crevasse splay or floodplain deposits, tend to occur laterally and vertically adjacent to the sand and gravel channel deposits. These fine-grained deposits typically consist of sands and silts with lower permeability. Like channel deposits, these materials are not necessarily laterally extensive or layered. Also at time of deposition, these fluvial deposits may be interspersed with debris flow deposits consisting of poorly sorted clay, silt, sand, and gravels. These debris flow deposits typically have lower permeability relative to other deposits due to poor sorting. Pedogenically-altered sediments may also develop, representing a prolonged hiatus in deposition. These pedogenically-altered sediments typically consist of fine-grained sand and clay, may be carbonate rich, and typically have very low permeability. Weissmann et al. (20023) described mature paleosol surfaces as laterally extensive units capping individual stratigraphic units. Though these units are laterally extensive, covering large portions of the fan system, it is possible that stream or rivers may cut these units creating conduits for fluid flow. Also, these units tend to vary in thickness over the alluvial fan. At the Lawrence Livermore National Laboratory (LLNL), California (Figure 1.1), the complex structural architecture of channel, floodplain, debris flow, and pedogenically-altered sediments, as well as the stratigraphic arrangement of these sediments, strongly influence the hydraulic response and distribution of contaminants within the alluvial fan aquifer (Blake et al., 1995; Blake et al., 1997; Noyes et al., 1997; Noyes et al., 2000). Remediation efforts at the LLNL site have led to an improved understanding of the subsurface conditions and contaminant distribution in the subsurface. Through a systematic methodology, the aquifer beneath the LLNL property limits and neighboring areas were divided into hydrostratigraphic units or HSUs (Blake et al., 1995; US. Department of Energy, 1998; Noyes et al., 2000; Figure 1.2). HSUs were developed by evaluating hydraulic tests, water levels, geophysical well logs, geological core descriptions, and chemical analysis of sediment and groundwater. HSU boundary delineation helped improve placement of extraction wells and implement a more cost effective clean up of site contaminants (U .S. Department of Energy, 1998). Figure 1.2 schematically shows the distribution of hydrostrati graphic units (delineated with black lines) in cross section across the southern portion of the LLNL property. Within HSUs, the locations of volatile organic chemical (V CC) plumes are reasonably resolved. It has also been shown that locally HSU boundaries correlate well to unconformity surfaces which are defined by laterally-extensive, pedogenically—altered sediments that mark possible sequence boundaries (Weissmann, 2001; Bennett and Weissmann, 2001). 1“ I‘\ g Figure 1.1: Lav central C aliibrr Department of ‘FJFE a. S “FL-Him Mr eh I It”: ' 0 v- 9318) 5L Seder Ilee a; 01020 Figure 1.1: Lawrence Livermore National Laboratory (LLNL) is located in west- central California approximately 50km east of San Francisco (modified from US Department of Energy, 1998). /f~‘-TT“““*”---33. 15> Nu Projection of Intersection of _, _ ——--~_-~q water table and HSU boundary , T’ “‘7 . /<££/ , . . I FF $4 / ._ \ .’ , “-l Cross section V I 2“" . ‘ ‘ 3: “ / ‘-.': /--~ r-—-.+’....-~ {I , . Iocatlon .;., /' (A: r _ a- “1:, / /. I ’NT [ha- ' HSU 1A M ‘ " e 9 If I“: g 1 A . a. 1 -- ‘ Legend (subsurface teetures) ‘50 lsoconoentretion r s " contour, ln ppb :33 HSU 5 Extraction well HSU 38 11 SU 6 HSU 7 5 “mm IMOWII \\ Total VOCs above MCLe "H—fim fl“ Elevated vadoee ‘6; zone totalVOCeoll Scale : Feet concentrations " " 100 to 999 ppb 0 500 1000 10.0pr I: 1t09ppb Figure 1.2: Schematic block diagram showing distributions of VOC contaminants within hydrostratigraphic units (HSUs) (Modified from US Department of Energy, 1998) Remediation efforts at the LLNL site have led to a better understanding of the subsurface and contaminant distributions. As a result, HSU boundaries have improved clean up efforts (U .S. Department of Energy, 1998). However, implementation of more detailed heterogeneities will most-likely provide a better description of HSU boundaries and the distribution of hydraulic properties within these boundaries. The need to more accurately simulate fluid flow has led to better heterogeneity characterization approaches through incorporation of more geologic information. Eschard et al., (1998) and Weissmann and Fogg, (1999) have used the geology and stratigraphy as a framework for geostatistical realizations in order to better define regions that are more closely stationary. Generating the spatial distribution of hydrofacies at this site, or heterogeneity within a geologic framework using transition probability geostatistics may provide a better understanding of flow within and across HSU boundaries. Resolving a detailed distribution of hydrofacies within a stratigraphic framework should also aid in remediation efforts at this site. This thesis explores two approaches to resolve the spatial distribution of the smaller scale heterogeneities, or more specifically, the spatial distribution of hydraulic properties. The first approach is an analytical solution analysis of pumping test results. The second method combines both stochastic realizations and numerical modeling approaches under different conceptual frameworks. Simulated drawdown is then compared to pumping tests for model validation. Analytical solutions, such as those first derived by Theis (193 5) are commonly used for evaluation and interpretation of pumping test results in confined aquifers. By applying these analytical methods, a spatial distribution of hydraulic properties and anisotropy in the aquifer may be estimated (Gloaguen et al., 2001). However, an approach of this type assumes (among other things) a homogeneous aquifer of constant thickness and infinite extent. There are few (if any) geologic processes that produce an aquifer with these qualities. Hantush and Jacob (1955) derived solutions for a leaky aquifer system, and Neumann and Witherspoon (1972) derived solutions for a multi- layered system. Kruseman and de Ridder (1990) present many equations that have been modified to accommodate confined, unconfined, leaky, anisotropic, and multi-layered aquifers. However, in order to apply each of these equations, assumptions must be made. The overriding assumptions of a homogeneous aquifer of constant thickness and infinite extent are still inaccurate within the context heterogeneous alluvial aquifer. Gravel and sand bodies that generally comprise the main aquifer are not of constant thickness or laterally extensive. In close proximity to the pumping well, a multi-layered conceptual model may be sufficient for evaluating pumping test with analytical solutions. Over a relatively short distance fi'om the pumping well, it may be assumed that the hydraulic properties within these layers are uniform in which assumptions for analytical solutions may be valid (Appendix E). However, evaluation of pumping test results at observation wells located farther away may be much more difficult. Between pumping and observation wells, there often exists a complex spatial pattern of drawdown which is primarily the result of physical heterogeneities that exist within an alluvial aquifer (Carle, 1996). Klingbeil et al. (1999) state that aquifer pumping tests yield hydraulic properties at a scale much larger than the typical length of structures in a heterogeneous aquifer. When characterizing an alluvial aquifer using analytical solutions, smaller scale features are often overlooked. More rec heterogeneiti' ”5 eta}- 3001; C0? drawdown 31 d: Statistical methot structures within influence on fluit‘: numerous pumpir obsen‘ed segment heterogeneous str. 85953213 to have generally uses (1:. 0bScl'vaiion wells. Ohm-311°“ POinis T0 better at mangraphic zone. . More recent methods are being developed to examine the impacts of local-scale heterogeneity using analytical solutions (Schad and Teutsch, 1994; Passinos, 2001; Zhan et al., 2001; Copty and Findikakis 2002). These methods include examining the transient drawdown at different phases (early, middle, and late) through semilog analysis and statistical methods. Interpretations of the transient drawdown are used to infer finer scale structures within the aquifer. These smaller scale features most likely have strong influence on fluid and contaminant migration. Hermann and Teutsch (1994) use numerous pumping tests to investigate different investigational scales. In this study, the observed segmentation of drawdown curves were use to interpret effecting lengths heterogeneous structures. However, they also conclude that numerical models are necessary to have a quantitative interpretation of measured drawdown. These approaches generally asses drawdown fiom a single pumping well or a pumping well with few observation wells. The work presented here describes an analysis of drawdown at several observation points. To better account for the smaller scale heterogeneities that are present within each stratigraphic zone, a different approach must be taken. Individual stratigraphic units, such as channel gravels and sands, floodplain deposits, and pedogenically—altered sediments, can be recognized by textures and bedding patterns that originated from the depositional processes. Furthermore, stratigraphic units of the same type tend to posses similar hydraulic properties. The term “hydrofacies” is used to refer to stratigraphic units with similar hydrogeologic properties (Ritzi et al., 1995; Klingbeil et al., 1999; Gaud et al., 2001). Within each individual sediment facies, or in this case, individual hydrofacies (e. g., sand, gravel, silty sand, and paleosol), the hydraulic properties may differ by 2 orders of magnitude. However, hydraulic properties between each hydrofacies may differ by 5 orders of magnitude or more (Freeze and Cherry, 1979). Resolving a detailed spatial distribution of hydrofacies requires geologic information. Most geologic information is obtained from boreholes. Core recovered from boreholes can be used to identify geologic structures and sediment facies. Geophysical logs are used to measure resistivity, conductivity, porosity, density, and other parameters. Geophysical logs and borehole descriptions can be used in unison to identify hydrofacies, which may include channel gravels and sands, floodplain sand and silts, and pedogenically—altered sediments within an alluvial fan setting. While borehole data provide excellent coverage in the vertical direction, architectural complexity and borehole spacing is generally too great to yield sufficient insight on sediment facies distributions in the horizontal direction. A stochastic approach (e.g., transition probability) geostatistics can be used to account for the distributions of sediment units in the horizontal direction (Carle and Fogg, 1996; Carle et al., 1998; Weissmann et al., 1999). Weissmann and Fogg (1999) and Weissmann et al. (2002b) use transitions probability geostatistics within a geologic framework to create geologically realistic realizations of alluvial aquifer systems that are conditioned to borehole information. Using this approach, they generated three-dimensional realizations of hydrofacies distributions that are geologically reasonable. Purpose of study The primary purpose for this work is to study the influence of stratigraphy on aquifer hydraulics. This is accomplished by first modeling the stratigraphy through a detailed £60103 conceptual of ti. simulated draw C 1 solution was use The Heli: selected for this : preliminary stud. probability geoste sell log data were aiiOWing for cross measured results \ contaminant rem e characterization m To inx'esti : Ieels of heter Oten abomogeneous-} FHLP), 3 ) 41) tra“$11101 “F 1998;1irpci at ”In“ 011;; (STPG) Inkii‘idual detailed geologic assessment, analyzing both geophysical and core data. Then, multiple conceptual of this stratigraphy are evaluated using numerical groundwater models were simulated drawdown can be compared to the observed. For completeness, an analytical solution was used to evaluate and interpret observed pumping test results. The Helipad Site at Lawrence Livermore National Laboratory (LLNL) was selected for this study. Selection of this site was based on the following criteria: 1) preliminary studies were conducted at this site that modeled the system using transition probability geostatistics, 2) a large amount of previously collected core and geophysical well log data were available at the site, 3) a dense spacing of wells at this site exists, allowing for cross-well correlation of facies, 4) a ten day pumping test conducted with measured results was available at this site, and 5) this site is currently undergoing contaminant remediation, thus development of a more detailed hydrologic subsurface characterization may further aid in LLNL remediation efforts. To investigate the influence of stratigraphy on aquifer hydraulics, four stratigraphic models were developed for the Helipad Site that incorporates different levels of heterogeneity. These models include 1) a homogeneous-layered model (HL), 2) a homogeneous-layered model incorporating relatively low permeable paleosol layers (HLP), 3) transition probability geostatistical simulations (Carle and Fogg 1996; Carle et al. 1998) (TPG), and 4) transition probability geostatistical simulation in a stratigraphic framework (STPG). This approach simulates the spatial distribution of hydrofacies for individual stratigraphic zones independently before merging them into a single realization (Weissmann and F ogg, 1999; Weissmann et al., 2002b). Evaluation of these models i were based on 3 (Harbaugh e! at litis document is 0 The first s of the slut: LLNL. Sp. remediatic- ' The secont conceptual site. This c Publication ' D16 third Sc Work Condu e I‘~()110\,‘.1-ngaE 0 App; 0 App: 0 AFDC Strait. O Amie disrm were based on a comparison of simulated pumping test results using Modflow-2000 (Harbaugh et al. 2000) to observed field results and utilized numerical modeling software. Thesis Outline This document is divided into three main sections with subsequent appendices: - The first section, covered by Chapter Two, describes the geology and hydrology of the study region. A brief review is given of previous hydrological studies at LLNL, specifically at the Helipad Site, and a brief history of contamination and remediation efforts at this site. 0 The second section, covered by Chapter Three, presents a comparison of the four conceptual models for success in modeling a long-term pumping test at the study site. This chapter also forms a draft of a manuscript that will be submitted for publication in Water Resources Research. 0 The third section, covered by Chapter Four, revisits conclusions reached for the work conducted for this thesis. 0 Following appendices are attached: 0 Appendix A contains well core descriptions used in site characterization. 0 Appendix B includes hydrofacies and geophysical log correlations. 0 Appendix C contains a collection of cross-sections showing well-to—well stratigraphic correlations. 0 Appendix D contains analytical solutions to estimate hydraulic property distributions at the Helipad Site. IL .\_ tea: me Appendix E contains a series of isopach maps of stream channels for each stratigraphic horizon. Isopach maps are centrally located around the central Helipad study area. Appendix F includes a brief description of the transition probability geostatistical approach which utilizes a Markov chain model to resolve a 3-D distribution of hydrofacies. This appendix also outlines the procedures taken in development of the transition probability geostatistical realizations for the study area. Appendix G describes generation of numerical groundwater models in which the four conceptual modeling approaches are evaluated. Appendix H is a complete record of drawdown results for the Helipad Study Site. This includes both observed and simulated drawdown results. Appendix I contains the FORTRAN codes used to transfer geostatistical realizations into a format that could be incorporated into numerical groundwater models. 10 The Heir; underlain by an .. lfe.g..si1t. sand. g distribution of 'Ji. —_—_ — oier relati\'€1}' 51"- irii‘iuerice of the a" assessment ofthis_ tiaraeierization \\ _ for ei'aluation. To accomp within the Liverm I investigation throu. Emlogy through a 1L Lll‘ertriore region. 7 Chapter Two Geologic Setting and Stratigraphic Assessment Introduction The Helipad Site at Lawrence Livermore National Laboratory (LLNL) is underlain by an alluvial aquifer where the complex structural architecture of sediments (e.g., silt, sand, gravel, etc.) strongly influences fluid flow. The heterogeneous distribution of these sediments produces a system where hydraulic properties vary greatly over relatively short distances. Since the main purpose of this thesis is to study the influence of the alluvial stratigraphy on aquifer hydraulics, a detailed geologic assessment of this site was made. This geologic assessment allows for a subsurface characterization where aspects of the geology can be incorporated into conceptual models for evaluation. To accomplish this, the regional and local geologic setting and the hydrology within the Livermore Basin was investigated. This chapter presents findings from this investigation through two main sections. The first section introduces the background geology through a description of the regional geology and the geologic evolution of the Livermore region. The second section outlines the local geology and hydrology observed at the Helipad Site. 11 Background Geology Regional Geologic Setting for the Livermore Valley The Livermore Valley is located within the California Coast Range Province. This province is composed of late Mesozoic through late Tertiary marine sedimentary rocks deposited on a complex basement of oceanic and continental crust (Figure 2.1). The Coast Range Province consists mainly of sub-parallel mountain ranges that are predominantly aligned with the majority of active faults in this region. While many of the faults are relatively minor, three major fault systems form the majority of tectonic history for western central California (Carpenter et al., 1984) —- the San Andreas, the Sur- Nacimiento, and the Coast Range Thrust. The regional alignment of geologic structures is a reflection of the deformation that has occurred. In this region, most basins have a north-northwest trending pattern. However, the Livermore Basin is an exception to this. It is an east-west trending structural basin that lies within the Diablo Range. The Livermore Valley is approximately 27 kilometers long (east to west) and 12 kilometers wide (north to south) (Figure 2.1). The Lawrence Livermore National Laboratory (LLNL) site is located in the southeastern portion of the Livermore Valley, near the base of the Diablo Mountain Range. The Livermore Basin has a relatively smooth valley floor surface. At the LLNL site, the basin floor gently dips approximately 1 degree to the northwest. 12 SAN EC;-. LI‘vERl Expalination LIVERMORE VALLEY Figure 2.1 Regional geologic map of western-central California. The Livermore Valley is outlined with a dashed line (Modified from USGS, 1966). Image is presented in color. Geologic Evolution and Stratigraphy of Middle California After the accretion of the Nevadan Orogeny (early Mesozoic time), subduction in central California shifted from the Sierra foothills 130 kilometers seaward to the present site of the Coast Ranges (Howard, 1979). At this new offshore subduction zone, the leading edge of the overriding plate was elevated to form a submarine ridge over the offshore oceanic plate (Figure 2.2a). Following these events a broad forearc basin developed landward where shallow marine and eroded sediments from the accreted orogeny started to accumulate giving rise to the Great Valley Sequence. This sequence is composed of stratified sedimentary elastic rocks that have generally experienced little deformation and mild burial metamorphism. During this same time (early to late Mesozoic), sediments were washed over the submarine ridge and deposited into a deep marine trench to the west, forming an accretionary prism at the leading edge of the submarine ridge (Hamilton, 1969; Ernst, 1970; Hsu, 1971) (Figure 2.2a). Sediments deposited here are known as the Franciscan Assemblage and are generally composed of greywacke, shale, chert, and limestone (Figure 2.3). By the early Cretaceous, sediments at the leading edge of the subducting oceanic plate had reached sufficient depth beneath the overriding continental plate that partial melt resulted. Magrrra rose to the surface resulting in widespread volcanic activity along the elevated ridge (Figure 2.2b). Volcanic rocks were deposited both east and west of this newly formed volcanic arc. By the late Cretaceous, continued subduction from the offshore oceanic plate resulted in crumpling and metamorphism of Franciscan sediments giving metagraywacke, argillite, blueschist, and greenstone rocks. Through continued deposition and subduction, 14 A) Magmatic Arc Figure 2.2: A) Generalized depiction of Late Jurassic to early Cretaceous landscape of western central California (modified from Howard, 1979). B) Generalized depiction of Late Cretaceous landscape of western central California (modified from Howard, 1979). 15 GENERALIZED COLUMNAR SECTION - TELSA OUADRANGLE Thickness Age Formation Column in n Description Aluvium Qal & '- . Ouatemary Terrace deposits . o.OtI . ? Gravels, sands. suits, clays. W‘ 3 ' 9" .° 100' + . Tulare and ' - Plio- Liverm ore ”9*" Continental deposits of gravels, Plotstooene gravels . TP-‘f , 4000. sands. clays. Unconformity~ ' ° ' Q - res-rs- 3 § :Ir‘s—Z 2000' + Shales, blue sandstone, tuffs. a Neroly a 1:: +.+ Upper o Miocene E 3 - Blue sandstone. andesitic conglamerates, g 3 . tuft. a) . . . Granuliferous white sands, bluff sands, 0‘3“” tufts, conglomerate. coal. Middle Oursan ? Tan sandstone. tuffaceous shales. Miocene .' : '. Bl d, white . l ' . Middle T I . . uff san sands cays (marine) E 9 $8 “8 . 2000' ocene -__-_:_-_-_ Bluff sands. chocolate shales, coal : 33:; (brackish-water). Moreno ' - Bluff sandstone t . siliceous. a illaoeous. Grande — Km 0.650 and sandy shale(s. ir)nestone concrrgetion. . . . - . sandstone beds. Conformity‘ . . ' . g ‘7“.— ‘ Upper 3' '. .- - ' - Concretionary and massive sandstone. Cretaceous s; Panache . .Kp :. 10,000' + argillaceous and silty shales. = . .' . ' ‘ conglamerate. g ..'.e . ‘6 ' '- , " °. 9 . e . .1 o ‘5 ’33—; .- Fault —"—' beggloeous Hoursetown Kh __ 0500‘ Dark shales, sandstones. Cbast Range Thrust Fault :——-,< Chart Sandstone. shales. chart lenses. ;. '. conglomerate. J ' Franc’ ”°’ P’llowb n amassm and rdfgtaend - - -‘ Jb ' asa . Explaination Cretaceous Intrusive rocks ° =, 15900.? Glaucophane schist. x Vertebrate fouls =‘..- k ‘ lmortabratefoeefls ._-.-_ . . serpentine. diabase. * Radiolere . -§—t~ - . diorite-gabbro. Foramlnifers i7- . fl Leaves .e .0 .e gs; M TU" Figure 2.3 Stratigraphic column for the Livermore Valley (modified from Huey, 1948) 16 younger sediments were wedged under the leading edge of the oveniding continental plate resulting in further elevation of this submarine ridge (Figure 2.2b). Also during this time, portions of the Franciscan Assemblage were forced beneath sediments of the Great Valley Sequence, the boundary between these units being a low angle thrust fault composed of mafic and ultramafic rocks (Howard 1979; Carpenter et al., 1984; Figure 2.3). However, the complex geology of this region makes it difficult to distinguish a depositional contact between the Franciscan Assemblage and the Great Valley Sequence deposits (Dickerson, 1965). Generally, the differentiation of these two formations is based on the juxtaposition of rock with different layering attitudes and degrees of metamorphism. During the late Cretaceous to early Miocene, the Kula Plate ceased subduction in this region and the leading edge of the continent was torn off by the northern movement of the Pacific seafloor (Howard, 1979). This event formed the proto-San Andreas Fault System. The northern movement of the Pacific seafloor eventually carried the actively subducting F arallon Plate into this region. At this time, faulting along the San Andreas ceased. Is was not until subduction of the Farallon Plate ceased that the present day San Andreas Fault System became active (early to late Miocene). At the close of the Cretaceous, there was a prolonged period of erosion of the folded western Great Valley Sequence over which tertiary sediments would be later deposited (Huey, 1948; Dibblee and Darrow, 1981). In the Great Valley, shallow marine sediments accumulated under conditions of fluctuating sea levels and differential subsidence of the basin floor. Today these sediments outcrop to the east in the Almont Hills (Figure 2.4) and to the west in along the East Bay Hills. Sediments deposited during 17 AL‘L, UpperUvemoreGnvele 4 River/Stream - One ’ LowerUvumeGr-vele N 3:... . m“ ”PM” w 2 == - Tertiary Norunerlne-Merine 0 I 2 Great Veley Sequence-Upper - Cretaceoue Freneieeen Aeeemblaoe Manolo-Cretaceous Figure 2.4 Generalized geologic map of the Lawrence Livermore National Laboratories and vicinity (Modified from Dibblee, 1980). Image is presented in color. this time conszr conglomerate. . 195.1). Early r. that accumulate- with Iii; 1h? depositional basin. From the dfjoosited in a n: T8555 SCdiments 3.3") eroded from my Diablo R, l‘CDuR 19.4,) a deformed Tmiar hemmed low In the 635 lots indicate 1113: this time consist primarily of sandstone and shale, with lesser amounts of tuff and conglomerate, and all show signs of significant deformation (Sweeny and Springer, 1981). Early Tertiary deposits also contained abundant amounts of volcanic sediments that accumulated in the forearc basin. With the reactivation of the San Andreas Fault System during the Pliocene time, the depositional setting environment transitioned from a forearc basin to a strike slip basin. From the Miocene through Holocene, lacustrine and fluvial sediments were deposited in a newly developed strike slip basin known today as the Livermore Basin. These sediments consist primarily of poorly consolidated coarse-grained debris (Figure 2.3) eroded from both the Franciscan Assemblage and the Great Valley Sequence of the nearby Diablo Range (Figure 2.4). Locally, these deposits exceed 1.2 km in thickness (CDWR, 1974) and were deposited in a series of alluvial fans that rest unconformably on deformed Tertiary rocks. These deposits exhibit minor folding and faulting, and are downwarped towards the valley center (Carpenter et al., 1980). In the eastern portion of the Livermore Valley (location of the LLNL site), well logs indicate that the alluvium is composed of interfingering gravel, sand, silt, and clay (CDWR, 1966). The upper Livermore formation ranges in thickness from a few meters to over 60 meters and it generally increases in thickness toward the west. From the complex heterogeneity of this system, sediment correlation from well logs is difficult over large distances. 19 Helipad Site Geology Core and Facies Descriptions For identification of the sediment architecture at the Helipad Site, several sources of geologic data were available. Subsurface data included geophysical logs from 32 wells with depths up to 61 meters and approximately 1220 meters of core recovered from 28 of these wells. Characterization of the Helipad Site involved a visual description of textural and sedimentologic characteristics in core. This included a visual estimation of grain size, shape, sorting and color, as well as description of sedimentary and pedogenic structures. In addition, geophysical logs were obtained from each well and were used for correlation of hydrofacies across boreholes. Four hydrofacies dominate the Helipad Site stratigraphic section (Appendix A). These include gravel and sand hydrofacies of channel deposits, sandy silt hydrofacies of overbank deposits, and paleosol hydrofacies facies. The gravel hydrofacies is typically dominated by Clast-supported gravels consisting of pebbles and cobbles. Deposits tend to be massive at a core scale. Clasts are commonly weathered and supported within a reddish, clay-rich matrix. Clay coats are common on clasts. Manganese oxides are also common on the clay coatings which may be an indication of a wetting and drying of sediments. This is also consistent with pedogenic alteration related to present and probable past climates. Gravel hydrofacies display the highest hydraulic conductivity. However, with the presence of thick clay coatings, hydraulic conductivities may be significantly reduced in relation to more typical gravel deposits. 