Geologic and Hydrologic Constraints on Fluid and Heat Flow in Overpressured Rocks of the Rio Grande Embayment, Gulf of Mexico Basin by Thomas Edward McKenna, B.S., M.S. Dissertation Presented to the Faculty of the Graduate School of TheUniversity ofTexasatAustin in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy The University of Texas at Austin August 1997 Geologic and Hydrologic Constraints on Fluid and Heat Flow in Overpressured Rocks of the Rio Grande Embayment, Gulf of Mexico Basin ToKim And Jillian Acknowledgments Thank for you your support and friendship. adviser: Jack Sharp committee: Phil Bennett, Earle Mcßride, Mark Miller, Charlie Kreitler family: Kim, Jillian, Patsy, Skip, Sheil, Heather, Bill, Debbie, Bill, Jr., Jason, Doey, Ron, Cheryl, Amanda, Justin, Michael, Lishah, and Jess Frank, Mary, Roy, Michael, Kathy, Wayne, Ryan, Nathan research assistants: J. Drummond, Carolyn Cooper, Brad Wolaver, Christi Cell funding: US Department of Energy, American Chemical Society, Geological Society of America, American Association of Petroleum Geologists, Gulf Coast Association of Geological Societies, Industrial Sponsors of the UT Gulf Coast Diagenesis Group, UT Geology Foundation, editor: Laurie Schuur; equipment: Rich Ketcham, Dave Blackwell samples: Lynton Land, Kitty Milliken, Leo Lynch, Tim Diggs friends: Phil, Chris, Mahk, Tony, Matt, Gay, Ashley, Ingrid, Fu Li, Mel, Leo, Fred, Betty, Scotty, Jim, Stacy, Todd, Li Xiang, Kurt, Carol, Danielle, Dan, Liu Qunling, Li Ning, Laurie, Brian, Cambria, Barb, Mark, Katherine, Marty, NJ, Florida, and Carolina friends, Rosedale neighbors discussions: Bill Powell, Bill Galloway, David Maidment inspiration: Ornette, Trane, Jerry and boys, Captain 8., Lao Tzu, Dick Feynmann “Goodnite Austin, Texas, wherever you are.” (fz) Geologic and Hydrologic Constraints on Fluid and Heat Flow in Overpressured Rocks of the Rio Grande Embayment, GulfofMexico Basin Publication No. Thomas Edward McKenna, Ph.D. The University of Texas at Austin, 1997 Supervisor: John M. Sharp, Jr. Fluids are episodically expulsed from extremely overpressured sediments during natural hydrofracturing events and discharge vertically along regional fault zones in the Rio Grande Embayment. The basal 9 km (29,500 ft) of the stratigraphic section is extremely overpressured with fluid pressures at 80 to 90% of the overburden close to the minimum needed for the onset of pressure, pressure hydraulic fracturing. Small increases in fluid pressure in this extremely overpressured regime trigger hydraulic fracturing and episodes of fluid discharge. The most recent pulse of fluid discharge was along the Wilcox Fault Zone where the largest thermal anomaly in the Gulfof Mexico Basin occurs coincident with a positive fluid pressure anomaly and salinity inversions. Heat conduction is not a viable mechanism for producing the anomaly. Heterogeneous salinity distributions along the Wilcox, Vicksburg, the and Frio Fault Zones represent cumulative effect ofexpulsion events in the fault zones. Neither distinct thermal nor pressure anomalies are evident in the Vicksburg or Frio Fault Zones, suggesting that fluid expulsion events had a shallower fluid source, were of smaller magnitude, were more diffuse than in the Wilcox Fault Zone, or that similar thermal and pressure anomalies have already dissipated. The thermal conductivity of 83 Wilcox and Frio sandstones from the Embayment were measured and conductivity was correlated to petrographic variables. Thermal conductivity ranges from 2.06 to 5.73 W/m/K over a porosity range of 2.4 to 29.6 %. For a given porosity, because of a higher quartz content, Wilcox sandstones are more conductive than Frio sandstones. Thermal conductivities of clean (< 25% clay) sandstones can be described by a multilinear and function of decreasing thermal conductivity with increasing porosity increasing thermal conductivity with quartz content. Radiogenic heat production within the sedimentary section of the Embayment is a significant source of heat, contributing up to 26% of the overall surface heat-flow density in the Embayment. Heat production rates range from a 33 lowof0.07±O.OlpW/m incleanStuartCitylimestonesto2.21±.24pW/m in Frio mudrocks. Mean heat production rates for Wilcox sandstones, Frio sandstones, Wilcox mudrocks, and Frio mudrocks are 0.88, 1.19, 1.50, 1.72, and 3 more pW/m respectively. In general, the mudrocks produce about 30-40% heat , than stratigraphically equivalent sandstones. Frio rocks produce about 15% more heat than Wilcox rocks per unit volume of clastic rock (sandstone/mudrock). Table of Contents Chapter 1: Introduction 1 Chapter 2: Subsurface Temperatures, Fluid Pressures, and Salinities in theRioGrandeEmbayment,GulfofMexicoBasin 8 Abstract 8 Introduction 9 Background 11 Methods 28 Results 32 Discussion 46 Conclusions 62 Chapter 3: Thermal conductivity of Wilcox and Frio sandstones in the Rio Grande Embayment, GulfofMexico Basin 63 Abstract 63 Introduction 64 Samples and Measurements 66 Results 68 Discussion 71 Conclusions 80 Chapter 4: Radiogenic heat production in sedimentary rocks in the Rio Grande Embayment, GulfofMexico Basin 82 Abstract 82 Introduction 83 Previous Work and Significance 85 Methodology 88 Results of Heat Production Calculations 91 Thermal Models 91 Implications 94 Conclusions 95 ChapterS: Summary 97 Chapter 6: Suggestions for Future Work 101 References 224 List of Tables Table 2.1. Temperatures picked from temperature-depth plots 201 Table 2.2. Fluid surfaces and fluid / pressures, tops of overpressure overburden pressure ratios picked from pressure-depth plots 206 Table 3.1. Thermal conductivity, porosity, and petrographic data 210 Table 3.2. Thermal conductivity measured parallel and perpendicular to stratigraphic bedding 214 Table 3.3. Mineral thermal conductivities 214 Table 4.1. Abundances of naturally occurring radioactive elements 215 Table 4.2. Details on sample and stratigraphic section locations 216 Table 4.3. Radiogenic heat production data 217 Table4.4. Summarystatisticsformeasurementsofradiogenicheat production 221 Table 4.5. Model equations, parameter descriptions, and input values 222 List of Figures Figure 1.1. Regional location map 105 Figure 2.1. Local location map 106 Figure 2.2. Geologic cross section (see Figure 2.7 for location) 107 Depth to basement (Sawyer et al., 1991) 108 Figure 2.3. 109 Figure 2.4. Bouger gravity anomaly map (SEG, 1982) 110 Figure 2.5. Magnetic anomaly map (Godson, 1982) 11l Figure 2.6. Salt diapir provinces (Ewing, 1991) Figure 2.7. Location map for Cretaceous margins, fault trends, and cross section 112 Figure 2.8. Stratigraphic chart with depositional episodes, depocenter locations, and tectonic events (Galloway et al., 1991) 113 Figure 2.9. Location of Wilcox Rosita and Rockdale Delta Systems (Edwards, 1981) 114 Figure 2.10. Location of Frio Formation depositional systems (Galloway et al., 1982) 115 Figure 2.11. Orientation of maximum horizontal stress (Zoback et al., 1991).. 116 117 Figure 2.12. Depth to the top of fluid overpressure Figure 2.13. Depth to the base of the Texas Gulf Coast aquifer systems (Ryder, 1988). This was mapped as the shallower of the top of fluid overpressure (irregular pattern in the east) or the Paleocene Midway Group (linear pattern in the west) 118 Figure 2.14. Depth to the top of extreme overpressure 119 Figure 2.15. Thickness of the overpressured regime 120 Figure 2.16. Fluid/overburden pressure ratio in the extremely overpressured regime 121 122 Figure2.17. DefinitionsofhydrologicregimesintheGulfofMexico Figure 2.18. Depth to the top of the oil industry’s “hard” overpressure 123 Figure 2.19. Thermal anomaly in map view (isotherm map for the 3 km (9,840 ft) depth-slice) 124 Figure 2.20. Grid overlay for location of data used in temperature-depth and pressure-depth plots 125 Figure 2.21. Isotherm for the 4.5 km (14,760 ft) depth-slice 126 map Figure 2.22. Isotherm mapfor the 4.0 km (13,120 ft) depth-slice 127 Figure 2.23. Isotherm mapfor the 3.5 km (11,480 ft) depth-slice 128 Isotherm 129 Figure 2.24. mapfor the 3.0 km (9,840 ft) depth-slice Figure 2.25. Isotherm mapfor the 2.5 km (8,200 ft) depth-slice 130 Figure 2.26. Isotherm for the 2.0 km (6,560 ft) depth-slice 131 map Figure 2.27. Isotherm mapfor the 1.5 km (4,920 ft) depth-slice 132 Isotherm 133 Figure 2.28. map for the 1.0 km (3,280 ft) depth-slice Figure 2.29. Isotherm map for the 0.5 km (1,640 ft) depth-slice 134 Figure 2.30. Temperature contoured on geologic cross section (see Figure 2.7 for location) 135 Figure 2.31. Temperature-depth plot for data northwest of the Wilcox Fault Zone along the cross section in Figure 2.30 (50-100 km [3l-62 mi] from northwestern side of section) 136 Figure 2.32. Temperature-depth plot for data northwest of the Wilcox Fault Zone along the cross section in Figure 2.30 (101-150 km [63-93 mi] from northwestern side of section) 137 Figure2.33. Temperature-depthplotfordataintheWilcoxFaultZonealong the cross section in Figure 2.30 (151-190 km [94-118 mi] from northwestern side of section) 138 Figure 2.34. Temperature-depth plot for data between the Wilcox and Vicksburg Fault Zones along the cross section in Figure 2.30 (191-225 km [ll9-140 mi] from northwestern side of section).... 139 Figure2.35. Temperature-depthplotfordatabetweentheWilcoxand Vicksburg Fault Zones along the cross section in Figure 2.30 (226-245 km [l4O-152 mi] from northwestern side of section).... 140 Figure 2.36. Temperature-depth plot for data in the Vicksburg Fault Zone along the cross section in Figure 2.30 (246-275 km [153-171 mi] from northwestern side of section) 141 Figure 2.37. Temperature-depth plot for data in the Frio Fault Zone along the cross section in Figure 2.30 (276-340 km [172-211 mi] from northwestern side of section) 142 Figure 2.38. Temperature-depth plot for combined data in the Wilcox and Frio Fault Zones along the cross section in Figure 2.30 143 Figure 2.39. Temperature-depth plot for the Wilcox Fault Zone in grid area 1007 in Figure 2.20 144 Figure 2.40. Temperature-depth plot for the Wilcox Fault Zone in grid area 1107 in Figure 2.20 145 Figure 2.41. Temperature-depth plot for the Wilcox Fault Zone in grid area 409 in Figure 2.20 146 Figure2.42. Temperature-depthplotfortheVicksburgFaultZoneingrid area 514 in Figure 2.20 147 Figure2.43. Temperature-depthplotfortheFrioFaultZoneingridarea 1012 in Figure 2.20 148 area Figure 2.44. Temperature-depth plot for the Frio Fault Zone in grid 1311 in Figure 2.20 149 815 Figure 2.45. Temperature-depth plot for the Frio Fault Zone in grid area in Figure 2.20 150 Isobar 151 Figure 2.46. map for the 3.5 km (11,480 ft) depth-slice Figure 2.47. Isobar map for the 3.0 km (9,840 ft) depth-slice 152 Figure 2.48. Isobar map for the 2.5 km (8,200 ft) depth-slice 153 Figure 2.49. Isobar map for the 2.0 km (6,560 ft) depth-slice 154 Figure 2.50. Isobar map for the 1.5 km (4,920 ft) depth-slice 155 Figure 2.51. Isobar map for the 1.0 km (3,280 ft) depth-slice 156 Figure 2.52. The ratio of fluid pressure to overburden pressure contoured on geologic cross section A-A’ (Figure 2.2) 157 Figure2.53. Pressure-depthplotfordatanorthwestoftheWilcoxFaultZone along the cross section in Figure 2.52 (40-100 km [25-62 mi] from northwestern side of section) 158 Figure 2.54. Pressure-depth plot for data northwest of the Wilcox Fault Zone along the cross section in Figure 2.52 (101-150 km [63-93 mi] from northwestern side of section) 159 Figure 2.55. Pressure-depth plot for data in the Wilcox Fault Zone along the cross section in Figure 2.52 (151-190 km [94-118 mi] from northwestern side of section) 160 Figure 2.56. Pressure-depth plot for data between the Wilcox and Vicksburg FaultZonesalongthecrosssectioninFigure2.52 (191-225km [ll9-140 mi] from northwestern side of section) 161 Figure 2.57. Pressure-depth plot for data between the Wilcox and Vicksburg Fault Zones along the cross section in Figure 2.52 (226-245 km [l4O-152 mi] from northwestern side of section) 162 Figure2.58. Pressure-depthplotfordataintheVicksburgFaultZonealong the cross section in Figure 2.52 (246-275 km [153-171 mi] from northwestern side of section) 163 Figure2.59. Pressure-depthplotfordataintheFrioFaultZone alongthe cross section in Figure 2.52 (276-340 km [172-211 mi] from northwestern side of section) 164 Figure 2.60. Pressure-depth plot for combined data in the Wilcox and Frio Fault Zones along the cross section in Figure 2.52 165 Figure 2.61. Pressure-depth plot for the Wilcox Fault Zone in grid area 1007 in Figure 2.20 166 Figure 2.62. Pressure-depth plot for the Wilcox Fault Zone in grid area 1107 in Figure 2.20 167 Figure 2.63. Pressure-depth plot for the Wilcox Fault Zone in grid area 409 in Figure 2.20 168 Figure 2.64. Pressure-depth plot for the Vicksburg Fault Zone in grid area 514 in Figure 2.20 169 1012 in Figure2.65. Pressure-depthplotfortheFrioFaultZone ingridarea Figure 2.20 170 1311 in Figure 2.66. Pressure-depth plot for the Frio Fault Zone in grid area Figure 2.20 171 Figure 2.67. Pressure-depth plot for the Frio Fault Zone in grid area 717 in Figure 2.20 172 173 Figure 2.68. Salinity map for the 3-3.5 km (9,840-11,480 ft) depth interval .. Figure 2.69. Salinity map for the 2.5-3.0 km (9,840-11,480 ft) depth interval 174 Figure 2.70. Salinity map for the 2-2.5 km (6,560-8,200 ft) depth interval 175 .... Figure 2.71. Salinity map for the 1.5-2 km (4,920-6,560 ft) depth interval 176 .... Figure 2.72. Salinity map for the 1-1.5 km (3,280-4,920 ft) depth interval 177 .... Figure 2.73. Salinity map for the 0.5-1 km (1,640-3,280 ft) depth interval 178 .... Figure 2.74. Salinity plotted on cross section A-A’ (Figure 2.2) 179 Figure 2.75. Chloride-depth plot for data northwest of the Wilcox Fault Zone along the cross section in Figure 2.52 (40-100 km [25-62 mi] from northwestern side of section) 180 Figure2.76. Chloride-depthplotfordatanorthwestoftheWilcoxFaultZone along the cross section in Figure 2.52 (101-150 km [63-93 mi] from northwestern side of section) 181 Figure 2.77. Chloride-depth plot for data in the Wilcox Fault Zone along the cross section in Figure 2.52 (151-190 km [94-118 mi] from northwestern side of section) 182 Figure 2.78. Chloride-depth plot for data between the Wilcox and Vicksburg Fault Zones along the cross section in Figure 2.52 (191-225 km [ll9-140 mi] from northwestern side of section) 183 Figure 2.79. Chloride-depth plot for data between the Wilcox and Vicksburg Fault Zones along the cross section in Figure 2.52 (226-245 km [l4O-152 mi] from northwestern side of section) 184 Figure 2.80. Chloride-depth plot for data in the Vicksburg Fault Zone along the cross section in Figure 2.52 (246-275 km [153-171 mi] from northwestern side of section) 185 Figure 2.81. Chloride-depth plot for data in the Frio Fault Zone along the cross section in Figure 2.52 (276-340 km [172-211 mi] from northwestern side of section) 186 Figure 2.82. Locations of silica knobs and uranium mines with ores of an epigenetic origin 187 Location Figure 3.1. map for samples with thermal conductivity measurements 188 - 189 Figure 3.2. QFR (Quart-Rock Fragment Feldspar) ternary diagram Figure 3.3. Thermal conductivity versus porosity 190 Figure 3.4. Thermal conductivity contoured in quartz-porosity space 191 Figure 3.5. Predicted versus measured thermal conductivity for clean sandstones, a.) linear regression model, b.) multi-component geometric mixing model, c.) a two-component geometric mixing model 192 Figure 3.6. Predicted versus measured thermal conductivity for clean sandstones: a.) arithmetic mixing model for grain-matrix conductivity and a 2-component geometric mixing model for effective thermal conductivity, b) multiple linear regression model with quartz and porosity as the independent variables 193 Figure 3.7. Predicted versus measured thermal conductivity for data from other sources 194 Figure 3.8. Domain (shaded) of multiple-linear regression model with calculated thermal conductivity (W/m/K) contoured in porosity-quartz space 195 Figure 4.1. Locations of samples with radiogenic heat production measurements and stratigraphic section 196 197 Figure 4.2. Graphical summary of radiogenic-heat production data Figure 4.3. Radiogenic-heat production versus depth 198 Figure 4.4. Model parameters with depth calculated from equations from