20 The sand hydrofacies typically consists of well to moderately well sorted, fine to medium sand with interspersed layers of coarse to very coarse sand. Grains are typically subrounded to subangular. Sand deposits are generally massive and show possible signs of cross-stratification. Within some sections, clay coatings are present indicating a degree of pedogenic alteration. The silty sand hydrofacies typically consists of silt that contains amounts of fine to medium sand. This hydrofacies commonly displays some evidence of pedogenic alteration. This includes the presence of root traces that are commonly clay-filled and surrounded by reduction halos. The silty sand hydrofacies typically have thin clay coats, dispersed manganese oxides, and a brown coloration. These sediments are most likely floodplain deposits. Pedogenically-altered deposits or paleosol hydrofacies typically consists of clay- rich silty sand deposits. They show evidence of significant pedogenic alteration that includes very thick clay coats, continuous manganese oxide coats, slightly prismatic to prismatic or blocky structure, occasional root traces, and reddish coloration. Due to the processes involved in the pedogenic alteration, most root races are destroyed. Additionally, some of these deposits contain significant calcium carbonate. Comparison of Core to Geophysical Well Logs As previously stated, well core from 28 out of the 32 wells in the Helipad study were recovered. Recovered core from these wells were generally incomplete. Sections of core as a result of drilling difficulty, handling, and sampling were missing or had been shifted such that elevations were no longer aligned. Also, core recovered from these wells 21 were used for various site investigations resulting in removal for sampling purposes. To account for sections of missing core and to correct for elevations, geophysical logs were used. Multiple geophysical logs including gamma, conductivity, caliper, resistivity logs were recorded for each well within the study area. Gravel, sand, silty sand, and paleosol hydrofacies were correlated to geophysical logs (Appendix B). Weissmann (2001) observed that gravel, sand, and sitly sand hydrofacies character was easily distinguished in geophysical logs. The gravel hydrofacies is distinguished by relatively high resistivity and relatively low gamma ray signature. The sand hydrofacies is distinguished by relatively high resistivity and relatively low gamma ray signatures, where the resistivity signatures for the sand hydrofacies are typically less than those observed with the gravel hydrofacies. The silty sand or sandy silt hydrofacies are distinguished by relatively low resistivity and relatively high gamma ray signature. On geophysical logs, the paleosol hydrofacies character is difficult to distinguish from the silty sand hydrofacies since it shows a relatively low resistivity and relatively high gamma ray signature. However, the gamma ray signature for paleosol hydrofacies often displays a slightly higher signature than the silty sand hydrofacies. Correlations and Cross Sections The correlation of well core to the geophysical logs allowed for sections of core to be adjusted back to their correct elevation. Also, damaged or otherwise missing sections of core were interpreted to one of the four defined hydrofacies. These adjustments and interpretation together gave more complete facies descriptions in wells that could then be 22 used for stratigraphic correlation of hydrofacies between wells. Across the Helipad study site, fourteen cross sections were generated (Appendix C). Channel sand and gravel bodies are correlated with confidence across short distances. However, these bodies do not correlate across larger distances (Figure 2.5). The only unit that could be correlated between most wells with a degree of confidence was the paleosol hydrofacies. Since the paleosol hydrofacies are laterally extensive. In places where the paleosols are missing, gravel and sand hydrofacies exist. It is interpreted that stream channels eroded the paleosol surface at these location. There are however, a few instances where paleosol units are missing and the silty sand hydrofacies exist where paleosol is expected. This is interpreted to be a location where conditions were not adequate for development of a paleosol or where fine-grained sediment filled the channel. Paleosols identified in cross section appear to be laterally extensive across the entire model area in a manner similar to that described by Weissmann et al. (2002a). Using this reasoning, paleosols could be correlated throughout the Helipad study site. Using paleosol correlations from cross sections, the tops and bottoms of paleosol surfaces were generated with Rockworks 2002 (Rockware Inc., 2002) using an inverse distance linear algorithm (Figure 3.2). Paleosol units are informally named Ab trough F and paleosol-bounded stratigraphic units are informally named Ab through F following Weissmann (2001). Surfaces generated with this type of algorithm were deemed reasonable for the purpose of this study. These surfaces separate stratigraphic units used in evaluating the influence of heterogeneity on aquifer hydraulics (described in more detail in Chapter 3). Only those sections that lie below the water table were analyzed in 23 South North WELL 1655 WELL 1852 WELL 1657 Wl657 184 m ./ . wrssz , w1655 ”7'" fl- Gravel D- Sand C]- Sandy Slltl - Farmland West East WELL 1205 WELL 1 252 WELL 125° Figure 2.5: Stratigraphic cross sections at the Helipad Site. Top paleosol surfaces are shown (dashed line). Paleosols are informally named following Weissmann (2001). Image presented in color. 24 detail for this study. Weissmann (2001) indicated that there is a striking similarity between paleosols and hydrostratigraphic units and suggested that HSU boundaries should be updated accordingly. Hydrology in the Livermore Basin The Livermore Valley Groundwater Basin lies within the Diablo Range, and encompasses approximately 17,000 hectares. Prominent streams for this basin (all of which are ephemeral) include the Arroyo del Valle, Arroyo Las Positas, Arroyo Seco, Arroyo Mocho, Alamo Creek, South San Ramon Creel, and Tassajara Creek. The Arroyo del Valle and Arroyo Mocho are the largest of these streams and drain the largest area. All of these streams flow north and westward toward the valley floor until converging with the Arroyo de la Laguna, which flows southward out of the Livermore Valley into the Sunol Valley Groundwater Basin (Figure 2.6). At the LLNL site, natural drainages have been altered by continuing construction activities. As a result, the current location of the Arroyo Seco and the Arroyo Las Positas streams are not representative of historical paths. These stream channels merge about 1.6 kilometers to the west of the LLNL property before continuing toward the Livermore Basin and merging with the Arroyo Macho. Historical (1940) aerial photographs of the Livermore property show locations of old stream channels for the Arroyo Seco and Arroyo Las Positas (Figure 2.7). Currently, Arroyo Seco is located in the southwest corner of the property. The Arroyo Las Positas was redirected north then west along the property line. Flood control andprevention of water infiltration were the main goals of the stream channel redirection. 25 Tr; 1' In. "-'.--I" - ~ Livermore Va .“~.-I-.-..' Figure 2.6: Illustrated in this Figure are major streams within the Livermore Valley. Streams are marked by dashed lines; the position of the LLNL property is outlined with a box and labeled. 26 h,- Old Arroyo Seco Channels ' Las Positas ‘ Creek . W . . Arroyo Seco Channel 4” '/4 mile Figure 2.7: Aerial photograph (from 6/8/1940) of the LLNL property area. Indicated are northwest trending river channels of the Las Positas and the Arroyo Seco. Also indicated is the more recent location of the Arroyo Seco stream channel. 27 Thel as a sequence system from I toward the CC‘. the Sunol Va'. l'pper and Let be an average 1995). Storage alluxium (L'pp: lhlclmess. H0“ less. The L'ppe WUDCable (Sat groundwater 5C The Livermore Valley ground water system is described by Harrach et al. (1995) as a sequence of semi-confined aquifers. Groundwater moves down gradient through the system from the valley perimeter (Franciscan and Great Valley Sequence Formations) toward the central valley and then flows westward toward the southwest of the basin into the Sunol Valley Basin. However, since 1945, heavy water usage from the Livermore Basin has eliminated subsurface flow out of the Livermore Basin. The groundwater system in this basin is composed of two main aquifers —- the Upper and Lower Members of the Livermore Formation (Figure 2.3). The lower member has an average thickness of 1,000 meters over an area of 250 kilometers (Harrach et al., 1995). Storage capacity for this aquifer is significantly greater then the overlying alluvium (Upper Livermore Formation), which is on average only 100 meters in thickness. However, hydraulic conductivity for the Lower Livermore Formation is much less. The Upper Livermore Formation is less consolidated and therefore is more permeable (San Francisco RWQCB, 1982). Consequently, this alluvium is the principal groundwater source within the Livermore Valley. The depth to groundwater varies from about 10 meters to about 45 meters beneath LLNL (Carpenter et al., 1984; Thorp et al, 1990). Prior to major remediation project at this site, groundwater flow was generally from southeast to northwest, however, current groundwater gradients have been significantly altered partially as a result of remedial pump and treat practices at the LLNL site. Figure 2.8 shows the groundwater gradient across the Livermore property in 1997. At this location, gradients are greatest in the northeast comer at about 0.15 meter/meter and decreases toward the southwest at a gradient of 0.002 meter/meter (Harrach et al., 1995). 28 Figure 2.8: Water table elevations beneath the LLNL property, recorded in 1997 (Modified from Harrach, 1998). 29 Chapter Three Pumping Test Evaluation in a Heterogeneous Alluvial Aquifer Wote: This chapter forms a manuscript that will be submitted to Water Resources Research and was authored by T rahan, R.S., Phanikumar MS.,, Hyndman, D. W., and Weissmann, G.S., Dept. of Geological Sciences, Michigan State University, East Lansing, MI, 48824) Abstract: Pumping test results at the Helipad Site, Lawrence Livermore National Laboratory (LLNL), are used to evaluate four conceptual models. Each of these models incorporates different aspects of the physical heterogeneity observed. At this site, relatively mature paleosols within the alluvial deposits are observed. These paleosols mark unconfonnities that separate stratigraphic zones within this aquifer. Top and bottom surfaces for each paleosol unit are identified in geophysical well logs and core. This provides the stratigraphic framework in which conceptual models are developed. By resolving heterogeneity at the hydrofacies level within each stratigraphic zone and paleosol unit (simulated using transition probability geostatistics, conditioned to geophysical well log and core data), a better match pumping to test results was possible relative to conceptual modeling approaches that do not resolve the heterogeneity at the hydrofacies level. This study also shows that analytical solutions for heterogeneous systems do not accurately capture the observed geology for this site. Introduction In an alluvial aquifer, the structural architecture of sediments is generally very 30 complex. In these types of systems, the distribution of hydrofacies (e.g., silt, sand, gravel, etc.) will strongly affect fluid flow through an aquifer. The heterogeneous distribution of these hydrofacies produces a system where hydraulic conductivity and storage coefficient values can vary by orders of magnitude over only a few meters. Several examples show how the spatial distribution of these hydraulic properties greatly influences the subsurface migration of fluids (Freyberg, 1986; Sudicky, 1986; Ritzi et al., 1994; Carle, 1996; Dominic and Ritzi, 1996; F ogg et al., 1998; Zhang et al., 2000; Weissmann et al., 2002b Biteman et al., in press; Weissmann et al., in press). These study sites show that subsurface heterogeneities strongly influence fluid migration. Therefore, accurate prediction of fluid flow and solute transport requires reliable and realistic subsurface characterization (Webb and Davis, 1998; Boutt et al., 2001). There are many approaches to characterize subsurface. Commonly, analytical or geostatistical methods are employed. In evaluating pumping test results, analytical methods may be used to solve for average hydraulic conductivity (K) and storage coefficient (88) values between well pairs. This is accomplished through Theis curve matching (or some variation on this analytical solution) of drawdown histories. However, interpretation of these solutions with respect to the distributions of hydrofacies within a system is difficult. Using analytical solutions to interpret transient drawdown histories, Schad and Teutsch (1994) and Copty and Findikakis (2002) examined the impacts of local-scale heterogeneity in a larger scale system. These studies attempt to use transient drawdown histories to infer finer scale structures within the aquifer than conventional analytical solutions which only provide effective parameters for a homogeneous aquifer. Geostatistical methods are also commonly used to develop a spatial distribution of 31 hydraulic conductivity (K) and storage coefficient (Ss) values. These simulations of the heterogeneity may be incorporated into numerical groundwater models for evaluation. Teles et al., (2002) use a combination of semi-empirical and statistical approaches to roughly reproduce a heterogeneous aquifer. Froukh (2002) and Yang et al., (2002) develop conceptual models based on large-scale geologic features (i.e., formation contacts) where reasonable hydraulic parameters based on composition are assigned. These large-scale models of the system heterogeneity are then calibrated to better match observed data (i.e., head distributions, geochemistry, etc.). However, common approaches for defining the structural architecture of an aquifer system often tend to overlook smaller scale geologic features. These smaller scale features may have a significant influence on fluid migration. Also, methods for generating reasonable geologic structures such as process-imitating methods cannot be conditioned to geologic data (Koltermann and Gorelick, 1996). Klingbeil et al. (1999) state that aquifer pumping tests yield hydraulic properties at a scale much larger than the typical length of structures. Likewise, analytical solutions for pumping tests estimate effective system parameters in which smaller scale structures are not well resolved. In this paper, we explore the influence of heterogeneity on aquifer hydraulics. We present a comparison of four different conceptual models of the geology from a simple homogeneous-layered model to a more complex model of the stratigraphy and facies. Also, we evaluate analytical solutions for measured drawdown histories. Numerical groundwater models are used to evaluate each conceptual model through a comparison of pumping test results. To generate more complex models of the spatial distributions of hydraulic parameters, a transition probability geostatistical approach within a 32 stratigraphic framework (Carl and Fogg, 1996; Carle et al., 1998; Weissmann and Fogg, 1999; Weissmann et al., 2002b). Conceptual models of the geology include a homogeneous-layered model (HL), a homogeneous-layered model incorporating relatively low permeable paleosol layers (HLP), a filll block volume transition probability geostatistical simulation (TPG) following methods of Carle et al. (1998), and a stratigraphic transition probability geostatistical simulation (STPG) following methods of Weissmann and Fogg (1999). Study Area The Helipad Study Site at Lawrence Livermore National Laboratory (LLNL) (Figure 3.1) was selected for this study because of the large amount of previously collected core and geophysical well log data that was available. There was also a dense spacing of wells that allowed for correlation of faces across boreholes. A long term pumping test conducted with measured results was available. Also, since this site is currently undergoing contaminant remediation, development of a more detailed hydrologic subsurface characterization may further aid in these efforts. The Helipad Site is regionally located along the southeastern margin of the Livermore Valley Basin within the Coast Range Province of California, approximately 55 kilometers east of San Francisco. As a result of weathering, eroded debris from the Diablo Mountain Range has produced a thick succession (over 1,000 meters) of unconsolidated Quaternary alluvial-fan deposits. Our study focuses on the upper 70 meters of a multi-layered alluvial fan deposit. The Helipad Study Site has an average 33 Livermore Valley San Francisco 50km Figure 3.1: Lawrence Livermore National Laboratory (LLNL) is located east of San Francisco, within the coast range province of California (modified from US Department of Energy, 1998). surface elevation of 190 meters (MSL) that gently slopes to the northwest. Locally the water table is approximately 29 meters below the ground surface and grades toward the southeast. Several plumes are recognized at the LLNL site, generally bounded within hydrostratigraphic units (HSUs) (Blake et al., 1995, Noyes et al., 2000). HSUs were defined on the basis of an interconnected network of relatively highly permeable deposits (i.e., sands and gravels) with lower permeable sediments (i.e., silts and clays) set around those deposits. Boundaries separating HSUs were defined as laterally correlating low permeable zones that significantly reduced vertical flow. For remediation of local plumes, the Helipad Site was developed for an electro-osmosis remediation project. This project goal was to locally remediate VOC contaminants within HSUs 3A and 3B. At this site nine closely spaced wells were used to facilitate this remediation plan (1550 and 1650 series wells). 34 Site Geology and Stratigraphy For development of the conceptual models, several sources of geologic data were available. Subsurface data at the Helipad Site included geophysical logs from 32 wells with depths up to 61 meters and approximately 1220 meters of core recovered from 28 of these wells. Subsurface characterization of the Helipad Site involved a visual description and evaluation of both textural and sedimentologic characteristics in core. Based on these descriptions (Appendix A; Weissmann, 2001), four hydrofacies were defined for this study area. The gravel hydrofacies has a relatively high permeability and is generally composed of coarse-grained, clast-supported gravel with thick reddish clay coats on clasts. The sand hydrofacies has a relatively moderate permeability and is composed of a moderately sorted, fine to medium-grained sand. The silty sand hydrofacies has a relatively low permeability and is predominately composed of silt containing minor fine to medium-grained sand. The paleosol hydrofacies has a relatively very low permeability and is composed of pedogenically—altered soils that are clay rich and generally contain significant calcium carbonate. It was observed that the gravel, sand, and sitly sand hydrofacies character was easily distinguished in geophysical logs (Weissmann, 2001; Appendix B). With the correlation of hydrofacies from well core to geophysical logs, sections of core were adjusted to the depth reported on the geophysical logs. Damaged or otherwise missing sections of core were interpreted over short intervals based on the geophysical logs. These adjustments and interpretations together provided a more continuous hydrofacies log that could then be used for stratigraphic correlation of hydrofacies between wells across the Helipad Study Site. 35 Fourteen cross sections were created for the Helipad Site correlating gravel, sand, and paleosol hydrofacies across boreholes (Appendix C). However, the paleosol hydrofacies was the only unit that could be correlated between wells with any degree of confidence. This was because paleosol units were observed to be laterally extensive representing periods of fan exposure (Weissmann et al., 2002a). These paleosol units separated the alluvial fan into a series of stratigraphic units which provided the geologic framework used for development of conceptual models. In cross section, paleosol units also closely correlate to hydrostratigraphic units defined by LLNL. However, paleosol units may not represent the only boundaries between HSUs. Based on these observations, a three-dimensional model of the stratigraphy was developed (Figure 3.2). Top and bottom surfaces of paleosol units were interpolated using an inverse distance algorithm. Both paleosol and stratigraphic units are informally named Ab trough F (Weissmann, 2001). Only those units that lie below the water table (C through F) were used for conceptual model development. Pumping Test Data For implementation of the electro-osmosis remediation project at the Helipad Site, eight wells (1650, 1651, 1652, 1653, 1654, 1655, 1656, and 1657) were arranged in a square array and spaced approximately 8 meters apart (Figure 3.3). All eight wells were screened across HSUs 3A and 3B approximately 30 to 39 meters below the ground surface. Continuous pumping from wells 1551 and 1552 (Figure 3.3) was conducted over a ten day interval during which groundwater elevations from surrounding wells were recorded. Observations from well 1651 were rejected due to a suspected poor well 36 Stratigraphy fl Ab paleosol i - Ab zone I Lower Livermore base Figure 3.2: Shown above is an illustrated conceptual model of the Helipad Study Site. Six paleosol surfaces (colored) have been identified which are used to separate depositional sequences (transparent) (Appendix C). Also indicated is the saturated zone (translucent blue shading). Image presented in color. 37 construction or poor annular seal. Monthly groundwater elevations for this well reflect monthly groundwater elevations for HSU 4 which is approximately 3 meter less than measured in HSU 3A and 3B. Drawdown histories from observation wells in the 1650 series wells are grouped into three different clusters (Figure 3.4). After one day, the hydraulic response from wells 1650, 1656, and 1653 show an average drawdown of 1.9 meters, well 1652 shows a drawdown of 1.4 meters, and wells 1651, 1655, and 1657 shows an average drawdown of 0.97 meters. These drawdown histories strongly correlate to the facies distributions at the site (Figure 3.5a). Central helipad cross sections indicate a gravel channel passing through those wells with a larger drawdown. The gravel hydrofacies has a relatively high permeability which may account for this larger drawdown. This gravel channel transitions into a sand channel that thins to the east of wells 1650, 1656, and 1653 (Appendix C). The sand hydrofacies has a moderate permeability which may account for the lower drawdown observed. For those wells screened outside of HSUs 3A and 3B (1250 series wells), there is little to no drawdown recorded (Figure 3.4). An analytical solution based on observed drawdown in 1650 series wells was developed (Appendix D). For solutions of hydraulic conductivity, transmissivity, and storage coefficients, individual drawdown histories for 1650 series wells were matched to leaky aquifer type curves (Hantush and Jacob, 1955; Appendix D). For these solutions, it was assumed that pumping was only from the 1551 well. This assumption was made because pumping from well 1551 was six times greater than from well 1552. Here the effects of pumping from well 1552 were assumed to be negligible. Likewise, a type curve matching approach is unable to perform a simultaneous analysis for drawdown. 38 HELIPAD \N-1251 KEY O- CORED WELL .- NOT coaeo c, SCALE (um; so O minim.) .5 Figure 3.3: Well and cross section locations within the Helipad Site study area. Observed Drawdown . Observed Drawdown Helipad Site Helipad Site 2.5 0.25 , 7 2 0.2 » _ E15 E 0.15 E 3 s ‘0 5 1 E 0.1 5 o 0.5 0.05 .. o o " l 1 l 0,2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Time (day) Time day —OBSw1250 - —OBSw1255 —t—OBSw1655 -— OBSw1251 --O- OBSMGGO -a— 068th - - OBS W1252 - -B— oesmeez -O- 085 W1657 ----- OBS W1253 --<)-- OBS W1663 ----- OBS W12“ --)(-- OBS W1664 Figure 3.4: Observed drawdown history for 1650 (left) and 1250 (right) series wells. Three different clusters of drawdown are observed due to the physical heterogeneity of the alluvial system. 39 A) LLNL Helipad Site Well Locations and Channel Distribution Fl. Trahan 6/2003 M2500 KEY 0 - CORED WELL 0 - NOT CORED SCALE 0 (but) as ll (me) 71 s LLNL Helipad Site Analytical Solution Hydraulic Conductivity (K) (m/day) Fl. Trahan 6/2003 KEY ° ' CORED WELL . ' NOT CORED CI 3 1 SCALE ('0'!) a o (m) u , ’ \ o w1657 ‘ \ ‘ \wissr. O t \ W1550°\ i \ \ , , w1652. \\ \ \ 1 \ \ w 552° \\ \ w1858. \ x \ \ w1655° ‘ , O \\ \31654 ‘ mesa \ . \ \ l \ o ‘\ x x \ I \ w1551 \ \ I [:I Gravel Hydrofacies (K=30m/d) E] Sand Hydrofacies (K=2m/d) Figure 3.5: A) Well locations and interpreted distribution for gravel and sand hydrofacies within the E3 channel. Pumping wells are indicated with black circles. B) Analytical solution to observed drawdown histories at the Helipad Site showing a contour map of hydraulic conductivity, values are measured in m/s. 40 Superposition may be used to account for the effects of multiple pumping wells. However, even with this solution, there are many assumptions about the aquifer that must be made. In a heterogeneous system, many of these assumptions are invalid. Results for the analytical solutions are placed at the observation wells and contoured. Since the 1250 series wells are screened across HSU 4 and out of the stratigraphic unit of interest, analytical solutions were not evaluated for these wells. The solution for hydraulic conductivity indicated a zone of higher values passing through wells 1655, 1552, and 1651 within the central Helipad study area (Figures 3.5b). Transmissivity and storage coefficient values are indicated in Appendix D. Distribution of hydraulic conductivities are contrary to the observed geology (Figure 3.5a) which shows a higher permeability gravel channel passing through wells 1650, 1656, and 1653. At this location, the geology correlates well to observed drawdown, however the analytical solution for this site does not. Modeling Pumping Tests The Conceptual Models: Four conceptual models that incorporate elements of the physical heterogeneity that exists within the alluvial fan system were developed for the Helipad Site (Figure 3.6a-d). For evaluation of each conceptual model, drawdown histories were simulated using Modflow-2000 (Harbaugh et al., 2000; Hill et al., 2000). These simulated results should reflect the observed differential hydraulic responses. The first conceptual model is a homogeneous-layered model (HL) (Figure 3.6a). Stratigraphic sections are separated according to top paleosol surfaces, with a geologic model that resembles the HSU model 41 A) B) Homogeneous-Layered Homogeneous-Layered Model (HL) Model with Paleosols (HLP) I D paleosol I F paleosol I D zone I F zone I E paleosol I Lower Livermore D E zone C) Transition Probability Geostatistical Stratigraphic Transition Simulation (T PG) Probablltly Geostatistical Simulation (STPG) I gravel D sand I sandy silt I paleosol Figure 3.6: A) Homogeneous-layered model. B) Homogeneous-layered model incorporating paleosol layers. C) Transition probability geostatistical simulation. D) Stratigraphic transition probability geostatistical simulation (stratigraphic zones are simulated separately and later merged). Image presented in color. 42 used by LLNL. The second conceptual model is a modified version of the HSU model in which solid, low hydraulically conductive paleosol units (C through F) are incorporated between homogeneous stratigraphic units (Figure 3.6b). The third conceptual model uses transition probability geostatistics to generate the spatial distribution of hydrofacies throughout the model domain (Carle, 1996; Carle and Fogg, 1996; Carle et al., 1998; Figure 3.6c). This conceptual approach incorporates more geologic information in terms of average channel orientation and widths, where each of the individual stratigraphic units was simulated simultaneously (Appendix E and F). The fourth conceptual model incorporates the most geologic information and uses transition probability geostatistics using a Markov chain model like the third. However, individual stratigraphic units and paleosol units are simulated individually and later merged to produce one final realization of the Helipad study area (Weissmann and Fogg, 1999; Weissmann et al., 1999; Appendix F; Figure 3.6d). Isopach maps for channel sands and gravels indicate that widths and orientations vary between stratigraphic units (Appendix E). This approach better preserved these characteristics by assigning mean channel widths and orientations to individual simulations from their corresponding stratigraphic unit. Merging each of these simulations results in a realization where the alignment of sand and gravel hydrofacies is different across stratigraphic zones. This approach is time consuming and requires significant geologic data. However realizations generated are more geologically realistic and accurate then approaches that do not incorporate this level of detail. 43 Building Conceptual Models: The layered stratigraphic conceptual model was developed following the HSU model developed by LLNL. Since HSUs were well correlated with paleosol units, these paleosol units were used as the basis to separate stratigraphic units. These surfaces were modeled across the study area with an inverse distance linear algorithm conditioned to the top of correlated paleosol units. Individual stratigraphic units were assigned values of hydraulic conductivity (K), specific storage (Ss), and vertical and horizontal anisotropy based from analytical solutions (Table 3.1, Appendix G). Vertical anisotropy values were assigned to contain drawdown within unit E (pumping layer), while horizontal anisotropy values were assigned to induce an elliptical drawdown cone to better match observed results. Values of K and Ss were calibrated for the E zone to better match the observed drawdown data. The second conceptual model is similar to the first. However, paleosol units with relatively low K were incorporated between stratigraphic units. Bottom paleosol surfaces were also generated using an inverse distance linear algorithm conditioned to the observed bottom of paleosol units at each well. Paleosol units were assigned to have a minimum thickness of 0.5m (vertical model resolution). Individual stratigraphic units were assigned the same initial values of K and Ss as in the first model. In this conceptual model, paleosol layers which had a relatively low permeability assigned to them, limiting fluid flow across stratigraphic units (Table 3.1). Therefore, vertical anisotropy values were assigned to one. Horizontal anisotropy values were assigned to induce an elliptical drawdown cone which provides a better match to observed results. Values of K and Ss 44 were calibrated for the E unit to better match observed drawdown (Table 3.1, Appendix G). Following the methods of Carle (1996), Carle and Fogg (1996), and Carle et al., (1998), the third conceptual model was generated with transition probability geostatistics, or T-PROGS (Carle, 1999). Imbedded transition probabilities for the horizontal dimensions incorporated mean channel lengths, widths and orientations from the E unit (Appendix E). This simulation was generated using a 2m x 2m x0.5m grid which was later refined around pumping wells to match the Modflow grid. In this type of conceptual model, the spatial distribution of hydrofacies accounts for any anisotropy in the system. Individual cells are assumed to be isotropic. One realization using this approach was generated. The fourth conceptual model was developed in the same manor as the third, however, geologic information in the form of mean channel lengths, widths, and channel orientations based off channel isopach maps corresponding to each individual stratigraphic unit were used (Appendix E). Also, this approach incorporated paleosol units. Hydraulic parameters were then assigned to each hydrofacies (Table 3.1, Appendix G). Multiple realizations for the distribution of hydrofacies were produced for this model type (Appendix F). Hydraulic properties from one random realization was then calibrated, these calibrated values of hydraulic parameters were used to simulate drawdown for each realization generated with transition probability geostatistics. Twenty-five realizations using this approach were generated. 45 Homogeneous-Layered Model: Layers Hydraulic conductrvrty Storage Coefficient (l/m) (In/d) C zone 1 0.00018 D zone 1 0.00021 E zone 0.25 0.00075 F zone 1 0.0002 Lower Livermore 1 0.0003 Homgeneous-Layered Model with Paleosols: Hydraulic conductivity Layers (m/d) Storage Coefficient (l/m) Paleosol layers 0.000046 0.0007 C zone 1 0.00018 D zone 1 0.00021 E zone 0.5 0.00005 F zone 0.0002 Lower Livermore 0.0003 Transition Probability Geostatistical Simulation: Hydrofacies Hydraulr(cr:/(:ir;ductrvrty Storage Coefficient (l/m) Gravel 30 0.00005 Sand 2 0.0002 Silty Sand 0.0009 0.0003 Paleosol 0.000046 0.0007 Lower Livermore 0.01728 0.0003 Table 3.1: Listed above are hydraulic parameters assigned to either layers or hydrofacies used in each of the modeling approaches. Listed values have been calibrated to best match observed drawdown histories. 46 Groundwater Modeling: Groundwater modeling software used for this study to calculate head distributions for both steady and transient states was MODFLOW-ZOOO (Harbaugh et al., 2000; Hill et al., 2000). The groundwater model consisted a 360m by 310m by 39m grid (casting, northing, and vertical respectively). Grid cells have a maximum horizontal dimension of 2m by 2m and are refined around the pumping wells (1551 and 15 52) to a minimum cell size of 0. 1 5m by 0.15m. Eastern and western boundaries were set as constant head based on a measured water table; northern and southern boundaries are perpendicular to flow and were assigned as no flow boundaries. Drawdown was observed from the 1650 series wells and pumping was from wells 1551 and 1552 at rates of 1634de and 2.72m3/d respectfully. Pumping wells are screened across HSU 3 which correlates to the E stratigraphic unit (~28m to 38m depth below the ground surface). Results from simulations are illustrated in Figures 3.7 and 3.8. Results and Discussion The simulated groundwater drawdown results for the Helipad Study Site shows that the conceptual models that incorporate the most geologic information (i.e., stratigraphic transition probability geostatistical simulations) best matched the observed drawdown histories for most wells (1650, 1653, 1654, 1655, and 1656) (Figure 3.7). However not all twenty five STPG realizations accurately reflected the observed drawdown histories, (Appendix H). The sum of squared residuals for the first STPG simulation was 3.48 (calibrated results), compared to the HL, HLP, and TPG models which had squared residual sums of 8.45, 7.95, and 43.64 respectively. The TPG 47 I" or 0 !° 01 Drawdown (m) N 9‘ O Drawdown (rn) __T_-__.__. O O Drawdown (m) 0 "Time(d"ay)ii' ’1' .L. . .. .. . 4 I ...... o --------------- O """""""" I '0. 1 mass. —— .d-- — -ooagsuo>a-O-o-o-n--.---- --.-co > 1 Time (day) i 1 Drawdown '(m) O w-1650 Viv-1651 w-1657 O' O O w-1656‘ W-l552 xiv-1652 a o c g, = -1 w-1653., - .w 655 W"654 O w-1551 2.5 1 i“ 01 O N tn Drawdown (m) Drawdown (m) Drawdown (m) “”552 ., .. a” """" ‘T’i‘ma‘rciayi' " “ 1 —- Observed - .— Homogeneous Layered — I- Homogeneous La ered with Paleosols - +-Transition Probabi ity Geostatistics .. + - Stratigraphic Transition Probability Geostatistics Figure 3.7: Drawdown results for modeling approaches are plotted for each observation well. Positions of plots are similar to there locations at the Helipad Site (a representation of well locations are illustrated in the center of this figure). For each graph, drawdown is in meters (y-axis) and time is in days (x-axis). Pumping wells are 1551 and 1552. Observed drawdown for well 1651 is not included due to poor well construction. 48 simulation used calibrated parameters from the calibrated STPG realization. Both the HL and HLP models were calibrated individually. The HL, HLP, TPG conceptual modeling approaches do not adequately capture the heterogeneous alluvial fan aquifer response (Figure 3.7 and 3.8). The hydraulic response of the aquifer produced three clusters of drawdown, wells 1650, 1656, and 1653 had a high drawdown, well 1652 had a moderate drawdown, and wells 1651, 1655, and 1657 had little drawdown (Figure 3.4 and 3.7). The calibrated STPG realization model shows two main clusters of drawdown where individual well drawdown is similar to the observed values. This characteristic is also observed in other STPG simulations. Simulated drawdown from well 1652 is high. North and south of this observation well, there are low conductive paleosol hydrofacies in the STPG realization. The close proximity of these paleosols around well 1652 creates a local zone of increased drawdown. This may account for the higher simulated drawdown. It is also possible that hydrofacies parameters need to be further calibrated to better match the observed drawdown. Simulated drawdown from conceptual models is presented in cross section and in map view in Figure 3.8. In cross section, each conceptual model is adequate for containing drawdown within the main pumping stratigraphic layer (unit E). With the first conceptual model type, vertical anisotropy was responsible for containing drawdown. In the other conceptual model types, the low hydraulic conductivity of paleosol units was responsible for containing drawdown. However, in map view, only those conceptual models which resolved heterogeneity at the hydrofacies level showed a drawdown cone that was similar to observed data (Figure 3.8c-d and 3.9). In these simulations the 49 Figure 3.8: Cross section and map views of drawdown propagation at time of 0.15 day A) Simulated drawdown for the homogeneous-layered conceptual model. B) Simulated drawdown for the homogeneous-layered model incorporating paleosol layers. C) Simulated drawdown for the transition probability geostatistics conceptual model. D) Simulated drawdown for the stratigraphic transition probability geostatistics conceptual model. Image presented in color. 50 [17‘- \ Drawdown I" \\ o \ o ‘ O \ I i. 1.1 \ \o.4 0'05") I | \ \ H o O O 0.5-1m i‘1.1’\ \oe] . \\ ' \.l .i _ 1-1.5m \ (SC-H \o.5l i] \ \‘ \ l l +1.5m \ \\ l l i \ \\ O‘ \ \ \\ \ l \ \‘ ll \ ‘— ‘ .1 Figure 3.9: Contour map of measured drawdown at a time of 0.15 day. Drawdown values for 1650 series observation wells are indicated. drawdown eclipse was aligned with wells 1650, 1656, and 1653. The drawdown ellipses for the HL and HLP models were aligned with well 1651, 1552, 1654, and 1551. These models were unable to adequately capture the heterogeneous hydraulic response of the system. Simulated drawdown for the 1650 series wells for the STPG simulation were calibrated to observed drawdown by adjusting the hydraulic properties of hydrofacies. Calibration of hydraulic properties resulted in drawdown curves that are similar to the observed. However, simulated drawdown from twenty-five realizations for this conceptual model type had variable results (Appendix H). Five of these realizations preformed well more closely matching the observed drawdown. Eight realizations preformed moderately, and twelve realizations preformed poorly. Calibration of individual simulations may produce better simulated drawdown that more closely match the observed drawdown. Common characteristics of hydrofacies distribution of those realizations that preformed poorly include over-elongated sand and gravel channel bodies 51 that extend across the model area. These channels generally have widths greater then 20 meters. Also realizations that have narrow channel widths, generally less then 8 meters, and short channel lengths, generally less then 20 meters, performed poorly. Only those realizations that show discrete gravel intervals (8m by 30m) within moderately elongated sand bodies (10-15m by 40-60m) had the best results. Given these general characteristics of channel bodies, it would be advisable to generate many realizations for the distribution of hydrofacies. Then filter through these realizations excluding those realizations that have over elongate or discrete channel characteristics. From the remaining realizations, only use those realizations that seem to accurately reflect the observed pumping test for modeling contaminant remediation. Conclusions Heterogeneity of sediments within an alluvial fan aquifer has a significant influence on the hydraulic response of the aquifer system. A long term pumping test at the Helipad Study Site shows that drawdown is along a preferential flow path. Results of this pumping test were used to evaluate four conceptual models. These included a homogeneous-layered model (HL), a homogeneous-layered model incorporating relatively low conductivity paleosol layers (HLP), a transition probability geostatistics simulation (TPG) for the model area following methods of Carle (1996), and a stratigraphic transition probability geostatistics simulation (STPG) following methods of Weissmann and Fogg (1999). Only those models that resolved heterogeneity at the hydrofacies level using transition probability geostatistics produced drawdown results that were consistent within the context of the heterogeneous alluvial fan system. 52 Simulated drawdown from the HL and HLP were able to confine drawdown within the main pumping layer, however the propagation of the drawdown cone was inconsistent with the observed results. In development of conceptual models for subsurface characterization, large scale geologic features, such as unconfonnities separating stratigraphic units or regional contacts, are commonly incorporated. However, in many cases, heterogeneities at scales smaller then regional contacts are simply not incorporated. Local evaluation of an aquifer using analytical solutions to pumping tests may not be valid as was observed in this study. For the development of conceptual models used for site investigations, it is important to include detailed geologic information. Since the hydraulic response of an aquifer is directly related to the physical heterogeneity of the system, incorporating aspects of the geology is necessary for reasonable and realistic pumping test modeling. Our results show that incorporating heterogeneities at the hydrofacies level reasonably matched the hydraulic response of a heterogeneous alluvial fan aquifer. 53 Chapter Four Conclusions Alluvial Fan Heterogeneity The general architecture of alluvial fans consists of sequences separated by unconfonnable surfaces. Within individual sequences, channel gravel and sand, which consist of highly permeable sand and gravel, form the main aquifer. Floodplain deposits which are generally composed of lower permeable sands and silts are deposited adjacent to these channel deposits. Pedogenically—altered sediments or paleosols which consist of fine-grained sand and clay may develop capping individual stratigraphic units. These units tend carbonate rich and typically have very low permeability and represent a prolonged hiatus in deposition. In alluvial fan development, rivers and streams may erode through these deposits. This provides conduits for fluid flow between sequences. At the Lawrence Livermore National Laboratory (LLNL), California, a similar structural architecture of channel, floodplain, debris, and paleosols exists. Conceptual Model Development Four conceptual models incorporating elements of the physical heterogeneity were developed. The first conceptual model is a homogeneous-layered model where stratigraphic sections were separated according to interpolated top paleosol surfaces. This type of model was developed to resemble the HSU model used by LLNL. The second conceptual model was a modified version of the first in that paleosol units were incorporated between stratigraphic units. The third conceptual model used transition 54 probability geostatistics with a Markov chain to generate the spatial distribution of hydrofacies where each of the individual stratigraphic units was simulated simultaneously (Appendix F). This approach incorporated geologic information in terms of an average channel length, width and orientation. The fourth conceptual model used transition probability geostatistics with a Markov chain model to generate simulations over the entire block volume for each stratigraphic and paleosol unit (Appendix F). Corresponding stratigraphic intervals bounded by paleosol surfaces were pulled out of these full block volumes and merged to produce a final realization for the distribution of hydrofacies. This approach incorporated the most geologic information in comparison to other models developed. In total twenty-five of these realizations were generated. Evaluation of Conceptual Models For evaluation of each conceptual model, drawdown histories were simulated using Modflow-2000 (Harbaugh et al., 2000; Hill et al., 2000). These simulated results should reflect the differential hydraulic response observed. The Grid used within the groundwater model was 360m by 310m by 39m (x, y, and 2 respectively). Grid cells were refined around pumping wells and consisted of approximately 5 millions cells. Eastern and western boundaries were assigned as constant head boundaries, northern and southern boundaries were assigned as no flow. These boundary conditions were assigned to generate a groundwater gradient that was representative of the true system (Appendix G). From numerical groundwater models, drawdown was observed from 1650 series wells. Evaluation of conceptual models showed that the heterogeneity within the alluvial aquifer had a significant influence on the hydraulic response of the system. 55 A long term pumping test at the Helipad Study Site shows that drawdown is along a preferential flow path. Only those models that incorporated finer scales of heterogeneity using transition probability geostatistics produced drawdown results that were consistent within the context of the heterogeneous alluvial fan system. Simulated drawdown from both homogeneous models were able to confine drawdown within the main pumping layer, however the propagation of the drawdown cone was inconsistent with the observed results. Calibration of hydraulic properties for hydrofacies within the stratigraphic transition probability geostatistical simulation was able to produce drawdown curves that were similar to the observed. Simulated drawdown from the other twenty-four realizations of this model type produced variable results (Appendix H). It was observed that there were common characteristics to those realizations that preformed well. Those realizations that showed discrete gravel intervals (8m by 30m) within moderately elongated sand bodies (10—15m by 40-60m) produces the best results. Analytical Solution Evaluation An analytical solution based on observed drawdown in 1650 series wells was developed (Appendix D). Results from this solution produced fields of transmissivity, storage coefficient, and hydraulic conductivity values within the central Helipad study. Results of this analytical solution were contrary to the observed geology. The geology indicated a gravel channel passing through wells located to the west of the Helipad Site and a sand channel passing through the other wells. Since gravels generally have a higher permeability then sands, it would be expected that the results of the solution would also 56 indicate this. However, the results indicate a zone of high conductance passing through the central wells at the Helipad Site. These results indicate analytical solutions are inaccurate within the context of this system. Future Considerations Given the general characteristics of transition probability geostatistical simulations that preformed well, future simulations generated with transition probability geostatistics should be analyzed to filter out realizations that have poorly performing characteristics. From the remaining realizations, only use those realizations that seem to reasonably reflect the observed pumping test when using this type of model for remediation efforts. Due to time limitations, other issues were not able to be analyzed in great detail within this study. These include the effects of defining hydrofacies based on sediment descriptions. Within individual hydrofacies, hydraulic properties are variable. In core, a gravel may be identified and is given a gravel hydrofacies classification. This classification has the highest hydraulic conductivity. However, a range of clay content was observed in described gravels. This ranged form little clay to very thick clay coats. So by defining gravel, which contains thick clay coats, as a gravel hydrofacies may be inaccurate. Assigning a hydrofacies classification that has lower associated hydraulic conductivities may be more reasonable. In future, realizations, it might be advisable to incorporate hydraulic variability within individual hydrofacies. Within this study, hydrofacies were assigned based on identified sediment type. 57 There were also concerns with the simulation quenching routine. Simulation quenching of large models did not reach the objective function assigned within the TPSIM parameter file. This may be due to the relatively high angle (30°+) between the grid orientation and the Markov chain model orientation (Weissmann, personal comment). This may account for some of the poor drawdown result for the transition probability geostatistical simulations. Future work may allow groundwater models be optimized with parameter estimation routines that can account for updating well files in highly heterogeneous systems as well as optimize to drawdown instead of head values. Finally, both tracer tests and pumping tests were conducted for the Helipad Site. However, time limitations prevented using tracer data for evaluating conceptual models. Running particle transport models is recommended to further evaluate conceptual models developed within this study. 58 C. conducts]l 1208. ll: 1353. DJ described ': F0; sand hydra hidmfaeies locals. si'. hydrofacies are exPOSec Web A.” Appendix A Core Descriptions Core description for wells located in the model area followed the previous work conducted by Weissmann (2001) with addition of wells 220, 653, 906, 1205, 1206, 1207, 1208, 1223, 1303, 1306, 1401, and 1416. Core descriptions for wells 1307, 1250, 1251, 1252, 1253, 1254, 1255, 1550, 1551, 1552, 1553, 1650, 1653, 1655, and 1657 previously described by Weissmann (2001) were also used for evaluation of this site. Four hydrofacies are denoted. These include the gravel hydrofacies (Figure A.l), sand hydrofacies (Figure A.2), silty sand hydrofacies (Figure A3), and paleosol hydrofacies (Figure A.4). Gravel and sand hydrofacies are interpreted to be channel deposits, silty sand hydrofacies are interpreted to be overbank deposits, and paleosol hydrofacies are interpreted to represent a depositional hiatus where alluvial fan surfaces are exposed allowing soil development. Core descriptions are presented in Figures A.5 through A.16. 59 Gravel Facies Figure A]: Gravel Facies o Clast supported. 0 Massive. 0 Typically contains thick, reddish, clay coats on grains. Sand Facies Figure A. 2. Sand Facies Typically fine to medium sand, with some coarse to very coarse sand and/or gravel, commonly silty. o Grains are typically subrounded. 0 Well to moderately sorted with some clay coats on grains. 60 Silty Sand Facies Figure A.3: Silty Sand Facies 0 Massive. 0 Common root traces. 0 Typically brown (lOYR) with mottling around root traces. 0 Light pedogenic alteration (thin to no clay coats, minor MnO, no visible pedogenic structures). Paleosol Facies 4 -D- .!:1£_-—-u-€-9.~»—u-::!! Figure A. 4: Paleosol Facies -Evidence for pedogenic alteration: 0 Thick clay coats on ped faces. 0 Preservation of soil structures (blocky to prismatic). o Reddish color (5 to 7.5 YR typical). 0 Common Mn- and Fe-oxides. 0 Common carbonate in upper paleosols (above 50 ft) and Lower Livermore. Formation; Carbonate is present in lower paleosols (below 50 it), however are less common. 0 Occasional root traces. 61 Michigan State University Department of Geological Sciences Well Name: w.220 Location: TFD LLNL Date Logged: 51150002 Graphic Description General Description Depth (meter) Silty sand to sandy silt. Mno along root traces above 43 it (possible paleosol) (10YR 614) blocky structure Figure A.5a: Core description of facies identified in well 220. 62 Michigan State University Department of Geological Sciences Well Name: w.220 Location: TFD LLNL Date Logged: 5/15l2002 Graphic . . Description General Descn ptr on Depth (meter) Clay rich sandy silt, prismatic ped structures, MnO on clay coats (2.5YR6/4) F ine to coarse massive sand, silt gading downward to medium - v. coarse sand. moderate to well sorted, sharp basal contact Clay rich fine silty sand, moderate clay coats, prismatic stuctures (10YR6/4) Clay rich sandy sitl. moderate clay coats on ped faces, prismatic structure, Mno - dendrites and coats on ped faces (10YR67/4) Massive fine sandy silt. fine Mno dendrites along plant traces, rare root traces (clay lined) (10YR6-7/4) Figure A.5b: Core description of facies identified in well 220 continued. 63 Michigan State University Department of Geological Sciences Well Name: W-220 Location: TFD LLNL Date Logged: 5ilSl2002 Graphic . . Description General Descnptr on Depth (meter) L ' Poorly sorted gravel and pebbles (up to 3cm), clay coats on clasts. clast suported (GLEY 7/106Y) Clay rich line silty sand, moderate clay coats. prismatic structures (10YR7/4 - 2.5Y 7/2 mottles) Gravel fining downward to v. fine to fune sand. diffuse Mno, massive (10YR614) Figure A.5c: Core description of facies identified in well 220 continued. 64 Michigan State University Department of Geological Sciences Well Name: W-653 Location: TFD LLNL Date Logged: 61512002 Graphic Description General Description 75 ‘65 3.5. 5 O. 0) o i Silt. v. fine sand to fine sand. MnO in root traces, little clay, grades downward to clayey ' ' gravel with pebbles (10 YR616) H Loose gravel up to 3 cm _ Thin clay coats, silt to fine sand, few pebbles up to 3 mm, Mno speck grades downward - ' to poorly sorted silty gravel (lOYR714) Silt, v. fine to coarse said, MNo specks. thin clay coats deacreasing downward (10YR714) Sand with few pebbles up to 7mm. Mno specks, v. fine to fine CaC03 (10YR614) Sand with CaC03 and Mno specks, grain size increases with depth, thin clay coats deacreasing with depth (10YR614) “"'-’-'""l V. fine to fine sand. CaC03 presentwith MnO specks (10YR714) V. fine to fine sand. CaC03 presentwith Mno specks. root traces visible (10YR714) Thick clay. CaC03 coating peds and in root traces. MnO specks. clay decreasing with depth ( 1 OYR 514) Moderate clay. CaC03 in root traces and ped faces Figure A.6a: Core description of facies identified in well 653. 65 Michigan State University Department of Geological Sciences Well Name: W-653 Location: TFD LLNL Date Logged: 61512002 E m . 1;, Graphic ‘ . 5 Description General Description 8 C] i .. V. fine to fine sand, few pebbles up to 4mm, few Mno specks, increasing grainsize E downward Silty sand, v. fine to fine, MnO specks CaC03 blebs, poorly sorted (10YR714) Silty sand, v. fine to fine, MnO specks, pebbles up to 3 mm, thick clay coats, poorly sorted (10YR714) V. fine to fine sand, small MnO specks, well sorted, thin clay coats, few root traces (10YR714) Sandy gravel, v. fine to medium sand matrix, pebbles up to 4 cm, thin clay coats (10YR 616) V. fine to medium sand, thick cal coats. Mno in root traces (10YR614) Gravel with fine sand matrix (10YR616) Sitly to fine gravel, poorly sorted, MNo specks (7.5YR 516) Sand on top of paleosol, v. fine to fine sand, Mno specks and blebs, clay in root traces with reduction halos (10YR616 to 10YR714) Silt to fine sand, few pebbles up to 4 mm, few roottraces (10YR714) Thin to moderate clay coats, silt of fine sand, MnO specks and blebs, few root traces with Figure A.6b: Core description of facies identified in well 653 continued. 66 Michigan State University Department of Geological Sciences Well Name: w-653 Location: TFD LLNL Date Logged: 61512002 Graphic Description General Description 5 a) E. 5 a o D Moderate to thick clay coats, Mno blebs and specks, root traces with clay, clay decreases with depth, sitly, ( 10Y R616) V. thick clay coats, MnO blebs, root traces, v. fine to coarse sand, poorly sorted, (10YR614) V. fine to v. coarse sand. Mno specks (10YR714) Fining upward gravel to sand overlying v. fine to v. coarse sand, pebbles up to 1 cm, lage root traces with halos filled with clay (10YR616) Thick clay coats, clay decreases with depth, MnO specks, blebs in root traces, little CaC03 (10YR616) Silt to v. fine said, mottled 2.5Y 812 and 10YR714, few MnO blebs, root traces with clay Thick clay coats, Mno blebs and specks, clay decreasing with depth (10YR714) V. thin clay coats decreasing with depth, small MnO blebs, v. fine to fine sand, few corase sands, clay in roottraces (10YR714) Figure A.6c: Core description of facies identified in well 653 continued. 67 Michigan State University Department of Geological Sciences Well Name: W-906 Location: TFD LLNL Date Logged: 61412002 Graphic nnqrrinfinnt General Description Depth (meter) “1' i ' Mud with gravel up to 35mm, pebbles with interbedded sands and gavel, Clast supported with silty sand (2.5Y5/4) lnterbedded sand and gravel up to 7 cm, clay coats, Clast supported (2.5Y 514) Weak paleosol, sharp contact, Mno and F90, minor prismatic structure (10YR616) Weak paleosol, silty to v. fine sand, minor v. cease sand, CaC03 web structure. thin clay coats (2.5Y 716) Weak paleosol, CaC03 web structure decreasing downward, MnO spews, fine to medium sand (10YR516) Weak paleosol, fine to medium sand, minor pebbles, disperse CaC03 and MnO. moderate clay coats (10YR516) lnterbedded sands and gravels, clast supported, clay coats on grains 10YR5/6) Silt grading downward into interbedded sand and gravel up to 3 cm, moderate clay coats ' and MnOspeacks decreasing dmnward, (10YR416) lnterbedded silty sand and gravels, thin caly coats on pebbles, trace MnO (10YR518 to , 10YR6/4) """ , Medium to coarse sand, some pebbles (2.5Y 613) . lnterbedded medium to coarse sand and gravel grading downward to fine sands (loose ‘ " ‘ core) 1m lnterbedded medium to coarse sands and gravels up to 4cm Silt and fine to coarse sand, minor Mno, scattered pebbles with thin clay coats (10YR616) Weak paleosol (61' to 66'), moderate clay, prismatic structure (10YR5/8), interspersed pebbles up to 4mm, grading downwa'd to silt with interspersed pebbles less then 3.5cm (10YR618) Figure A.7a: Core description of facies identified in well 653 continued. 