Table 4.5 199 Model results 200 Figure 4.5. Introduction Chapter 1: The Gulf of Mexico Sedimentary Basin is the greatest energy resource region in North America and of the foremost petroleum provinces in the one world. It is a hydrocarbon megaprovince with a known ultimate recovery of over 220 billion barrels of oil equivalent and is comparable in size to the Arabian- Iranian province of the Middle East (Nehring, 1991). It also hosts reserves of lignite, uranium, and metallic-sulfide ore deposits (Riggs et al., 1991). There has been extensive exploration for and production of these for the last 100 resources years as exemplified by the development of over 4,500 oil and gas fields. the Although the origins of these resources are dependent on hydrodynamic and/or geothermal histories of the basin, these aspects of the basin’s evolution are still poorly characterized. Numerical models of the basin's burial, fluid flow, and thermal histories are available (Sharp and Domenico, 1976; Bethke, 1986; Blanchard, 1987; Pantano et al., 1990; Harrison and Summa, 1991; Lerche and McKenna, 1991; Mello et al., 1994; Mello and Kamer, 1996; Nunn, 1996) and have contributed significantly to our conceptualization of basin processes, but, still, these are only crude representations of reality until they are adequately constrained by some of the fundamental data required for regional geothermal and hydrogeological studies. If we want to get beyond the “broad-brush” the evolution of approach to modeling the basin’s hydrogeology and it’s geothermal systems, we need to quantify these fundamental data. These data include the magnitudes and distributions of fluid pressure and hydraulic potential, formation-water salinity, formation temperature, and the thermal and hydraulic properties of the rocks. These data need to be evaluated in a holistic approach and worked into a testable conceptual model that couples the interwoven processes of sediment deposition, compaction, fluid flow, heat and mass transfer, some and diagenesis. This is an attempt to provide and analyze of these data in the context of evaluating the hydrodynamic and geothermal regimes of the onshore south Texas part of the Rio Grande Embayment (Embayment) (Figure 1.1). Results are presented from analyses of the thermal properties of rocks from the Embayment and from the development and analysis of a large geographic­information-system database of subsurface temperatures, fluid pressures, and salinities. These data, along with the available regional geological and are are geophysical interpretations, used to evaluate the hypothesis that fluids episodically expulsed from overpressured rocks, discharging vertically along the regional Wilcox and Frio Fault Zones in the Embayment. Results indicate that the hypothesis is viable; it is consistent with geologic, geochemical, hydrologic, and geothermic observations. The results are presented as a set of three independent papers organized as three chapters (Chapters 2,3, and 4). Chapter 2 provides a regional view of fluid and heat flow in the Embayment, while Chapters 3 and 4 focus on laboratory-scale measurements of the thermal properties of rocks from the Embayment. Chapters 5 and 6 summarize the results and give suggestions for future research, respectively. The results of temperature, pressure and salinity mapping are given in 2 Chapter along with a synthesis of the present-day hydrodynamic and geothermal regimes of the Embayment. A conceptual model is presented for the episodic flow of fluids in the overpressured rocks and discharge along regional fault zones. Regional hydrogeologic studies of overpressured basins that are based on fluid observations (like this study) are sparse. The use of pressure pressure data for mapping has received relatively widespread use in hydropressured and underpressured basins (Western Canadian [Alberta] Basin, Rocky Mountain Basins) where gravity driven flow is dominant. Most previous studies of overpressured basins simply map the “top of soft and hard overpressure” (see definitions below) which gives only a first-order view of the some hydrodynamics. Existing studies typically use sort of proxy information which (e.g. well logs, drilling-mud weights) for mapping the top of overpressure is proving to be inaccurate (Hermanrud, 1994; Leftwich and Engelder, 1994). To my knowledge, this is the first regional study in the Gulf of Mexico Basin to use Kreitler et al. (1988, actual pressure data for mapping of virgin fluid pressures.. data set fluid a 1990) used part of this same to map present-dav pressures on regional scale for the Frio Formation in coastal Texas. Scientists at Louisiana State University and the Louisiana Geological Survey have provided sub-regional and reservoir-scale work (e.g. Manor and Bailey, 1983; McCulloh, 1985, 1988; Bray and Manor, 1990; Manor and Sassen, 1990). In the Embayment, trends in fluid observed flow are on pressures a regional scale implying a relatively open system within the overpressured regime. initiated by Bodner Detailed temperature mapping in the Embayment was and Sharp (1988) and Pfeiffer and Sharp (1989) and was used as the starting point for this study. Finer-scale mapping and analyses of additional data in this study constrain the shape and position of a thermal anomaly along the Wilcox Fault Zone and clearly show that it is centered on the most-basinward of the faults. Forced convection of heat by upwardly discharging fluids is the preferred explanation for the thermal anomaly. Salinity mapping augments the extensive work of Land and others at The University of Texas (e.g. Morton and Land, 1987; Land and MacPherson, 1992; Land, 1995) and confirms the extremely heterogeneous distribution of salinity. General trends of higher salinity in the Oligocene and younger formations relative to waters in the Paleocene/Eocene formations is confirmed. In general, there to be higher salinities along appear some parts of the fault zones. Salinity inversions also are noted. Abridged versions of this chapter are currently in press in conference proceedings (McKenna and Sharp, in press [a]; McKenna, in press). Chapter 3 is the thermal conductivity of Wilcox and Frio sandstones on and was published in the American Association of Petroleum Geologists Bulletin (McKenna et al., 1996). The measurements are the first reported for this area and supplement the extremely sparse information for the Gulf of Mexico and in the geothermal and petroleum literature in general. Arguably, the data provided in this are the most well-documented of any currently available in the paper literature for testing mechanistic models of the thermal of conductivity sandstones. Researchers in this field have largely been constrained to using the few measurements from the work of Woodside and Messmer (1961a, 1961b) when they wanted a high-accuracy data set that included detailed petrographic descriptions. Results indicate that, at a given porosity, Wilcox sandstones have higher conductivities than Frio sandstones. Most variance in thermal conductivity can be accounted for by variations in quartz content and porosity. Chapter 4 is on the radiogenic heat production of sedimentary rocks in the area and is in press in the American Association of Petroleum Geologists Bulletin (McKenna and Sharp, in press [b]). Again, the measurements are the first for this reported area and supplement the extremely sparse information for the Gulf of Mexico. Results indicate that radiogenic heat is a significant source of heat and should be included in thermal modeling. It contributes up to 26% to the overall surface heat flow in south Texas. The magnitude and distribution of the heat source is not amenable to being a contributor to the observed thermal anomaly. In the course of this study, two problems stood out and deserve an open- minded re-evaluation of our concepts of fluid flow in the basin. The first is the fluid-volume problem. Where do the large volumes of water come from that are needed to do the pervasive diagenetic work that is clearly documented in the literature (e.g. Wood and Hewett, 1984; Sharp et al., 1988; Weedman et al., 1996; Land et al., 1997)7 How are the observed extreme fluid overpressures maintained if these large volumes of fluids are flushing through the basin? Is there a source of recharge “down there” (Land et al., 1995) where the basin fill and underlying recirculate via free crust are undergoing prograde metamorphism? Do fluids convection in the hydropressured and extremely overpressured zones? At this answers point, the are not clear. The other outstanding problem (Dutta, 1987; Blackwell and Steele, 1989; Land, 1994; McKenna and Sharp, 1996) is our meager understanding of the thermal and hydraulic properties of the mudrocks that make up 80-90% of the basin fill. How do fluids move in mudrock within a “hostile environment” (Fowler 1970, p. 419) where fluid pressures appear to be maintained near those required for the onset of hydraulic fracturing of the rock? Are the concepts of permeability and Darcy’s law, that we usually take for granted, valid in this type of regime fluid primarily upwards in or does flow “devious paths” (Jones, 1969, 809) in pulses within evolving fractures p. (Walther, 1990) at extremely fast velocities approaching 1000 m/yr (3,280 ft/yr) (Nunn, 1996)? Again, the answer is not clear. Given the fact that there are so many wells drilled in the Gulf of Mexico and the existence of an immense literature base, it is to be lulled into easy thinking that the properties of the basin fill are well-characterized and our concepts of basin processes are well-understood. This is certainly not the case. While we have a relatively good understanding of the geology of the basin fill and of some of it’s shallow hydrogeologic characteristics, the available data for describing and analyzing the bulk of the basin’s hydrogeology and geothermal systems is minuscule and has an embarrassingly low accuracy (particularly for the mudrocks). a At the time of this writing, the Gulf of Mexico is again “hot” prospect for oil and gas exploration. My hope is that this research will provide some thermal and hydrologic constraints for research and exploration problems during this period and beyond. problems include sedimentary basin evolution, These petroleum maturation and migration, sedimentary diagenesis, and the formation of inorganic mineral deposits. As we try to understand these problems and we must develop quantitative descriptions of the processes involved, overcome the “cruel lack of data” (Bumis, 1986, 2) on the thermal and hydraulic p. properties of the rocks and recognize the “glaring omission” (Dutta, 1987, p. viii) of characterizing the physical and chemical properties of 80-90% of the basin fill. Chapter 2: Subsurface Temperatures, Fluid Pressures, and Salinities in the Rio Grande Embayment, Gulf of Mexico Basin Abstract Fluids are episodically expulsed from extremely overpressured sediments during natural hydrofracturing and discharge vertically along the regional Wilcox and Frio Fault Zones in the Rio Grande Embayment. The basal nine kilometers of the stratigraphic section are extremely overpressured with fluid pressures at 80 to 90% of lithostatic close to the minimum needed for the onset pressure, pressure of hydraulic fracturing. Small increases in fluid pressure caused by recharge to the system from hydrocarbon generation, diagenetic/metamorphic fluids, or other areas where the limiting pressure of hydraulic fracturing is higher, trigger hydraulic fracturing and episodes of fluid discharge. The Wilcox and Frio Fault are zones a more Zones both fluid discharge but the Wilcox Zone appears to be focused discharge area for fluids from a deeper fluid source. The most recent pulse of fluid discharge was along the Wilcox Fault Zone where the largest thermal anomaly in the Gulf of Mexico Basin occurs coincident with a positive fluid pressure anomaly and salinity inversions. The thermal anomaly is centered on the most basinward and deepest of the Wilcox faults. Heat conduction is not a viable mechanism for producing the anomaly. Hydraulic fracturing in the fault zone creates a relatively permeable vertical pathway that focuses fluid discharge from the extremely overpressured rocks along the deep-seated faults. Heterogeneous salinity distributions along the Wilcox, Vicksburg, and Frio Fault the cumulative effect of expulsion in the fault zones. events Zones represent Neither distinct thermal nor anomalies are evident pressure in the Vicksburg or Frio Fault Zones, suggesting that fluid expulsion events had a shallower fluid source, were of smaller magnitude, were more diffuse than in the Wilcox Fault Zone, or that thermal and pressure anomalies that once existed there have already dissipated. The deep source of fluids, hydraulic fracture, and episodic fluid expulsion hypothesis is consistent with the geologic, temperature, fluid pressure, salinity, and thermal properties data, with diagenetic observations, and with basic principles of hydrogeology and rock mechanics. The Rio Grande Embayment be may a regional discharge zone for the thick sedimentary section in the western Gulf of Mexico Basin. Introduction is the The Gulf of Mexico Sedimentary Basin greatest energy resource region in North America. It hosts huge proven reserves ofoil, natural gas, lignite, uranium, and mineral deposits. Although there has been extensive exploration for and production of these resources, the basin's fluid flow and thermal histories are still poorly characterized. Numerical modelsof the basin's burial, fluid flow, and thermal histories (Sharp and Domenico, 1976; Bethke, 1986; Blanchard, 1987; Pantano et al., 1990; Harrison and Summa, 1991; Lerche and McKenna, 1991; Mello et al., 1994; Mello and Kamer, 1996; Nunn, 1996) lend insight into the histories, but these models are not adequately constrained by fluid pressure, and data. A flow and model is temperature, salinity conceptual transport presented based on analysis of a large data set of formation temperatures, fluid and fluid salinities from the Rio Grande Embayment (the Embayment) pressures, on the northwestern margin of the basin (Fig. 1.1). In the model, fluids are expulsed episodically from extremely overpressured sediments during natural hydrofracturing and discharge subvertically along the Wilcox, Vicksburg, and Frio Fault Zones that rim the thick clastic section of the Embayment. These two fault zones have been fluid discharge zones at different times throughout geologic history. The most recent pulse of fluid discharge was along the Wilcox Fault Zone where positive thermal and pressure anomalies are observed. Salinity inversions along all of the fault zones show the cumulative effect of expulsion events in both fault zones. The largest oil and gas plays in the Embayment also occur along the fault zones. This is a mature hydrocarbon province, with production occurring since the late 19305, and the conceptual model provides insight into the charging of these important reservoirs. Study area 42 42 The study area covers an area of about 8.7 x 10 km(3.4 x 10 mi ) and is located in the coastal plain of Texas (Figures 1.1 and 2.1) south of San Antonio. It borders Mexico and the Gulf of Mexico. Topography in the area is relatively flat with the surface sloping at about one to two degrees towards the coast (southeast). The maximum relief in the area is only about 800 m (2,600 ft). The study focuses on analysis of subsurface data from depths of 0.5 to 4.5 km (1,640 to 14,800 ft) with particular emphasis on the section deeper than the top of fluid (where fluids in the rocks are at a than overpressure pressure greater hydropressure). Data and/or interpretations from shallower and deeper depths also are evaluated as available/needed. The study area includes the northern part of the geologically-defined (see below) Rio Grande Embayment. The southern a part of the Embayment (Burgos Basin) is in Mexico where only few published geological syntheses are available. Subsurface data from the Burgos Basin in Mexico would be a great asset but were not available for the study as they have not been released by Mexico’s National Petroleum Company (PEMEX) for public use. Background The Gulf of Mexico Basin is one of the