68 Michigan State University Department of Geological Sciences Well Name: woos Location: TFD LLNL Date Logged: 61412002 Graphic ' . Dnerrintinn General Descnpnon Depth (meter) 1:;4. fin. ' i: ' Clast supported with interbedded sand, clay coating on pebbles (up to 7cm), MnO on -- "'- gravel (7.5YR5/6) _, _ . Clast supported with interbedded sand, clay coating on pebbles (up to 7cm) (7.5YR5/6) Weak paleosol, sitly sands with interspersed pebbles, matrix supported, clay coats on pebbles (10YR618) Silty sand, v. fine to fine, clay lamellae, pebbles interspersed (2.5Y713) LW ' ‘ ' \m\ ‘ ‘1‘ ‘ . EM lnterbedded sands and gravels, medium to vary coarse (10YR614) I lnterbedded sands and gravels, medium to vary coarse (10YR6/4) Silty sand, medium to fine, F90 and MnO decreaseing downward, clay lamellae ,, lnterbedded sand and gravels up to 3.5cm, clay coating (10YR518) Fine to medum silty sand, clay lamellae (10YR613) Silty sand. trace F90 and MnO, medium to fine sand, clay lamellae Weak paleosol, silty sand, v. fine to medum, thin clay coats, increasing grain size with depth (10YR5/4) Weak paleosol at 117', moderate clay coats, Mno and F90 patches, diffuse CaC03 and CaC03 nodules up to 4cm at 122‘ (10YR613 to 10YR7/6) Silty sand. v. fine to medium, sparse F90 and MnO, mottled, moderate clay coats, minor prismatic structure (10YR616 to 2.7T714) Figure A.7b: Core description of facies identified in well 906 continued. 69 Michigan State University Department of Geological Sciences Well Name: w.906 Location: TFD LLNL Date Logged: 6l4l2002 Graphic _ . Description General Description Depth (meter) ‘ V. fine to fine sand and silt, CaC03 common, Mno patches, mottled color 2.5Y7.4 (2.5Y8l3) V. fine to fine sand and silt, CaC03 common, MnO patches, mottled color 2.5Y7.4 (2.5Y8l3) V. fine to fine sand and silt, CaC03 common, Mno patches, clay lamellae, minor prismatic structure (2.5Y7I2) Fine to v.fine sand and silt, disperse Mno and CaC03, coarse sand interbeds (10YR7I3) MnO patches, CaCo3 filaments, minor prismatic structure (10YR4/6) Figure A.7c: Core description of facies identified in well 906 continued. 70 Michigan State University Department of Geological Sciences Well Name: W.1205 Location: TFD LLNL Date Logged: 6150002 Graphic Description General Description E a) E, E o. a) O Vfine to fine silty sand, thick clay coats (2.5Y4/3) Fine to medium sand (2.5Y6/8) Clean gravel, pebbles up to 4cm Silty sand, fine to vfine, diserse CaCOS (2.5Y7r5) Fine to medium sand, abundant CaC03, decreases with depth (2.5Y4/6) Fine to medum sand; silt, pebbles up to 1 cm Medium to coarse sand, pebbles up to 2cm (10YR7/6) V. fine to fine sitly sand, spase pebbles, CaCOS filaments to diffuse (2.5Y7I2) V.fine to fine silty sand, CaC03 filaments, MnO specks (10Y R6l6) V. fine to coarse sand, diffuse CaC03 (2.5Y7/3 to 2.5Y514) Fine to medum silty sand, CaC03 present (2.5Y614) Silty sand to coarse sand. pebbles common up to 8 cm, interbedded sands and gravels, minor CaC03 present (2.5Y7/2) Silt with v. fine to fine sand, minor coarse sand, diffuse MnO, patches of CaC03, clay lanellae (10YR6I4) Figure A.8a: Core description of facies identified in well 1205. 71 Michigan State University Department of Geological Sciences Well Name: W-1205 Locan’on: TFD LLNL Date Logged: 61512002 Graphic . ‘ nncrrinfinn General Descnptron Depth (meter) Fine to medium sand and silty sand, clay coats, Mno patches (10YR58) Thin clay coats, fine sand with sparse coarse sand, pebbles up to 7mm, trace MnO (10YR 5/8) Thin clay coats, fine to medium sand with sparse coarse sand, few pebbles up to 2mm, trace MnO Figure A.8b: Core description of facies identified in well 1205 continued. 72 Michigan State University Department of Geological Sciences Well Name: W-1206 Location: TFD LLNL Date Logged: 6/6/2002 Depth (meter) Graphic . _ Description General Descripti on 5.3.93 Fine to medium srlty sand, trace CaC03 decreasrng wrth depth, clay coats on pebbles up {+3 to 3.5cm (2.5Y7/4) :::I Fine sand, disperse MnO, thin clay coats, few C3003 nodules (10YR5I8) E23: Fine to medum sand, prismatic structure, diffuse CaC03, minor Mno, few pebbles up to “7+3 8 cm near base (10YR714) Silty sand, CaC03 filaments, siterspersed pebbles 10YR6I4) Silty sand, CaCOS filaments, siterspersed pebbles Sand and gravel, pebble up to 1cm, matrix supported (10YR5/6) Prismatic structure, CaC03 abundant, sparse pebbles up to 3mm Fine to medium sand with clay coas, Mno blebs (2Y7/6) Prismatic structure, moderate clay coats, MnO patches, CaC03 webs (10YR516) Vfine to fine silty sand, disperse CaC03, sparse pebbles (2.5YR713) ‘ _ lnterbedded sand and gravel, pebble up to 2.5cm _ Weak paleosol, moderate clay coats, CaC03 and MnO present Prismatic structure Fine to medum sand, pebbles up to 4mm, clay lamellae Medium to v.curse sand, prismatic structure, moderae clay coats, trace MnO and F90 (10YR6/4) V.fine to fine silty sand, spase CaCOS, clay coats decrease with depth (10YR518) Figure A.9a: Core description of facies identified in well 1206. 73 Michigan State University Department of Geological Sciences Well Name: W—1206 Location: TFD LLNL Date Logged: 6(612002 Graphic Description General Description :5 (D E, 5 O. (D o Weak paleosol, moderate caly coats, CaC03 nodules, prismatic structure (10YR6/6) Silty sand, pebble up to 4mm (10YR614) Fine to medum silty sand, few coarse sands, Mno blebs, CaC03 nodules up to 4mm, few pebbels up to 4mm (10YR5I8) Weak paleosol, moderate clay coats, prismatic structure, CaC03 filaments (10YR6/6) V.fine to fine silty sand, sparse MnO blebs, sparse FeO (10YR7/6) V.fine to fine silty sand, moderate clay coats decreasing with depth, minor prismatic structure (10YR518) V.fine to fine silty sand, few pebbles, diffuse Mno (10YR7/4) Weak paleosol, incresed MnO, sparse pebbles (10YR6/8) V. fine to fine silty sand, few medium to coarse sands (10YR7I8) Fine to coarse sand, silt lenses present, (10YR6/3) CaC03 cement, Mno specks, minor prismaic structure, moderate clay coats decreasing with depth (10YR5/8) V. fine to fine silty sand, sarse MnO (10YR7/4) V.fine to fime sand, moderate clay coats decreaseing with depth, minor prismatic struture (10YR618) Fine to v. corarse silty sand, few pebbles up to 7 cm (10YR5/4) Figure A.9b: Core description of facies identified in well 1206 continued. 74 Michigan State University Department of Geological Sciences Well Name: W-1206 Location: TFD LLNL Date Logged: 6I612002 Graphic _ . nncrrinnnn General Description Depth (meter) Prismatic structure, moderate caly coats, Mno patches, (10YR6I4) .. lnterbedded sandy gravel, pebbles up to 2.5 cm, grading down to clean sands with few ', pebbles up to 7 cm (10YR6/3) V. fint to fine silty sand (10YR616) mee to fine silty sand, prismatic structure, moderate clay cuts, increased MnO with depth (10YR616) Silty sanddisperse Mno patches, interspersed pebbles up to 1.2 cm (10YR5/4) Weak paleosol, minor prismatic structure, reddish color, moderate clay coats (10YR6/6) ' Fine to v, coarse silty sand (10YR6/6) lnterbedded sand and gravel, pebbles up to 4 cm Fine to medum sitly sand, increasing pebble content with depth up to 2.5 cm, prismaic structure, moderate caly content (10YR616) V. fine to fine silty sand, sparce Mno and F90 (10YR7I2) Silty sand grading downward to interbedded sands and gravels, dffuse Mno and F90 specks, rough laminations visible (10YR7/6) ‘1' - lnterbedded sands and gravels, pebbles up to 8cm (loose core) (10YR7I4) Figure A.9c: Core description of facies identified in well 1206 continued. 75 Michigan State University Department of Geological Sciences Well Name: W-1206 Location: TFD LLNL Date Logged: BIG/2002 Graphic . . Description General Descripti on Depth (meter) 5+: Wine to fine silty sand, diffuse CaC03, MnO patches, sparse pebbles (10YR‘H4) E12: Silty sand, v.fine to fien sand, sparse coarse sand grains, prismatic structures between _.;.::- 194' and 198', CaC03 filaments, Mno patches (10YR7I4) so I | ....:.‘ l Figure A.9d: Core description of facies identified in well 1206 continued. 76 Michigan State University Department of Geological Sciences Well Name: W-1207 Location: TFD LLNL Date Logged: 614/2002 Graphic _ _ Dnsrrinfinn General Description Depth (meter) V.fine to fine silty sand, few pebbles up to 5mm, moderate clay coats (10YR412) Loose gravel, borhole size pebbles V.fine to fine silty sand, moderate clay coats, few MnO specks, CaC03 blebs in root traces (10YR6/60 V.fine to fine silty sand, moderate caly coats decreasing with depth, CaC03 common (10YR616) V.fine to fine silty sand, thick clay coats decreasing with depth, MnO in roottraces (10Y R5/4 downward to 10Y R7/4) . an revel fining Epward vfine to fine sandy silt, thin clay coats, MnO blebs in root traces (10YR6/6 to 10YR7I4) Figure A.10a: Core description of facies identified in well 1207. 77 Michigan State University Department of Geological Sciences Well Name: W-1207 Location: TFD LLNL Date Logged: 674l2002 Graphic . . Description General Descnptr on Depth (meter) \ Sand and gravel, pebble up to 2 cm grading down to pebble up to 4cm, cla,l coats on grains (10YR674) \ Figure A.10b: Core description of facies identified in well 1207 continued. 78 Depth (meter) Michigan State University Department of Geological Sciences Well Name: w.1208 Location: TFD LLNL Date Logged: 6f6l2002 Graphic Description General Description V.fine to medium silty sand, pebble layers at 100' and 101.8 ' up to 40mm (clast suported), Mno specks (10YR576) V.fine to coarse sand and gravel fining upward to sandy silt, common pebl es up to 55mm, ' ' Mno specks common (10YR5IS) V.fine to fine silty sand, root traces and Mno specks present (10YR516) V.fine to fine silty sand, few root races and CaCOB blebs, thick clay coats, common MnO, prismatic structures (10YRSI4) Fine to coarse sand fining upward, loose, few pebbles up to 3mm (10YR6/4) Silt, thick clay coats, prismatic structure, MnO common (10YR5I4) Fine to corase sand fining upward, rare pebbles up to 25mm (10YR614) Fine to coarse sand and gravel, loose, pebbles up to 25mm (10YR 6/4) Fine to medum sand, thick clay coats, MnO specks common, rare pebbles up to 5mm (10YR6/4) V.fine to fine sandy silt, prismatic structures, CaC03 common, rare root traces (10YR614) Figure A.11a: Core description of facies identified in well 1208. 79 Michigan State University Department of Geological Sciences Well Name: W-1208 Location: TFD LLNL Date Logged: 676l2002 Graphic . _ Description General Descripti on It? a) E, E a a) D V.fine to fine sandy silt, MnO specks common (10YR616) Figure A.11b: Core description of facies identified in well 1208 continued. 80 Michigan State University Department of Geological Sciences Well Name: W-1223 Location: TFD LLNL Date Logged: 5/14I2002 Graphic Description General Description E a) E, 5 :1 <1) 0 V.fine silty sand, tabular vugs present, thin clay coats (10YR5/8) Weak paleosol, vfine to fine silty sand, pebbles up to 4mm, moderate clay coats (10YR5/6) V.fine to fine silty sand, filament CaCOS, prsimatic structure, minor MnO, (10YR676) - Sand and gravel, clast supported, pebbles up to 5cm, thick clay coats 7.5YR516) V.fine to fine silty sand, thick clay coats, MnO and CaC03 in root traces, prismatic structure (7.5YR676) V.fine to coarse sand and gravel, pebbles up to 5cm, (10YR6/4) Vfine to fine silty sand, pebbles up to 1cm, thick clay coats, MnO in root traces (color grades from 10YR6/3 to 10YR616) Clast supported said and gravel, moderate caly coats (10YR676) , . V.fine to fine silty sand, sharp contactwith gravel, moderate clay coatspossible root l traces or burrows (10YR7/6) - _ Matrix suported gravel, pebbles up to 4cm, thin clay coats,v.fine to medium grained matirx (10YR676) V.fine to fine silty sand, moderate clw coats decreasing with depth, minor MnO, minor root traces (7.5YR5I6) Wig etciSligiasgit ?§%r?3,dsfi%ar‘6ec'b :38'still%l%éwaiBe%aéYe%°tfitai371§¥nR3l9i caly coats (1 OY R6} 6) Fine to medium silty sand, few pebbles up to 2cm, thin caly coats (10YR676) V.fine to fine silty sand, Mno nodules possible filling root traces, thin caly coats (7.5YR5I6) Figure A.12a: Core description of facies identified in well 1223. 81 Depth (meter) Michigan State University Department of Geological Sciences Well Name: W.1223 Location: TFD LLNL Date Logged: 571472002 Graphic ' ' Description General Description :23. i ' ‘ Fine to coarse sand, prismatic structures, moderate clay coats, Mno specks (2.5Y773) Silty fine sand, few pebbles up to 1 cm, cross laminations, thin caly coats, few MnO blebs (10YR774) V.fine to fine silty sand, moderate clay coats, MnO in root traces (10YR676) V.fine to fine silty sand, moderate clay coats, few MnO blebs, few pebbles up to Srnm, grades downward to fine and coarse sand (10YR 674) l“l‘l“‘l"i‘l"l‘l"l"'l"A V.fine to fine silty sand, MnO speck and in roottraces, few pebbles up to 8mm, decreasing clay content with depth, CaC03 between 128' and 129' (10YR774) 1411st l'l'l'l'l‘ ‘l‘l,‘l'lfl_'l'l'l'l‘lfl'l'l‘l‘ i l) l .l. Fi - V.fine to fine sand grading downward to calst supported gravelS. M00 speck and blebs, gure A. 12b: Core description of facies identified in well 1223 continued. 82 Michigan State University Department of Geological Sciences Well Name: W-1223 Location: TFD LLNL Date Logged: 571472002 Depth (meter) Graphic _ . Description General Descripti on 5:1,; V.fine to fine silty sand, increasing caly content downward, MnO specks (2.5Y 774) ::::: >—-—-. w - - - fl] ,2 l l‘ ‘l‘ 'l ‘l .41 gr "l“l’i'l :. ill: .1, 'i "H" ill I Hi- gill; Fining upward sand and gravel to vfine silty sand, cross laminations, MnO blebs, pebbles up to 4cm (10YR674) Vfine to fine silty sand, moderate clay coats, MnO specks and blebs increasing with depth, few root traces, decresing clay contentwith depth (10YR774) l $009 Clast supported sandy gravel, pebbles up to 2cm, fine sand matrix (10YR674) Figure A.120: Core description of facies identified in well 1223 continued. 83 Michigan State University Department of Geological Sciences Well Name: W-1303 Location: TFD LLNL Date Logged: 5717/2002 E (1) 3; Graphic ‘ . .5 Dnerrintinn . General Descnthon 8 D D M” t Vfine silty sand, thin clay coats, Diffuse CaC03 (10YR5/4 to 10YR774) '0 .‘O . . “5n? t :Sandy gravel, pebbles up to 5cm, fine to medium sand matrix (10YR5/6) .39“? to .’ m 1 Fine to coarse sand (10YR5/4) ‘1 See above 0 See above 0 Fine to coarse sand, pebbls up to 5mm, thin clay coas (10YR676) V.fine to fine silty sand, moderate clay coats, FeO along ped faces, clay decreasing with E depth, pebbles up to 1 cm (2.5 Y673) . Sand and gravel, pebbles up to 4 cm, thin clay coas, clast supported (7.5YR576) Fine sand, well sorted (10YR676) Silt, thick clay (10YR5/6) Fine to medum sand, well sorted (10YR674) ‘°’ Clast supported gavel v fine to fine matrix, pebbles up to 6 cm, thin clay coats ‘ ’ (75Y 514) Fining upward sequence, Mno specks, thin clay coats C3003 coaing prismatic peds, vfine to fine matrix (7.5YR7/6) Figure A.13a: Core description of facies identified in well 1303. 84 Michigan State University Department of Geological Sciences Well Name: w.1303 Location: TFD LLNL Date Logged: 571772002 Graphic Description E a) .5, E G <1) 0 General Description Clayey silt, pebbles up to 2 cm (10YR674) . Fining upward sequence from gravel to sand, clast supported gravel, pebbles up to 3 cm, ' thin clay coats (10YR576) V.fine to medium said, pebbles up to 1 cm, MnO and FeO specks (10YR774) Silty clay, pebbles up to 8mm, MnO specks (10YR576) V.fine to fine silty sand, few coarse, thin clay coats MnO specks and in roottraces (10YR576) V.fine to fine sand, MnO specks, thin cla/ coats (10YR774) Fine to v.coarse sand (10YR674) Sandy gravel, clasts supported, fine to v.coarse sand matrix, pebbles up to 5 cm 10YR674 Gravely said, vfine to medium matrix, thin clay coats, pebbles up to 2 cm, cross lanination (10YR 574) Fine to v.coarse sand (10YR576) V.thick to fine clay coats, vfine to fine silty sand, Mno blebs in root traces, CaC03 common (10YR674) V.fine to fine silty sand, MnO specks, few pebbles up to 1 cm (10YR676) Figure A.13b: Core description of facies identified in well 1303 continued. 85 Michigan State University Department of Geological Sciences Well Name: w.1303 Location: TFD LLNL Date Logged: 571772002 It? 0) g Graphic . . g Description General Descripti on 0) O -rfiffzij .23: l " Fining upward sequence, vfine to coarse sand grading up to vfine to fine sand, MnO ‘ ' ' ' ' bleds decreasing with depth, clay in root traces (10Y R674) 2:25 V.fine to fine sand, few v.coarse sand, thin clay coats, MnO specks and bleds (10YR676) ~'-"-:—:-i S—iiE-‘i 15:2: “M _.__:'.._ ”'l Sandy gravel, slit to fine sand matrix, pebbles up to 6 cm, clast supported (7.5YR576) Lara; V.fine to fine silty sand, thin clay coats, MnO blebs and nodules, clay in root traces l‘?‘_'_"."'..r .L_._.._. (10YR476) '—'-;—:3 V.fine to fine silty sand, MnO specks and bleds, thin clay coats, CaC03 blebs (10YR776) S22"..- +j—:-—;—'1= V.fine to fine silty sand, MnO specks and bleds, thin clay coats, CaC03 blebs (10YR776) -Fii‘i iii: 8 if." a . 3+: Sand lens, vfine to coarse, poorly sorted (10YR676) . .rsrqysrc 'l'l'l -l-l‘l ilili V.tine to fine silty sand, MnO specks and bleds, thin clay coats, CaC03 blebs (10YR776) Figure A.13c: Core description of facies identified in well 1303 continued. 86 Michigan State University Department of Geological Sciences Well Name: w.1303 Location: TFD LLNL Date Logged: 571772002 Graphic . . Description General Descnptron Depth (meter) Fining upward sequence, vfine to fine sand grading down to fine to coarse sand, pebbles up to 3 cm (5Y871) Figure A.13d: Core description of facies identified in well 1303 continued. 87 Michigan State University Department of Geological Sciences Well Name: w.1306 Location: TFD LLNL Date Logged: 571472002 a ‘6 . 35, Graphic . _ 5 Dnerrintinn‘ General Description 8 O ‘ Fine to coarse sand and gravel, moderate clay coats, pebbles up to 5 cm (2,5Y574) - ~ - Sand and gravel pebbles up to 4 cm, thin clay coats ':: Gravel with sand interbeds, fine to medum sand, pebbles up to 3 cm, matrix supported ~ _ (2.5Y673) V.fine to fine silty sand, thick clay coats, prismatic structure,diffuse MnO, CaC03 present (10YR676) V.fine to v.coarse sandy silt, CaCOS present, MnO in root traces (2.5Y773) I». , - '., v 1% Wine to v.coarse sand and gravel, pebbles up to 4 cm, cross bedding visible (10YR674) Weak paleosol, vfine to fine silty said, MnO 7 Cacoa in root traces, moderate clay coats lmfi Gigiyesgioarse sandy gravel, pebbles up to 6 cm, interbedded sands and gravels ‘ - - (l1 YRS/4) . ,- . ineto coarse sandy gravel, spase CaCOS, pebbles up to 24 mm (10YR576) V.fine to fine silty sand, moderate clay coats, 06003 and MnO in root traces, prismatic structure (10YR574) ' Moderate caly coats, increasing clay content downward, sparse pebbles, root traces with Mno (10YR576) Figure A.14a: Core description of facies identified in well 1306. 88 Michigan State University Department of Geological Sciences Well Name: w.1306 Location: TFD LLNL Date Logged: 571472002 a ‘6 . 3 Graphic . . 5 Description General Description 8 O Medium to vary coarse sand, thin caly coats, MnO, specks, few root traces (2.5Y574) Moderate clay coats, few pebbles, moderate root traces, MnO specks, prismatic structure (10Y R674) lnterbedded sand and gravel, pebbles up to 2.5cm, thin clay coats, minor Mno and F90, sparse root traces (10YR578) l l ... C O l A. Wine to fine silty sand, thin clay coats, MnO blebs, sparse root traces (2.5Y774) ll 1 I; V.fine to fine silty sand, moderate clay coats, few root traces (2.5Y 574) V.fine to fine silty sand, FeO aid Mno in root traces, thin clay coats, grades downward to fine and coarse sand (2.5Y774) r ilililil ll Fine to coarse sand, thin clay coats , V.fine to medium snlty sand, speckled MnO, MnO in sparse root traces, few cobbles up to 7cm at 93.7' (2.5Y7/3) 'l' ll. - _,., "l".|."l.| l-l.l.l-.l-l. if‘ V.fine to medium sand overlying gravel, pebbles up to 4.5cm (2.5Y773) lnterbedded sand and gravel, Pebbls up to 2.5cm, thin clay coas (10Y R678, 10YR776, Silvio/’gifine sand, sparse pebbles up to 3cm, thin clay coats (2.5Y 673) Weak paleosol, moderate caly coats, sparse root traces, MnO specks (10YR5/6) V.fine to fine silty sand, MnO specks and MnO in root traces, thin clay coats decreasing downward (10YR 773) illlllll Weak paleosol, sparse CaC03 nodules up to 4cm, minor MnO in root traces (10YR576) V.fine to fine silty sand, sparse MnO specks, FeO in root traces, clay coats deacreasing " Figure A. 14b: Core description of facies identified in well 1306 continued. 89 Michigan State University Department of Geological Sciences Well Name: W-1306 Location: TFD LLNL Date Logged: 571472002 s 5 . g Graphic . . 5 Description General Description 8 O with depth (10YR773) in, Wine to v.coarse sand and gravel, pebbles up to 2.5cm (10YR578) V.fine to fine silty sand, moderate clay coats, halo around root traces, decreased root traces with depth (10YR 674) V.fine to coarse sand, sparse pebbles up to 5mm, sparse MnO and root traces, thin clay coats (10YR574) V.fine to fine silty sand, moderate clay coats, few root traces, MnO specks «7“. . ."j k V.fine to coarse sand grading downward into sandy gravel, pebbles up to 2.8cm :2. ‘ YR57 . , 13:;— Uiine tofillne srlty sand, MnO blebs, FeO in sparse root traces, moderate clay coats “;j—: . (2.5Y 774) +2-9- -l--- .- 5%: ' ' ' Sandy gravel, moderate clay coats, MnO specks and in root traces, pebbles up to 4.5cm 4.1-4.3 (10YR676) sass 772:1: V.fne to fine sandy silt, increasing root traces and MnO with depth, FeO increasing with :2:2; depth, moderate clay coats (10YR774) :23. a Vfine to fine silty sand, possible paleosol, moderate caly coats, MnO in root traces Figure A.14c: Core description of facies identified in well 1306 continued. 90 Michigan State University Department of Geological Sciences Well Name: W-1401 Location: TFD LLNL Date Logged: 671072002 Graphic . . Description General Descn pti on Depth (meter) Fine to medium sandy silt, thin clay coats, rare pebbles (10YR6-774) Fine sandy silt grading down to medium sand (10YR773) Fine to medium sandy silt, minor prismatic structure, thick clay coats, MnO specks (10YR674) Fine to medum sandy silt, massive (10YR676) Siltys sand to sandy silt, clay and CaC03 rich, Prismatic to blocky structure, thick caly coats (10YR673) Same as above, lacking CaC03, less clay Same as wove with CaC03, prismatic structure (10YR673-4) V.tine to fine silty sand, minor CaC03, rare pebbles (10YR 674) Sandy gravel, moderate clay coats Sandy silt, prismatic structure, thick clay coats, MnO (10YR573) V.fine to fine sandy silt, massive, CaC03 patches, thin to moderate clay coats (10YR673- 4) Sandy silt, thick clay coats, prismatic structure, rare CaC03 and MnO (7.5YR574) V.fine to fine sandy silt, rare CaC03 (10YR674) - Sandy gravel, thick clay coats, sandy clay matrix, clast supported, pebbles up to 30mm Fine to medium sandy silt, thin clay coats, root traces with reduction halos, (10YR672 to 10YR674) Silty sand, thick clay coats, prismatic structure, minor Mno (7.5YR674) V.fine to fine sandy silt, fininrg upward from a fine to medium sand (10YR674EJ6) Weak paleosol, Silty sand, hick clay coas, blocky structure, minor MnO (1 R574) Fine to medium sandy silt (10YR674) Figure A. 1 5a: Core description of facies identified in well 1401. 91 Michigan State University Department of Geological Sciences Well Name: w.1401 Location: TFD LLNL Date Logged: 671072002 Graphic _ _ Description General Descripti on '8' ‘5 E, E Q a) O Vfine to fine sandy silt, CaC03 nodules, elongate MnO (10Y R674) Fining upward v.coarse sand to fine and medium said (10YR67/6) V.fine sandy siilt (10YR776) silty sand, thick clay coats, prismatic to blocky structure, minor MnO (10YR574) Fining upward gravel to sand, massive (10YR6774-6) Sandy silt, thick clay coats, prismatic and blocky structure, minor MnO (10YR774) Sandy silt, thick clay coats, strong prismatic structure, MnO present (10YR676) Sandy silt to silty sand, moderate clay coats (10YR7734) Fining upward from medium to coarse sand and gravel, thin clay coats (10YR674) Silty sand, thick clay coats, prismatic to block stucture, MnO specks (10YR-2.5Y674-6) Silty sand, thin caly coats, minor MnO (10YR674) Weak paleosol, vfine to fine sandy silt, some root traces (2.5Y-10YR6-774) Fining upward from fine to coarse sand to send, some Mno blebs Weak paleosol, vfine to fine saidy silt, moderate clay coats, prismatic to blocky structure (10YR574 to 2.5Y774) Sandy silt, disperse MnO (10YR674-6) V.fine to fine sany silt, prismatic structure, thick clay coats (10YR672) Fine to medum sandy silt, prismatic structure, thick clay coats, some CaC03 (10YR574) Medium to v.coarse sand, thick clay coats decreasing with depth (7.5YR674) Sandy gravel, thick clay coats, clast supported Figure A. 1 5b: Core description of facies identified in well 1401 continued. 92 Depth (meter) Michigan State University Department of Geological Sciences Well Name: w.1401 Location: TFD LLNL Date Logged: 671072002 Graphic Description General Description lilrlili '. _n nu— .u—‘c o.— .—. a--. o——‘ a... O —. . —- a... -. .—-o 9—. e. ‘-l.|.| Vfine sandy silt, root traces, massive, increased CaCO3 content at 170' (10YR676 to 2.5Y773) Same as daove, some Mno blebs Same as move, dispersed MnO (257673) Sandy gravel, thin caly coats, broken core Sandy silt, Mno specks, (10YR774) Fine to medum sand (10YR676) Sandy silt, thin clay coats, some MnO blebs (10YR6773) Sandy gravel, moderate clay coats Sandy silt, Some CaC03 and MnO blebs (10YR774) Medium to coarse sand, thin to moderate caly coats Sandy silt, thick clay coats, prismatic structures, CaC03 and MnO present (2.5Y672) Figure A. 1 5c: Core description of facies identified in well 1401 continued. 93 Michigan State University Department of Geological Sciences Well Name: w.1416 Location: TFD LLNL Date Logged: 67372002 Graphic ' . Description General Descripti on '8‘ ‘65 E 5 cm a) D Clast supported sand and gravel up to 45mm, moderate clay coats Clast suported medium to coarse sand and gravel, pebbles up to 4cm (10YR573) fine to coarse silty said, MnO specks, increased clay content downward (10YR674) Vfine sandy silt, moderate clay coats, MnO specks and blebs (10YR676) V.fine to fine sandy silt, thick clay coats, few root traces (1 0Y R476) V.fine to fine sandy silt, CaC03 common (10YR476) V.fine to fine sandy silt, moderate clay coats (10YR574) Medium to coarse sand with interbeds of gravel, pebbles up to 20mm (10YR476 to 10YR674) Fine to v.coarse sand, pebbles up to 12mm, sharp basal contact V.fine to fine sandy silt, thick clay coats, CaCOS common (10YR574) V.tine to fine sandy silt, thick clay coats, Mno specks (10YR574) V.fine to fine sandy silt, thick clay coats, Mno specks and blebs (10YR473) Same as above, no CaCO3 Same as move with CaCO3 (10Y R574) V.fine to fine sandy silt, rare pebbles up to 20mm, disperse Mno specks (10YR574) Same as drove, CaC03 present at 61' (10YR674) Figure A.16a: Core description of facies identified in well 1416. 94 Depth (meter) Michigan State University Department of Geological Sciences Well Name: w.1416 Location: TFD LLNL Date Logged: 67372002 Graphic Description General Description Medium to v.coarse sand, pebbles up to 20mm, moderate caly coats (10YR574) V.fine to fine sandy silt, rare pebbles up to 20mm, few root traces, MnO specks (10Y R574) . .. Clast supported said and gravel, pebbles up to 30mm, moderate clay coats (10YR574) vs . -. b': L Same as move, pebbles up to 50mm ' - . Same as above V.fine to fine sandy silt, few pebbles, MnO specks (10YR574) V.tine to fine sandy silt, fining upward from medium to coarse sand, MnO specks (10YR574) Clast supported said and gravel, pebbles up to 20mm Fine to v.coarse sand, few pebbles up to 20mm (10YR574) Fine to medum sand, few pebbles up to 6mm (10YR574) Fine to v.coarse sand, rare pebbles up to 50mm (10YR574) V.fine to fine sandy silt, thick clay coats, MnO specks (10YR772) Same as above Same as above Same as diove (10YR674) V.fine to fine sandy silt, MnO specks (10YR774) Same as above, root traces, clay lamellea V.fine sandy silt, thick clay coats, disperse Mno blebs (2.5Y774) Same as move, MnO blebs common Figure A.16b: Core description of facies identified in well 1416 continued. 95 Michigan State University Department of Geological Sciences Well Name: W-1416 Location: TFD LLNL Date Logged: 67372002 Graphic Description Depth (meter) General Description .‘2‘13 . . . he ~l‘lilililil . 'lil'l‘l‘l'l‘l‘ 'l "l _. l' l vvvvv Same as move V.fine sandy silt, MnO specks and Cacos common, prismatic structure (10YR674) V.fine to fine sandy silt, thick clay coats, Mno specks, disperse CaC03 (10Y R674) V.fine sandy silt, MnO specks, disperse CaCO3, prismatic structure (10YR674) V.tine to fine sand, Mno specks, thin clay coats V.fine to fime sandy srlt, disperse CaCO , MnO specks common (10YR874) V.fine to fine sandy silt, pebbles up to 12mm, few MnO blebs, diffuse CaCOS, few root traces (10YR774) Vfine to fine sandy silt, few root traces, Mno specks common (2.5Y774 to 10V R574) V.tine to fine sandy silt, few root traces with halos, few MnO blebs (10YR674) Fining upward medium to coarse sa'id, pebbles up to 4mm, thin caly coats (10YR574) V.tine to fine sandy silt, Mno specks common, dsperse CaC03 (10Y R774) Same as above, prismatic structures Fining upward medium sand to fine sand, rare pebbles up to 10mm, thin clay coats YR57 Wine tollne sandy silt, thick clay coats, prismatic structure, CaC03 and Mno common (10YR674) V.fine to coarse sand, fining upward (10YR674) Figure A.16c: Core description of facies identified in well 1416 continued. 