most-cited ‘“classic” examples of a divergent margin. There is an extensive literature base describing the basin’s geologic and fluid-overpressure environments. The geothermal regime of the basin is not as well characterized. Brief overviews of the geology, hydrogeology, and geothermics of the Embayment are given below. Geology The Gulf of Mexico Basin is thick a sequence of sediments deposited on a divergent continental margin resulting from Late Triassic rifting of continental crust. The geology is generally characterized by extensive evaporite deposition in theJurassicduring atimeofbroadregionaltectonicsubsidence, developmentofa basin-rimming carbonate margin in the Cretaceous, and progradation of clastic sediment nearly 400 km (250 mi) basinward of the Cretaceous carbonate shelf margin in the Cenozoic (Figure 2.2). Extensive growth faults developed during deposition of the Cenozoic section due to mobilization of underlying salt and mudrock in overpressured response to the large depositional load. Extensive literature describes the evolution, geology, and structure of the basin; good reviews are available in Winker (1982), Salvador (1987), Curtis (1987), Sharp et al. (1988), Worrall and Snelson (1989), and Salvador (1991b). The following a description gives general overview of the geological development of the onshore south Texas part of the Rio Grande Embayment. The Embayment is the area located northeast of the Peyotes, Picachos, and Tamaulipas arches in Mexico, southeast of the basin-marginal Balcones Fault Zone (structural boundary on Figure 1.1) and southwest of the San Marcos Arch in Texas (Figure l.l). It is open to the Gulf of Mexico to the southwest. The section of the Embayment in northeast Mexico is typically referred to as the Burgos Basin. The northwest-trending frontal folds of the late Mesozoic- Cenozoic fold-thrust belt (Sierra Madre Oriental) occur about 100 km (60 mi) southwest of the Embayment. Basement The basin was initiated in the Late Triassic via crustal attenuation and concomitant sea-floor spreading associated with the breakup of the Pangea supercontinent (Pindell, 1985; Salvador, 1991a). The specific nature and distribution of the crust underlying the Mesozoic rocks in the area is not well known (Sawyer et al., 1991: Muehlberger and Land, 1988). Deep seismic- reflection data are not readily available because the thick pile of overlying clastic sedimentary rocks attenuates seismic while carbonates and possibly energy evaporites provide a large impedance contrast that is hard to penetrate (Sawyer et al., 1991). Sawyer et al. (1991) used reflection seismic, refraction seismic, gravity, magnetics, subsidence calculations, and velocity analysis to define the character of crust for the entire Gulf of Mexico Basin. gross the underlying Unfortunately, their data are very sparse in south Texas so the interpretation for this area is speculative. In south Texas, the depth to basement is generally to increasing towards the basin (southeast) with depths from 2 to 12 km (6,600 A local 39,400 ft) (Figure 2.3). northeast-trending basement high with an amplitude of about 2 km (6,600 ft) occurs just eastward of the present-day coastline. The basement surface steepens significantly to the northwest close to the axis of the San Marcos Arch in northern McMullen, northern Live Oak, Karnes, and Dewitt Counties and to the southwest towards the frontal thrusts of the Sierra Madre Oriental. The Embayment is underlain by “transitional continental crust.” on Primarily, this is based the assumption of the crustal thinning that is required to produce the needed accommodation space for the abrupt thickening of Tertiary sediments basinward of the Cretaceous shelf margin. Thin transitional crust underlies the area basinward (southeast) of the Cretaceous shelf margin. It is highly extended, and subsided faster and to greater depths than the adjacent thick transitional crust occurring northwestward of the shelf margin. The boundary between thick and thin transitional crust was mapped by Sawyer et al. (1991) based on the position of the Cretaceous shelf margin and appears to correspond to a major tectonic hinge zone in the basement. Thick transitional crust is characterized by well defined basement highs (e.g., San Marcos Arch) with intervening lows (e.g., the Embayment). The thinning of the transitional crust also is inferred by the general increase in gravity (Figure 2.4) and magnetic from northwest to southeast. (Figure 2.5) responses Proprietary seismic reflection profiles in Zapata County (in the southwest of the study area) show a fairly flat basement surface under the Wilcox Fault Zone (Ewing, 1986). Muehlberger and Land (1988) present several possibilities for the composition of the transitional crust. It may consist of rifted Grenville basement as exposed 160 km (100 mi) to thenorthwestin theLlanouplift,buriedrocksof the Ouachita trend(Flawn et al., 1961), or some exotic continental basement related to a proto-South American continent. Whatever the composition of the basement, it’s evolution appears relatively clear (Sawyer et al., 1991). Rifting in the basin began in the Late Triassic to Early Jurassic and by the Middle Jurassic the Embayment crust was being attenuated. Based on tectonic-subsidence modeling, the crust was thinned 1.5 to 2.5 times it’s original thickness with the stretching factor increasing to the southeast. Subsequently, the crust underwent thermal subsidence that was amplified by a factor of two to three by sediment deposition. Presently, the upper part of the crust is within greenschist-amphibolite facies metamorphic conditions at temperatures exceeding 350 °C (660 °F) and pressures greater than 300 MPa (43,500 psi). Concomitant with formation of the northwest-trending frontal folds of the late Mesozoic-Cenozoic fold-thrust belt to the southwest (Sierra Madre Oriental), a zone of northwest trending basement-cored, low-amplitude folds were produced on the western edge of the Embayment (Peyotes, Picachos, and Tamaulipas Arches; Figure 1.1). Smaller low-amplitude folds (e.g. Cinco de Mayo and Lopeno Anticlines in Zapata County) with the same trend occur within the southwest part of the Embayment. A similar origin is suggested for the Embayment-bounding San Marcos Arch to the northeast (Laubach and Jackson, 1990). Sedimentary Fill The evolution of sedimentary fill in the Embayment is generally characterized by the deposition of salt in the Jurassic, development of a distinct carbonate margin in the Cretaceous, and progradation of clastic sediment nearly 400 km (250 mi) basinward of the Cretaceous carbonate shelf margin in the Cenozoic (Figure 2.2). Extensive growth faults developed along the Cenozoic clastic margins due to mobilization of the underlying salt and overpressured mudrock. Terrestrial syn-rift deposition may have occurred in the Embayment coincident with Late Triassic to Early Jurassic crustal attenuation. While there is substantial evidence for subsurface Triassic grabens in other parts of the basin (Salvador, 1991a), there is only limited evidence for their existence in the 2.4 and Embayment. Gravity and magnetic intensity maps (Figures 2.5) show Appalachian-style anomalies (Raring, 1986; Sheridan et al., 1988) indicative of possible deep seated and elongate rift basins parallel to the present coast in south Texas (Raring, 1986), but no Jurassic sediments have been observed in outcrop nor penetrated by wells. Extensive evaporite deposition within the Gulf of Mexico Basin occurred in the Late Middle Jurassic during a time of broad regional tectonic subsidence. The extent of the originally-deposited salt within the Rio Grande Salt Basin of the Embayment is indefinite (Ewing, 1991; Stricklin, 1994) (Figure 2.6) and is commonly described as having significantly thinner salt than the Houston Diapir Province to the north where the original thickness of salt is estimated to be 1.5 to 2.5 km (4,900 to 6,900 ft) (Worral and Snelson, 1989; Seni and Jackson, 1984). At the start of the Early Cretaceous, the basic geographic framework of the present-day Gulf of Mexico established with deep central gulf was a surrounded by stable platforms with open marine connections (Salvador, 1987). In the Cretaceous, the Gulf was connected to the Western Interior Seaway and a large carbonate margin formed around the rim of the basin. Two distinct gentle carbonate shelves developed in the Early Cretaceous in the northwest Gulf: the LowerCretaceous Sligo (or Cupido) and Stuart City Reef Trends. In most of the northwestern Gulf of Mexico these margins occur in the same geographic position with the younger Stuart City trend overlying the Sligo Trend, but they diverge in the Embayment with the Stuart City trend occurring northwestward of the Sligo trend (Figure 2.7). After development of a major mid-Cretaceous unconformity, the reefs were drowned and chalks, marls, and shales of the Eagle Ford, Austin, Taylor, and Navarro Groups were deposited (Sohl et al., 1991). During this time well defined. the position of the shelf margin was not Multiple depositional episodes in the Cenozoic (Figure 2.8) resulted in progradation of clastic sediment nearly 400 km (250 mi) basinward of the Early Cretaceous carbonate shelf margin and formation of distinct clastic shelf margins (Winker, 1982; Worral and Snelson, 1989; Galloway et al., 1991). Multiple regressive and transgressive cycles occurred that were controlled by a complex sea interplay of sediment supply, level, and the instability of underlying salt. Deposition shifted between three major embayments, the Rio Grande Embayment, the Houston Embayment, and the Mississippi Embayment. In the Cenozoic, to 8 km (26,000 ft) of sandstones, siltstones, and up mudrocks were deposited in the Embayment. The major depositional episodes were in the Early Eocene (Upper Wilcox Group) and Oligocene (Vicksburg and Frio Formations). Growth fault zones (Wilcox, Vicksburg, andFrio Fault Zones) with large offsets mark the location of these shelf margins. In the Early Paleocene, prior to Wilcox progradation, the mud-dominated clastic sediments of the Midway Group prograded over the Cretaceous shelf margin. This initial progradational event did not result in the growth faults übiquitous to the major depositional episodes. In the Late Paleocene, the first major progradational episode of terrigenous sedimentation in the basin occurred with sediment resulting from uplift to the west during the Laramide Orogeny. A distinct shelf margin formed basinward of the Cretaceous shelf margin. The locus of this Wilcox Lower deposition was in the Houston Embayment to the north of the study area in the Rockdale Delta System (Figure 2.9). In the Embayment, this as a narrow depositional episode is recognized strike-oriented strandplain/barrier island system (Fisher and McGowan, 1967). The Lower Wilcox in the Embayment is extensively faulted but faulting did not seem to occur contemporaneous with deposition as there is no expansion of the Lower Wilcox across the faults (Bebout et al., 1982; Dickerson et al., 1995). After deposition of mudrock in the Early Eocene (Middle Wilcox) during a regional transgression, the Upper Wilcox was deposited in the Rosita Delta System of the Embayment (Edwards, 1981; Bebout et al., 1982) (Figure 2.9). Growth faulting in a 10-20 km (6.2-12.4 mi) wide strike-oriented band developed where the sediments prograded basinward over the stable Cretaceous shelf margin onto unstable an basinal muds and underlying salt (Worral and Snelson, 1989). This resulted in abrupt thickening of the sediment package on the downthrown sides of the faults. Just updip of the faults the top of the Wilcox is at depths of about 1,800 m (5,900 ft) and downdip of the faults the top is at 2,700 to 3,700 m (8,900 to 12,100 ft). Correlation of selected intervals of the upper part of the Wilcox indicates expansion of the section by 6 to 10 times across the faults (Bebout et al., 1982). Sandstone volume increases basinward across the faults but the sandstone to mudrock ratio is not appreciably altered (Edwards, 1980). The deepest-seated Wilcox growth faults extend from the land surface down through Tertiary and Upper Cretaceous strata and sole out in salt or possibly on Paleozoic (Ouachita) rocks (Worral and Snelson, 1989; Fiduk and Hamilton, 1995). In the Middle to Late Eocene, regressive sequences (Claiborne [Queen City, Weches, and Yegua Formations], and Jackson Groups) were deposited with the development of only minor growth faults. The Oligocene section (Vicksburg and Frio Formations) attains it’s greatest thickness in south Texas. The Embayment was the main depocenter in the basin during this period with sediment derived from intense volcanism and uplift of the Southern Cordillera of Trans-Pecos Texas and northern Mexico. The Vicksburg Formation consists of marine shales with interbedded sands that greatly thickens across a major northeast trending growth fault (the Vicksburg Fault). The Frio (Catahoula) Formation represents a progradational system that is over 4.5 km (14,800 ft) thick in south Texas (Galloway et al., 1982). The Norias Delta System (Galloway et al., 1982) in the south and the Greta/Camauchua Barrier/Strandplain System to the north were the major depositional systems (Figure 2.10). The Frio Formation also is cut by major listric growth faults (Frio Fault Zone). These faults occur over a relatively wide (50-60 km [3l-37 mi]) strike-oriented band and the deepest faults sole out above the Cretaceous section in Eocene mudrocks (Worral and Snelson, 1989). Progradation of the shelf margin continued throughout the Neogene and Quaternary, but the study area is characterized by the depositional and structural features of the Sligo carbonate reef margin and the large depocenters and growth faults of the Wilcox and Frio clastic margins. Salt The extent of the original salt deposited in the Rio Grande Salt Basin within the Embayment is indefinite (Ewing, 1991). Low-relief, anticlinal salt features are observed on seismic reflection profiles landward of the Cretaceous are shelf edge (Raring, 1986), but only a few isolated salt structures present as diapirs in the Tertiary section (Figure 2.6). Areas of extensive salt domes (e.g. the Houston Diapir Province in the Houston Embayment) are typically interpreted as being where the thickest salt accumulated. This implies that less salt was deposited in the Embayment or that there were similar thicknesses of salt but more efficient lateral withdrawal of salt in the Rio Grande Embayment. The latter could be caused by a relatively evenly distributed depositional load in the Embayment during deposition of the Paleocene/Eocene Wilcox Group (Worral and Snelson, 1989). The few salt structures in the Embayment, unlike in the major diapir provinces, do not disrupt the trend and style of the Paleogene growth fault (Worrai and Snelson, 1989). Faults in the Embayment define a linear zones trend unlike those in the Houston Embayment to the north (Figure 1.1). There is more littleevidenceforsignificantthicknessesofunmobilizedsalt. Theremaybe extensive salt within the Embayment in Mexico (Dickerson et al., 1995), but this is not well-documented in the literature. Proprietary data and Raney et al. (1991) suggest the occurrence of turtle structures in the Wilcox Group, indicating extensive withdrawal of salt in the Paleocene/Eocene. Salt withdrawal may be coincident with the Laramide thrusting event that created the Sierra Madre Oriental and other northwest-trending, low-amplitude folds in the northwestern part of the Embayment. Salt withdrawal also may be responsible for the extensive Lobo gravity slide in the Lower Wilcox section of south Texas (Dickerson et al., 1995) which has a different style (no expansion across the faults) than other faults in the Wilcox trend (Ewing, 1987). The large areal extent and consistent faulting and erosional across the Lobo basin patterns suggest triggering by one relatively short event soon after Lobo deposition (Long, 1986), consistent with salt mobilization and the occurrence of turtle structures in the Upper Wilcox section. StateofStress The state of stress in the Embayment is extensional with the maximum principal stress oriented vertically and the minimum principal stress oriented northwest-southeast (Zoback et al., 1991), perpendicular to the regional Wilcox, Vicksburg, and Frio Fault Zones (Figure 2.11). Hydraulic fractures would also preferentially be oriented parallel to these faults. The state of stress is compressional in the Sierra Madre Oriental to the southwest of the Embayment. The southern extension of the Embayment in Mexico (Burgos Basin) has been described by some workers as having the characteristics of a foreland basin (Dickerson et al., 1995). 200 km (125 mi) offshore to the southeast are the Perdido Foldbelt and the Mexican Ridges, compressional features formed by accommodating extension of the sedimentary pile it slides towards as an open lateral boundary in the Gulf of Mexico. Hydrogeology The basin can be divided into three hydrodynamic regimes based on natural breaks in pressure gradients observed in pressure-depth plots for the the Embayment: hydropressured, overpressured, and extremely overpressured regimes. The hydropressured regime is the shallowest regime and is defined as the interval where the fluid pressure gradient is nominally hydropressured (fluid pressure resulting mostly from the weight of the overlying water). The top of the overpressured regime is easily defined as the depth where the pressure gradient begins to increase significantly above the nominal hydropressure gradient. Pressure gradients in this regime are highly variable. The overpressured regime is the transition zone between the hydropressured and extremely overpressured regimes. The top of the extremely overpressured regime is defined as the depth below which the constant for area. a pressure gradient is approximately given This constant gradient is typically 80-90% of the overburden gradient (pressure gradient due to the weight of the overlying rock and fluid). The boundaries between these regimes vary in both space and geologic time. The hydropressured regime is predominantly a gravity-driven flow system occurring in the shallower parts of the Embayment with flow resulting from elevation differences on the water table. Buoyancy-driven flow (free convection) occur are also may where there large fluid-density contrasts. This regime has relatively low fluid pressures, low lateral fluid-pressure gradients (and hydraulic-potential gradients) and a vertical fluid-pressure gradient (nominal hydropressure gradient) of 9.8 to 10.52 kPa/m (0.433 to 0.465 psi/ft) for fluid densities of 1000 3 to 1073 kg/m and total dissolved solids concentrations (as [NaCl]) of 0 to 85,000 zone includes water with a meteoric ppm. The hydropressured origin, and/or fresh/saline waters discharged from deeper in the basin (formation waters derived sediment Distinct from compaction and/or metamorphic/diagenetic waters). hydrocarbon accumulations also exist in this zone. In the study area, the base of the hydropressured regime is spatially variable and occurs at depths from 0.5 to 3.5 kilometers (1,640 to 11,480 ft) (Figure 2.12). The base of this shallow flow system in the clastic rocks is defined by Ryder (1988) as either the top of the Midway (Paleocene) confining unit or the top of the overpressured High vertical pressure-head gradients (indicating zone. in the upward flow) underlying overpressured regime preclude present-day meteoric recharge to the overpressured regime. Ryder’s (1988) implied top of overpressure (Figure 2.13) is similar in both shape and magnitude to the one mapped in this study (Figure 2.12). In this study, the hydropressured regime is defined as the shallow regime where fluid pressure gradients are consistently less than 10.52 kPa/m (0.465 psi/ft). This is the gradient resulting from the weight of an NaCl brine with [NaCl] of 85,000 ppm at 25 °C (77 °F) with a corresponding 3 . density of 1073 kg/m The overpressured and extremely overpressured regimes have relatively high fluid-pressures with flow generally being upward. It is generally accepted that compaction disequilibrium is/was the primary cause of fluid overpressure in the basin (Dickinson, 1953, Bredehoeft and Hanshaw, 1968; Chapman, 1972; Sharp and Domenico, 1976; Bethke, 1986). Potential secondary causes include hydrocarbon generation (Barker, 1990; Luo and Vasseur, 1996), clay dehydration (Powers, 1967; Burst, 1969; Bruce, 1984), aquathermal pressuring (Barker, 1972; or waters Sharp, 1983), and metamorphism of basin fill influx of metamorphic from the basement (Rubey and Hubbert, 1959; Bredehoeft and Ingebritsen, 1990; Land, 1991). These secondary causes are not well documented but may be very over The important in maintaining fluid pressures geologic time. two regimes include formation waters derived from sediment compaction (meteoric or reactions seawater origins) and/or waters derived from metamorphic/diagenetic within the sedimentary section or underlying basement. Water with a meteoric chemical signature could also be in these zones if at some time there were later meteoric incursions deeper into the section and overpressure developed without totally flushing out meteoric waters (Harrison and Summa, 1991), but this is not well-documented. Local hydrocarbon accumulations also exist in these two regimes, but are more commonly found in the upper part of the overpressured regime. The of the in the is top overpressured regime Embayment spatially variable and occurs at depths from 0.5 to 3.5 km (1,640 ft to 11,480 ft) (Figure 2.12). The base (top of the extremely overpressured regime) also is spatially variable and to occurs at depths from 2.5 to 4.5 km (8,200 14,760 ft) (Figure 2.14). The thickness of this transitional zone from hydropressured to extremely overpressured regimes ranges from 0.5 to 2 km (1,310 to 6,560 ft) (Figure 2.15). The depth to the top of the overpressured regime is a function of the lithology and architecture of the depositional systems (Dickinson, 1953; Rubey and Hubbert, Harkins with the 1959; and Baugher, 1969) top of the overpressured regime strongly correlated to the depth where there is a decrease in the sandstone/mudrock ratio. Leftwich’s (1993) study of lithofacies and fluid confirmed this pattern for Leftwich pressures overpressures in the Embayment. occurs in the lower part of a (1993) found that the top of overpressure generally sandier section lying above the interfingering sandstone/mudrock interface between deltaic and marine shales. driven by sequences Flow in this regime is very large vertical and lateral pressure-head gradients, with the elevation Flow also component of hydraulic head and buoyancy forces being negligible. may occur in pulses with fluid moving along with evolving, upward-moving fractures originating in underlying extremely overpressured regime (Fyfe, the 1978; Nunn, 1996). zone zone where Below the overpressured is the extremely overpressured flow is driven by high vertical and moderate lateral pressure gradients and via the propagation of hydraulic fractures (Fyfe, 1978, p. 259; Nur and Walder, 1990; Walther, 1990; Nunn, 1996). The top of the extremely overpressured regime is spatially variable and at depths from 2.5 to 4.5 km (8,200 to 14,760 ft) occurs (Figure 2.14). Assuming that this regime extends to the basement, then up to 9 kilometers of the basin fill is extremely overpressured. Although pressure gradients are relatively constant with depth (at least in the mappable, upper part of the regime), they are spatially variable and mappable (Figure 2.16). The relatively constant gradient, presumably, defines the upper limit of fluid pressure for a given depth. If fluid pressures increase slightly, then the rocks will hydraulically fracture, permeability will increase, and the pressure will be released (Fyfe et al., 1978). This limiting pressure is the fluid pressure that exceeds the least principal stress by an amount equal to the tensile strength of the rock (Rubey and Hubbert, 1959; Secor, 1965). Thermobaric conditions conducive to metamorphism occur in the deeper parts of this regime. Very little is known about this thermobaric zone, but simple calculations of and pressure indicate that it about 3 km temperature using typical gradients encompasses (9,840 ft) of the extremely overpressured regime in metamorphic greenschist and possibly amphibolite facies. The expulsion of volatile fluids (water, C0 ) during 2 be metamorphism of the sedimentary section and the underlying basement may significant in maintaining fluid pressures over geologic time and inducing hydraulic fractures (Fyfe et al., 1978; Bredehoeft and Ingebritsen, 1990). The above definitions of hydrodynamic regimes differ from those in the et literature. Sharp al.’s (1988) definitions (meteoric, compactional­overpressured, and thermobaric) have connotations of the origin of the water. Kreitler’s (1989) definitions (shallow freshwater, hydrostatic saline, and overpressured) have connotations of fluid composition. Leftwich and Engelder’s (1994) definitions are based on pressure-depth profiles for single wells or individual fields. in the oil The terminology developed for engineering purposes industry (hydrostatic, top of soft overpressure, top of hard overpressure) are pervasive in the literature and deserve further clarification. In oil industry terms, the top of the soft and hard are arbitrarily set at the depths overpressure regimes where the ratio of fluid pressure to the assumed overburden pressure of 22.62 kPa/m (1.0 psi/ft) are 0.5 and 0.7, respectively. These descriptive numbers (0.5, 0.7, 1.0) are often misconstrued in the literature as pressure gradients. In fact, the first two are ratios of fluid to overburden at a given depth as pressure pressure defined by Rubey and Hubbert (1959). The assumed overburden pressure gradient of 22.62 kPa/m (1.0 psi/ft) is the calculated gradient resulting from the water-saturated rocks with an weight of the overlying assumed constant bulk 3 density of 2310 kg/m . In reality, the overburden pressure gradient is dependent on porosity and the density of the rock matrix and pore-filling fluids in the overlying section. The gradient is less than 22.62 kPa/m (1.0 psi/ft) at shallow depths and greater than 22.62 kPa/m deeper in the section. Over the mapped from about 20­ interval of 0.5 to 4.5 km (1,640-14,760 ft) in this study, it ranges 23 kPa/m (0.89-1.02 psi/ft) with a gradient of 22.62 kPa/m (1.0 psi/ft) at a depth of about 2.5 km (8,200 ft). This calculation assumes a porosity-depth profile 3 (Dickinson, 1953) and matrix and fluid densities of 2500 and 1025 kg/m , respectively. The relationship between these different definitions is shown in Figure 2.17. The hydropressured regime is equivalent to Sharp et al.’s (1988) meteoric freshwater regime, Kreitler’s (1989) shallow and hydrostatic saline regimes, Leftwich and Engelder’s (1994) pressure leg 1, and most of the oil industry’s hydrostatic regime. The boundary between the hydropressured and overpressured is regimes equivalent to the top of Sharp et al.’s compactional-overpressured regimes, the top of Kreitler’s overpressured regime, the top of Leftwich and Engelder’s pressure leg 2 and falls within the oil industry’s hydrostatic regime. Leftwich and Engelder’s pressure leg 3 is entirely encompassed in the base of the overpressured regime. Their leg 3 is defined as the between the depth where zone there is a sharp increase in the pressure gradient and the top of their pressure leg 4 (see below). upper boundary is interpreted to be coincident with the top of The mudrock undercompaction, but sufficient data documentation is not available to confirm that supposition. The large range of pressure at a given depth in the data set from in cells in this resulting spatial averaging grid paper precludes observation of distinct linear trends as suggested by Leftwich and Engelder (1994). The boundary between the overpressured and extremely overpressured regimes Sharp et compactional-overpressured regime is within al.’s in the Embayment. Dependent on the ambient thermobaric conditions, it is conceivable that this boundary could occur within Sharp et al.’s thermobaric regime. The boundary is within Kreitler’s overpressured regime, coincident with the top of Leftwich and Engelder’s pressure leg 4, and below the oil industry’s top of hard overpressure (Figure 2.18). Geothermics Geothermal gradients throughout most of the northwestern Gulf of Mexico Basin are about 20 to 35 °C/km (1.1 to 1.9 °F/100 ft) (AAPG and USGS, 1976). In general, gradients are highest in the thinner onshore sediments and decrease basinward (Sharp et al., 1988). Gradients also tend to increase with depth (Bebout et al., 1982; Sharp et al., 1988). Higher geothermal gradients exist in the Embayment relative to other parts of the basin (AAPG and USGS, 1976). These relatively high geothermal gradients were recognized soon after hydrocarbon production commenced in the early 19405. A large positive thermal an anomaly occurring in arcuate zone inland and subparallel to the coastline along on the Wilcox Fault Zone (Figure 2.19) is superimposed the relatively higher geothermal gradients (Pfeiffer, 1988; Pfeiffer and Sharp, 1989). Geothermal gradients in this zone exceed 60 °C/km (3.3 °F/100 ft) with immediately adjacent areas having geothermal gradients as low as 29 °C/km (1.6 °F/100 ft). Jones (1975) identified this thermal anomaly during regional mapping in the early 1970’5. It was further discussed by Wesselman (1983) who mapped subsurface temperatures in a cross section through the Embayment. The anomaly also is evident (but is overly generalized) on a geothermal gradient map for North America (AAPG and USGS, 1976), where it appears to continue into the southern Rio Grande Embayment in Mexico (Burgos Basin). Detailed subsurface Bodner and temperature mapping in the Embayment by Bodner (1985), Sharp (1988), Pfeiffer (1988), and Pfeiffer and Sharp (1989) determined the general shape and position of the anomaly. Methods Subsurface temperature, fluid pressure, and salinity were compiled within a geographic information system for analysis. Temperature and pressure were mapped at selected depths, while salinity was mapped over depth intervals. The tops of fluid overpressure, extreme overpressure, and the oil industry’s “hard” overpressure were mapped along with the distribution of the fluid/overburden ratio within the extremely overpressure regime. Temperature Bottom-hole temperature data from well-log headers were available from the work of Bodner (1985) and Pfeiffer (1988) and were augmented for this study from the files of the Texas Bureau of Economic Geology for a total of 5,735 data points. The additional temperature data were collected near the geologic cross- section (Figure 2.7) to better constrain the shape of the northwest limb of the anomaly and the position of its peak. All temperature data were corrected for thermal disturbances during drilling by the Kehle (1971) method (Pfeiffer, 1989). Temperature-depth plots were prepared for 20 x 20 km (12.4 x 12.4 mi) grid-cells 22 with an area of 400 km(154 mi) (Figure 2.20). Representative temperatures (Table 2.1) were picked from these plots at 0.5 km (1,640 ft) intervals. Isotherm maps were prepared for depth slices from 0.5 to 4.5 km (1,640 to 14,760 ft). Isothermal contours represent where isothermal surfaces cut the depth-slice. Error in temperature picks from the temperature-depth plots is estimated to be about 5 °C (9 °F). About 90-95% of the data points were honored when hand- contouring isotherms resulting in confidence that the mapped thermal patterns reflect true subsurface patterns. Additionally, individual data points within 25 km (15.5 mi) of a geologic cross-section (Figure 2.7) were projected along regional strike onto the section for analysis. Fluid Pressure The object of this pressure analysis is to map virgin fluid pressures as they existed before extensive hydrocarbon production. Results of bottom-hole pressure and drill-stem tests were available from the work of Akhter and Kreitler (1990) and Kreitler et al. (1990) and were augmented for this study from Petroleum Incorporated’s (Houston) database for a total of 23,840 data points. The data set is very noisy with respect to estimating virgin fluid pressures. Noise a results from errors in the original measurements, compilation errors, spread in pressure resulting from grouping data over