96 Appendix B Geophysical to Well Core Correlations This appendix contains 32 Geophysical logs for wells located in the Helipad Site study area, LLNL California. Four major hydrofacies were described from core recovered within the study area — gravel, sand, silty sand and paleosol (Weissmann, 2001). These hydrofacies show relatively unique geophysical character, high resistivity spikes relative to SPR are an indication of gravel or sand Hydrofacies (Weissmann, 2001). Low resistivity spicks relative to SPR are an indication of silty sand and paleosol hydrofacies (Weissmann, 2001). Based on these observations, hydrofacies elevations in core were adjusted to match geophysical logs. Gravel and sand hydrofacies are interpreted as channel deposits; silty sand hydrofacies are interpreted as floodplain deposits. Paleosol hydrofacies may overprint any of these facies (through weathering and by addition of clay and carbonate). Indicated in these figures (B.l through B32), are correlation of gravel, sand, silty sand, and paleosol hydrofacies to a combination of single point resistance (SPR), resistance (RES, LRES, RI), conductivity (Cond), 16—inch short normal resistivity (16]N), 64—inch long normal resistivity (64IN), natural gamma ray (Gamma), and caliper. The stratigraphy shown in these logs (B.1 through B31) was used to develop cross well correlations (Appendix C). 97 Michigan State University Department of Geological Sciences Well Name: w.220 Location: TFD LLNL Date: 772002 Lower leannore E Callpar 3 Sin 15in M0786” RES __ E Gamma 8 0 1m 4m 0 D 1 3 53-} :lt': B: = 3:5 91‘ 1 83.. 1 8: Figure 3.1: Geophysical well logs and core descriptions for well 220. 98 Michigan State University Department of Geological Sciences Well Name: w.653 Location: TFD LLNL Date: 772002 * Gravel Silty Sand a Lmruvormore .9 Sand - Palaoaoi a r a a I aoo ... onao ———_———a— Am a . . a o c a c c v - :5 WW" SPR _...... o ....................... 5 Win 50'“ Hydrofacies LRES — E Gamma 8100 275 1m 5m 0 C) 2D 10 llJJlllllllllllllliJlllJlIIIllllllu1111lllllllllLlllllllllJJ 30 40 5] ED Figure B.2: Geophysical well logs and core descriptions for well 653. 99 Michigan State University Department of Geological Sciences Well Name: W.906 Location: TFD LLNL Date: 772002 o ....................... E Sln 15h Hydrofacies 16 iii — E Gamma 54m ........... 8 0 100 p 10 ion rim 0 D 1 -2‘ 9..“ i .. ‘r 8: 1 . - i B— ,=' 2 i 2%,! :1 J . 91 E i I i 8- i 1'5 2i :( ': 8.]; Figure 33: Geophysical well logs and core descriptions for well 906. 100 Michigan State University Department of Geological Sciences Well Name: W-1205 Location: TFD LLNL Date: 772002 - :E- Gravel SiltySand Lmrlmflnm ' "3 Sand - Palaoaol g Cond _....._. g Gamma Hydrofacies RI — g i—‘—ra 0) fl 1 10 till ‘IIIII C) 0 10 lIllllllllIllllllllllllllllllll[lllllllllllllllLlllllllllllllllI 20 3] 40 50 60 I:igure B.4: Geophysical well logs and core descriptions for well 1205. 101 Michigan State University Department of Geological Sciences Well Name: w.1206 Location: TFD LLNL Date: 772002 For": Gravel 3m, 5.“ Lower Livarmora Sand - Palaoanl arm ”3.” FM. '.. MW” Tim-{:93 l:igure B.5: Geophysical well logs and core descriptions for well 1206. 102 Michigan State University Department of Geological Sciences Well Name: w.1207 Location: TFD LLNL Date: 772002 '- [:33- Gravel smy Sand 3 I-mrllvemm :I-If' Sand Palaoaol E Callpar CE) 5|" 15'" Hydrofacies Cond E Gamma 3 ii run 400 o c, _ 1 9: a: 8% 1 1 91 1 .i 8.. I 8% I‘7igure B.6: Geophysical well logs and core descriptions for well 1207. 103 Michigan State University Department of Geological Sciences Well Name: w.1208 Location: TFD LLNL Date: 772002 g .......‘E.'.'.'.'!!.‘3.'. ..... SPR —..—. E Sln 15h Hydrofacies 16 iii —— E Gamma 5“" ........... g 0 100 p in ion till) 0 ES $3— "-. ‘ ". _ ,- .. ..... I Fl: 1: 9-3 8 8 Irigure B.7: Geophysical well logs and core descriptions for well 1208. 104 Michigan State University Department of Geological Sciences Well Name: w.1223 Location: TFD LLNL Date: 772002 ._.-... Lower Livermore E 93"?" SPR _.._. E 5|" 15"- Hydrofacies 16 iii E Gamma 54m ........... 8 0 100 in ion mu 0 D 10 20 fl lJllllllllllllllllllillillllllllllllllllllllllllllllLlllLllIJL 40 H] 60 Figure 38: Geophysical well logs and core descriptions for well 1223. 105 Michigan State University Department of Geological Sciences Well Name: W-1250 Location: TFD LLNL Date: 772002 E “"9“ SPR _.._. "5’5"! 11"» Hydrofacies 16 iii —— E Gamma 54m ........... 8 0 100 10 1m 11m 0 D 1.. m. -r. 7' 10 V2?! “a a a 20 I] 5] 40 llllllLlllglIllIIILILIIIULLJLLJILIIIILLLLJLJIIIllllllllllllllll \ny'fi.*N.JMM~Mauv-,WWI~M‘M.' wr’flo‘.w‘¢’~ l Figure B.9: Geophysical well logs and core descriptions for well 1250. 106 Michigan State University Department of Geological Sciences Well Name: W-1251 Location: TFD LLNL Date: 772002 - j - Gravel Slity Sand - Lower Livermore Sand Palaoaol E “"9" SPR -.._. E 5'" 15h Hydrofaelaa 16 iii — E Gamma 54m ........... 8‘ 0 '00 p 10 ion rum 0 D 5 *r . l :l 8 It . l - i. 9.: I -i I : i'" 617 :i 1i 17 I ! Fiji: «r :i 'll 3! “i 91“ :.-‘ :3 e-j‘ 8-1 on":1 Figure B.10: Geophysical well logs and core descriptions for well 1251. 107 Michigan State University Department of Geological Sciences Well Namei W—1252 Location: TFD LLNL Date: 712002 Gravel Siltys.“ Etmrllvannora ':"""lSand Palaoaol E ...... Estes: ..... spa -..... E 5|" W» Hydrofacies 16 in — E Gamma 54m ........... 8 0 '00 p 10 run «inn 0 o 7 I r v- ,‘tm‘ snefr - MQJWW Z] ‘V‘f‘ : 40 “VWWMW‘. grit-MWW I" M ,i Q i l '3 = "ll A A ' I u I . . ,‘ , ' ' 5331‘; is" .‘. l : '\ -. .1“. r4“ lb." ’1 ' 4 {fix ., [W l ”-fl 5 c r: I d ‘W'WW .a ‘.’ I O. I I . ‘ . "'-"*\ab .‘Ahs A- Figure B.11: Geophysical well logs and core descriptions for well 1252. 108 Michigan State University Department of Geological Sciences Well Name: W-1253 Location: TFD LLNL Date: 772002 s ...... 9:!3231 ..... spa _.._. E Sin 15in Hydrofaclaa 15m _— E Gamma 54 m ........... g 0 “'0 p in ion ‘lllll D 10 Z] I] 40 11111111111llllllilllllllllllllllllllllllll Figure B. 12: Geophysical well logs and core descriptions for well 1253. 109 Michigan State University Department of Geological Sciences Well Name: w.1254 Location: TFD LLNL Date: 772002 - :.:_:-. Gravel Silty Sand — Lower Livermore Palaoaol E cal'l’“ SPR _.._. GE) Sin 15In Hydrofacies ‘6". E Gamma 54m ........... g 0 100 p 10 100 um D ' gifiimm Im- ' .a.aaaaaa 10 LlllllLJlllIJLLJlllLllllllllLJlellllllllllllJLlJllllllllllLlJllll . 23 I] 40 50 fl Figure B. l 3: Geophysical well logs and core descriptions for well 1254. 110 Michigan State University Department of Geological Sciences Well Namel W-1255 Location: TFD LLNL Date: 772002 . :23} Gravel Slity Sand - Lower Livermore :3: Sand Palaoanl E ......S.'..'.'.'?.‘.'.'. ..... SPR _.._.. “5’ 5|" 15"! Hydrofacies 16 III —— E Gamma 5‘ m ........... 8 0 100 p in ion till) a C3 Figure B. 14: Geophysical well logs and core descriptions for well 1255. 111 Michigan State University Department of Geological Sciences Well Name: w.1303 Location: TFD LLNL Date: 772002 E WW" SPR _.._.. “5’ 5"! 15M Hydrofaclaa 16 iii --— E Gamma 54 IN ........... g 0 100 1|] inn um 0 10 lllllllllllllIllIllllllllllLLlLJlllllJlll111111llLLJLLlJlllllll l‘r‘a 1.9" _ . . cup.- « - » tawny? I.» r A, ' n .i. mix. ac". st “w ~ ‘w .u- .' . I ' ' , .\ a W} l“ Jr ‘A 20 II “we-WM “s - s’i’i: W‘s?“ "- 2‘" ‘c 40 0.; 0. av C 5o I N 3 E] WM.4’~M. ,v.“‘WVVW\~”vw—w%o~ Figure B. 1 5: Geophysical well logs and core descriptions for well 1303. 112 Michigan State University Department of Geological Sciences W911N8m93W—1306 Location: TFD LLNL Date: 7000? - f};- Gnml Silty Sand - Lmrle-nnm :I': If Sand Paleosol E ...... $333235. ..... SPR —..—. E 5'" 15'» Hydrofacies 16 m E German 5“" ........... 8 0 “I" 10 1m 1m 0 D 5' i’ 10 lilllllllllllllllllIlllllllllllIllllLJllllJJlJJlJJLLJlLllllllLl fish”. 20 I] } oWWh-of'fiwww’rfi. 40 1 5] “W‘s-“J‘Mkh«MW/”‘5'“.rw/‘N’W‘W’wavi *s--M-vv".-w-V\ Ni“ N'w ED Figure 3.16: Geophysical well logs and core descriptions for well 1306. 113 Michigan State University Department of Geological Sciences Well Name: w.1307 Location: TFD LLNL Date: 712002 Gravel Silty Sand - Lower leormora Palooaol E “"9“ SPR _.._. 3 Sin 15"! Hydrofacies 16m — E Gamma a". ........... 3 a 100 J: 1a 100 «m 0 CI " hm~wrngmg ;_ _ 134:; so ’fi‘evtq. 10 20 .‘v‘Mu/wvmn nus-J fl J 40 llllllLLlLlllllllllllllllLlllJlLJllllllllllllLlLleLllljlllLlll Mali-"‘3‘ wflk. 50 so Figure B.17: Geophysical well logs and core descriptions for well 1307. 114 Michigan State University Department of Geological Sciences Well Name: w.1401 Location: TFD LLNL Date: 712002 - {if-A Gravel Silty Sand - Lmrllvemore Sand Palaoaol E ...... 5:!!231 ..... SPR _.._- 0E) 5|!- 15|n Hydrofaclas 16 IN — E Gamma 54 m ........... g 0 100 L to 1m 1m 0 _- ‘35 J’ j .r“ a—' E I a 3 5" a—j g 3 a. u r .1 ¢' ' C : 'l ‘ i : '8 53‘. a - ) 3 1. E 3) 81 f : S Figure B. 1 8: Geophysical well logs and core descriptions for well 1401. 115 Michigan State University Department of Geological Sciences Well Name: W-1416 Location: TFD LLNL Date: 7000') - lama Silty Sand - Lower Livermore '3 Sand - Paleosol E “"9" SPR _.._.. E 5'" 15'» Hydrofacies 16 m — ‘5" Gamma 64m ........... 8 0 100 p 1o 1011 11m 0 o -l 3 4‘21.“ c" I) " Jun“ 0‘ 1 ' ~. a; .E) C v- d 1". 2 E ‘1‘? I: I 1 32. z i -1 “:7 " A :3, i 2.3 i 8‘: "a”; 2 I}, 1 I . , ., .. ’ ' ‘2? i ( 9i?” i :L, :3 =i "I B:{ 1 i ii i ‘ '1 : -_: 81 .J Figure B.19: Geophysical well logs and core descriptions for well 1416. 116 Michigan State University Department of Geological Sciences Well Name: W-1550 Location: TFD LLNL Date:712002 ,- if} Gravel Silty Sand - Lmrtlvormore 21:3": Sand Palaoaol E ...... 933231 ..... SPR _.._. “5’5"! 15in Hydrofacies 16111 —— E Gamma 54m ........... 8 0 100 p 111 1011 11ml 0 D C v- 1;“, l a '4'3’ 1.. 40 s.-.-_,v~..-..’r.- wwfiw-‘Ww‘d‘asaqwué Figure 3.20: Geophysical well logs and core descriptions for well 1550. 117 Michigan State University Department of Geological Sciences Well Name: w.1551 Location: TFD LLNL Date: 712002 Gravel Silty Sand - Lmrleormore - Paiaoaol E ...... 931333: ..... SPR _.._. g 5111 15in MMdB 15m _— E Gamma 64m ........... 8 0 100 p 111 11111 11m] 0 o q 31 :1 e; v; .1 E {1: 3.1 . 1' 9': l . 1? 11 .1 £2“. 1 j 1:” Fl: { ‘1 :f : 112111111111- 9{ § 5 2., - ‘3 8'3 8% Figure 82]: Geophysical well logs and core descriptions for well 1551. 118 Michigan State University Department of Geological Sciences Wall Name: w.1552 Location: TFD LLNL Date: 70002 . Gravel Silty Sand - Lmruvonnoro Palaoaol E “"1’“ SPR _.._. “5’51" 15"! Hydrofacies 16111 — E Gamma 54m ........... 9, n 1110 111 1110 11m 0 CI .1 : '5 . L -1 (J' 1:); 7 Z L... ‘ mi} I «c... 9.: 1': 3 ‘1... i 1. 111-5 g : S... . 1).; 3 1. 9‘ a : ‘1 . 85 8-2 ; Figure 8.22: Geophysical well logs and core descriptions for well 1552. 119 Michigan State University Department of Geological Sciences Well Name: w.1553 Location: TFD LLNL Date: 70002 ' :..;' Gravel Silty Sand a Lmrlearmore ski-:ZZSand Paleosol E ..... 5331235. ..... SPR _.._. E 5|» 15in Hydrofacies 16111 — E Gamma 54m ........... 1% 0 '00 1o 11m 11m 0 D 9 8 We‘ve. :3.- - ~' Fl 9 8 8 Figure B.23: Geophysical well logs and core descriptions for well 1553. 120 Michigan State University Department of Geological Sciences Wall Name: w.1650 Location: TFD LLNL Date: 712002 137% Gravel Silty Sand - Lmrllvemore Sand - Paleosol E “"9" SPR _.._.. E 5|» 15in Hydrofacies 16111 — E G.mm. “0" nnnnnnnnnnn 3 11 1110 111 11111 11m 0 C) 1 .1 1 ea 9‘. 1 Fl: 1 .l 9: 1 fl .1 53: 8; ad Figure B24: Geophysical well logs and core descriptions for well 1650. 121 Michigan State University Department of Geological Sciences Well Name: W-1651 Location: TFD LLNL Date: 7I2002 m Gravel 5111,, Sand - lmrlmflnm Sand - Paleosol E C'“P°' SPR _.._. {'5’ 5"- 15|n Hydrofacies 16111 — E Gamma 54m ........... 8 0 "I“ 111 11111 11m 0 D 10 .- X's-1'1». ‘14" M‘y-w-IWK-MA’ 20 40 50 llllllllllllllllIlllllllLJlellllLllLllLllllllllllllllLLlllllJl 60 Figure 8.25: Geophysical well logs and core descriptions for well 1651. 122 Michigan State University Department of Geological Sciences Well Name: w.1652 Location: TFD LLNL Date: 712002 ' Gravel Silty Sand 5 Lower Livermore ' " :: Sand - Palaoaol E c611,... SPR _.._. g 5ln 151n We!“ 16!" E 5......“ 5.... . _ ...... ' . ' . ' .. 8 11 11111 "1 "III 1111111 0 o , 9- . . R 1 .>7. . ' 3'43? ' 9 ................ f' B 8 Figure 3.26: Geophysical well logs and core descriptions for well 1652. 123 Michigan State University Depa1tment of Geological Sciences Well Name: W-1653 Location: TFD LLNL Date: 712002 Lower Livermore E “"9" SPR _.._. 25'" 15"! Hydrofacies 16111 — E Gamma 54m ........... 8 0 100 p 111 11111 1111111 0 C) .l g; : fi‘:’¥7“'¥$¥sl§f" 9: R: E - em- : \",K¢;1' 9% 53-: 8: Figure B.27: Geophysical well logs and core descriptions for well 1653. 124 Michigan State University Department of Geological Sciences Well Name: W-1654 Location: TFD LLNL Date: 712002 - f}; Gravel Silty Sand - Lmrllvonnoro ' Sand - Paleosol 3 WW" SPR _.._. E 5|» 15'" Hydrofacies 16111 -— E Gamma 54m ........... 8 0 10" 10 11111 11m 0 C) i O ’0- {if 1'3 } :{' 8 vi l ‘0‘ i I ’1 1 .3 l '"1 1 'l :1; L 40 va Figure 8.28: Geophysical well logs and core descriptions for well 1654. 125 Michigan State University Department of Geological Sciences Well Name: w.1655 Location: TFD LLNL Date: 712002 " Gravel Silty Sand - LMI’ lenoro 5: Sand - Paleosol E c"'9‘" SPR _.._. 2 Sin 15in Hydrofacies 15m _— E Gamma 54 Ill ........... 8 0 100 111 11111 1111111] 0 D 10 Z] I] 40 50 ED lllllllllllllllllllIn][JJWLIJLLILLIILIIIIIllllllLlllllLllllll Figure 8.29: Geophysical well logs and core descriptions for well 1655. 126 Michigan State University Department of Geological Sciences Well Name: w.1656 Location: TFD LLNL Data: 712002 ’ Gravel Silty Sand - Lower Livermore 3: Sand - Paleosol 5 “mm SPR _.._. E 5'" 15'» Hydrofacies 15111 — E Gamma 54m ........... g 0 100 p 111 11111 1111111 0 10 4 5"“. MW e 13“»: 20 llllllllllllllllllllllllllljIlllllLLllLll_l_LLlJlllLJlllllllllJll 3 hung: : 3] 111-"at" kwva w 0 40 50 ED Figure B.30: Geophysical well logs and core descriptions for well 1656. 127 Michigan State University Department of Geological Sciences Well Name: w.1657 Location: TFD LLNL Data: 712002 002C Gravel Silty Sand l-M' ”WWW" Sand - Paleosol 3 “mm SPR _.._.. 3 Sin 15in Hydrofacies 15111 — E Gamma 54 m ........... 8 0 100 p 111 11111 111m 0 C) 2 "3 1 ") a.“ .) " . '4 1 R- r 3 ...... .3 3 f. _ C a: , - > : ‘1 i S : 8' 91 J 1 l 31 8': Figure B.31: Geophysical well logs and core descriptions for well 1657. 128 Appendix C Cross Sections This appendix contains fourteen cross-sections through the Helipad Site study area, LLNL California. Well locations through which cross sections are generated are shown in Figure Cl and C8. The stratigraphy shown in these cross sections (C.2 through C7, and C9 through C.16) was used to define sequence boundaries used in the geostatistical characterization described in Chapter 3 and Appendix F. Paleosol surfaces are informally named Aa through F following Weissmann (2001). 129 .3329: 03 Bee 3223 Ho .23 2.03 95:3 35 39:03 05 as 303 me 55804 ”_.0 0.5me 089% 089.... 089.: 1 ,052... 082.: 1; . bananas}; one], -. .1 .. w .0373 093...: 0.3a; , v.3“ o I; .8738? 052.; , 063.3 0.873 up .2825 a 8 :8: o m.._5. cum“?! 0 23.: , ~8ka enact. .m mcozoomnnoco .9235 can 8&8: ._z..._ 130 .88 89:0: 3580 5883 05 "3:85 588-820 wagon“ 5.8: 9 Snow "NO onE comm a s .E..E E 3 E can. E 3 c «.30 in a $68 in = «.30 .im on? 44m? one .395 «new 44m? 5.52 530w _omogaa - “mum 23 2.6 2.8 Scam- _U. 21. _U .920 - I Noah cuts: .m 28 Res: .2: zotomm z-m 25.595 131 .33...- I 23 3.5 2.8 288 - U 2..- D .35 - n< r 2 0.8. E to .— 33 in c 38 .iu 52. 44m; $9 44m; 5.52 .88 was»: 35:3 05 .3588 8308.398 mums—ob 5.8: 8 538 8.0 85E a. fig E 3. «GONE a... 538;. .m 25 “59.2... 1.2.... 209—bum z-w 1_25 . D as. _U .25. i a h E 0.8. E 3 E in. E .6 NSNK 8 06—0 film C hi—o 3‘” G :5 a CWCEF .m S2 in; 32 .8; Bu. .8: 35 nus—o: .._zj 20:.0mm z-m O O 5.. z 5: m 2mmhwm¥ NQQN\N , K 111/f0 $3.3 CNS“: .m Bu; 8o... mcozowmlmwoho , .. w. . , . ,, . 1 . . , _NCLQ—F: 1%; e 9% 28...»... ..z.... 137 .88 >23 vane: 05 £9.85 8.88-398 wag—ob 5.6: 2 530m 5.0 Rama ”48.2. . I 2 «8% 2.8 _.m . , . 552.. .m 2.5 >25 . _IIL «gm 32.»... ._z._._ 25. D .... 20.5mm z.m ., 25595 35.0 . I "Mays, , J. , z 8.23 .iu c 5.8» .im = 8;; .im 3o 44m? 8m in; wear in? Saw in? 5.52 5:8 138 :38; 33 44m; 532 a. :8 E won a «.0; .3 «map in? .33 >95 tam—o: 05 .3585 8303-380 9.?an 5.8.. 9 £30m .30 ii 0.0 .. . ‘U... ., . . ,1... ~00ka :38... .m 25 82.2. .2... 5.3.53» ZOFOmw Zuw : as 3 sum 22...; .2523 . 25.595 58» 139 Sufism 89 flu! 5.02 .83 .33». 9250: 05 smack: 5.83-398 macaw.“ 58: 9 550m ”:0 95me figs: . .88 .m :..m 5.8 . D 28. . . . . ... 4h I g . . . v. .c . n... -- ,..,. Q... .I Q. . O. ,, a... . .01: E 5:. EN: NSNR 538... .m 3......” 5...... 36 82.2. .5... "a3; 29 a; 20.5mm z.m £30m 2mm#m» ”S .U 03wE Ngwk .3th .t 85 68:2.— 42.... ZOFOmw z.m .1 zmmhw25 . D ,. 26 3&3: ._z._._ . .. 295mm Es 2.3. D .,.,, ”1...,“ _ ZEWIEOZ .380. I 5.. . E of: E 13 a. 33. s 33 emfiooiofi sci—Sim «map flu; up: 44m? 80 jug 5: in: «mm; .— pd; .iw 8 9.“; 53w «mam 142 .83 35% 98:0: 05 swap—5 8:08-398 9:96.: 33 3 60>» H: .0 059m _Sosa.! . _ 233.5 , _..w «I. .15]: 2.53:8.D ._ ,.. . w H...” , I A L w .5 _U 1+ .._ F . w. . h .22?! , . .. s m. .m u p .w. m... a ; o _ .1 , 1.. “A. - .. 6 0 ....... ....- -- :1..- u a M 6 h a: _. p M . 9 6 -0 9‘... . a... .. rig g E Susana £35598 Noowk quaim 3.35 8.53m .65ng m 8“. in; «ma. 3m; 3a. du; 25 03:0: 42.... $8 .83 20.5% mi ._26 . D ”a: .,... \ cum 3&3: ._z._._ :5. D 7W . M \ 22.5% mi {51 . \.......1.. «Mi. . J95 @320: 05 .3585 5308-968 waméob “mac 8 603 ”3 .0 053m ~8ka 538» .m 25 82.2.. ._z._._ zoFomm Es 25:58 a :5 in c 33 .im 39 in? con— flu? “mm; 145 Appendix D Analytical Solution Introduction Pumping test results are commonly evaluated using analytical solutions following those first derived by Theis (1935). Using the Theis equations and following a set of assumptions, a field of hydraulic parameters (i.e. transmissivity, storage coefficient, and hydraulic conductivity) may be resolved. Hantush and Jacob (1955) first developed an analytical solution for transmissivity and storage coefficient in a leaky aquifer. This type of solution was used to evaluate observed drawdown at the Helipad Site. Within the context of an alluvial fan aquifer, it is probable that there is to a degree, vertical fluid flow between stratigraphic units. Analytical solutions for the Helipad Site include transmissivity (T), storage coefficient (85) and hydraulic conductivity (K). Solutions of T, 83, and K fields were compared to the geology of the Helipad Site. The type curve with a 0.01 leakage factor had the best fit to that data. For the Helipad Site, there are eight closely spaced observation wells (1650 series wells) that range from 30 to 110 feet from the main pumping well (1551). Pumping from this site is from wells 1551 and 1552 at 3gpm and 0.5gpm, respectfully, however only pumping from well 1552 was used for this analytical solution. Listed in table D.1 is a set of drawdown histories for the wells at this sight. Results of the analytical solutions for each observation well are shown in Figures D.1 through D.8. Analytical solutions used Ground Water Analysis Package (GWAP) software. These results are then plotted in 146 a map view of the central Helipad Site and contoured for geologic assessment (Figures D.10 through D.12). Assumptions for Analytical Solutions There are several assumptions that must be made in calculation of analytical solutions. For calculation of solutions at the Helipad Site, a confined leaky aquifer was assumed. The following is a list of assumptions that must be made for analytical solutions to be valid (Batu, 1998): 0 Confined leaky aquifer assumptions — o Aquifer properties — Homogeneous and isotropic. Horizontal and has a constant thickness overlain by a confining layer having a constant vertical hydraulic conductivity and thickness. This confining layer is assumed to have a constant head plain. Infinite and laterally extensive. l Compressible and completely elastic. 0 Well properties - Has an infinitely small diameter. Fully penetrating. Discharges at a constant rate. Head loss through the screen is negligible. 147 ' Discharge to well is derived exclusively from storage in the aquifer. o Other properties— ' Water is immediately released from storage in the aquifer. ' Storage in the aquifer is proportional to head. ' Water has uniform density and viscosity. ' Darcy’s equation describes fluid flow. I Flow is horizontal and is directed toward the pumping well. Methods In order to determine T, Ss, and K for each observation well, type curve matching was constructed using the Ground Water Analysis Package (GWAP) sofiware. An analytical solution for each observation well required: 0 An aquifer thickness — o For the aquifer thickness, the maximum thickness for any sand or gravel channel visible across the well screen interval was assumed to be the main aquifer. This thickness of these channels was recorded as the aquifer thickness for calculation of the analytical solution. 0 Pumping well discharge — o Pumping within the alluvial fan aquifer over the observation period was from wells 1551 and 1552. Well 1551 had a pumping rate of 3gpm while well 1552 only had a pumping rate of 0.5gpm. For calculation of a solution, it was assumed that pumping from well 1552 was negligible and 148 a pumping rate of 3 gpm from well 1551 was assumed to be the primary source of pumping. o Radius of pumping well — 0 Wells drilled at LLNL use a standard six inch radius well tubing. A well radius of 0.5 feet was used for the analysis. 0 Distance from observation well to pumping well — o Distances from pumping well to observation wells were calculated using the Pythagoras thorium. Given the northing and casting coordinates for each of these wells, distances between pumping (1551) and observation wells (1650 series) was calculated. 0 Time and drawdown — o A table of time and drawdown was inserted into GWAP. Time is measured in minutes and drawdown is measured in feet. An abbreviated drawdown set is listed in table DJ for 1650 series wells used. 0 Curve matching ~— 0 A type curve for a confined leaky aquifer with a leakage factor of 0.01 was used to match drawdown data for each observation well. Early time drawdown was matched with the type curve. Results of these solutions are listed in figures D.1 through D.8. Results The following are a series of type curve matches for wells 1650, 1651, 1652, 1653, 1654, 1655, 1656, and 1657 (Figures D.l through D.8). Listed within each figure 149 is a solution for transmissivity (T), storativity (Ss), and hydraulic conductivity (K). Also listed within each figure are aquifer thickness (b), type curve (r/B), and the leakage factor (B). A summary of this information is listed in Table D.2. Using observed drawdown histories (Appendix H) located at the Helipad Site (Figure D.9), values listed in Table D2 are mapped. Figures D.10 through D.12 illustrate the distribution of hydraulic properties (T, S, and K) across the central Helipad study area. The distributions of hydraulic properties show a centrally located north — south trending zone of increased T, S and K. This zone may translate to a channel sand or gravel unit passing through this area. To the east and west of this zone are smaller values of T, S, and K. These zones may translate to overbank deposits lateral to a main channel deposit. However, results of the analytical solution are contrary to the observed geology for the Helipad Site. The geology of this site (Appendix C) shows a higher conducting gravel channel passing through wells 1650, 1656, and 1653 which also is in correlation to a higher observed drawdown. This gravel channel transitions to a lower conductive sand channel that thins to the east of wells 1650, 1656, and 1653 corresponding to a moderate to low observed drawdown. Conclusions Evaluation of pumping test results for determination of the distributions of hydraulic properties for the Helipad Site may be solved using an analytical solution. However, there are several assumptions that must be made for this type of solution to be valid. If these assumptions are made, then distributions of transmissivity, storativity, and hydraulic conductivity may be mapped. This distribution of hydraulic properties is related 150 to the heterogeneity at this site. Heterogeneity is illustrated through a north-south trending zone of increased hydraulic properties which may relate to a channel deposit passing through this site. From previous geologic investigations (Chapter 2), the Arroyo Seco shows signs of a northern migration across the LLNL site. The distributions of hydraulic propertied (Figures D.10 -— D.12) may be evidence of this channel. However, the distributions of T and K fields are contrary to the geology of this site. Assumptions for a confined leaky aquifer were made in the analytical solution. In a heterogeneous alluvial aquifer, many of these assumptions are broken. This may account for the discrepancy between results and the observed geology at the Helipad Site. In an analytical solution, T is inversely proportional to drawdown such that high drawdown is associated with low T. However, geologically high drawdown is related to location within the channel and not necessarily a low T. 151 DRAWDOWN Start date - 4-27-00 at 14:20 Time min) W1650 W1651 W1652 W1653 W1654 W1655 W1656 W1657 0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 50 0.45 0.05 0.16 0.66 0.60 0.19 0.54 0.07 100 0.73 0.11 0.33 0.97 0.97 0.33 0.81 0.17 150 0.90 0.18 0.47 1.16 1.19 0.43 0.97 0.25 200 1.03 0.23 0.58 1.28 1.33 0.51 1.10 0.33 250 1.13 0.28 0.67 1.38 1.44 0.57 1.20 0.39 300 1.21 0.32 0.75 1.46 1.53 0.62 1.28 0.45 350 1.28 0.37 0.81 1.52 1.60 0.67 1.34 0.49 400 1.35 0.40 0.87 1.58 1.67 0.71 1.41 0.54 450 1.41 0.43 0.93 1.63 1.73 0.74 1.46 0.58 500 1.46 0.46 0.98 1.67 1.78 0.77 1.51 0.62 550 1.50 0.49 1.02 1.71 1.82 0.80 1.55 0.66 600 1.54 0.52 1.06 1.74 1.86 0.83 1.58 0.69 650 1.57 0.54 1.09 1.77 1.89 0.84 1.61 0.72 700 1.60 0.56 1.12 1.78 1.91 0.85 1.64 0.74 750 1.62 0.57 1.15 1.80 1.93 0.87 1.66 0.77 800 1.64 0.59 1.17 1.82 1.95 0.88 1.68 0.79 850 1.67 0.61 1.19 1.83 1.97 0.89 1.70 0.80 900 1.69 0.63 1.21 1.85 1.99 0.91 1.72 0.82 950 1.71 0.65 1.23 1.86 2.01 0.92 1.74 0.84 1000 1.73 0.67 1.25 1.87 2.03 0.93 1.75 0.86 1050 1.74 0.69 1.26 1.89 2.04 0.94 1.77 0.88 1100 1.76 0.70 1.28 1.89 2.05 0.95 1.78 0.89 1150 1.77 0.71 1.29 1.90 2.06 0.96 1.78 0.90 1200 1.77 0.71 1.29 1.89 2.06 0.96 1.78 0.91 1250 1.77 0.73 1.30 1.89 2.05 0.96 1.78 0.92 1300 1.77 0.73 1.30 1.88 2.05 0.96 1.78 0.93 1350 1.77 0.74 1.30 1.88 2.05 0.96 1.78 0.93 1400 1.78 0.75 1.30 1.87 2.05 0.96 1.78 0.93 1440 1.78 0.75 1.30 1.87 2.05 0.96 1.78 0.93 Table D. 1: Listed above are accumulative drawdown readings at a specified time afier pumping commencement. Drawdown histories for wells 1650, 1651, 1652, 1653, 1654, 1655, 1656, and 1657 are recorded. 152 Figure D.1: Curve match and analytical solution for well 1650. < eve ~ 7 «= l'l. " ' ‘ 22+ let): u . '. . o 9.99 1.99 2.99 9 3.99 4.99 5.99 9.89 %W.M.J.QI T lleo 991W“ 1.1“?» .9" -9.29 +.*’ 9.99 log 0 109 N(U,r/B) ' + s -1.29 + Type Curve: r/B : 9. 91 .-1.99 Transmissivity: T : 3. 117E- 885 sq M/sec Storativity: SS : 1.571E- 884 + Aquifer Th1ck: b : 1. 829E+888 n Hydraulic Cond: R : 1.?85E- -885 n/sec Leakage Factor: B : 2.594E+ 883 n -2l20 .‘2199 -1.15 -8.15 8.85 1.85 2.85 3.85 91:11:91) (Mata (Durves (Hlatch Em (Drint (IDnits (8)9119 (11}in Figure D.2: Curve match and analytical solution for well 1651. 153 8.88 1.88 2.88 3.88 4.88 5. 88 8.98 1. 88 a.'* + ‘+ + + MWOiM‘AH. -9.92 f" 9.99 log a" log H(H,r/B) ¢“’ 5 -1.82 Type Curve: r/B : 8.81 -1.88 Transmissivity: T : 4.?18E- -885 sq n/sec Storativity: SS : 1.329E- 883 Aquifer Th1ck: b : 1. 524E+888 n Hydraulic Cond: K : 3. 897E 885 n/sec Leakage Factor: B : 2. 513E+883n -2.82 .-2.88 -1.87 -8.87 8.13 1.13 2.13 3.13 SELECT) (Mata (Durues (Matchfliflflflhint (anits (8)9111 (Exit 9.92' + , + + + 1‘ ,, Lo fe-«EWMMT ”" "J”, -9018 .0 h 8 log ’ N(U,r/B) .1 -1.18 Type Curve: r/B : 8. 81 Transmissivity: l : 3. 264E- 885 sq n/seo Storat iuity: Ss : 8. 338 -884 Aquifer Th1ck: b : 1.829E+888 n Hydraulic Cond: H : 1. 785E 885 n/sec 2 18 Leakage Factor: B : L 898E+883 n '-1.59 —9.59 9.42 1.42 2.42 31 8.88 log ‘ -1008 -2.88 42 Figure D.3: Curve match and analytical solution for well 1652. 09. 8.88 1.88 2.88 3.88 4.88 5. 8.76 INJJWL; m .H’wa." r or“ .+' -8.24 +* log I "(H.r/B) + -1.24 Type Curve: r/B : 8. 81 Transmissivity: T : 2. 843E— 885 sq n/sec Storativity: Ss : 5.141E- 884 Aquifer Th1ck: b : 2. 438E+888 n Hydraulic Cond: H : 1.166E- 885 n/sec 2 24 Leakage Factor: B : 1.152E+883 n '-1.99 9.99 1.99 2.99 3.99 4. mm} (Mata (Ourves (Match—EM {Plrint (IDnits (H)eIp (Exit 88 L 88 8.88 log ‘ -1980 1-2.88 Figure D.4: Curve match and analytical solution for well 1653. 154 W) (Dlata (Ourves (Match Him-(Phint (ll)nits (H)efp (Exit -8.28 lo HtU.g/B) -1.28 8.88 log -2.28 Type Curve: Transmissivity:r Storatiuity: Aquifer Thick: Hydraulic Cond: Leakage Factor: .81 .59 .5858- 884 21 .12 .81 3E- 885 sq n/sec 9E+888 M TE- 885 n/sec 8E+883n 4-1.88 -1.89 -8.89 8.91 1.91 2.91 .-2.88 3.91 Figure D.5: Curve match and analytical solution for well 1654. m} (Maia (Duryes (BREW (Erint (1Dnits (Help (Exit Figure D.6: Curve match and analytical solution for well 1655. 155 , o g- 8.88 1.88 2.88 g 3.88 4.88 5.88 1.88 1 1.88 + + +' + + mmhw .4” -9.99 " 9.99 log .4. log H(U,r/B) .+ s -1.88 Type Curve: r/B : 8. 81 4-1.88 Transn1ssiuity: T : 4.948E- 885 sq n/sec Storativity: Ss : 1.?588- 883 Aquifer Thick: b : 1.219E+888 n + Hydraulic Cond: R : 4.853E-885 n/sec Leakage Factor: B : 1.268E+883 n -2.88 .-2.88 -1.3? -8.37 8.63 1.63 2.63 3.63 991111511 (Eata (Eurues (Eatchlfiwwwint (IDnits (Eelp (Exit Figure D.7: Curve match and analytical solution for well 1656. 8.88 1 88 2.88 3 88 4.88 5.88 8.84 ”J“ +¢”1,* 1.88 ",gww . ‘ ° .114“ 10M.” 0 2 .‘ -8.16 , 8.88 + 10 log 919,3/91 1 s -1.16 * Type Curve: r/B : 9. 91 .-1.99 Transnjssivity: T : 3. 418E- 885 sq M/SEO Storativity: Ss : 2.283E- 884 Aquifer Th1ck: h : 1.829E+888 n Hydraulic Cond: H : L 869E- 885 n/sec Leakage Factor: B : 1.988E+883 n -2.16 .-2.88 -8.99 8.81 1.81 2.81 3.81 4.81 SELECT} (Diata (C>urves (H)atch IBEbfiéflééH