a large grid-cell, and from due to depressurization extensive hydrocarbon production (Hill and Sharp, 1993). errors. Depressurization is the most significant of these A first-pass culling procedure to reduce the noise of depressurization due to hydrocarbon production was accomplished by removing all pressure data that were below the hydropressure gradient of 10.52 kPa/m (0.465 psi/ft). This resulted in a culled x data base of 6,460 points. Pressure-depth plots were prepared for the 20 20 km (12.4 x 12.4 mi) grid-cells (Figure 2.20) as for the BHT data. Pressures were picked (Table 2.2) from these plots at 0.5 km (1,640 ft) intervals to isobar prepare maps for depth slices from 1 to 3.5 km (3,280 to 11,480 ft). Maximum pressures were picked from a line drawn on the pressure-depth plot representing the outer (high-pressure side) envelope of the fluid pressures. This results in a high estimate of fluid pressure, but is the only consistent method available due to the extensive noise in the data. Error in maximum-pressure picks from the pressure- depth plots at a given depth is estimated to be less than 5 MPa (725 psi). Even withthislargeerror,trendsaremappable. About90-95%ofthedatapointswere honored when contouring isobars thus giving confidence that the mapped pressure patterns reflect true subsurface patterns on a regional scale. Additionally, individual data points within 25 km (15.5 mi) of a geologic cross-section (Figure Pressures 2.7) were projected along regional strike onto the section for analysis. may be used as proxies to hydraulic head in the overpressured and extremely overpressured regimes as the elevation component of head is negligible. Fluid Salinity Salinity data were available from Petroleum Incorporated’s (Houston) database(AkhterandKreitler, 1990)aswellasanotherdatabaseattheUniversity = of Texas (Morton and Land, 1987; Land, personal communication, 1995) (n 2,382 points). In this study, salinity is defined as chloride concentration; it is the dominant anion in waters of the Embayment and a linear relationship between [Cl] and total dissolved solids is available (Morton and Land, 1987; Land, personal communication, 1995) as [TDS] = 1788 + 1.638 [Cl] where both [TDS] and [Cl] have units of parts per million. Extreme heterogeneity in salinity concentrations precluded using a mapping method similar to that for pressures and temperatures. Salinity was mapped in 0.5 km (1,640 ft) depth intervals from 0.5-3.5 km (1,640-11,480 ft). Additionally, individual data points within 25 km (15.5 mi) of a geologic cross-section (Figure 2.7) were projected along strike onto the section for analysis. The data did not always include information on what formation the samples were from but always included the depth of the sample. In analyzing the depth-interval maps, estimates of the host formation were taken from Dodge and Posey (1981). Results are as Results presented descriptions of the temperature, pressure, and salinity maps and cross sections. Regional patterns are readily mappable on temperature and pressure maps. Salinity is very heterogeneous, but local areas of evident. are vs. vs. high salinity Graphs of temperature depth, pressure depth, cross section. Additional and salinity vs. depth are presented for data along the vs. north temperature vs. depth and pressure depth graphs are presented for areas andsouthof thecross section. Temperature an The thermal anomaly is subparallel to the Wilcox Fault Zone and has anticlinal shape with the highest temperatures in the core, the vertical axial surface striking to the northeast-southwest, and the axis plunging to the northeast. Therefore, temperatures increase from northeast to southwest and outward from the core of the anomaly. The southwestern boundary is undefined because of a - lack of data south of the United States Mexican border. Isotherm maps are presented in Figures 2.21-2.29. The anomaly is evident on depth slices from 1.5 to 4.5 km (4,920 to 14,760 ft) (Figures 2.21-2.27). Maps are described below from deepest to shallowest depth-slices. For descriptive and comparative the width of the anomaly is defined as the distance from the isotherm purposes, with a value 20 °C lower than the maximum temperature across the hotter core to the same isotherm on the other limb. The measurement is taken perpendicular to axial trace in northern Duval County. of the the The temperature amplitude the core of anomaly is described by finding the temperature difference between the anomaly and a point 100 km (62 mi) to the southeast. Again, the measurement is taken perpendicular to the axial trace in northern Duval County. The nose of the anomaly is defined as the axis of the isotherm “anticline” with a value 20 °C lower than the maximum temperature. The isothermal for a depth of 4.5 km (14,760 ft) (Figure 2.21) shows map the nose of the anomaly in the north central part of the Embayment, centered on northeast Duval, northern Jim Wells, and southern Live Oak Counties. The width of the anomaly is 85 km (53 mi) and the temperature amplitude is about 50 °C (90 on °F). The shape is asymmetrical with a steeper temperature gradient the northwest limb (although the shape of the southeast limb is not well defined near the center of anomaly). The anomaly only occurs eastward of the Cretaceous occurs shelf margin. A well-defined, semicircular, low-temperature anomaly centered Kenedy County in the southeast in the location of the Norias Delta on System in the Frio Formation. The core of the anomaly occurs within the area with a relatively low fluid/overburden pressure ratio (Figure 2.16) in the the northern terminus of the shallowest extremely overpressured regime and at occurrence of extreme overpressure (Figure 2.14). The isothermal map for a depth of 4.0 km (13,120 ft) (Figure 2.22) shows the nose of the anomaly in northern Duval County. The width of the anomaly is 65 km (40 mi) and the temperature amplitude is about 50 °C (90 °F). Again, the shape is highly asymmetrical with a much steeper temperature gradient on the northwest limb and the anomaly only occurs eastward of the Cretaceous shelf margin. The semicircular, low-temperature anomaly occurs centered on the Norias Delta System in Kenedy County in the southeast. Again, the core of the anomaly occurs within the area with a relatively low fluid/overburden pressure ratio (Figure 2.16) in the extremely overpressured regime and at the northern terminus of the shallowest occurrence of extreme overpressure (Figure 2.14). The isothermal for a depth of 3.5 km (11,480 ft) (Figure 2.23) shows map the nose of the anomaly farther north in northeast Live Oak and Bee Counties. is The width of the anomaly is 85 km (53 mi) and the temperature amplitude about 40 °C (72 °F). The shape is still asymmetrical with a steeper temperature gradient on the northwest limb. The anomaly is constrained eastward of the Cretaceous shelf margin south of McMullen County but occurs westward of the margin to the north. The semicircular, low-temperature anomaly still exists centered on Kenedy County in the southeast but the lateral temperature gradients The core near the are less than in the deeper depth-slices. of the anomaly occurs top of extreme overpressure (Figure 2.14). The isothermal for a depth of 3.0 km (9,840 ft) (Figure 2.24) shows map the nose of the anomaly farther north in northeast Bee and southern Goliad Counties. The width of the anomaly is 90 km (56 mi) and the temperature amplitude is about 30-40 °C (54-72 °F). Again, the shape is asymmetrical with a steeper temperature gradient anomaly is mostly on the northwest limb. The constrained eastward of the Cretaceous shelfmargin except fornorthofLive Oak County. The low-temperature anomaly is still centered on Kenedy County in the southeast but it no longer has a semicircular shape. The core of the anomaly occurs in the overpressured regime (Figure 2.12). The isothermal map for a depth of 2.5 km (8,200 ft) (Figure 2.25) shows the thermal anomaly with a different pattern than in the deeper depth slices. The of the anomaly is farther south and bends towards the east in northern Duval nose and Jim Wells Counties. The bend lies above the area with relatively low fluid/overburden ratio in the extremely overpressure regime (Figure 2.16) and just north of the shallowest occurrence (Figure of the top of extreme overpressure The width of the increased to 140 km (87 mi) and the 2.14). anomaly has temperature amplitude has decreased to about 20 °C (36 °F). The shape is slightly asymmetrical in the opposite of the deeper depth-slices and the anomaly is sense no longer constrained eastward of the Cretaceous shelf margin. The low temperature anomaly in the southeast is dying out. The isothermal map for a depth of 2.0 km (6,560 ft) (Figure 2.26) shows a much wider anomaly with a much smaller temperature amplitude. The nose of the anomaly is in northern Duval, southern Live Oak and southern McMullen Counties. The width of the anomaly (as defined) has increased to 215 km (134 mi) and the temperature amplitude is only about 10 °C (18 °F). The anomaly no in Zapata and Webb Counties to the southwest but with the low longer appears temperature amplitude and wide width of the anomaly it is becoming harder to constrain it’s shape within the resolution of the data. The anomaly (as defined) extends westward of the Cretaceous shelf margin and the low temperature anomaly in the southeast is no longer evident. The isothermal map for a depth of 1.5 km (4,920 ft) (Figure 2.27) shows an extremely damped out anomaly and is close to the limits of data resolution (note the smaller contour interval). The nose of the anomaly is no longer defined. The width of the anomaly as defined cannot be measured as the temperature amplitude is too small (about SHC [9 °F]). Again, the anomaly is weak in Zapata and Webb Counties to the southwest. The anomaly extends well westward of the Cretaceous This is above the of shelf margin. depth-slice top overpressure (Figure 2.12). on The anomaly is no longer definable depth-slices for 1.0 and 0.5 km 3,280 and 1,640 ft). The isothermal for a depth of 1.0 km (3,280 ft) (Figure map 2.28) possibly shows evidence for a damped anomaly (note the smaller contour interval) but is close to the limits of data resolution. The isothermal map for a depth of 0.5 km (1,640 ft) (Figure 2.29) shows evidence of the thermal no in in anomaly. Slightly higher temperatures occur the semicircular pattern Kenedy County. In cross-section (Figure 2.30), the asymmetric shape of the anomaly is evident with steeper dipping isotherms on the northwest side of the anomaly in the deeper part of the section (150-200 °C [302-392 °F]). Shallower isotherms (100-125 °C [212-257 °F]) have the opposite sense, with steeper isotherms on the southeast side. The peak of the anomaly is coincident with the farthest-basinward and apparently deepest-seated Wilcox Faults. The anomaly tends towards an isoclinal shape at greater depths as described by the decrease in width in the depth-slice descriptions. A better quantitative description of the true profile shape is possible from the cross-section (compared to the defined width used for for the descriptive purposes depth-slices). The anomaly is up to 150 km (93 mi) wide, has a temperature amplitude at a given depth of up to 50 °C (90 °F), and a depth amplitude for a given isotherm of up to 2 km (6,560 ft). Temperature-depth plots for binned data along the cross section (Figures 2.31-2.37) show a relatively higher geothermal gradient in the Wilcox Fault Zone (up to 63 °C/km [3.5 °F/100 ft]) and a concave downward shape to the temperature profile. The geothermal gradients northwestward of the Wilcox Fault Zone (30 °C/km [1.65 °F/100 ft]) and southeastward in the Frio Fault Zone a (33 °C/km [l.Bl °F/100 ft]) are relatively linear. This is strikingly evident on temperature-depth plot combining data for the Wilcox and Frio Fault Zones (Figure 2.38). Additional at a temperature depth plots (Figures 2.39-2.45) prepared lower spatial resolution (on the 20 x 20 km grid) show similar trends. Two of the plots (Figures 2.39 and 2.40) are for areas north of the cross section in the Wilcox Fault Zone and show the concave-downward shape. On another (Figure 2.41), located south of the cross section in the Wilcox Fault Zone, the concavity is not evident but temperatures are much higher relative to those along the Frio Fault Zone. Figure 2.42 is in the Vicksburg Fault Zone and shows intermediate concave temperatures between those in the Frio and Wilcox Fault Zones and no shape. Figures 2.43-2.45 are for the Frio Fault Zone. Figures 2.43 and 2.44 are for areas north of the cross section and exhibit much lower temperatures than in the Wilcox Fault Zone. These do exhibit a kink in the geothermal gradient (gradient increases with depth) at about 2.5 to 3 km (8,200 to 9,840 ft). Figure 2.45 shows a very linear gradient to 4 km (13,120 ft). Fluid Pressure Similar to the thermal pattern, the isobaric pattern has an anticlinal shape with the highest pressures in the core, the axial surface striking to the northeast- southwest, and the axis plunging to the northeast. The anticlinal shape is not nearly as well-defined (sometimes it is more like a ridge) as for the thermal anomaly and the axial trace is slightly offset to the southeast from the axial trace of the thermal anomaly. Fluid pressures increase from northeast to southwest and outward from the core of the anticlinal shape. Again, the southwestern boundary - is undefined because of a lack of data south of the United States Mexican border. Maps are described below from deepest to shallowest depth-slice. a The isobar map for a depth of 3.5 km (11,480 ft) (Figure 2.46) shows broad anticlinal shape with pressures increasing from northeast to southwest. The in fluid is relatively low with a maximum relief of 8 MPa (1,160 range pressure psi). A broad high is centered on western Jim Hogg, Brooks, southern Duval, and southern Jim Wells Counties. This high is centered on the area with a relatively high fluid/overburden ratio in the extremely overpressured regime (Figure 2.16). Most of this depth-slice (except for Live Oak and Bee Counties in the north and Kenedy County in the southeast) is in the extremely overpressured regime (Figure 2.14). The isobar map for a depth of 3.0 km (9,840 ft) (Figure 2.47) shows an anticlinal shape in the southwest centered on Jim Hogg, western Brooks, and southern Duval Counties. The of the anticlinal shape west core is offset to the from the relative high in the fluid/overburden ratio in the extremely overpressured regime (Figure 2.16). Pressures increase from northeast to southwest. The plunge of the anticlinal shape declines to the north and becomes more of a high The relief in fluid-pressure has more than doubled from the 3.5­ pressure ridge. km (11,480 ft) depth-slice to 20 MPa (2,900 psi). A semicircular, low pressure area is evident in the southeast centered on Kenedy County. This depth-slice is mostly in the overpressured and extremely overpressured regimes (Figure 2.14) for the areas in the extreme north and east that are in the shallow except hydropressured regime (Figure 2.12). The isobar map for a depth of 2.5 km (8,200 ft) (Figure 2.48) shows a high pressure ridge trending northeast-southwest with it’s axis centered on Starr, Jim Hogg, southeastern Duval, and northern Jim Wells Counties. The southeast limb of the ridge is relatively steep with lower occurring in the Frio pressures Fault Zone. bulge trending northwest-southeast occurs A small high pressure from southern Live Oak to western Nueces Counties. This overlies the relatively low fluid/overburden pressure ratio in the extremely overpressured regime (Figure 2.16). Another bulge extends westward into southeastern Webb County towards the Lower Wilcox Lobo Trend. The fluid-pressure relief is the same as for the 3.0-km (9,840 ft) depth-slice (20 MPa [2,900 psi]). This depth-slice is mostly in the overpressured regime (Figure 2.12) except for the southern part of the high pressure ridge in the extremely overpressured regime (Figure 2.14) and north and east that the areas in the extreme are in the hydropressured regime (Figure 2.12). The isobar map for a depth of 2.0 km (6,560 ft) (Figure 2.49) shows an anticlinal shape trending northeast-southwest with it’s axis centered on Starr, Jim Hogg, southeastern Duval, and northern Jim Wells Counties. The shape of the ridge is relatively symmetrical. The two bulges are evident with the northern one (above the relatively low fluid/overburden ratio [Figure 2.16]) extending pressure further west into McMullen County. The fluid-pressure relief has decreased from in the 2.5-km (8,200 ft) depth-slice to 15 MPa (2,200 psi). This depth-slice is the overpressured regime (Figure 2.12) along the central high-pressure ridge in the south and in the hydropressured regime elsewhere. an The isobar map for a depth of 1.5 km (4,920 ft) (Figure 2.50) shows anticlinal shape in the south centered on Starr and Jim Hogg Counties and is relatively flat in the north. Pressures increase from northeast to southwest. The fluid-pressure relief has decreased significantly to 6 MPa (870 psi). This depth-slice is mostly in the hydropressured regime (Figure 2.12) except for in the extreme southwest in the overpressured regime. The isobar for a depth of 1.0 km (3,280 ft) (Figure 2.51) still shows map pressures increasing from northeast to southwest. The fluid-pressure relief has decreased significantly to 2 MPa (290 psi). This depth-slice is entirely in the hydropressured regime (Figure 2.12). Figure 2.52 depicts hydrodynamic regimes represented by contours of the ratio of the fluid-pressure to the assumed overburden pressure of 22.62 kPa/m (1.0psi/ft). Thisismoreusefulthancontouringfluidpressureasitgivesanidea of the magnitude of fluid overpressure and good delineation of the regimes. A ratio of less than 0.465 is in the hydropressured regime. The overpressured the from 0.465 to 0.8 and is the transition zone between the regime spans range hydropressured zone and extremely overpressured regime. Below the 0.8 contour, the sediments are extremely overpressured. From our understanding of zone the basin hydrodynamics, the ratio in the extreme overpressure should be relatively constant from its onset to at least the base of the clastic section and most likely to the basement at depths of 10-12 km (32,800-39,400 ft) but no data are availablefor the deeper section. Therefore, up to 9 km (29,500 ft) of section is extremely overpressured. occur at shallower depth in the Wilcox Fault Zone than in Overpressures the Frio Fault Zone. A larger vertical gradient in the fluid/overburden pressure ratio is evident in the Frio Fault Zone relative to the Wilcox Fault Zone indicating a thinner transition zone (overpressured regime) from the hydropressured to the extremely overpressured regimes. The peaks in the anticlinal shapes of the top of the overpressured (ratio = 0.465) and extremely overpressured (ratio = 0.8) regimes occurs at the same position (230 km [143 mi]) along the section in the area between the Wilcox and Frio Fault Zones coincident with the mud-rich section between these major fault zones. The shape of the top of the overpressured zone is asymmetric with the steeper limb on southeast side (Frio Fault Zone). The top of the extremely overpressured zone also is asymmetric but the steeper limb is on the northwest side (Wilcox Fault Zone). The overpressured regime exists in the Cretaceous carbonate section but no data are available to indicate that the extremely overpressured regime also occurs there. Pressure-depth plots (Figures 2.53-2.59) along the section also show the higher pressures in the Wilcox Fault Zone relative to the Frio Fault Zone in the 2­3 km (6,560-9,840 ft) depth range. This is best seen in a direct comparison in Figure 2.60. Also evident is the much thinner transition zone from the hydropressured to extremely overpressured regimes in the Frio Fault Zone relative to the Wilcox Fault Zone. Vertical pressure gradients in the are overpressured regime about 25 kPa/m (1.11 psi/ft) in the Wilcox Fault Zone and 50 kPa/m (2.21 psi/ft) in the Frio Fault Zone. Additional pressure-depth plots (Figures 2.61-2.67) prepared at a lower spatial resolution (on the 20 x 20 km grid) than the plots along the cross section show similar trends. These plots are for the same areas as the temperature-depth plots in Figures 2.39-2.45. Three of the plots are for the Wilcox Fault Zone. Two (Figures 2.61 and 2.62) are for areas north of the cross section, while the third (Figure 2.63) is for an area located south of the cross section. These plots zone show the relatively thicker transition (overpressured regime) compared to those in the Frio Fault Zone. Figure 2.64 is in the Vicksburg Fault Zone and also has a thick transition zone. Figures 2.65-2.67 are for the Frio Fault Zone. Figures 2.65 and 2.66 are for areas north of the cross section and exhibit the much thinner transition zone relative to the Wilcox Fault Zone and the greater depth to area the top of the overpressured zone. Figure 2.67, nearby the for the temperature-depth plot in Figure 2.45, shows a hydropressure gradient to greater than 3 km (9,840 ft). Salinity The most striking feature about the salinity of the ground water in the on Embayment is it’s spatial heterogeneity. This is evident all of the maps. local Despite this general observation, areas with some higher salinity waters are observed. are described below from the to shallowest Maps deepest depth- intervals. The salinity map for the interval from 3-3.5 km (9,840-11,480 ft) (Figure 2.68) includes only minor amounts of data for the Frio Formation in the Frio FaultZone (mostly centeredoncoastalNuecesCounty)andfortheWilcoxGroup along the Wilcox Fault Zone. These measurements are all within the overpressured regime. The salinity in the Wilcox Group is relatively low (generally < 10,000 ppm [Cl]), but data are sparse. In the Frio, salinity is heterogeneous, but, in general, less than 35,0000 ppm [Cl]. The salinity map for the interval from 2.5-3 km (8,200-9,840 ft) (Figure 2.69) better exemplifies the heterogeneity in salinity. Salinity in the Wilcox Group is relatively low and homogeneous. These data from within the are overpressured regime. Salinities in the Frio and Vicksburg Formations in the Vicksburg Fault Zone are also relatively low and homogeneous. These data are from the overpressured regime, at depths close to the extremely overpressured regime. Higher salinity waters occur in the Frio Formation in the Frio Fault Zone relative to the Vicksburg Fault Zone. An area of highly saline fluids occurs in the coastal Nueces and San Patricio Counties within the Frio Formation (above less saline fluids in the interval below). Data from the Frio Fault Zone are from depths near the boundary of the hydropressured and overpressured regimes. The salinity map for the interval from 2-2.5 km (6,560-8,200 ft) (Figure 2.70) again exemplifies the heterogeneity in salinity. Waters in the Wilcox and Lower Claiborne Groups along the Wilcox Fault Zone are generally low salinity, but salinity increases along strike to the north. All data from the Wilcox Fault Zone are fromtheoverpressuredregime. SalinityintheVicksburgandFrioFault Zones is heterogeneous but several distinct areas with some high salinity waters are evident. One is oriented along the most landward of the faults in the Frio Fault Zone in Kleberg, Nueces, and San Patricio Counties. As in the 2.5-3 km depth-slice, a zone of high salinity also is located in coastal Nueces County. Another zone occurs in the Vicksburg Fault Zone in extreme western Kleberg County. Data from the Frio and Vicksburg Fault Zones are near the overpressured/hydropressured boundary (except in the southern Vicksburg Fault Zone in the overpressured regime). The salinity map for the interval from 1.5-2 km (4,900-6,560 ft) (Figure 2.71) again shows the heterogeneity in salinity and many interesting features. Salinity in the Lower Wilcox Group in Lasalle and McMullen Counties is very low and the data are from the hydropressured The salinity in the Wilcox zone. andClaiborne Groups along the Wilcox Fault Zone is still relatively low. These data are near the overpressure/hydropressure boundary. A strike-oriented band of high salinity in Jim Hogg and Duval Counties is in the Claiborne Group is coincident with minor growth faults that cut the top of the Yegua Formation. These data also are Data in the near the overpressure/hydropressure boundary. Vicksburg Fault Zone show extreme heterogeneity. In the south (Starr/Hidalgo near County boundary), the data are from the Vicksburg and Frio Formations the are overpressure/hydropressure boundary. Salinities generally high in the hydropressured zone in the Frio Fault Zone within the Frio and Miocene Formations. The high salinity trend recognized in the 2-2.5 km interval also is seen oriented near the most landward of the faults in the Frio Fault Zone in Kleberg, Nueces, and San Patricio Counties. The salinity map for the interval from 1-1.5 km (3,280-4,920 ft) (Figure 2.72) again shows the heterogeneity in salinity. A zone of high salinity is seen along the Wilcox Fault Zone in northwest Duval County in the Claiborne Group. These data are from the hydropressured zone. The high salinity trend in the Claiborne Group along the minor Yegua faults is still evident. South of southern Jim Hogg County, these data are from the overpressured regime. To the north, they occur in the hydropressured regime. Relatively low salinities occur in the area between the Yegua and Frio Fault Zones. The high salinity trend recognized in the 2-2.5 and 1.5-2 km intervals also is seen the most landward near of the faults in the Frio Fault Zone in Kleberg, Nueces, and San Patricio Counties in Miocene units within the hydropressured regime. The salinity map for the interval from 0.5-1 km (1,640-3,280 ft) (Figure 2.73) shows that the updip Claiborne Group and Frio Formation have lower at salinities than deeper depths downdip. All data are in the hydropressured regime. section. Most of the on Figure 2.74 depicts salinity the geologic cross zone waters below the top of overpressure have relatively low salinity. A of high salinity is observed above the Wilcox Fault Zone in the hydropressured regime. The high salinity trend along the most landward of the Frio Faults is readily evident. A zone with some relatively high salinity water occurs just below the top of overpressure in the Vicksburg Fault Zone. Chloride-depth plots (Figures 2.75-2.81) along the section better indicate the magnitude of the salinity. Except for the most westward plot (Figure 2.75) salinity is variable at all depths in all plots. Highest salinity (up to 120,000 ppm [Cl]) is evident above the Wilcox (Figure 2.77) and Frio (Figure 2.81) Fault Zones. There also are some high zones just basinward of the Wilcox Fault Zone (Figure 2.78). In the Frio Fault Zone (Figure 2.81) there is the distinct decrease in salinity at about 2 km (6,560 ft) depth as seen the cross section. on Discussion Fluids are episodically expulsed from extremely overpressured sediments during natural hydrofracturing and discharge vertically along regional fault zones in the Embayment. The basal 9 km (29,500 ft) of the stratigraphic section is with fluid at 80 to 90% of the overburden extremely overpressured pressures close to the minimum needed for the onset of pressure, pressure hydraulic fracturing. At least 3 km (9,800 ft) of the extremely overpressured section is within the greenschist metamorphic facies and possibly part of the amphibolite facies. Small increases in fluid pressure in this extremely overpressured regime, caused by recharge to the system from hydrocarbon generation, diagenetic/metamorphic fluids, or other areas where the limiting pressure of hydraulic fracturing is higher, trigger hydraulic fracturing and episodes of fluid discharge. The Wilcox, Vicksburg, and Frio Fault Zones all fluid discharge are but the Wilcox Fault Zone is a more recent discharge area for fluids from a zones, deeper fluid source. Flow along the fault zone appears to be upwards from the southwest from areas of higher fluid pressure towards an area with a relative minimum in the fluid/overburden pressure ratio in the extremely overpressured regime. There is no apparent lithologic reason why this area should maintain a lower fluid/overburden pressure ratio. It may occur in this position as a result of an increase in the dip of the (relatively impermeable) basement along the San of hot fluids Marcos Arch forcing fluids to discharge upward. This discharge could result in increased quartz and calcite cementation in this zone that would make these rocks able to maintain the higher differential stress necessary for hydraulic fracturing. This cementation, however, has not been documented. The most recent pulse of fluid discharge was along the Wilcox Fault Zone where the largest thermal anomaly in the Gulf of Mexico Basin occurs coincident with a positive fluid pressure anomaly and salinity inversions. The thermal anomaly is centered on the most basinward and deepest of the Wilcox faults. Heat conduction is not a viable mechanism for producing the anomaly. Hydraulic that fracturing in the fault zone creates a relatively permeable vertical pathway focuses fluid discharge from the extremely overpressured rocks along the deep-seated faults. Heterogeneous salinity distributions along both the Wilcox and Frio Fault Zones represent the cumulative effect of expulsion events in both fault zones. Neither distinct thermal nor anomalies are evident in the Frio pressure Fault Zone, suggesting that fluid expulsion events had shallower fluid source, a were of smaller magnitude, were more diffuse than in the Wilcox Fault Zone, or that similar thermal and pressure anomalies have already dissipated. The deep source is of fluids, hydraulic fracture, and episodic fluid expulsion hypothesis consistent with the and discussed temperature, fluid pressure, salinity data as below. The Rio Grande Embayment may be a regional discharge zone for the thick sedimentary section in the western Gulf of Mexico Basin. Temperature Geothermal gradients in the Rio Grande Embayment are higher relative to those in the Houston Embayment to the north of the San Marcos Arch (Figure 1.1) (Bebout et al., 1982; Loucks et al., 1986) and other parts of the basin (AAPG and USGS, 1976). The mapped thermal anomaly along the Wilcox Fault Zone is the largest in the Gulf of Mexico Basin with width of 150 km (93 mi), a a temperature amplitude at a given depth of up to 50 °C (90 °F), and a depth amplitude for a given isotherm of up to 2 km (6,560 ft). Geothermal gradients within the anomaly exceed 60 °C/km (3.3 °F/100 ft) with immediately adjacent low 29 as as areas, including the Frio Fault Zone, having geothermal gradients °C/km (1.6 "F/100 ft). Based interpretations of regional subsurface on temperature maps in the early 19705, Jones (1975) suggested that this positive thermal anomaly results from deep basinal fluids advecting heat upwards along the Wilcox Fault Zone. The advection hypothesis was further supported by the work of Wesselman (1983), Bodner and Sharp (1988) and Pfeiffer and Sharp (1989). Detailed subsurface temperature mapping in the Embayment by Bodner and Sharp (1988), and Pfeiffer and Sharp (1989) constrained the shape and position of the anomaly and numerical modeling by Bodner and Sharp (1988) indicated the viability of the advection hypothesis. Temperature mapping in this study confirms the general anticlinal shape of the anomaly with the highest temperatures in the core, the vertical axial surface striking to the northeast-southwest, and the axis plunging to the northeast. Temperatures increase to the southwest to the edge of data availability (the Texas-Mexico border) and the anomaly dies out towards the San Marcos Arch to the north. The anomaly is evident from depths of 1.5 km (4,920 ft) to at least 5 km (16,400 ft). The core of the anomaly appears to be offset basinward from the Wilcox Fault Zone in the depth-slices (Figures 2.21-2.27) and this offset was noted by Bodner (1985) and Pfeiffer (1989) as a possible problem with the hypothesis of advection of heat upwards along the fault zone. Wilcox faults on the depth-slice maps taken from the Tectonic Map of Texas (Ewing, 1990), were a regional scale map where the faults are mapped only if they cross the top of a key stratigraphic horizon. The easternmostWilcox faults on the map occur where the key