rint (H)nits (H)eip (Eixit -8.86 198“”, 9.99 log 3" log H(U,r/B) “1’ s -1.86 Type Curve: r/B : 8. 81 4-1.88 Transn1ssiyity: T : 4. 383E- 885 sq n/sec Storativity: Ss : 7. 9738- 884 Aquifer Th1ck: b : L 524E+888 n Hydraulic Cond: H : 2. 824E- -885 M/sec Leakage Factor: B : 2. 668E+883 M -2.86 1-2.88 -1.74 -8.?4 8.26 1.26 2.26 3.26 Figure D.8: Curve match and analytical solution for well 1657. 156 EELECT) (Nata (Curves (Eatchflflfl’hint (Enits (H)elp (Exit Well C2115: “13311:?” Storativity 1:123:29. cgiydl::t‘il\l'ii:y #:3121312 (meter) (m/day) (meter) 1650 0.01 2.693 1.571e-4 1.829 1.473 2.594e3 1651 0.01 4.076 1.329e-3 1.524 2.675 2.513e3 1652 0.01 2.820 8.338e-4 1.829 1.542 1.890e3 1653 0.01 2.456 5.141e-4 2.438 1.007 1.152e3 1654 0.01 2.240 7.505e-4 1.219 3.502 1.260e3 1655 0.01 4.268 1.750e-3 1.219 1.837 1.010e3 1656 0.01 2.9531 2.203e-4 1.829 1.615 1.908e3 1657 0.01 3.717 7.973e-4 1.524 2.433 2.668e3 Table D2: Summary table of hydraulic parameters from analytic solutions, result are recorded in meter and day units. 157 9.3.2 as 8.99:. .o 8.8.96 8.593 5 8.. .88 «a $2 as. .22 .22 $2 .92 .Nm2 ..2 .82 2.63 a 8..an spouses so; beam 88...... .888 65 a 2.63 8.3.88 .o 8.83 ad 2am... 3111...: (12.1.1 .1 .11. an Ag. 0 .Iflullfl 8 :8: o m._5. .82; 083: 83% , 05:, :38... .m . . 95:80.. =m>> 1%; . 25 23...»... ..z._.. 158 émf was 33 m2?» swab: @5an .2523 m 950% 8:362:5wa mo .895.me 2F .moumimflfimfib mo 92: 39:00 “2.9 ogwi Dumoo ._.Oz . . jug ammoo . o moomE 538... .m E 2.28.85? 8:28 .8382 2.6. 82...: ._z.... 159 63— 23 33 £63 swag—t 9288 658:0 a 9.65 256568 692on .«o 2.255.528 2:. .83?» 286508 09208 mo 92: 5880 A Md oSwE 3 .8815 c an at: o 35% mead u _o omcoo 82 - . 44w; ammou . 0 «BONE 533 .m 39 EofiEwoo 69205 cozaom _mo_§_mc< 25 68:6... ._z.... 160 .me =63 swsofiu $5me 98 58: @2255 358:0 a @505 55:26:00 215.63 mo cacao—52c 25. .835» €283.28 21:26.3 we 92: 58.80 N._. D oni ammoo hoz - . ._._m>> ammoo . o 8on :mcmfi .I 0: 33:03:00 0.323... co_S_ow _mo=>_mc< 25 82.8: 42.... 161 Appendix E Isopach Maps Isopach maps were developed within the central model region to estimate mean channel lengths, widths, and channel orientation for each stratigraphic unit. These parameters were used in the Markov chain model development and geostatistical simulations (Appendix F). Each stratigraphic unit (Aa through F) may contain multiple channels; therefore, an average orientation and width from each stratigraphic unit was used for geostatistical realizations. Paleosol units used average channel orientations and widths from overlying stratigraphic unit since it was assumed that channels formed in these units would cut down into or through underlying paleosol layers. Isopach maps are illustrated in figures E.1 through E.14. 162 2.8 3.. o... 8.. 226% 8.8 8.8... 8.3 8.8.88 8.8.28 .8 .8... 8.282 88.80 ...m 08E... m. .9385 o S .8: o m.._m¥ .. «8% o :32. .m m<=c3 91%; gas. .838. 163 £5. n< 05 mo newton. 5.5: 05 £515 226% Ba .58 d... 8.83.. .2520 mo an... 538. .5380 ”Nm 23$". 2 .288... .. .Iflilul 8 .8... o ":3 331......583 =03. . >mv. . «8% 0.. SEE. .m F 92...: 7%; ans. 50802 164 £5. p< 2.. .0 53.8.. .526. 2.. £55. £263 98 9.3 .0... 3.8%.. .282... .0 9.8 £088. 58:00 ”Wm 05»; m. .281... o 8 .58. o 35$ «08 n. 1:8... 58:00 >mv. Noewh 52.8.. .m N n< .ED 922 comaoa. 165 4.5. m. 05 £515 205% 0:0 0:8 .0... 3.02.0“. .0530 mo 92: £038. 59:00 ”1m. 05%.". on mp .233... 0 fi .8... a w..mv_ 08%.. 53.2. .m m. =5 00.). 200002 166 0.0: O 05 .0 00.98.. 00...... 05 0.5.3 903% 000 .0500 .0... 3.02.0.0 .0000£0 .0 008 £00000. 50.000 ”Wm. 00%... a. .921... 0 8 .8... o 33%.. gnu-2:85.583 =03. . >5. o NSNK O 20:00... .m . 0 .E: 7%; v 00.2 500002 167 0.0: U 0£. .0 00.0.00 .030. 05 0......» A.0>0.w .000 008 .0... 3.00000 .0000£0 .0 00:. £0008. 50.000 6m. 00%... m. .282... a “$1 8 .3... o m4<0m «on. m u .028...— 53:00 >m¥ 0800 00:0... .t N 0 .ED 005. £00000. 168 0.00 D 0... .0 00.000 .0000 0.... 0.5.3 90.6% 000 .0000 .0... 000000.. .000000 .0 00.0 £00000. 0.0.000 "Nam. 0.5m... 0. .281... o S .31. o w..mx 00000 0000.... .m P o .E... 00.). 500000. 169 0.00 D 05 .0 00.000 0.00.0. 05 0.5.3 ..0>0..w 000 0000 .0... 0000000 .0000£0 .0 00.0 £00000. 0.0.000 ”win. 0.5m... 2 .281... o 8_ .8... o m...mv. .1.|.I.. 1...... ....I . . 0800 0 $2.00 .0 m n. :03 7%; 00.). 000000. 171 0.00 m 05 00 00.000 00000 05 0.00.3 C0>8w 00.0 0000 .0..v 00.00000 .000000 ..0 008 000000. 0000000 50m 000M..... «\1\ O. \ a ./ . / _ / .00 ,/ / ace. 0 / x ./ o. l. / ix / / o / Q. './. / o / \ C / / \0 0. .2835 0 00 08: a m._mv_ 008R :20: .m P w :0: 00—2 200000. 172 0.00 m. 05 .0 00.000 0.00.8 05 0.5.3 90>00w 000 0000 . o . a .. . .|I|.|u||Nm|I\ _0 b. \ III, 0. \\ //n. \0 \ o. (a. . o. o. o. G 0 AD 9.. o . o . .0... 00.00000 60000.0 00 000. 000000. 0000000 U. _.m. 00030 0. .2825 0 8 .$5 a u4m¥ weak 0000.2. .0. N m. .05 002 500002 173 0.00 m. 05 .0 00.000 .030. 05 0.5.3 900% 000 0000 . O . O C \/ . \ o\m / \ . / A\ v / / / a O 0 . .V / . o /o. u / / 0.. / / 0 / / a. / — ‘ / / / \\ / a \ / . \o / \ .0... 00.00000 .0805. .0 008 000000. 800000 ”a... 0.0%.... 0. .2825 o _Wfi S 08: o ma<0m .8. 0 0 1:8... 52:8 >9. NQQNR 00:00... .m m m :03 00.2 200002 174 0.00 n. 05 .0 00.000 .0000 05 0.5.3 900% 000 0000 .0... 00.00000 .0805. .0 008 000000. 0.0.000 ”2m. 0.0m... 0. 7.8!... o o... 08.. o m..mx 00000 0000.... .t F n. :03 00.2 200000. 175 0.00 0 05 .0 00.000 .030. 05 0.5.3 c0>00w 000 0000 O o . 0.. o \x/ A. / m / .0 x . .O/ .\oC/ /o. N / 4 8500/ a. o / I. / ./ o/n./ 0.. .U / / o / x O. / \. O a O .0... 0000000 .000000 .0 008 5.0000. 80.000 ”in“. 0.0m... 0. .2685. a on .82. o 53.6 810.5:85.zéao >m¥ 0000.0 2000... .m N n. :03 00.). 200000. 176 Appendix F Transition Probability Geostatistics Introduction Transition probability geostatistics is an indicator geostatistical method. This method can be used to estimate the spatial distribution of user defined categories which are based on geologic information (i.e. well core and geophysical logs). At the Helipad Site transition probability geostatistical models were generated for two conceptual model types — a full transition probability geostatistical model and a stratigraphic transition probability geostatistical model. Outlined in the following sections are steps taken in developing each of these model types. This includes, Markov chain model development, defining stratigraphic units, incorporating geologic information in the form of channel widths, lengths, and orientations, and combining individually modeled stratigraphic units to produce a stratigraphic transition probability geostatistical realization. This outline is divided into two sections based on conceptual model type. Transition Probability Model Development Listed below are the procedures that were taken in developing both transition probability geostatistical simulations. It is important to note that there is a T-PROGS interface within GMS 4.0 that allows you to develop transition probability geostatistical models. However, during the initial development of these model types, GMS 4.0 was not available. Also, development of transition probability models outside of GMS allows the user to have more flexibility in generating realizations in terms of matching Markov 177 chain models to measured data, generating larger realizations, and was necessary for development of our stratigraphic simulations. In model development GAMEAS, GRAFXX, MCMOD, TPSIM, and CHUNK programs were used. Listed below is an outline for procedures in model development. Carle (1999) fully describes this software and data formats required. Procedures for transition probability geostatistical simulation — Hydrofacies categories from core / geophysical well log data were initially discretized to 0.5 meter spacing. From these measurements, the vertical (2- direction) transition probabilities between hydrofacies and proportions of individual hydrofacies were calculated using GAMEAS. The results of this calculation showed that the silty sand hydrofacies had the highest overall proportion (Table F.1). In the conceptual model development, the silty sand hydrofacies was interpreted to fill in around all other categories (cross-section interpretations, Appendix C). Based on this, the silty sand hydrofacies was designated as the background category. Using GRAFXX, the 1-D transition probabilities were plotted as a matrix of graphs. This was done to asses the data quality and interpret juxtaposition relationships between hydrofacies. MCMOD was used to develop a vertical Markov chain model fit to measured transition probabilities. GRAFXX was then used again to plot both the Markov chain model and the measured data. This Markov chain model was then adjusted to best fit the measured data through manipulation of the 178 embedded transition probabilities (Figure F1). The resulting vertical embedded transition probabilities are listed in Table F .2. The horizontal (x and y-directions) Markov chain models were deve10ped using geologic reasoning and the application of Walther’s law, in that the lateral juxtaposition tendencies are assumed to have the same statistics as the vertical upward juxtaposition tendencies. Lateral facies distributions were also assumed to be symmetrical. Horizontal juxtaposition values were then adjusted so that embedded transition probabilities were no longer negative. At this point, mean channel lengths and widths were incorporated. Lengths and widths of channel hydrofacies were measured from isopach maps for the E stratigraphic unit (Appendix E) and were incorporated into the horizontal embedded transition probabilities at an azimuth of 340 degrees (Table F .2). Within the TPSIM parameter file, the model dimensions as well as the grid dimensions were assigned to match the grid dimensions developed for the groundwater numerical model (described in Appendix G). TPSIM was run to generate multiple realizations for the spatial distribution of hydrofacies. There were 26 realizations generated, one transition probability geostatistical simulation and 25 stratigraphic transition probability geostatistical simulations. Cell dimensions used to produce this realization were 2 x 2 x 0.5 meter (north- south, east-west, and vertical directions respectively) in dimension. Full simulation dimensions are 362 meters east to west, 310 meters north to south, and 39 meters vertically. In total each realization contains 2,188,290 cells. 179 Procedures for the stratigraphic transition probability geostatistical simulation — Using laterally-extensive paleosol markers, the aquifer at the Helipad Site was subdivided into several stratigraphic units or zones (described in chapters 2 and 3, Figure 3.2). Top and bottom surfaces were generated from these correlations and were modeled across the study area using an inverse distance linear algorithm. As a result, thirteen zones were identified. These included six paleosol zones (Ab through F) and seven stratigraphic zones (Aa through F). However, only those zones that were below the water table were modeled (C through F ). For this modeling approach, individual stratigraphic and paleosol zones were modeled using different Markov chain models. Hydrofacies categories from core / geophysical well log data were discretized to 0.5 meter spacing. However, the vertical (z-direction) transition probabilities were calculated over a discrete interval for each zone. These intervals were defined by top and bottom elevations from modeled paleosol surfaces. Transition probabilities and proportions for each zone were calculated with GAMEAS. The results of these calculations showed that within each stratigraphic zone the silty sand hydrofacies consistently had the highest proportions (Table F. 1). For stratigraphic zones (C, D, E and F), the silty sand hydrofacies was assigned as the background category. In paleosol zones, the paleosol hydrofacies consistently had the highest proportions (Table R1) and was set as the background category for each of these zones (D, E, and F). 180 GRAF XX was used to plot the 1-D transition probabilities as a matrix of graphs for each zone. Again, this was done to asses the data quality and interpret juxtaposition relationships between hydrofacies. MCMOD was used to develop vertical Markov chain models for each individual zone. GRAFXX was then used again to superimpose these Markov chain models over the measured data for corresponding stratigraphic and paleosol zones. Markov chain models were then adjusted for each zone to best fit the measured data through manipulation of embedded transition probabilities (GRAFFXX plots for each zone resemble Figure RI). The resulting vertical embedded transition probabilities are listed in Tables F .3 through F.9. Using the same reasoning as stated above, off-diagonal vertical upward embedded transition probabilities were assigned to horizontal embedded transition probabilities for each zone. For these models, symmetry was assumed. These values were then adjusted to so that embedded transition probabilities were no longer negative. Mean channel lengths and widths were incorporated into each zone (paleosol zones used mean channel lengths and widths from overlying stratigraphic zones). Lengths and widths of channel hydrofacies were measured from isopach maps for multiple channels within each zone (Appendix E). These were incorporated into corresponding horizontal embedded transition probabilities (Tables F.3 through F .9). 181 For this modeling approach, each zone was simulated over the entire simulation block volume (even though conditioning data only existed over a limited portion of this block). TPSIM was used to generate these simulations. Following methods of Weissmann and Fogg (1999), simulations generated for each of the individual zones (C through F) were merged to create a final whole aquifer realization of the Helipad Site (Figure F .2). Out of each simulated zone, only those specific cells bounded between corresponding paleosol top and bottom surfaces were used. This was accomplished using a FORTRAN Code (Appendix I). Following this methodology, 25 total realizations were generated to resolve the spatial distribution of hydrofacies for the Helipad Site. Each of these realizations is illustrated in Figures F.3 through F.8. Conclusions Each of the resulting realizations (Figures F.2 through F .8) reflects aspects of the physical heterogeneity that exist in an alluvial fan system. These geostatistical realizations reflect juxtaposition tendencies, proportions and channel length, width, and orientations of the real system. Also, this modeling approach tends to preserve a fining upward character observed in well logs. Gravel hydrofacies are observed to be located beneath sand hydrofacies, similarly, sand hydrofacies are observed to be below silty sand hydrofacies. Using a stratigraphic transition probability geostatistical approach, channel orientations associated with different stratigraphic zones are better preserved. Also, by 182 simulating each stratigraphic zone separately, unconfonnities are preserved where hydrofacies are discontinuous across stratigraphic boundaries preserving the larger scale heterogeneity of the system. In total, one transition probability geostatistical realization and twenty-five stratigraphic transition probability geostatistical realizations were developed. Distributions of hydrofacies for each of these realizations were used in numerical groundwater models (described in Appendix G). 183 Global Proportions: Gravel Sand Silty Sand Paleosol Hydrofacies Hydrofacies Hydrofacies Hydrofacies Sinfufigion 0.24 0.12 0.52 0.13 C zone 0.26 0.12 0.59 0.04 D paleosol 0.40 0.03 0.08 0.50 D zone 0.15 0.19 0.62 0.036 E paleosol 0.22 0.04 - 0.74 E zone 0.08 0.19 0.64 0.10 F paleosol 0.11 0.04 - 0.85 F zone 0.22 0.08 0.65 0.05 Table F. 1: Proportions of hydrofacies for the TPG simulation and for each stratigraphic zone of the layered simulation. E and F paleosol zones lack the silly sand hydrofacies and are modeled excluding this category. 184 Vertical Markov Chain Model Unlayered Simulation Helipad Site, LLNL Gravel Sand Silty Sand Paleosol Gravel Sand Jransition Probabili 111111 fi‘l‘i v ‘ LA ‘ '3 § . U3 ° > -+ £3 < , 1 ‘3 if. . . . 3° - ’ ~ 0 . . Q < .. ‘3 i 3 j 3 0 5': j _. 0 000°, 0 q . . ° 9‘ 00 -;/’OWUOvvv-5-T ioooo iooofi, + 00 0000 , 0 3 6 L89 ("7) Measured modeled 0000000 Figure F .1: Vertical Markov chain model fit to measured facies data for the transition probability geostatistical simulation modeling approach. 185 Vertical (z-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=1 .013 0.1000 0.7000 0.2000 Sand 0.0010 L=1 .200 0.8660 0. 1000 Silty Sand 0.4632 0.1716 L=1.490 0.3651 Paleosol 0.4000 0. 1 000 0.5000 L=0.700 Horizontal (x-direction) embedded transition probabilities Gravel Sand Silly Sand Paleosol Gravel L=5.000 0.0974 0.8747 0.0278 Sand 0.1900 L=5.000 0.7958 0.0142 Silty Sand 0.6601 0.3079 L=8.223 0.0319 Paleosol 0.3600 0.0940 0.5460 L=35.000 Horizontal (y-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=50.000 0.0974 0.7074 0.1950 Sand 0.1900 L=50.000 0.7106 0.0993 Silty Sand 0.5172 0.2663 L=79.660 0.2163 Paleosol 0.3600 0.0940 0.5460 L=50.000 Table F.2: Embedded transition probability matrices for the transition probability geostatistical simulation. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) 186 Vertical (z-direction) embedded transition probabilities Gravel Sand Silly Sand Paleosol Gravel L=.600 0.0620 0.8970 0.041 0 Sand 0.0001 L=0.620 0.9998 0.0001 Silty Sand 0.6507 0.2402 L=0.987 0.1089 Paleosol 0.5 1 00 0.2600 0.2300 =O.500 Horizontal (x-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=7.000 0.0597 0.9379 0.0022 Sand 0.1 300 L=7.000 0.8651 0.0048 Silty Sand 0.6816 0.2890 L=1 1.598 0.0293 Paleosol 0.0500 0.0500 0.9000 L=25.000 Horizontal (y-direction) embedded transition probabilities Gravel Sand Silly Sand Paleosol Gravel L=60.000 0.0597 0.9338 0.0064 Sand 0. 1300 L=60.000 0.8560 0.0139 Silly Sand 0.6472 0.2728 L=94.824 0.0798 Paleosol 0.0500 0.0500 0.9000 L=75.000 Table F.3: Embedded transition probability matrices for the C zone. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) 187 Vertical (z-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=3.000 0.0500 0.0500 0.9000 Sand 0.8000 L=3.000 0.0500 0.1500 Silty Sand 0.2666 0.0166 L=3.000 0.3651 Paleosol 0.8623 0.0898 0.5000 L=3.593 Horizontal (x-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=7.000 0.0500 0.0262 0.9237 Sand 0.8000 L=7.000 0.0210 0.1790 Silty Sand 0.0400 0.0020 L=2.000 0.9580 Paleosol 0.5907 0.0072 0.4020 L=35.596 Horizontal (y-direction) embedded transition probabilities Gravel Sand Silly Sand Paleosol Gravel L=50.000 0.0350 0.0375 0.9275 Sand 0.2800 L=25.000 0.04875 0.67125 Silty Sand 0.0800 0.0130 L=20.000 0.9070 Paleosol 0.6456 0.0584 0.2959 L=43.506 Table F.4: Embedded transition probability matrices for the D paleosol. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) 188 Vertical (z-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=0.950 0.2100 0.7899 0.000001 Sand 0.0510 L=0.800 0.9489 0.000001 Silty Sand 0.3397 0.4304 L=1.505 0.1527 Paleosol 0.0570 0.2630 0.6800 L=0.3 80 Horizontal (rt-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=5.000 0.2141 0.7859 0.000001 Sand 0. l 700 L=5 .000 0.8300 0.00000 Silly Sand 0.4199 0.5585 L=11.086 0.0214 Paleosol 0.000002 0.000001 0.9999 L=30.000 Horizontal (y-direction) embedded transition probabilities Gravel Sand Silly Sand Paleosol Gravel L=50.000 0.2141 0.7859 0.000001 OSand 0.1700 L=50.000 0.8299 0.000001 Silty Sand 0.3793 0.5047 L=l00.124 0.1162 Paleosol 0.000005 0.000007 0.9999 L=50.000 Table F.5: Embedded transition probability matrices for the D zone. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) 189 Vertical (z-direction) embedded transition probabilities Gravel Sand Paleosol Gravel L=3.000 0.0500 0.9500 Sand 0.2750 L=3.000 0.7250 Paleosol 0.8781 0.1218 L=9.327 Vertical (rt-direction) embedded transition probabilities Gravel Sand Paleosol Gravel L=5.000 0.0491 0.9509 Sand 0.2700 L=5.000 0.7300 Paleosol 0.8775 0.1225 L=15.520 Vertical (y-direction) embedded transition probabilities Gravel Sand Paleosol Gravel L=50.000 0. 1000 0.9600 Sand 0.1 100 L=25.000 0.8900 Paleosol 0.7478 0.2521 L=131.019 Table F.6: Embedded transition probability matrices for the E paleosol. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) 190 Vertical (z-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=0.530 0.9000 0.0200 0.0800 Sand 0.0070 L=0.5 50 0.9430 0.0500 Silty Sand 0.2739 0.3848 L=1.721 0.3412 Paleosol 0.2800 0.4000 0.3200 L=0.650 Horizontal (rt-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=8.000 0.8789 0.0229 0.0981 Sand 0.3700 L=8.000 0.5680 0.06196 Silty Sand 0.0165 0.9741 L=47.168 0.0093 Paleosol 0.3800 0.5700 0.0500 L=40.000 Horizontal (y-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=50.000 0.7126 0.2335 0.0538 Sand 0.3000 L=50.000 0.6637 0.0362 Silty Sand 0.0850 0.5741 L=148.694 0.3408 Paleosol 0.0500 0.0800 0.8700 L=60.000 Table F .7: Embedded transition probability matrices for the E zone. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) 191 Vertical (z-direction) embedded transition probabilities Gravel Sand Paleosol Gravel L=3.000 0.0500 0.9500 Sand 0. 1 501 L=3 .000 0.8498 Paleosol 0.7704 0.2295 L=18.269 Vertical (x-direction) embedded transition probabilities Gravel Sand Paleosol Gravel L=8.000 0.0599 0.9400 Sand [ 0.1800 L=8.000 0.8200 Paleosol 0.7748 0.2251 L=49.516 Vertical (y-direction) embedded transition probabilities Gravel Sand Paleosol Gravel L=50.000 0.0466 0.9533 Sand [ 0.0700 L=25.000 0.9300 J Paleosol 0.6061 0.3938 L=238.70 Table F.8: Embedded transition probability matrices for the F paleosol. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) 192 Vertical (z-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=1 .35 0.1300 0.8600 0.0100 Sand 0.0010 L=0.520 0.8990 0.1000 Silty Sand 0.4550 0.3434 L=1.770 0.2015 Paleosol 0.001 0 0.0690 0.9300 L=0.560 Horizontal (x-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=7.000 0.1277 0.8622 0.0100 Sand 0.3600 L=7.000 0.5405 0.0995 Silly Sand 0.8123 0.1807 L=18.976 0.0069 Paleosol 0.1900 0.6700 0.1400 L=30.000 Horizontal (y-direction) embedded transition probabilities Gravel Sand Silty Sand Paleosol Gravel L=50.000 0.093 1 0.8955 0.01 13 Sand 0.2100 L=40.000 0.6423 0.1476 Silty Sand 0.6736 0.2143 L=108.212 0.1 121 Paleosol 0.0500 0.2900 0.6600 L=50.000 Table F .9: Embedded transition probability matrices for the F zone. These matrices are read as transition probabilities from the row hydrofacies to the column hydrofacies. (Labels: L, mean length. Bold numbers indicate background category with computed values listed in the table.) 193 TPG Realizaton Helipad Site gravel sandy silt [:1 Sand I paleosol Figure F2: Full realization following methods for an TPG simulation approach. Four categories are used (gravel, sand, silty sand, and paleosol) to characterize the geology at the Helipad Study Site, LLNL. 194 .830 E 68:82“ owns: .25 39—03 05 Ha mofifiofims mo :oaanwE Evan 05 SH 533:3. 20:3 :25 use 3329 8 ~050on uowBE 2a Awe—Hon m smack: UV mecca 3.088 38833 ”mm Emmi / _852 d m— 195 Realization 1 Realization 2 LLNL LLNL Realization 3 Realization 4 LLNL LLNL Realization 5 LLNL gravel E] sand sandy silt I paleosol Figure F .4: Full realization (1 through 5) following methods for a zoned simulation approach. Four categories are used (gravel, sand, silty sand, and paleosol) to characterize the geology at the Helipad Study Site, LLNL. 196 Realization 6 Realization 7 LLNL LLNL Realization 8 Realization 9 LLNL LLNL Realization 10 LLNL gravel D sand sandy silt I paleosol Figure F5: Full realization (6 through 10) following methods for a zoned simulation approach. Four categories are used (gravel, sand, silty sand, and paleosol) to characterize the geology at the Helipad Study Site, LLNL. 197 Realization 12 LLNL Realization 11 LLNL Realization 14 Realization 13 LLNL LLNL Realization 15 LLNL sandy silt I paleosol Figure F6: Full realization (11 through 15) following methods for a zoned simulation approach. Four categories are used (gravel, sand, silty sand, and paleosol) to characterize the geology at the Helipad Study Site, LLNL. 198 Realization 16 Realization 17 LLNL LLNL Realization 1 8 Realization 19 LLNL LLNL Realization 20 LLNL gravel ., I] sand sandy silt I paleosol Figure F.7: Full realization (16 through 20) following methods for a zoned simulation approach. Four categories are used (gravel, sand, silty sand, and paleosol) to characterize the geology at the Helipad Study Site, LLNL. 199 Realization 21 Realization 22 LLNL LLNL :P Realization 23 Realization 24 LLNL LLNL Realization 25 LLNL gravel [j sand sandy silt I paleosol Figure F8: Full realization (21 through 25) following methods for a zoned simulation approach. Four categories are used (gravel, sand, silty sand, and paleosol) to characterize the geology at the Helipad Study Site, LLNL. 200 Appendix G GMS Model Development Introduction A three-dimensional groundwater flow model, MODFLOW-2000 (Harbaugh et al., 2000), was applied for computation of steady state and transient heads for the Helipad study area, Lawrence Livermore National Laboratory (LLNL). The following sections are a discussion of groundwater flow model development for each of the four numerical models used in evaluation of pumping test results at the Helipad Site. This includes an outline for the steps taken in groundwater flow model development, boundary condition assumptions made, and a brief discussion of optimization routines, model calibration, and hydraulic properties incorporated for solutions to head distributions. Model development Listed below are procedures used for grid development and incorporation of geostatistical information in the form of a three-dimensional distribution of hydrofacies (discussed in Appendix F). In model development, Groundwater Modeling Software version 4.0 (Brigham Young University, 2000) was used to write files needed by MODFLOW-2000 for computation. Due to the large size of these models, and the inability of GMS to calculate heads at observation wells, computation of head distributions was run from a command prompt outside of GMS 4.0. Modifications were made to some of these files as needed to acquirer head solutions. Procedures for groundwater flow model development — 201 Grid — 0 Grid and cell dimensions were set equal to transition probability geostatistical simulations. Cell dimensions have a maximum size of 2 x 2 x 0.5 meters across an area of 362 meters (east-west) by 310 meters (north-south) by 39 meters (vertical direction). The origin for the model area was set at 4239.5 (x), 2083.5 (y), and 124.5 (2). This origin placed well 1654 at the center of the grid. Wells locations — 0 Prior to inputting well locations, well coordinates within the study area wells were rotated around an origin. Well locations were rotated 20 degrees clockwise around the origin. This was done to align the groundwater flow model with the principle groundwater flow direction observed at this location (Figure 2.8). Grid refinement — O The grid cell size was refined around pumping wells to a minimum cell size of 0.15 meters with a bias of 1.2 to better resolve head distributions near pumping wells. This was also done to minimize numerical error that is generally associated with pumping wells. In this refined grid, hydrofacies were assigned to each cell based on the location of cell center. If the coordinate of a cell center fell within the boundary of a hydrofacies, that hydrofacies would be assigned to that cell. 202 0 Geologic information (Homogeneous-Layered and Homogeneous-Layered with paleosol) — 0 Top and bottom paleosol surfaces were generated with Rockware 2002 software (RockWorks Inc., 2002) as a result of paleosol correlations made across the study area (appendix C). o A tabular 2-D scatter set was exported from Rockworks 2002 for each paleosol top and bottom surface. These scatter sets were then imported into GMS and interpolated using an inverse distance method algorithm. ' For the Homogeneous-Layered model, top paleosol surfaces were incorporated to segregate homogeneous layers. ' For the Homogeneous-Layered model with paleosol layers, both top and bottom paleosol surfaces scatter point sets were interpolated to separate model layers. 0 Geologic information (transition probability geostatistics) — 0 Using GMS to generate a generic .mfs file for the Helipad Site grid domain, material ID’s (hydrofacies categories) were inserted under the DMAT heading in place of the 1 (this heading must also be changed to MAT). Material [D’s are inserted as one continuous column of numbers (output from FORTRAN code). Note, output file from FORTAN code assigned a negative value to material ID’s that were used as conditioning points for transition probability geostatistical simulations. This file is then saved and GMS reopened (for this 203 procedure to work properly, the .mst file must be deleted, this will not affect the overall model). 0 Using a FORTRAN code (Appendix I), simulation .bgr files were converted to a list of material ID’s (accounting for sequential order). 0 Pumping — o Pumping from wells 1551 and 1552 was set over a screened interval between appropriate top and bottom elevations. 0 Wells 1551 and 1552 had a pumping rate of 1633de and 2.27m3/d, respectfully. These pumping rates were held constant throughout the duration of the pumping test at the Helipad Site. 0 For solutions to head distributions, a two stress period simulation was run. The first stress period was manually set to steady state within the .dis file. The second stress period was set to transient with a 1.2 multiplier over 25 time increments 0 Observation wells — 0 Due to the large size of these models, an observation well file was developed following GMS observation-process documentation (Hill et al., 2000) for each of the 1200, 1500, and 1600 series observation wells. This file was set to record head at observation locations for each time step of each stress period. 0 Observation points were placed within the highest conductive material with the screened interval (the greatest amount of drawdown was observed within these units). 204 0 Model Boundaries 0 Eastern and western boundaries were assigned as constant head boundaries with northern and southern boundaries assigned as no-flow boundaries. This created a west trending flow gradient similar to that observed at the study area. 0 Modifications to generic model - 0 As a result of scaling material ID’s to fit the refined grid, the location of conditioned points used in generation of geostatistical realizations were slightly shifted (a few cells in to the right-left or up-down). Observation point locations were adjusted to account for this discrepancy. 0 At the constant head boundaries, cells became dry while Modflow calculated heads. This resulted in the simulation failing to converge. To correct for this problem, constant head cells were manually assigned to a material 3 ID (silty sand hydrofacies). This prevented cells from drying out during calculation of heads. Assumptions . In development of any groundwater model, there are certain assumptions about the system that must be made for simplification of the problem. The following is a list of assumptions that were made in development of groundwater models used for evaluation of pumping test conducted at the Helipad Site: 0 Boundaries conditions — 205 o The climate associated with this region is arid; the annual precipitation recorded at this location is 34 cm/year (U .S. Department of Energy, 1998). The pumping test was conducted fi'om late Aril to early May. Given a relatively low recorded average precipitation and the pumping test was conducted during the middle to late spring months, precipitation as a recharge into the groundwater model is set to zero for the steady state simulation and over course of the observed pumping test period. 0 The two streams located in the vicinity of the LLNL facility are the Arroyo Seco and Arroyo Las Positas. The Arroyo Seco is located well to the south west of the LLNL property (approximately 1 mile from the Helipad Site). The Arroyo Las Positas has been redirected following the eastern and northern portion of the LLNL property approximately 0.5 miles from the study site. Both of these streams are intermittent and are assumed to have no major influence to the ten day pumping test conducted at the Helipad Site. Drainage streams across the LLNL property are cement lined and assumed to be disconnected from the aquifer system. There is a large retention pond in close proximity to the Helipad Site. However, this retention pond is lined and assumed to be disconnected from the aquifer system. 0 The LLNL site is currently undergoing remediation, primary remedial efforts at this site involve pump and treat methods. Across the LLNL property there are a series of pumping wells extraction water out of 206 multiple stratigraphic horizons at relatively low rates. For the purpose of pumping test evaluation at the Helipad Site, drawdown histories were analyzed. It was assumed that pumping from wells other then 1551 and 15 52 would not affect these results. Drawdown histories for observation wells at the study site show an oscillation. This oscillation is attributed to night and day power consumption. More power is available during night hours allowing pumping rates to be slightly elevated. Drawdown histories were therefore evaluated during the first 24 hours of operation, after this it was assumed that the system came to equilibrium where drawdown oscillated around a mean value. Cell dimensions used in modeling efforts were assumed to be small enough to capture finer scale heterogeneities associated with alluvial fan architecture. Optimization Two parameter estimation routines were evaluated in optimizing our groundwater models. These included the MODFLOW-2000 PES (Hill et al., 2000) and the PEST Doherty et al. (2000) routines. Both of these methods rely on the observation process which calculates head values for comparison with measured or observed head values. To quantify this comparison, a variety of statistics are calculated including weighted least- squares objective fimction. A series of output files are written for graphical analysis of these statistics. While running the observation process, sensitivities are simultaneously 207 calculated using a sensitivity—equation. This process diagnoses inadequate data, identifies parameters that may not be optimized, and evaluates the utility of proposed new data (Hill et al., 2000). Finally, the parameter estimation process used a modified Gauss- Newton method to adjust variable values defined by the user. Executables for both the PBS and PEST routines require a set of input files which are described in detail in the User Guide to the Observation, Sensitivity, and Parameter- Estimation Processes documentation (Hill et al., 2000). As a result of running these programs, user defined variable parameters, such as hydraulic conductivity and storage coefficients are adjusted following a set of statistics to minimize the weighted sum of squares. For our numerical groundwater models, both routines were evaluated for optimization of material properties of geostatistical realizations. Hydraulic conductivity and storativity values for channel gravel and sand, and floodplain silty sands were set for the optimization routines. However, there were problems with using the PES and PEST routines. Due to the large size of the model, the optimization process required between a few days to weeks to run. Between every iteration, user defined variable parameters are updated. The model is rerun for a solution to the weighted sum of squares. Between iterations, the pumping well file was not updated for new values used for material ID’s. Since the well file is calculated for material properties within each cell, failing to update this well file resulted in false weighted sum of square values rendering the optimization routines invalid. This error was observed for both PES and PEST routines. Because of this error, PBS and PEST could not be used. Therefore groundwater models were manually calibrated to match observed drawdown values. 208 Calibration Initial values (hydraulic conductivity and storativity) used for gravel, sand, silty sand, and paleosol materials were obtained from calibrated model results from a nearby study site. Carle et al. (1998) used similar material types (channel, debris, overbank deposits) in a transition probability geostatistical model from a nearby study site at LLNL, these values are listed in Table G. 1. Using the spatial distribution of hydrofacies from realization one, hydraulic conductivity and storativity values for channel, sand, and silty sand materials were then calibrated to match drawdown histories observed in the 1650 series wells. Final calibrated values from this simulation were then used in the transition probability geostatistical simulation. Homogeneous-Layered models were calibrated individually by assigning reasonable generic hydraulic properties to each model layer. Anisotropic values as well as hydraulic properties were adjusted to best fit drawdown histories for the Homogeneous -Layered model. Only hydraulic properties were adjusted for the Homogeneous-Layered model with paleosol layers, paleosol layers are assigned a relatively low conductivity value which negates the need to assign anisotropy to prevent vertical fluid flow across stratigraphic zones. Table G.2 lists final calibrated K and S values used in each modeling approach. Conclusions A three-dimensional groundwater flow model was applied for computation of steady state and transient heads for the Helipad study area, Lawrence Livermore National Laboratory (LLNL). Each modeling approach used the same methods for groundwater 209 model development. Following a set of assumptions, solutions for head distributions for each modeling approach was returned. To better match drawdown histories observed at the study site, both parameter-optimization and calibration techniques were used. However, optimization routines (PBS and PEST) failed to update pumping well files which greatly impacted end results. Therefore, this method was only used as an evaluation tool for calibration of these models. Those materials that are most sensitive were adjusted. The results of calibration produced drawdown histories that are a relatively close match to observed values (see appendix H and discussion in chapter 3). Due to time limitations, optimization routines could not be further explored. 210 Homgeneous-Layered Model: Layers Hydraulic Horizontal Vertical Specific Storage Conductivity (m/d) Anisotropy Anisotropy (l/m) C zone 1 1 0.1 0.00018 D zone 1 1 0.1 0.00021 E zone 1 1 0.1 0.00026 F zone 1 l 0.1 0.0002 Lower Livermore l 1 0.1 0.0003 Homogeneous-Layered Model with Paleosol Layers Layers Hydraulic Horizontal Vertical Specific Storage Conductivity (m/d) Anisotropy Anisotropy (1/m) Paleosol layers 0.00004628 1 1 0.0007 C zone 1 1.8 1 0.00018 D zone 1 1.8 1 0.00021 E zone 1 1.8 1 0.00026 F zone 1 1.8 1 0.002 Lower Livermore l 1.8 1 0.0003 Transition Probability Geostatistical Simulation: Hy do facies Hydraulic Horizontal Vertical Specific Storage Conductivity (in/d) Anisotropy Anisotropy (l/m) Gravel 5.18 l 1 0.0001 Sand 2 1 1 0.0002 Silty Sand 0.01728 1 1 0.0003 Paleosol 0.00004628 1 1 0.0007 Lower Livermore 0.01728 1 1 0.0003 Table G. 1: Listed above are initial hydraulic pr0perties assigned to either layers or material (hydrofacies) used in each of the four modeling approaches. Values listed are initial values incorporated into each modeling approach. 211 Homogeneous-Layered Model: Layers Hydraulic Horizontal Vertical Specific Storage Conductivity (m/d) Anisotr0py Anisotropy (1/m) C zone 1 1 0.1 0.00018 D zone 1 1 0.1 0.00021 E zone 0.25614 4.47 0.2546 0.0007523 F zone 1 1 0.1 0.0002 Lower Livermore 1 1 0.1 0.0003 Homogeneous-Layered Model with Paleosol Layers Layers Hydraulic Horizontal Vertical Specific Storage Conductivity (m/d) Anisotropy Anisotropy (1/m) Paleosol layers 0.00004628 1 1 0.0007 C zone 1 ' 1.8 1 0.00018 D zone 1 1.8 1 0.00021 E zone 0.5 1.8 1 0.00005 F zone 1.8 1 0.002 Lower Livermore 1.8 1 0.0003 Transtion Probability Geostatistical Simulation: Hy do facies Hydraulic Horizontal Vertical Specific Storage Conductivity (tn/d) Anisotropy Anisotropy (l/mL Gravel 30 1 1 0.00005 Sand 2 1 1 0.0002 Silty Sand 0.009 1 1 0.0003 Paleosol 0.00004628 1 1 0.0007 Lower Livermore 0.01728 1 1 0.0003 Table G.2: Listed above are hydraulic properties assigned to either layers or material (hydrofacies) used in each of the four modeling approaches. Values listed have been calibrated for to best match drawdown histories observed (for comparison purpose, geostatistical models all used the same values calibrated from realization one). 212 Appendix H Drawdown Results This appendix contains drawdown histories for the observed pumping test (Figure H. 1) as well as pumping test results from each of the four conceptual models (drawdown at initial time of pumping are normalized to zero) — the Homogeneous-Layered conceptual model, the Homogeneous-Layered conceptual model that incorporated low hydraulically conductive paleosol layers, the transition probability geostatistical conceptual model, and the stratigraphic transition probability geostatistical conceptual model. These conceptual models were evaluated within GMS 4.0 (Brigham Young University, 2000) and used Moflow-2000 (Harbaugh et al., 2000) to solve for head distributions at observation wells. Calculations of head distributions for the models were solved in a two step manor. The first step solved for a steady state head distribution across the model area without pumping. The second step calculated head distributions for 25 time steps using a 1.2 multiplier for a steady pumping rate (described in Appendix G). The difference between steady state heads and head values during the transient Modflow simulation resulted in drawdown histories at each observation point. Figures H.2 through H.3O are drawdown histories simulated with Modflow-2000 (Harbaugh etal., 2000). Also, map views are presented for comparison of drawdown result for each conceptual model type. Overall, realizations XX and XX preformed the best; realizations xx xx did not perform as well. For those realizations that preformed well, it was characteristics of discrete gravel bodies (~8m by 20m) within elongate sand channels (~15m by 40m) were 213 observed. Channel bodies that were over elongate resulted in lower drawdown, those realizations that show very discrete channel bodies resulted in high drawdown. 214 Observed Drawdown Helipad Site 2.5 1 I 1 l 2 _....: .................. .0...}; ........................ “:1 4...... ..:.-.d"0--.---§------.----." w ,1 1:61-r-bww A a...:‘ 0. . o 5"": ' ‘3‘ : E1 5 r- ............. ..::.:.‘......’. 5.:{f ............. ................................. _4 C 0: : ’. ' : /’ : i -- —- ‘- g ' ’.s’ s , ..—:«- “‘3' 3 A 2, -' a = a: ................ 1 ................................. _ g 1* """ 11 " """ i """"" Jar; _____. Q ”P 5” _.. _' 4— .. O — . [.3 a , ’ t l . 0’ E 0.5 _ "HIM”. ........ ................................................... J > I z s 0’ é 3 "IF‘ I I ‘J E _- -_ _ 0 . ........... s- ...... .-..- ..... .-....-...? ....... '.' ..... a 1 1 1 1 0 0.2 0.4 0.6 0.8 1 Time (day) —OBS w1250 - —OBS w1255 —+—OBS w1655 -— OBS w1251 --O- OBS w1650 -a— OBS w1656 — - OBS w1252 --8- CBS w1652 -O— OBS w1657 ----- OBS w1253 --<>-- OBS w1653 ----- OBS w1254 «~96- OBS w1654 Figure H.l: Observed drawdown histories for 1650 series wells and 1250 series wells are depicted for the Helipad Site (1650 series wells are screened within HSU 3, 1250 wells are Screen across HSU 4). As a result of pumping (1650 series wells) there is little to no observed drawdown in1250 series wells. 215 Drawdown Homogeneous-Layered 2.5 Drawdown (m) Observed Drawdown Helipad Site 1 ' i ’ 0.4 0.6 0.8 1 Tums (day) Time (day) — w1251 -— w1253 — - w1255 --e-- w1650 - - El- - w1662 - o— w1653 -->(- w1654 --+— w1655 «an w1656 «Ou- W1657 Figure H. 2: Simulated drawdown histories for 1650 series and 1250 series wells within the Homogeneous-Layered conceptual model. Simulated results show minor drawdown in1250 series wells. 216 Drawdown Homogeneous-Layered with Paleosols 2.5 2.5 Observed Drawdown Helipad Site Drawdown (m) Drawdown (m) Time (day) 0 A“ . —‘_‘“ "...—rt m 1.". l l J 0 0.2 0.4 0.6 0.8 1 Time (day) —— w1251 - +— -— w1253 -->(- — — w1255 --+- --e-- w1650 "er-- - - EJ- - w1662 "-0-" w1653 w1654 w1655 w1656 W1657 Figure H.3: Simulated drawdown histories for 1650 series and 1250 series wells within the Homogeneous-Layered conceptual model with paleosol layers. Simulated results show minor drawdown in1250 series wells. 217 Drawdown Observed Drawdown Transition Probability Geostatistics 2 5 Helipad Site _I (I! _I 0'! .A Drawdown (m) Drawdown (m) .0 01 .° 01 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - +— w1653 -— w1253 --><- w1654 — — w1255 --+- w1655 --e-- w1650 "an w1656 - -a- - w1662 m... w1657 Figure H.4a: Simulated drawdown histories for 1650 series and 1250 series wells within the transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Transition Probability Geostatistical Simulation gravel E] sand sandy silt I paleosol Figure H. 4b: Map view of the spatial distnbution of hydrofacies for the transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 218 Drawdown Observed Drawdown Simulation 01 ' ' 2 5 , f 2 5 Helipad site 2 _-R‘ :J ----- , ...... - ’0 ; _ ';;‘ ’ w .0— -' " ' w-" $1.5 '- I’V-H ’f'f-:.:_..-F' o----,---..--.,-..-. _l E1. g ; ..l . v s - ' E a 1 g 5 a o 0'5 0.5 0 1" 1 1 ° °‘ 014 0'6 0'8 1 o 0.2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - .— w1653 -— w1253 -->(- w1654 — — w1255 --+— w1655 --e-- w1650 no" w1656 . .51. . w1662 --.-— w1657 Figure H.5a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical I gravel E] sand sandy silt I paleosol he m-- .i Figure H.5b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 219 Drawdown Observed Drawdown 2.5 Simulation 02 2 5 Helipad Site 2 2 51'5 31.5 5 s 3 s E 1 a 1 O 5 .° 01 .0 on 0 0.2 0.4 0.6 0.8 1 Time (day) 0 0.2 0.4 0.6 0.8 1 Time (day) — w1251 - .— w1653 -— w1253 -—>(- w1654 - — w1255 --+— w1655 --e-- w1650 “on w1656 a - w1662 ---O-- w1657 Figure H.6a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 02 gravel [j sand sandy silt I paleosol Figure H. 6b. Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24"1 model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 220 Drawdown Observed Drawdown 2-5 1 s'mr" m 03 f Helipad Site ; . 2.5 2 -. _ _ 2 51.5 - a , _ E175 g 5 g '0 . >,__.._,_—*—~..—i—~—— ..-—i '8 E 1 / ’F' .- -" g 1 . __ ___ ... ... -.-gu .- 0 / ..__.-..a-"f'":.~«__u__o.__.__‘m‘_ a .-"' ..J‘T' ________ 0 _____________ ._ 1'. 1‘. I .. , .. 4 o 5 o 5 {:r3 ' o 1 ’ - - -_, __ _ o 0 0'2 0'4 0'6 0'8 1 o o 2 o 4 o 6 0.8 1 Time (day) Time (day) — w1251 - e— w1653 -— w1253 --X- w1654 — — w1255 --+- w1655 --e-- w1650 "ts-- w1656 - -El- - w1662 -+- w1657 Figure H.7a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 03 gravel D sand sandy silt I paleosol 31., .... Figure H.7b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24 model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 221 Drawdown Observed Drawdown Simulatio' 4 ' - 2.5 n 0 , Helipad Site 2 _. ...... . .vw—fi-. J: A ".4" twfw' In 2 g / . s 5 1’... ----------- .- _____--e-- ' ---- '0'"--‘--- 0 A i 1 O 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - .— w1653 -— w1253 -->(- w1654 — — w1255 --+- w1655 --e-- w1650 "an w1656 - -El- - w1662 --O-— w1657 Figure H.8a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 04 a gravel [:I sand sandy silt I paleosol Figure H.8b: Map view of the spatial distribution of hyd facies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 222 Drawdown Observed Drawdown 2 5 Simulation 05 Helipad Site ' ’1' 1' ' x." 7 2-5 In. '5’ 2 ii. .r'Z 2 £15 hfé‘: .53.. .. . ._ .. .. - E15 S I: // g 5 ‘11" "’6’", g 1 . .-" ..I; 5 I ..F'. .X o 5 -. , ' 5 .. .-' - , 0 5 x .-’ I” i..."- ‘p/ E 0 . . 1 o 0 0-2 0-4 0-6 0'3 l o 0.2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - o— w1653 -— w1253 -->(- w1654 — — w1255 --+— w1655 --e-- w1650 “Ifiw- w1656 . .5. . w1662 ...... w1657 Figure H.9a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 05 I gravel D sand sandy silt I paleosol Figure H.9b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24“I model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 223 Drawdown Observed Drawdown 2.5 Simulation 06 2.5 Helipad Site 2 2 E15 E15, 5 g 1 C1 / 5 .0 01 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - .— w1653 -— w1253 --)(- w1654 — — w1255 --+— w1655 --e-- w1650 "an w1656 . -El- - w1662 —O-- w1657 Figure H.10a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 06 u gravel E] sand sandy silt I paleosol Figure H.10b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24'h model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). K 224 Drawdown Observed Drawdown 2 5 Simulation o7 Helipad Site 2.5 0’ E1. E1 E V ' E a 0 5 0.5 0 0.2 0.4 0.6 0.8 1 Time (dav) 0 0.2 0.4 0.6 0.8 1 Time (day) — w1251 - .— w1653 -— w1253 --)(— w1654 - — w1255 --+- w1655 --e-- w1650 "an w1656 - -El-- w1662 --O-- w1657 Figure H.l la: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 07 I gravel E] sand sandy silt I paleosol Figure H.11b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map View is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 225 Drawdown Observed Drawdown Simulation 08 Helipad Site Drawdown (rn) Drawdown (m) 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - e— w1653 -— w1253 -->(- w1654 - -— w1255 --+- w1655 —-e-- w1650 «an w1656 - -1-:1- - w1662 —-O-— w1657 Figure H. 12a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 08 I gravel E] sand sandy silt . .' I paleosol Figure H.12b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 226 Drawdown Observed Drawdown 2 5 Simulation 09 Helipad Site ‘ ’ ' 2.5 2 2 £15 E15 § 8 1 g i E1 0 0:5 0.5 0 . 0 4L 4i; 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - .— w1653 -— w1253 --)(— w1654 — — w1255 --+— w1655 --e-- w1650 neg-- w1656 - -El- - w1662 -O-- w1657 Figure H.13a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 09 gravel [:I sand sandy silt I paleosol i- Figure H.13b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24 model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 227 Drawdown Observed Drawdown Simulation 10 ' ' 2'5 Helipad Site 2 E15 .5 E1 5 0.5 0 ° °'2 °-4 °-6 0‘8 1 o 0.2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - 9— w1653 -— w1253 --)(- w1654 — - w1255 --+— w1655 --e-- w1650 no" w1656 - -El- - w1662 "-0-— W165? Figure H. 14a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 10 gravel [:1 sand sandy silt I paleosol Figure H. 14b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24‘h model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 228 Drawdown Observed Drawdown 2.5 Simulation 11 Helipad Site 2 E15 » . j g .__)(,__...:----,_ . 'D a 1 -: , . 5 05 {1’ 0 [ l l 0 0.2 0.4 0.6 0.8 1 Time (day) 0 0.2 0.4 0.6 0.8 1 Time (day) — w1251 - .— w1653 -— w1253 -->(— w1654 — — w1255 --+— w1655 --e-- w1650 «a» w1656 . .g. . w1662 —-.-- w1657 Figure H.15a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation ll gravel E] sand sandy silt I paleosol 1. '1 Figure H.15b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24"I model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 229 Drawdown Observed Drawdown Simulation 12 Helipad Site 2.5 '01 Drawdown (m) 1 ° 0'2 °“1 0'6 0-8 1 o 0.2 0.4 0.6 0.8 1 “me (63") Time (day) — w1251 - o— w1653 -— w1253 --)(- w1654 — — w1255 --+— w1655 --e-- w1650 "an w1656 - -El- - w1662 -O- w1657 Figure H.16a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 12 I gravel D sand [:3 sandy silt I paleosol Figure H.16b: Mapiew of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 241h model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 230 PM?“ Observed Drawdown Slmulation 13 Helipad Site 2.5 _3 01 1 _L Drawdown (m) Drawdown (m) .0 01 i l l 0.2 0.4 0.6 0.8 1 0 Time (day) 0.2 0.4 0.6 0.8 1 Time (day) — w1251 - .— w1653 -— w1253 --)(- w1654 - — w1255 --+— w1655 --e-- w1650 “rs-- w1656 . .5). . w1662 -.-- W1657 Figure H.17a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation l3 gravel E] sand [:1 sandy silt I paleosol transition probability geostatistical conceptual model. Map view is from the 24 model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 231 Drawdown Simulation 14 Observed Drawdown Helipad Site _I Drawdown (m) 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - o— w1653 -— w1253 -->(- w1654 - — w1255 --+— w1655 --e-- w1650 "an w1656 - -E|- - w1662 --O-- w1657 Figure H. 1 8a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 14 ‘. J . .a , . a gravel I:l sand sandy silt I paleosol '1 . ' - '11... ‘5. k . _ Figure H. 1 8b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map View is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 232 Observed Drawdown Drawdown 2 5 Slmulatlon 15 Helipad Site 1 1" ’41“ ' .' A4" 1 2.5 >1 3 1' A?” 2 “fl/'1 ,.....1... I 2,.- . " a. o" I]; 1'? 1'10". A ix! I": "It”I é1.5 #1116311" ..1 Ill." ........... , ...... _ E1 § :11 i I”; E g 1 1; 1 ; 4.11 . h 111'" ,, 4.. ,.,._ g D i / 5 6' . 0.5 ‘f/‘z 0.5 0 i . O 0.2 0.4 0.6 0.8 1 0 0‘2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - .— w1653 -— w1253 -->(- w1654 - — w1255 --+— w1655 --e-- w1650 «ts-- w1656 - -El- - w1662 -o-- w1657 Figure H. 1 9a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 15 I gravel [:1 sand sandy silt I paleosol -. '1: 3: is t 1% - Figure H.19b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit) 233 Drawdown Observed Drawdown Simulation 16 Helipad Site 2.5 , I Wear 2.5 2 P i / 219*???" / ..... ,‘ .;:,‘ 2 / ...; ’,' / #:‘49’ ”.../’1 g " : g 13 ' 8 E a g 1 o 5 0.5 n 4; 0 0.4 0'6 0‘3 1 o 0.2 - 0.4 0.6 0.8 1 Time (day) Time (day) __ w1251 - .— w1653 -— w1253 -->(- W1554 — — w1255 --+- w1655 "9-- w1650 no" w1656 --a-- w1662 —o- w1657 Figure H.20a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Figure H.20b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). Simulation 16 A. a gravel E] sand sandy silt I paleosol 234 Drawdown Simulation 17 2.5 r 2 I .2"). 1".'E". - E15 _ ..... .I/.- ...-9'" C . I. . B .- O "U E 1 o l l 0.4 0.6 0.8 1 Observed Drawdown Helipad Site 0.2 0.4 0.6 0.8 1 0 Time (day) _ w1251 - o— w1653 -— w1253 --)(- W1654 — - w1255 --+— w1655 --e-- w1650 «a» w1656 . -E|- - w1662 --O-— w1657 Time (day) Figure H.21a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 1 7 a gravel E] sand sandy silt I paleosol Figure H.21b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24 model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 235 Drawdown Observed Drawdown 2.5 Simulation 18 . Helipad Site . 2.5 2 ~- 2 $1.5 E15 E s v 8 E 1 a 1 o I— o 0.5 0.5 0 l r l 1 0 0 0.2 0,4 0.6 0.8 1 0 02 04 0.6 as 1 Time (day) Time (day) —— w1251 - .— w1653 -— w1253 --)(- w1654 — — w1255 --+— w1655 --e-- w1650 no" w1656 . .51. . w1662 ...... w1657 Figure H.22a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 18 a gravel [J sand sandy silt I paleosol Figure H.22b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 236 Drawdown Observed Drawdown Simulation 19 Helipad Site 2.5 154?. ,2", _z’. 2.5 Igil-JI / ' [I 2 .' I I x .... / . , 2 ’4'. [In E15 #5 / ,I — E15 : . .- 3 -- § ‘0 I. '0 g .f . ‘ g 1 5 i 5 I" . 0.5 l ..... o 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - 0— w1653 -— w1253 -—>(- w1654 - — w1255 --+— w1655 --e-- w1650 «an w1656 --El-- w1662 -O-- w1657 Figure H.23a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation l9 fl gravel [I sand sandy silt I paleosol . "S" ’13 ‘ Figure H.23b: Map view of the spati distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 237 Drawdown Observed Drawdown Simulation 20 H eli pad Site 2.5 , ‘ 2 5 I : .I" I are" _..---"" "’1 --_-.-P‘-- ----- 2 /‘:.-'.,.: """ 2 E15 _ . I/ff'L—‘f .._ a-"'.'" _ E1 5 E ’15:?" '1/ § § 1 vii" ”ff-1’7... . A u 5 tall :' I g 1 0 lines. .1)" 5 1'1 0 0.5 {/66 o 5 0 0 1 l .. . O 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 Time (day) Time (day) — w1251 - .— w1653 -— w1253 -->(— w1654 - — w1255 --+— w1655 --e-- w1650 "an w1656 - -a- - w1662 -o- w1657 Figure H.24a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 20 I gravel E] sand sandy silt I paleosol Figure H.24b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24 model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 238 DdeW“ Observed Drawdown Simulation 21 Helipad Site 2.5 2.5 _a (II _n Drawdown (m) Drawdown (m) P u: 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - e— w1653 -— w1253 -—>(- w1654 — — w1255 --+— w1655 --e-- w1650 «an w1656 El - w1662 ~0— w1657 Figure H.25a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 21 I gravel [:I sand sandy silt I paleosol Figure H. 25b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 239 Drawdown Observed Drawdown Simulation 22 Helipad Site ‘ .' -’ I 1‘ 3'... fi 25 - 0...; 1"} 2 E . ‘ E15 g ‘c' u 5 E " 5 1 o 5 0.5 i o 0-4 0-6 0-8 o 0.2 0.4 0.6 0.8 1 Time (day) Time (day) — w1251 - .— w1653 -— w1253 --><- w1654 — — w1255 --+— w1655 "9-- w1650 "an w1656 . -a- - w1662 -o- w1657 Figure H.26a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 22 gravel I:] sand D sandy silt I paleosol Figure H.26b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24‘h model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 240 Drawdown Observed Drawdown Simulation 23 - - 2.5 2 5 Helipad Site 2 él- E15 g c s g 1 Q 5 .0 01 0 0.2 0.4 0.6 0.8 1 Time (day) 0 0.2 0.4 0.6 0.8 1 Time (day) — w1251 - o— w1653 -— w1253 -->(- w1654 — — w1255 --+— w1655 --e-- w1650 "cg-- w1656 - -El- - w1662 "-0- w1657 Figure H.27a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 23 gravel 1:] sand sandy silt I paleosol Figure H.27b: Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24th model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 241 Drawdown Observed Drawdown 2 5 Simulation 24 Helipad Site ' T f 2-5 2 E “in 2 adanfi ----- . A .4533: - :3:-::':.:'::N“M E15 _ . ,,,,, I} _. .;.-.=-‘--"" . . ........... - E15 g /. .953.- / ‘u 1.1-""5 / E g §1_ _,’ ,I” .. .,._ g1 1:1 If ”a" . 5 l "I 5 o 5 1’ I" . ,r 1 ; ________ 0.5 ’ :3 __..---""'?' f ___§___.