mapping horizon changes from the Wilcox Group to the shallower Yegua Formation towards the east. Therefore, more basinward (eastward) Wilcox faults are not explicitly mapped as “Wilcox” faults and are not mapped at all if they do not cut the shallower Yegua horizon. The faults on the cross section (Figure 2.2) are from Galloway et al. (1994) and are inferred from detailed local stratigraphic relationships. This interpretation includes a more-basinward, 20-km (12-mi) wide zone of deep faults. The thermal anomaly is centered on these faults. In order to calibrate his advection model, Bodner (1985) postulated a high the Wilcox as permeability zone basinward and deeper from what he recognized section Fault Zone. The more detailed geologic cross available for this study (Galloway et al., 1994) shows that Bodner’s (1985) postulated high permeability area coincides with the deeper, more basinward Wilcox faults. The profile of the thermal anomaly in cross section (Figure 2.30) is distinctly different from the shape in Bodner (1985) and Pfeiffer (1989). The asymmetric shape was constrained with additional data collection near the cross section. Steeply dipping isotherms (150-200 °C [302-392 °F]) occur in the deeper section on the northwest side of the anomaly. Shallower in the section, the steeper dipping isotherms (100-125 °C [212-257 °F]) are on the southeast side of occurs the anomaly. This same trend along the strike of the anomaly in depth- slices with the change in asymmetry evident between the 3.0 and 2.5 km (9,840 and 8,200 ft) depth slices (Figures 2.24 and 2.25) along with an increase in the width of the anomaly in the shallower depth-slices. The anomaly tends towards an isoclinal shape at greater depths, centered on a 20 km (12 mi) wide band of the farthest-basinward Wilcox faults. The shape of the anomaly is consistent with upward advection along the deeper Wilcox faults where the distinct thermal peak is located. The steeply dipping isotherms on the northwest side in the deeper part of the section occur below the depth of the distinct nickpoint in the carbonate margin (Figure 2.30). The asymmetric shape may be caused by the relative impermeability of the carbonate section to the west and the lack of extreme overpressures in the carbonates. Flow in this deeper section is restricted to the clastic section where high fluid pressure in the extremely overpressured regime (Figure 2.52) results in hydraulic fracturing permeability, that enhances particularly along the fault zone. Above the carbonate nickpoint, the dip of the isotherms decreases as fluids can flow updip in the clastic section and raise the This is consistent with the observations discussed below. temperature. pressure The anomaly also persists upward into the hydropressured zone as seen in the 1.5­ km (4,920 ft) depth-slice (Figure 2.27) indicative of discharge of hot fluids from the overpressured regime into the hydropressured regime. The tight shape of the anomaly peak is most evident in the extremely overpressured regime in the 4-km (13,120 ft) depth-slice (Figure 2.22) where the anomaly is constrained eastward of the Cretaceous carbonate margin. The nose in of the anomaly the extremely overpressured regime occurs where the fluid/overburden pressure ratio is at a minimum (Figure 2.16). It that appears flow occurs from the southwest and from depth towards this discharge area that fractures at lower fluid pressures than the surrounding areas. A bend in the nose occurs in this of the anomaly in the 2.5 km (8,200 ft) depth-slice (Figure 2.25) same area indicative of upward flow of hot waters. This bend in the nose also is coincident in depth and area with the northern terminus of the shallow ridge in the top of the extreme overpressure (Figure 2.14). The high in the top of extreme occurs where there is a maximum in the fluid/overburden overpressure pressure ratio (Figure 2.16). Although fluid pressure is higher in this area, these rocks can maintain these higher pressures without fracturing. Another strong argument for recent advection along the Wilcox Fault Zone is the shape of the temperature-depth curve defined by data from the fault zone (Figure 2.38). This concave downward shape is the “classic” for shape identifying relatively recent or ongoing advective heat transport (Bredehoeft and are Papadopolous, 1965). The trends in the Vicksburg and Frio Fault Zones relatively linear. Fluid Pressure The is divided into three hydrodynamic the Embayment regimes: hydropressured, overpressured, and extremely overpressured regimes. This discussion focuses on the overpressured and extremely overpressured regimes and particularly the evidence supporting the advection hypothesis. The top of the overpressured regime (base of the hydropressured regime) in the Embayment is spatially variable and occurs at depths from 0.5 to 3.5 km (1,640 ft 11,480 ft) to The of the (Figure 2.12). top of the extremely overpressured regime (base overpressured regime) also is spatially variable and occurs at depths from 2.5 to to 4.5 km (8,200 14,760 ft) (Figure 2.14). The thickness of the overpressured regime (transition zone from hydropressured to extremely overpressured regimes) ranges from 0.5 to 2 km (1310 to 6,560 ft) (Figure 2.15). Trends in fluid pressure are readily mappable over this regional area, even with the large associated with the individual measurements. This ability to errors trends indicates and map a relatively open flow system in the overpressured extremely overpressured regimes. This is especially true in the extremely overpressured regime where lateral fluid pressure gradients low (Figure 2.46) are and the rocks are at pressures very close to those required to initiate hydraulic fracturing. A comparison of Figures 2.12, 2.14, and 2.18 shows that the shallowest surfaces forms a northeast trending depth to the top of the mapped overpressure ridge that occurs farther eastward with depth to the top of overpressure, “hard” overpressure, and extreme overpressure. This is interpreted to result from flow upward out of the extremely overpressured regime westward towards the Wilcox Fault Zone forming a pressure high. This pressure high and characteristic shape of the surfaces also is also evident in cross section (Figure 2.30). The shallow high (2.5-3 km [8,200-9,840 ft]) in the extreme surface (Figure overpressure 2.14) occurs within mudrocks of the Claiborne and Jackson Groups and within the Vicksburg Formation along the Vicksburg Fault Zone. This is caused by the rocks indicated as ductile nature of these muddy and volcanic-rock-fragment-rich by the relatively high maximum attainable pressure gradient in the area (Figure 2.16). This maximum attainable gradient is commonly called the fracture gradient and is thought to define the maximum that that can be pressure can maintained in the rock before hydraulic fracturing increases the permeability fractures allowing pressure dissipation (Fyfe, 1978). Hydraulic will preferably form outside of this zone of maximum fracture gradients in areas more conducive to fracturing with lower fracture gradients (Figure 2.16). Fractures will preferentially form perpendicular to the least minimum stress (parallel to the regional fault zones) in areas with low fracture gradients and structural weaknesses (fault zones). Fluids flowing through or moving with propagating fractures will tend to be in these zones. Therefore fluid is moving from high pressure areas in the south and at depth upwards along the Wilcox Fault Zone. Flow is constricted to areas west of the extreme overpressure ridge, and is towards an area with a minimum in the fracture gradient (0.8 of the lithostatic This flow from the gradient; Figure 2.16). extremely overpressured regime causes a pressure high seen in the 3-km (9,840 ft) depth-slice (Figure 2.47) above the top of extreme overpressure where the high pressure in the center of the anticlinal shape is offset to the west of the maximum fluid/overburden pressure ratio (Figure 2.16) Discharge along the Wilcox Fault Zone also is evident by the increased depth to the top of extreme overpressure but a decreased depth to the top of the overpressured regime (Figure 2.30) towards the fault zone. This in the Wilcox Fault dispels the notion that the shallower depth to overpressure Zone relative to the Frio Fault Zone is simply due to a lower vertical permeability. If this was the case, then the depth to the top of extreme should also be shallower in the Wilcox Fault Zone than in the Frio overpressure Fault Zone. Salinity Salinity is very heterogeneous but a few definite trends are evident. There are trends of higher salinity coincident with individual Yegua faults, the Vicksburg Fault, and the most landward of the Frio faults. This is especially evident in cross section (Figure 2.74) and in several depth-interval maps (1-2.5 km [3,280-8,200 ft]; Figures 2.70-2.72) where high salinity waters occur along the most landward of the Frio faults above the top of overpressure. There also is a more dispersed zone of high salinity above the Wilcox Fault Zone evident in the the cross section (Figure 2.74) and on 1-1.5 km (3,280-4,920 ft) depth-interval map (Figure 2.72). In the 1-1.5 km (3,280-4,920 ft) interval, within the hydropressured zone, bands of higher salinity are evident along both the Wilcox/Yegua and Frio Fault Zones relative to the interfault area. This is consistent with the model of expulsion of deeper saline waters upwards along the fault zones. The salinity shows the cumulative effect of expulsion events in both fault zones; the thermal anomaly indicates a stronger and/or more recent event in the Wilcox Fault Zone. Areas of saline waters overlying lateral to relatively or fresher waters result in fluid density inversions. Fluids in these areas are inherently unstable. The conditions may be conducive for free convection to be occurring where saline fluids flow up along the faults (forced convection); the pressure andthermalanomaliesdissipate;andthedenserwaters thenbegintosink (free convection). This model is consistent with the observed heterogeneous salinity distributions. After fluids discharge from the overpressured zone and sink, they are limited from sinking deeper by large pressure gradients encountered near the top of This is consistent with the observation that most of overpressure. the high salinity waters are above the overpressured zone. There seems to be a greater volume of saline waters in the Frio Formation and Miocene units relative to the Wilcox Group. This is especially evident at depths greater than 1.5 km (4,920 ft) (Figures 2.68-2.71) which is generally within the overpressured regime in the Wilcox Group and in the hydropressured regime in the Frio and younger formations (Figure 2.12). This is partly due to the much thicker hydropressured zone in the Frio and younger sediments. It is also possible that there is more salt under the Frio Fault Zone relative to the Wilcox Fault Zone, so that deep waters may interact with more salt and become more saline under the Frio section. This is implied because of the evidence given above for massive withdrawal of salt of the Wilcox lack of salt during deposition Group (Wilcox fault geometry, diapirs, turtle structures in the Wilcox Group, Lobo gravity slide). areas Comparison between The data discussed above is convincing evidence for the advection hypothesis but why are there no similar thermal anomalies along the Wilcox Fault Zone in the Houston Embayment and the Frio Fault Zone in the Embayment when other evidence indicates that there is also extensive flow in these areas? The primary control for the difference with the Wilcox Fault Zone in the Houston Embayment is difference in the structural style of the faults. The of map the Wilcox Faults (Figure 1.1) demonstrates that north of the San Marcos arch and into the Houston Embayment a markedly different fault geometry exists. To the south of the arch, the faults are very continuous along strike, relatively parallel, and occur over narrow To the north, the fault pattern is more a zone. erratic and the faults occur over a wider zone. These differences are attributed to different styles of deposition of the Wilcox Group as it prograded over the mechanically stable carbonate margin onto unstable basinal muds and salt (Worral and Snelson, 1989). In the Rio Grande Embayment, Lower Wilcox deposition was mostly strike oriented in barrier-island/strandplain environments. in This exerted an evenly distributed load on the underlying salt resulting relatively uniform withdrawal of the salt; the resultant faulting occurred in a pattern parallel to the carbonate margin and depositional strike. To the north, in was more the Houston Embayment, deposition in dip-oriented, deltaic environments that placed uneven loads on the salt. This resulted in the many salt diapirs and the erratic fault pattern. If we conceptualize the faults as vertical zones zones more permeable and/or conducive to hydraulic fracturing, closely a narrow zone spaced faults along would more efficiently focus upward moving fluids and heat. This is consistent with the occurrence of thermal anomalies along the Wilcox Fault Zone in the Rio Grande Embayment. Why is there no thermal anomaly along the Frio Fault Zone, compared to the Wilcox Fault Zone, in the Rio Grande Embayment? This relate to the may structural styles of the two fault zones. Wilcox faults cut the complete on sedimentary section and sole out in Jurassic salt top of the basement (Fiduk and Hamilton, 1995). The faults occur over a narrow zone of 10 to 15 km (6-12 mi) and individual faults are closely spaced. These faults in overpressured clastic a rocks formed just basinward of a mechanically stable carbonate shelf so there is large contrast in mechanical properties that enhances faulting. The Frio Faults, on the other hand, sole out above the Cretaceous carbonates in Eocene mudrocks. They occur over a much wider zone of 50-60 km (31-37 mi) and individual faults are more widely spaced. Faults of the Wilcox Zone tap the fluids deeper in the basin and focus the upward flow along a narrow zone. Therefore, the Wilcox Zone is more efficient in transporting heat from deeper in the basin. Other Hypotheses Some simple arguments are presented that show the inconsistency of the observations with conductive and/or heat-source causes for the anomaly. Based on a widely accepted model for the thermal evolution of divergent margins (McKenzie, 1978), any thermal anomaly resulting from the initial rifting event has long since dissipated (Pfeiffer, 1989). There is no geologic nor seismic evidence for any magmatic activity in the Embayment that could be a local source of heat. The thermal anomaly can not simply be due to contrasts in basal heat- local flow density, thermal conductivity contrasts, nor a source of heat. A band can of higher heat flow in the basement is a possibility and it’s existence not be disproved with existing limited data on basement composition. Even so, it is highly unlikely that the tight and asymmetric profile of the isotherms (Figure 2.30) could result from contrasts in basal heat-flow density in basement at depths 10 km. greater than The anomaly would have to be more broadly shaped. There are over 2 km (6,560 ft) of relief on the basement in the Embayment (Figure 2.3) and if these crustal rocks have higher thermal conductivity than the overlying basin fill, heat could be refracted into the basement highs causing a higher basal heat flow density in these areas. Mapped basement highs occur below the carbonate shelf margin and on the San Marcos Arch where observed thermal are an gradients relatively low, inverse pattern than what would be expected via heat refraction. New data (McKenna et al., 1996) show that Wilcox sandstones have higher thermal conductivities than Frio sandstones at the same porosity. This would be consistent with heat refracting towards a sand-rich Wilcox Fault Zone and causing a thermal anomaly, but the ratios of sandstone to mudrock are the same on either side of the fault zone (Bebout et al., 1982) so bulk thermal conductivity would not be markedly different across the faults. Measurements also indicate that the well-cemented carbonates to the west have higher thermal conductivities than expected for the