-o--' """"" 0 i g 0 0 0.2 0.4 0.6 0.8 1 O 0.2 0.4 0.6 0.8 1 Time (day) Time (day) —— w1251 - e— w1653 -— w1253 -->(- w1654 — — w1255 --+— w1655 --e-- w1650 "an w1656 --Ei-- w1662 ~0— w1657 Figure H.28a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 24 gravel [:1 sand sandy silt I paleosol Figure H.28 Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24 model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 242 Drawdown Observed Drawdown Simulation 25 - - 2.5 . ! 2 5 Helipad Site 2 2 E15 » E1 5 c V ' g E g 8 g ‘“ i ‘ 5 0.5 0 5 0 1 i l l 0 0 0.2 0.4 0.6 0.8 1 0 0 2 0 4 0 6 o 8 1 Time (day) Time (day) __ w1251 - e— w1653 -— w1253 --><- w1654 — — w1255 --+— w1655 --e-- w1650 "cc-- w1656 . ~El- . w1662 --.-— w1657 Figure H.29a: Simulated drawdown histories for 1650 series and 1250 series wells within the stratigraphic transition probability geostatistical conceptual model. Simulated results show minor drawdown in1250 series wells. Stratigraphic Transition Probability Geostatistical Simulation 25 fl gravel [:I sand sandy silt I paleosol Figure H.29 Map view of the spatial distribution of hydrofacies for the stratigraphic transition probability geostatistical conceptual model. Map view is from the 24 model layer (lower portion of the sand and gravel channel for the E stratigraphic unit). 243 Appendix I FORTRAN Code There were two FORTRAN codes developed to reformat output binary grid files (bgr) generated with T-PROGS (Carle 1999) to formats that could be read into GMS 4.0 (Brigham Young University, 2002) and Rockworks 2002 (Rockware Inc., 2002). The stratigraphic TPSIM code was developed to read each of the full block simulations (C through F) and top and bottom paleosol surfaces. This code outputs a category file (corresponding to materials read by GMS) and Rockworks files (for visualization of realizations, Appendix F). The transition probability geostatistical TPSIM code was developed to read in realizations generated by TPSIM and output files for GMS and Rockworks in the same manor as the previous code. These codes are listed below. 244 Stratigraphic TPSIM Code program sequences c Program to create an output file to be used in a 3D realization c of the LLNL study area - lists sequence number for C, D, E, and F. c Created for 9/02 simulations. C DO NOT USE WITHOUT ADJUSTMENT FOR SPECIFICS. . .NOTE: SOME PALEOSOL C ZONES ONLY HAVE 3 CATEGORIES, l-GRAVEL, 2—SAND, 3- PALEOSOL. THESE C ARE ADJUSTED IN THIS PROGRAM TO GIVE PALEOSOL AS CAT = 4. c revisions 11/8/02: read in telescoping grid from GMS and output heterogeneity c Declaration of variables parameter nx=181, ny=155, nz=79, base=124.25 parameter ix=206, iy=194 logical*1 iseq(nx*ny*nz),icat(nx*ny*nz) logical*l ccat(nx*ny*nz),dpcat(nx*ny*nz) logical*1 dcat(nx*ny*nz),epcat(nx*ny*nz) logical*1 ecat(nx*ny*nz),fpcat(nx*ny*nz) logical*1 fcat(nx*ny*nz) real sc(nx*ny),sd(nx*ny),se(nx*ny),sf(nx*ny) real sbc(nx*ny),sbd(nx*ny),sbe(nx*ny),sbf(nx*ny) real scl(nx,ny),sd1(nx,ny), se1(nx,ny),sf1(nx,ny) real sbcl(nx,ny),sbd1(nx,ny),sbe1(nx,ny),sbf1(nx,ny) real sll(nx*ny),slll(nx,ny) real elev real cond(nx*ny*nz) real xg(ix),yg(iy) integer idim(3) character*40 parfl character*40 cbfil character*40 dtfil character*40 dbfil character*40 etfil character*40 ebfil character*40 ftfil character*40 fbfil character*40 llfil character*40 seqfil character*40 catfil character*40 cbgr character*40 dpbgr character*40 dbgr 245 character*40 epbgr character*40 ebgr character*40 fpbgr character*40 fbgr character*40 gmsfil character*40 rckfill character*40 rckfi12 character*40 rckfilB character*40 rckfil4 character*40 rckfil character*40 gmsin character*40 gmsref nxy=nx*ny nxya=(nx-1)*(ny—l) nxyz=nx*ny*nz dx=2. dy=2. dz=0.5 c Where are the data??????? print*,'input par file name?:' c Open input and output files read(5,'(a40)') parfl print*,'input file name:', parfl open(l,file=parfl,status='old') read(1,'(a)') cbfil read(1,'(a)') dtfil read(1,'(a)') dbfil read(1,'(a)') etfil read(1,'(a)') ebfil read(1,'(a)') ftfil read(1,'(a)') fbfil read(1,'(a)') llfil read(1,'(a)') cbgr read(1,'(a)') dpbgr read(1,'(a)') dbgr read(1,'(a)') epbgr read(1,'(a)') ebgr read(1,'(a)') fpbgr read(1,'(a)') fbgr read(1,'(a)') seqfil read(1,'(a)') catfil read(1,'(a)') gmsfil read(1,'(a)') rckfill read(1,'(a)') rckfi12 read(1,'(a)') rckfil3 read(1,'(a)') rckfil4 read(1,'(a)') rckfil 246 read(1,*)sx,sy,sz read(l,*)cond0,condl,cond2,cond3,cond4,cond5 .read(1,'(a)') gmsin read(1,'(a)') gmsref close(1) open(2,file=seqfil,status='unknown',form='unformatted' open(3,file=catfil,status='unknown',form='unformatted' open(4,fi1e='dbg.txt',status='unknown') irank=3 write(2) irank write(2) nx,ny,nz write(3) irank write(3) nx,ny,nz c Read data from ASCII file to fill arrays. Sequence boundaries! print*,'reading Rockware grids' open(l,file=cbfil,status='old') do i=1,nx do j=1,ny read(1,*) sbc1(i,j) enddo enddo close(l) open(1,file=dtfil,status='old') do i=1,nx do j=1,ny read(1,*) sd1(i,j) enddo enddo close(1) open(l,file=dbfil,status='old') do i=1,nx do j=1,ny read(1,*) sbd1(i,j) enddo enddo close(1) open(l,file=etfil,status='old') do i=1,nx do j=1,ny read(1,*) se1(i,j) enddo enddo close(l) open(1,file=ebfil,status='old') 247 do i=l,nx do j=1,ny read(1,*) sbel(i,j) enddo enddo close(1) open(l,file=ftfil,status='old') do i=l,nx do j=1,ny read(1,*) sfl(i,j) enddo enddo close(1) open(l,file=fbfil,status='old') do i=1,nx do j=1,ny read(1,*) sbfl(i,j) enddo enddo close(1) open(l,file=llfil,status='old') do i=l,nx do j=1,ny read(1,*) slll(i,j) enddo enddo close(1) c read data from bgr files for each sequence print*,'reading bgr files' open(8,file=cbgr,status='old',form='unformatted') read(8) irank print*,'irank=',irank read(8) (idim(i),i=1,3) nxyz=idim(1)*idim(2)*idim(3) read(8) (ccat(i),i=l,nxyz) close(8) open(8,file=dpbgr,status='old',form='unformatted') read(8) irank print*,'irank=',irank read(8) (idim(i),i=1,3) nxyz=idim(l)*idim(2)*idim(3) read(8) (dpcat(i),i=1,nxyz) close(8) open(8,file=dbgr,status='old',form='unformatted') read(8) irank print*,'irank=',irank read(8) (idim(i),i=1,3) 248 OOOOOOOOOOOOOOOO nxyz=idim(1)*idim(2)*idim(3) read(8) (dcat(i),i=1,nxyz) close(8) open(8,file=epbgr,status='old',form='unformatted') read(8) irank print*,'irank=',irank read(8) (idim(i),i=1,3) nxyz=idim(l)*idim(2)*idim(3) read(8) (epcat(i),i=l,nxyz) close(8) open(8,file=ebgr,status='old',formz'unformatted‘) read(8) irank print*,'irank=',irank read(8) (idim(i),i=1,3) nxyz=idim(1)*idim(2)*idim(3) read(8) (ecat(i),i=1,nxyz) close(8) open(8,file=fpbgr,status='old',form='unformatted') read(8) irank print*,'irank=',irank read(8) (idim(i),i=1,3) nxyz=idim(1)*idim(2)*idim(3) read(8) (fpcat(i),i=1,nxyz) close(8) open(8,file=fbgr,status='old',form='unformatted') read(8) irank print*,'irank=',irank read(8) (idim(i),i=1,3) nxyz=idim(l)*idim(2)*idim(3) read(8) (fcat(i),i=1,nxyz) close(8) print*,'now converting 3 cat to 4 cat models' change 3 category zones to 4 category models do k=1,nz do j=1,ny do i=1,nx ijk=i+((j-1)*nx)+((k-1)*nx*ny) if(cpcat(ijk).eq.3)cpcat(ijk)=4 if(cpcat(ijk).eq.-3)cpcat(ijk)=-4 if(epcat(ijk).eq.3)epcat(ijk)=4 if(epcat(ijk).eq.-3)epcat(ijk)=-4 if(fpcat(ijk).eq.3)fpcat(ijk)=4 if(fpcat(ijk).eq.-3)fpcat(ijk)=-4 enddo ‘ enddo enddo print*,'finished doing that...‘ 249 c fill in surface arrays write(4,'(a40)') 'warnings to report' print*,'filling it all in...bored yet?‘ do j=1,ny do i=1,nx ixy=i+((j-1)*nx) sc(ixy)=sc1(i,j) sbc(ixy)=sbc1(i,j) sd(ixy)=sd1(i,j) sbd(ixy)=sbdl(i,j) se(ixy)=se1(i,j) sbe(ixy)=sbe1(i,j) sf(ixy)=sf1(i,j) sbf(ixy)=sbf1(i,j) sll(ixy)=slll(i,j) if(sbc(ixy).le.sd(ixy)) write(4,'(a40)') * 'WARNING: sbc>sd: ',i,j if(sd(ixy).le.sbd(ixy)) write(4,'(a40)') * 'WARNING: Sd>sbd: ',i,j if(sbd(ixy).le.se(ixy)) write(4,'(a40)') * 'WARNING: sbd>se: ',i,j if(se(ixy).le.sbe(ixy)) write(4,'(a40)') * 'WARNING: se>sbe ',i,j if(sbe(ixy).le.sf(ixy)) write(4,'(a40)') * 'WARNING: sbe>sf ',i,j if(sf(ixy).le.sbf(ixy)) write(4,'(a40)') * 'WARNING: sf>sbf ',i,j if(sbf(ixy).le.sll(ixy)) write(4,'(a40)') * 'WARNING: sll>sbf: ',i,j c check for thin paleosols...increase to keep it present. if(sd(ixy)- sbd(ixy).le.0.5)sd(ixy)=sbd(ixy)+0.6 if(se(ixy)- sbe(ixy).le.0.5)se(ixy)=sbe(ixy)+0.6 if(sf(ixy)- sbf(ixy).le.0.5)sf(ixy)=sbf(ixy)+0.6 enddo enddo c fill in bgr final grid array with proper code do k=l,nz e1ev=(k*dz)+base-(O.5*dz) do j=1,ny do i=1,nx ij=i+((j-l)*nx) ijk=((k-1)*nx*ny)+i+((j-l)*nx) if(e1ev.gt.sbc(ij)) then icat(ijk)=5 250 then iseq(ijk)=5 endif if(elev.gt.sd(ij).and.elev.1e.sbc(ij)) then icat(ijk)=ccat(ijk) v=ccat(ijk) if(ccat(ijk).lt.0) v=ccat(ijk)*(-l) iseq(ijk)=v+20 endif if(elev.gt.sbd(ij).and.elev.le.sd(ij)) then icat(ijk)=dpcat(ijk) v=dpcat(ijk) if(dpcat(ijk).lt.0) v=dpcat(ijk)*(—l) iseq(ijk)=v+30 endif if(elev.gt.se(ij).and.elev.le.sbd(ij)) then icat(ijk)=dcat(ijk) v=dcat(ijk) if(dcat(ijk).lt.0) v=dcat(ijk)*(-1) iseq(ijk)=v+40 endif if(elev.gt.sbe(ij).and.elev.le.se(ij)) then icat(ijk)=epcat(ijk) if(icat(ijk).eq.3) icat(ijk)=4 if(icat(ijk).eq.-3) icat(ijk)=-4 v=epcat(ijk) if(epcat(ijk).lt.0) v=epcat(ijk)*(-1) if(v.eq.3) v=4 iseq(ijk)=v+50 endif if(elev.gt.sf(ij) .and. elev.le.sbe(ij)) icat(ijk)=ecat(ijk) v=ecat(ijk) if(ecat(ijk).lt.0) v=ecat(ijk)*(-l) iseq(ijk)=v+60 endif if(elev.gt.sbf(ij).and.elev.le.sf(ij)) then icat(ijk)=fpcat(ijk) if(icat(ijk).eq.3) icat(ijk)=4 if(icat(ijk).eq.-3) icat(ijk)=-4 v=fpcat(ijk) if(fpcat(ijk).lt.0) v=fpcat(ijk)*(-l) if(v.eq.3) v=4 iseq(ijk)=v+70 endif if(elev.gt.sll(ij).and.elev.le.sbf(ij)) then icat(ijk)=fcat(ijk) 251 v=fcat(ijk) if(fcat(ijk).lt.0) v=fcat(ijk)*(-l) iseq(ijk)=v+80 endif if(elev.le.sll(ij)) then icat(ijk)=6 iseq(ijk)=90 endif enddo enddo enddo write(2) (iseq(k),k=l,nxyz) write(3) (icat(k),k=1,nxyz) close(2) close(3) c Prepare the gms and rockware input files print*, 'printing gms and rockware files' open(l,file=rckfill,status='unknown') open(2,file=rckfi12,status='unknown') open(3,file=rckfi13,status='unknown') open(4,file=rckfil4,status='unknown') open(S,file=rckfil,status='unknown') open(6,file=gmsfil,status='unknown') open(7,file=gmsin,status='old') open(8,file=gmsref,status='unknown') c writing Rockware files do k=1,nz do j=1,ny do i=1,nx ijk=i+((j-l)*nx)+((k-1)*nx*ny) x=sx+(2.*(i-l)) y=sy+(2.*(j-l)) z=sz+(0.5*(k-1)) if (icat(ijk).eq.l) then cond(ijk)=cond1 write(1,14) x,y,z,icat(ijk) write(5,l4) x,y,z,iseq(ijk) endif if (icat(ijk).eq.2) then cond(ijk)=cond2 write(2,l4) x,y,z,icat(ijk) write(5,14) x,y,z,iseq(ijk) endif if (icat(ijk).eq.3) then cond(ijk)=cond3 write(3,14) x,y,z,icat(ijk) write(5,l4) x,y,z,iseq(ijk) 252 endif if (icat(ijk).eq.4) then cond(ijk)=cond4 write(4,14) x,y,z,icat(ijk) write(5,14) x,y,z,iseq(ijk) endif if (icat(ijk).eq.0) then cond(ijk)=cond0 write(5,l4) x,y,z,iseq(ijk) endif if (icat(ijk).eq.5) then cond(ijk)=cond5 write(5,14) x,y,z,iseq(ijk) endif if(k.eq.20) print*,'youre now on layer 20...hang there' if(k.eq.50) print*,'youre now on layer 50...hang there' if(k.eq.70) print*,'youre now on layer 70...hang there' if(k.eq.90) print*,'youre now on layer 90...hang there' 14 format(3(f7.1),1x,i4) format(i8,1x,3(f7.1),1x,e10.3) enddo enddo enddo c writing GMS full grid file do k=nz,1,-l do j=ny,1,-1 do i=1,nx ijk=i+((j-1)*nx)+((k-l)*nx*ny) write(6,'(i2)') icat(ijk) enddo enddo enddo c writing GMS telescoping grid file C read refined grid x,y locations read(7,*) sx,sy print*,'filling in refined grid' print*,ix,nx,iy,ny do i=l,ix read(7,*) xg(i) enddo do j=1,iy read(7,*) yg(j) 253 enddo do k=nz,l,-1 do j=iy,1,-1 y9a=(yg(j)-sy)/dy nycell=(int(yga))+1 do i=1,ix xga=(xg(i)-sx)/dx nxcell=(int(xga))+1 ijk=nxcell+((nycell—l)*nx)+((k- l)*nx*ny1 write(8,'(i2)') icat(ijk) enddo enddo enddo print*,'gms and rockware files completed' print*,'Goodbye!‘ close(1) close(2) close(3) close(4) close(S) close(6) print*’l****************************l print*,'GMS and RockWare files written.‘ print*l1****************************I print*,'(:' stop end 254 Transition Probability Geostatistical TPSIM Code program Transition Probability Geostatistical TPSIM c Program to create an output file to be used in a 3D realization c of the LLNL study area - lists sequence number for C, D, E, and F. c Created for 9/02 simulations. C DO NOT USE WITHOUT ADJUSTMENT FOR SPECIFICS...NOTE: SOME PALEOSOL C ZONES ONLY HAVE 3 CATEGORIES, l-GRAVEL, 2-SAND, 3- PALEOSOL. THESE C ARE ADJUSTED IN THIS PROGRAM TO GIVE PALEOSOL AS CAT = 4. c revisions 11/8/02: read in telescoping grid from GMS and output heterogeneity c file. c Declaration of variables parameter nx=181, ny=155, nz=79, base=124.25 parameter ix=206, iy=194 logical*1 iseq(nx*ny*nz),icat(nx*ny*nz) real siml(nx,ny) real cond(nx*ny*nz) real xg(ix),yg(iy) integer idim(3) character*40 parfl character*40 simbgr character*40 gmsfil character*40 rckfill character*40 rckfi12 character*40 rckfi13 character*40 rckfil4 character*40 rckfil character*40 gmsin character*40 gmsref nxy=nx*ny nxya=(nx-1)*(ny-1) nxyz=nx*ny*nz dx=2. dy=2. dz=0.5 c Where are the data??????? 255 print*,'input par file name?:' ! Open input and output files read(5,'(a40)') parfl print*,'input file name:', parfl open(l,file=parfl,status='old') read(1,'(a)') simbgr read(1,'(a)') gmsfil read(1,'(a)') rckfill read(1,'(a)') rckfilZ read(1,'(a)') rckfil3 read(1,'(a)') rckfil4 read(1,'(a)') rckfil read(1,*)sx,sy,sz read(l,*)condo,cond1,cond2,cond3,cond4,cond5 read(1,'(a)') gmsin read(1,'(a)') gmsref close(1) c Read data from ASCII file to fill arrays. Sequence boundaries! print*,'red in par file' c read data from bgr files for each sequence print*,'reading bgr files' open(8,file=simbgr,status='old',form='unformatted') read(8) irank print*,'irank = ',irank print*,'irank=',irank read(8) (idim(i),i=1,3) nxyz=idim(1)*idim(2)*idim(3) read(8) (icat(i),i=l,nxyz) close(8) print*,'red in simbgr' c Prepare the gms and rockware input files print*, 'printing gms and rockware files' open(l,file=rckfill,status='unknown') open(2,file=rckfi12,status='unknown') open(3,file=rckfi13,status='unknown') open(4,filezrckfil4,status='unknown') open(S,file=rckfil,status='unknown') open(6,file=gmsfil,status='unknown') open(7,file=gmsin,status='old') open(8,file=gmsref,status='unknown') c writing Rockware files do k=1,nz do j=1,ny do i=1,nx ijk=i+((j—1)*nx)+((k—1)*nx*ny) x=sx+(2.*(i-l)) 256 y=sy+(2.*(j-l)) z=sz+(0.5*(k-l)) if (icat(ijk).eq.1) then cond(ijk)=condl write(l,14) x,y,z,icat(ijk) write(5,14) x,y,z,icat(ijk) endif if (icat(ijk).eq.2) then cond(ijk)=cond2 write(2,14) x,y,z,icat(ijk) write(5,14) x,y,z,icat(ijk) endif if (icat(ijk).eq.3) then cond(ijk)=cond3 write(3,l4) x,y,z,icat(ijk) write(5,l4) x,y,z,icat(ijk) endif if (icat(ijk).eq.4) then cond(ijk)=cond4 write(4,14) x,y,z,icat(ijk) write(5,14) x,y,z,icat(ijk) endif if (icat(ijk).eq.0) then cond(ijk)=cond0 write(5,14) x,y,z,icat(ijk) endif if (icat(ijk).eq.5) then cond(ijk)=cond5 write(5,14) x,y,z,icat(ijk) endif if(k.eq.50) print*,'youre now on layer 50...hang in there' 14 format(3(f7.1),1x,i4) 15 format(iB,lx,3(f7.1),1x,e10.3) enddo enddo enddo c writing GMS full grid file do k=nz,1,-1 do j=ny,1,-1 do i=1,nx ijk=i+((j-l)*nx)+((k-1)*nx*ny) write(6,'(i2)') icat(ijk) enddo enddo enddo c writing GMS telescoping grid file 257 c read refined grid x,y locations read(7,*) sx,sy print*,'filling in refined grid' print*,ix,nx,iy,ny do i=1,ix read(7,*) xg(i) enddo do j=1,iy read(7,*) yg(j) enddo do k=nz,1,—l do j=iy,1,-l Y9a=(Yg(j)-sy)/dy nycell=(int(yga))+l do i=l,ix xga=(xg(i)—sx)/dx nxcell=(int(xga))+1 ijk=nxcell+((nycell-l)*nx)+((k- l)*nx*ny) write(8,'(12)') icat(ijk) enddo enddo enddo print*,'gms and rockware files completed' print*,'Goodbyel' close(1) close(2) close(3) close(4) close(S) close(6) close(7) close(8) print*'I****************************1 print*,'GMS and RockWare files written.‘ print*,|****************************I print*,'(:' stop end 258 Bibliography Batu, V., 1998, Aquifer Hydraulics : A Comprehensive Guide to Hydraulic Data Analysis: New York / Chichester / Weinheim / Brisbane / Singapore / Toronto, John Wiley & Sons, Inc., 727 p. Bennett, G., and Weissman, GS, 2001 , Geostatistical and deterministic modeling of alluvial fan hydrostratigraphy at the Lawrence Livermore National Laboratory: North Central GSA Meeting, p. 1. Biteman, S., Hyndman, D.W., Phanikumar, MS, and Weissmann, G.S., in press, Integration of sedimentologic and hydrologic properties for improved transport simulations: detailed characterization of a glacial outwash aquifer at the Schoolcrafi Bioremediation Site, Michigan: Submitted for an SEPM special publication on Hydrogeophysics and Hydrostratigraphy (edited by John Bridge and David Hyndman). Blake, R. G., 1997, Hydrostratigraphic analysis-ground water cleanup at Lawrence Livermore National Laboratory, Lawrence Livermore National Laboratory, p. 1. Blake, R. G., Noyes, C.M., and Maley, M.P., 1995, Hydrostratigraphic analysis - the key to cost-effective ground water cleanup at Larwence Livermore National Laboratory, Larwence Livermore National Laboratory, p. 12. Boutt, D. F., Hyndman, D.W., Pijanowski, BC, and Long, D.T, 2001, Identifying Potential Land Use-derived Solute Sources to Stream Baseflow Using Ground Water Models and GIS: Ground Water, v. 36, p. 224—34. Carle, S. F ., 1996a, A trransition probability-based approach to geostatistical charaterization of hydrostratigraphic architecture: Dissertation thesis, University of California, Davis, 233 p. Carle, S. F., 1999, T-PROGS: Transition Proability Geostatistical Software Version 2.1, Davis, CA, University of California, Davis, p. 84. Carle, S. F., and Fogg, G. E., 1996b, Transition probability-based indicator geostatistics: Mathmatical Geology, v. 28. Carle, S. F., LaBolle, E. M., Weissman, G. S., VanBrocklin, D., Fogg, G. E., 1998, Conditioned simulation of hydrofacies architecture; a transition probability/Markov approach,, in G. S. a. D. Fraser, J. M., ed., Hydrogeology Models of Sedimentary Aquifers, v. 1, p. 147-170. 259 Carpender, D. W., Puchlik, K.P., ramirez, A.L., Wagoner, J .L., Knuass, K.G., and Kasmeyer, P.W., 1980, Status Report on the Geology of Larwence Livermore National Laboratory Site and Adjacent Areas, Larwence Livermore National Laboratory. Carpender, D. W., Sweeney, J. J ., Kasameyer, P. W., Burkhard, N. R., Knauss, K. G., and Shlemon, R. J ., 1984, Geology of the Lawrence Livermore National Laboratory Site and Adjacent Areas, Livermore California, Lawrence Livermore National Laboratory, University of Califomia, p. 150. CDWR, 1966, Evaluation of Ground Water Resources: Livermore and Sunol Valleys, Geology, Bulletin 118-2, Sacramento, CA, State of Califonia, Department of Water Resources. CDWR, 1974, Evoluation of Ground Water Resources: Livermore and Sunol Valleys, Bulletin 118-2, Geology, State of California, Department of Water Resources, Sacramento CA. Copty, N. K., and Findikakis, A.N., 2002, Stocastic Analysis of Pumping Test Drawdown Data in Heterogeneous Geologic Formations: Briding the gap between measurement and modeling in heterogeneous media. Dibblee, T. W., 1980, Preliminary geologic map of the Altamont Quadrangle, Alameda County, California, US Geological Survey. Dibblee, T. W., and Darrow, R.L., 1981, Geology of the Northern Diablo Range and Livermore Valley Area, in F. E.V., ed., Geology of Central and Northern Diablo Range, California, Pacific Section of SEPM, Los Angeles, p. 77-112. Dickinson, W. R., 1965, Folded Thrust Contact Between Franciscan Rocks adn Panoche Group in the Diablo Range of Centeral California: Geological Society of America, Golden, CO, Special Paper 82, p. 248-249. Doherty, J ., Brebber, L., and Whyte, P., 2000, PEST: Model-Independent Parameter Estimation, Watermark Numerical Computing, p. 249. ' Dominic, D. F ., Ritzi, R.W., Jr.,, 1996, Geostatistical analysis of facies distributions in a buried-valley aquifer: American Association of Petroleum Geologists 1996 annual convention, p. 36. Energy, U. S. D. o., 1998, Environmental restoration at Larwence Livermore National Laboroatory, US. Department of Energy, p. 24. Ernst, W. G., 1970, Tectonic Contact Between the Franciscan Melange and the Great Valley Sequence - Crustal Expression of a Late Mesozoic Benioff Zone: J ournel of Geophysical Research, v. 75, p. 886-901. 260 Eschard R., E. a., 1998, Combining Sequence Stratigtraphy, Geostatical Simulations and Production Data for Modeling a Fluvial Reservoir in the Chaunoy Field (Triassic, France): AAPG Bulitin, v. 82, p. 545-586. Fogg, G. E., Carle, SF, and Green, C., 1998, Connected-network paradigm for the alluvial aquifer system, in D. Zhang, Winter, L.C., ed., Theory, modeling, and field investigation in hydrogeology; a special volume in honor of Shlomo P. Neuman's 60th birthday, Special Paper - Geological Society of America: Meeting: Shlomo P. Neuman; recent advances after 30 years of exceptional contributions to well hydraulics, numerical modeling, and field investigations, Tucson, AZ, United States, Oct. 17, 1998., p. 25-42. Freeze, R. A., Cherry, J. A., 1979, Groundwater: Englewood Cliffs, Prentice Hall, 604 p. Freyberg, D. L., 1986, A natural gradient experiment on solute transport in a sand aquifer; 2, Spatial moments and the advection and dispersion of nonreactive tracers: Water Resources Research, v. 22, p. 2031-2046. Froukh, L. J ., 2002, Groundwater modeling in aquifers with highly karstic and heterogeneous characteristics (KHC) in Palistine: Bridging the gap between measurement and modeling heterogeneous media. Gaud, M. N., Smith, Gary A., 2001, Evaluating the transition of lithofacies to hydrofacies using outcrop data: Geological Society of America, v. 33, p. 16. Gloaguen, E., Chouteau, M., Marcotte, D., and Chapuis, R., 2001, Estimation of hydraulic conductivity of an unconfined aquifer using cokriging of GPR and hydrostratigraphic data: Journel of Applied Geophysics, v. 47, p. 135-152. Hamilton, W., 1969, Mesozoic Califomia and the Underflow of the Pacific Mantle: Geologic Society of America Bulletin, v. 80, p. 2409-2430. Hantush, M. S., and Jacob, C. E., 1955, Nonsteady radial flow in an infinite leaky aquifer: Trans. American Geophysical Union, v. 36, p. 95-100. Harbaugh, A. W., Banta, E.R., Hill, M.C., and Mcdonald, MG, 2000, Modflow-2000, The US. Geological Survey Modular Ground-Water Model-~User Guid to Modularization conceptsand the Ground-Water Flow Process, Reson, Virginia, US. Geological Survey, p. 121. Harrach, R. J ., Gallegos, G.M., F ailor, R.A., Christofferson, E., Tate, P.J., Brandstetter, E.R., Larson, J.M., McIntyre, J .R., Fields, B.C., Brown, R.A., Garcia, L.M., and Grayson, A.R., 1995, Environmental report 1994, Livermore, CA, Lawrence Livermore National Laboratory, p. 357. 261 : .- .I d ., ...d. r. .0.) .‘ l ...A I. 2.. I): :14. F. .93. ‘ , ‘3 ...-n. i . uva.4. . I . tau-‘10. ..r,.(1 Lu»... . a. L v. 1 [:9 . 3.5.3 . ... .....Cf . 11!. Hill, M. C., Banta, E.R., Harbaugh, A.W., and Anderman, ER, 2000, Modflow-2000, The US. Geological Survey Modualr Ground-Water Model--Userguid to the Observation, Sensitivity, and Parameter-Estimation Process and Three Post- Processing Programs, Denver, Colorado, US. Geological Survey, p. 209. Howard, A. D., 1979, Geologic History of Middle California: California Natural History Guides, University of California Press, 113 p. Hsu, K. J ., 1971, Franciscan Melanges as a Model for Eugeosynclinal Sedimantation and Underthrusting Tectonics: J oumeal of Geophysical Research, v. 76, p. 1162-1170. Huey, A. S., 1948, Geology of the Telsa Quadrangle, California: California Division of Mines, Bulletin, v. 140, p. 52. Inc., R., 2002, Rockwork52002, Golden, Colorado, RockWare Incorparated, p. 328. Klingbeil, R., Kleineidam, S., Asprion, U., Aigner, T., Teutsch, G., 1999, Relating Lithofaces to hydrofacies; outcrop—based hydrogeologic charaterization of Quaternary gravel deposits: Sedimentary Geology, v. 129, p. 253-264. Koltermann, C. E. a. G., SM, 1996, Heterogeneity in sedimentary deposits; a review of structure-imitating, process-imitating, and descriptive approaches,: Water Resources Research, v. 32, p. 2617-2658. Kruseman, G. P., and de Ridder, N. A., 1990, Analysis and evaluation of pumping test data: Wageningen, ILRI Publication 47. Newman, S. P., and Witherspoon, Paul A., 1972, F iels Determination of the Hydraulic Properties of Leaky Multiple Aquifer Systems: Water Resources Research, v. 8, p. 1284-1298. Noyes, C. M., Maley, M.P., and Blake, R., 1997, Hydrostratigraphic analisis- implimenting cost-effective superfund ground water cleanup, Larwence Livermore National Laboratory, p. 7. Noyes, C. M., Maley, M.P., and Blake, R., 2000, Defining Hydrostratigraphic Units within the Heterogeneous Alluvial Sediments at Lawrence Livermore National Laboratory, Lawrence Livermore National Laboratory, p. 33. Passinos, P. V., 2001, Well test to characterize idealized lateral heterogeneities: Maters thesis, Clemson Univerity, Clemson, SC, United States, 159 p. Ritzi, R. W., Jr., Dominic, D.F., Brown, N.R., Kausch, K.W., McAlenney, P.J., and Basia], M.J., 1995, Hydrofacies distribution and correlation in the Miami Valley aquifer system: Water Resources Research, v. 31, p. 3271-3281. 262 Ritzi, R. W., JR., Jayne, D.F., Zahradnik, A.J., Field, A.A., and Fogg, GE, 1994, Geostatistical modeling of heterogeneity in glaciofluvial, buried-valley aquifers: Ground Water, v. 32, p. 666-674. (RWQCB), S. F. B. R. W. Q. C. B., 1982, Water Quality Control Plan, Oakland, CA, San Francisco Bay Basin, State of California. Schad, H., and Teutsch, G., 1994a, Effects of the investigation scale on pumping test results in heterogeneous porous aquifers, in H. S. Wheater, Raats, P.A.C., and Armstrong, A.C., ed., Field, laboratory and modelling studies of flow and transport processes: Meeting: 17th general assembly of the European Geophysical Society symposium on Field, laboratory and modelling studies of flow and transport processes, Edinburgh, United Kingdom, Journal of Hydrology. Schad, H., and Teutsch, G., 1994b, Effects of the investigation scale on pumping test results in heterogeneous porus aquifers: Journal of hydrology, v. 159, p. 61-7 7. Sudicky, E. A., 1986, A natural gradient experiment on solute transport in a sand aquifer; spatial variability of hydraulic conductivity and its role in the dispersion process: Water Resources Research, v. 22, p. 2069-2083. Sweeney, J. J ., and Springer, J .E., 1981, Geology of Southeastern Livermore Valley, Alameda County, California, Livermore,CA, Larwence Livermore National Laboratory. Teles, V., Perrier, E., Delay, F., and Marsily, Gh. de, 2002, Generation of alluvial aquifers with a new gentic/stochastic sedimentation model: comparison with geostatistical approaches by means of groundwater flow simulations: Bridging the gap between measurement and modeling heterogeneous media. Theis, C. V., 1935, The relation between the lowering of piezometric surface and the rate and duration of the discharge of a well using groundwater storage: Trans American Geophysical Union, v. 16, p. 519-524. Thorpe, R. K., Isherwood, W.F., Dresen, M.D., and Webster-Scholten, CR, 1990, CERCLA Remedial Investigation Report for the LLNL Livermore Site, Livermore, CA, Lawrence Livermore National Laboratory. University, B. Y., 2002, Department of Defence Groundwater Modeling System (GMS), Version 4.0, Salt Lake City, UT. Webb, E. K., and Davis, J .m.,, 1998, Simulation of the spatial heterogeneity of geological properties: an overview, in G. S. a. D. Fraser, J. M., ed., Hydrogeologic Models of Sedimentary Aquifers, SEPM Concepts in Hydrogeology and Enviromental Geology No. 1, p. 1-24. 263 Weissmann, G. S., 1999a, Toward new models of subsurface heterogeneity: An alluvial fan sequence framewqork with transition probability geostatistics: Dissertation thesis, University of California, Davis, 279 p. Weissmann, G. S., 2001, Geologic Characterization of the Helipad Site (TFD), Lawrence Livermore National Laboratory, California, East Lansing Michigan, Michigan State University, p. 62. Weissmann, G. S., in press, Influence of incised valley fill deposits on hydrogeology of a glacially-influenced, stream-dominated alluvial fan, in J. B. a. D. Hyndman, ed., SEPM special publication on Hydrogeophysics and Hydrostratigraphy. Weissmann, G. S., and Fogg, G. E., 1999b, Multi-scale Alluvial Fan Hetrogeneity modeled with transition probability geostatistics in a sequence stratigraphic framework: Journel of Hydrology, v. 226, p. 48-65. Weissmann, G. S., Carle, SA, and Fogg, G.E., 1999c, Three-dimensional hydrofacies modeling based on soil survey analysis and transition probability geostatistics: Water Resources Research, v. 35, p. 1761-1770. Weissmann, G. 8., Mount, J .F., and Fogg, G.E., 2002a, Glacially-driven cycles in accumulation space and sequence stratigraphy of a stream-dominated alluvial fan, San Joaquin Valley, Califonia, U.S.A: Joumeal of Sedimentary Research, v. 72, p. 240-251. Weissmann, G. S., Yong, Z., LaBolle, E.M., and Fogg, G.E., 2002b, Modeling environmental tracer-based groundwater ages in heterogeneous aquifers: Water Resources Research. Yang, Y. s., Cronin, A.A., Elliot, T., and Kalin, R.M., 2002, Characterizing heterogeneous hydro geological systems through inverse groundwater flow and geochemical modeling: Bridging the gap between measurement and modeling heterogeneous media. Zhan, H., Wang, L.V., and Park, E., 2001, On the horizontal-well pumping test in anisotropic confined aquifers: Journal of hydrology, v. 252, p. 37-50. Zhang, Y., Weissmann, G.S., Fogg, G.E., Williamson, R.J., and Labolle, E.M., 2000, Effect of multi-scale heterogeneity on predictive uncertainty and groundwater age dates: AGU 2000 fall meeting, p. 434. 264 IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII Illlljllllzliflyllllfll.flflylll