bulk mudrocks/sandstones and higher temperatures are not observed above the carbonate section as expected in conductive heat-refraction. Measurements of radiogenic heat production (McKenna and Sharp, 1996; McKenna and Sharp, in press [b]) indicate heat more and in more production in the mudrocks than in the sandstones heat production the Frio than in the Wilcox. The differences are not enough to form an anomaly of this magnitude, but if such anomaly produced, it would be the even an was inverse of that observed. Ruling out these conduction and heat-source scenarios zone. leaves the hypothesis of advection of heat along the fault Other Evidence The advection hypothesis is supported by a number of independent observations. Many of these were previously discussed by Land (1991). These data suggest that fluids have moved along the Wilcox and Frio Fault Zones in the past: the observed thermal anomaly is possible evidence or recent for present episodes of upwelling of hot fluids in the Wilcox Fault Zone. The idea of faults in the Gulf of Mexico Basin being permeable migration pathways for fluids from deep in the basin is widely cited (Fowler, 1970; Land and Prezbindowski, 1981; Berg and Habeck, 1982; Galloway, 1982; Hanor and Bailey, 1983;McCulloh, 1985;MortonandLand, 1987;Sharpetal., 1988;Hanor and Sassen, 1990; Land, 1991; Summa et al., 1993). The enormous volume of reservoired hydrocarbons occurring above the depth of the thermal maturation window for potential source rocks (Dow, 1978: Galloway, 1982; Land, 1991) and above the deeper-buried more probable rocks in much of the basin is source over strong evidence for significant vertical migration of hydrocarbons several kilometers (Lopez, 1990). Abundant hydrocarbons occur along the Wilcox, Vicksburg, and Frio Fault Zones in the Embayment (Geomap, 1982). Most of the Frio reservoirs are above the depth of the hydrocarbon maturation window (Galloway et al., 1982), while the Wilcox rocks are thermally mature. The suspected source rock for Wilcox and Frio reservoirs is lean (less than 0.5% very total organic carbon) so focusing of hydrocarbons into a small area along the fault zones is required (Lopez, 1990). Linear trends of hydrocarbon accumulations in the Embayment (Salvador and Nehring, 1991) and the reservoirs are coincident These linear trends are not with the Wilcox, Vicksburg, and Frio Fault Zones. evident elsewhere in the basin suggesting more fault-controlled vertical migration in the Embayment. Siliceous knobs and calcite veins mapped in outcrop on the surface have been attributed to fluids upwelling along the Wilcox Fault Zone (Mcßride et al., 1968; Freeman, 1968). These features occur in southeastern McMullen County (Figure 2.82) just updip of the area with the minimum fluid/overburden pressure ratio in the extremely overpressured regime (the discharge area). In addition, over 50 scattered hills of circular or elliptical shape that be siliceous knobs may occur in southern McMullen and Duval Counties (Mcßride et al., 1968). Clay dikes are also observed in this area and also in Starr and Karnes Counties with strikes similar to the regional strike of the underlying Wilcox Fault Zone (Mcßride et al., 1968). Uranium deposits above and updip of the Wilcox Fault Zone formed during multiple, epigenetic alteration events when reducing fluids from deep in the basin discharged upward along deep-seated faults, reacted with uranium-rich oxidizing waters, and precipitated uranium roll-front deposits (Galloway, 1982). Reducing and sulfidic geochemistries characterize deep-basin ground waters with hydrocarbons and H S being the reductants. Oxidizing waters, resulting from 2 meteoric recharge, flow downdip dissolving uranium from the volcanic rock fragments in the formation. Waters from the deep basin flow upward along the faults and mix with meteoric waters in the environment. Specific supergene alteration zones, easily defined by the oxidation state of iron in the deposits and textural relationships of oxidized and reduced iron minerals indicate multiple episodes of basinal discharge of reducing fluids (Galloway and Hobday, 1996). Sulfur isotope data in the deposits indicate that the sulfur derived from was deeper evaporite-bearing strata (Goldhaber, 1983; Land, 1991) The fluid chemistry of some Frio Formation waters with Br/Cl ratios, high boron concentrations and “B-depleted isotopic compositions are similar to brines sampled from Mesozoic reservoirs (Land, 1991). It's suggested that these fluids and hydrocarbons have migrated upwards from deep in the basin via growth faults. (Land, 1991). Fisher and Land (1986) also state that strontium isotopes in calcite cement in Wilcox Group sandstones indicate an Eocene or earlier marine carbonate sources for the calcite cement in the Wilcox Group. Among several other possibilities for the source of this anomaly is local input of fluids of Mesozoic age. It's suggested that these fluids and hydrocarbons have migrated upwards from deep in the basin via growth faults. All of these observations are consistent with episodically discharged fluids zones. from the extremely overpressured regime discharging along the fault Conclusions 1. Pressure, temperature, and salinity data are all consistent with episodic, zones upward fluid expulsion from extremely overpressured along hydraulic fractures in regional fault zones. 2. Forced convection of heat is the preferred explanation for the thermal anomaly along the Wilcox Fault Zone where there is also a positive fluid pressure anomaly and salinity inversions. 3. The Wilcox Fault Zone is a preferred discharge zone for hot fluids from deep in the basin because of the deep-seated nature of the faults that cut the entire sedimentary section to depths of 10 km (32,800 ft) within the extremely overpressured regime. The relative structural weakness of the fault zone makes it a preferred location for the formation of hydraulic fractures. 4. The thermal anomaly is centered on the most basinward of the Wilcox faults. 5. Fluid flow appears to be upwards from the southwest (area of higher a area with a fluid pressure), along the Wilcox Fault Zone, towards discharge relative minimum in the fluid/overburden pressure ratio in the extremely overpressured regime. 6. The Frio Fault Zone has salinity inversions and anomalous fluid chemistry, but no thermal anomaly at present. This suggests that fluid expulsion events here were of smaller magnitude, more diffuse or from a shallower source than in the Wilcox Fault Zone, or that the thermal anomalies have already dissipated. Chapter 3: Thermal conductivity of Wilcox and Frio sandstones ofMexico Basin in the Rio Grande Embayment, Gulf Abstract Thermal conductivity and petrographic data are presented for verifying mechanistic models of sandstone thermal conductivity. We measured the thermal conductivity of 83 Wilcox and Frio sandstones from south Texas in the Gulf of Mexico sedimentary basin and correlated conductivity to petrographic variables. Thermal conductivities of water-saturated sandstones at 20 °C (68 °F ) and 3 MPa (435 psi) were measured on core-plugs using a divided-bar apparatus. Thermal conductivity ranges from 2.06 to 5.73 W/m/K over a porosity range of 2.4 to 29.6 %. For a given porosity, because of a higher quartz content, Wilcox sandstones are more conductive than Frio sandstones. A grain-matrix conductivity of 5.9 W/m/K is estimated for Wilcox sandstones; matrix conductivity is adequately described with an arithmetic mixing model. Thermal conductivities of clean (< 25% clay) sandstones can be described by a multilinear function of decreasing thermal conductivity with increasing porosity and increasing thermal conductivity with quartz content. For clean, quartzose (> 35% of the solids) sandstones, the be dependence on quartz content can be dropped and thermal conductivities can predicted with a linear decrease in conductivity with increasing porosity. These sandstones appear isotropic with respect to thermal conductivity. Introduction Hydrocarbon maturation and diagenesis are functions of the thermal one history of the host sediments/sedimentary rocks. Numerical modeling is of the primary tools used to reconstruct this thermal history (Sharp and Domenico, 1976; Burrus, 1986; Deming and Chapman, 1989; Naeser and McCulloh, 1989; Lerche, 1990; Hermanrud, 1993). While considerable effort has gone into developing sophisticated thermal-history models (Hermanrud, 1993), little effort has gone into quantifying the fundamental thermal rock-properties that are required as model input (Blackwell and Steele, 1989). The thermal conductivity of sedimentary rocks is one of these properties that has generally been neglected (Bachu et al., 1987; Blackwell and Steele, 1989; Demongodin et al., 1991; Williamson, 1992; Hermanrud, 1993). While there is extensive literature describing different ways to predict the thermal conductivity of sedimentary rocks (e.g., Hashin and Shtrikman, 1962; Lewis and Rose, 1970; Beck, 1976; Palciauskas, 1986; Zimmerman, 1989; Brigaud et al., 1990; Bachu, 1991; Nimick and Leith, 1992; Somerton, 1992; Mohanty, 1993; Luo et al., 1994), raw data are rarely presented. Some measurements of the thermal conductivity of sandstones are and Van der reported along with petrographic descriptions (e.g., Zierfuss Vliet, 1956; Woodside and Messmer, 1961a, 1961b; Goss and Combs, 1975; Brigaud and Vasseur, 1989; Somerton, 1992). There remains, however, a "cruel lack of data" (Burrus, 1986, p. 2) so that “input parameter accuracy... (is not) sufficient in the thermal modeling of sedimentary basins” (Hermanrud, 1993, p. 11). This is disconcerting because thermal conductivity plays a primary role in determining subsurface temperatures (Kappelmayer and Haenel, 1974; Lerche, 1990; Williamson, 1992; Hermanrud, 1993; Deming, 1994). In the simple case of steady-state conduction, the geothermal gradient (dT/dz) is inversely proportional to the thermal conductivity (X), (3.,) dz X where Q is the heat-flow density (heat flow unit area). A decrease in per thermal conductivity from 3 to 2 W/m/K, increases the thermal gradient from 20 to 30 °C/km (1.10 to 1.65 °F/100 ft) and raises the temperature at a depth of 3 km (9840 ft) by 30 °C (54 °F). The thermal maturation rate of organic material doubles with every 10 °C (18 °F) rise in temperature (Waples, 1980; Lerche, 1990). Therefore, if we want to accurately predict thermal and hydrocarbon generation histories, it is important to quantify thermal conductivity for a particular study area. Thermal history calculations for the Gulf of Mexico Basin (Sharp and Domenico, 1976; Bethke, 1986; Yukler and Dow, 1990; Pantano et al., 1990; Harrison and Summa, 1991; Lerche and McKenna, 1991) are based on thermal conductivity estimates derived from other workers for other basins. Fjeldskaar et al. (1993) demonstrates the errors in subsurface temperatures caused a similar extrapolation of thermal conductivity data. The effect of thermal by conductivity variations on subsurface temperature also is effectively demonstrated by Jones and Oxburgh (1979), O’Brien and Lerche (1984), and Ungerer et al. (1990). When calculating the thermal history of a basin, thermal conductivity is as typically estimated some function of porosity and/or lithology (Lewis and Rose, 1970; lingerer et al., 1990; Lerche, 1990; Bethke et al., 1993). There are few thermal-conductivity measurements in the literature that include rock descriptions any more detailed than simply "sandstone” or “shale,” making it difficult to develop reliable functions for thermal conductivity based on commonly reported petrographic/lithologic descriptions. Empirical functions have been developed for estimating thermal conductivity from petrographic data et (e.g., Brigaud and Vasseur, 1989; Brigaud al., 1990; Somerton, 1992), but these functions have not been tested for rocks in the Gulf of Mexico Basin. Only about 100 average thermal-conductivity values are available for the Gulf of Mexico Basin (Smith and Dees, 1982; Nagihara et al., 1993). Smith and Dees’ (1982) data do not include petrographic descriptions and Nagihara et al.’s (1993) data are limited to shallow (< 3 m [lO ft]), unconsolidated, hemipelagic muds. No published data exist for the major rock units of onshore Texas, such as the Wilcox Group or the Frio Formation. This paper supplies the first "ground­ truth" data on the thermal conductivity of Wilcox and Frio sandstones (important reservoir rocks in the Gulf of Mexico Basin) and provides a documented data set for verifying mechanistic models of sandstone thermal conductivity. Samples andMeasurements Wilcox and Frio sandstone samples were collected from petroleum-well cores in the South Texas portion of the Gulf of Mexico Basin (Figure 3.1), a thick sequence of sedimentary rock deposited on a divergent continental margin (Winker, 1982; Curtis, 1987; Salvador, 1991b). Wilcox sandstones were deposited during the Paleocene/Eocene by a large fluviodeltaic system and were partly reworked into strandplain and barrier island deposits (Galloway et al., to 1991). Sampled Wilcox sandstones are very-fine fine-grained subarkoses, sublitharenites, lithic arkoses, feldspathic litharenites, and litharenites (Figure 3.2) that were derived from the central interior of North America (Loucks et al., 1986). The Frio Formation is an Oligocene deltaic complex derived from west Texas and northern Mexico (Loucks et al., 1986; Galloway et al., 1991). Frio are samples very-fine to medium-grained lithic arkoses, feldspathic litharenites, Both units and litharenites with abundant volcanic rock-fragments (Figure 3.2). are relatively uniform in the study and become more quartzose towards east area Texas (Loucks et al., 1986). All samples had detailed petrographic descriptions (Fisher, 1982; Lynch, 1994; T. Diggs, 1992, personal communication) (Table 3.1). Porosity was measured using a water-saturation vessel (Cas and Wright, 1987) at a water pressure ofapproximately 28 MPa(4000 psi). Porosities were verified on several samples using mercury porosimetry. As recommended by the International Heat Flow Commission (Beck, 1988), thermal conductivity was measured on water- saturated core plugs (diameter: 2.5 cm / 1 in) on a standard divided-bar apparatus at Southern Methodist and University (Birch, 1950; Goss Combs, 1975; Blackwell and Spafford, 1987). Crystalline quartz and fused silica glass were used as the calibration standards using the thermal conductivities of Ratcliffe (1959). An insulator was placed around the samples to reduce the effects of moisture movement and evaporation so as to keep inaccuracy below the 5% level (Beck, 1988). About 2-4 MPa (290-580 psi) of uniaxial pressure was applied to the samples, parallel to the axis of the cylindrical core. Based on previous measurements on this divided bar (D. Blackwell, 1993, personal communication) and other studies (e.g.. Beck, 1988), this is adequate for performing pressure reproducible measurements. Higher pressures may cause the sample to collapse. Measurements are precise within 2% on this divided bar and interlab comparisons suggest inaccuracy of less than 4% (Blackwell and Spafford, 1987). A temperature drop of 10 °C (18 °F) was imposed across the apparatus. Results are presented as effective thermal conductivity of water-saturated core plugs at 20 examine °C (68 °F) and 3 MPa (435 psi); no attempt was made in this analysis to the effects of formation temperatures and on in-situ effective thermal pressures conductivity. Results Samples are divided into six categories based on rock-unit and petrography as listed in Table 3.1. “Clean” or “muddy” indicates less or more than 25% of the solids are clays, respectively. Quartzose indicates more than 35% of the solids “Calcite-rich” indicates more than 20% of the solids are quartz. are calcite. All clean Wilcox sandstones (ID #l-52) are quartzose, while clean Frio sandstones (ID# 53-77) are subdivided into quartzose (ID# 53-68) and nonquartzose sandstones (ID# 69-77). The thermal conductivities of clean Wilcox (ID# 1-52) and Frio (ID# 53­ 77) sandstones range from 2.06 to 5.73 W/m/K over a porosity range from 2.4 to 29.6 %, (Table 3.1; Figure 3.3). The high end of this range (> 4.5 W/m/K) is limit for sandstones and falls in higher than most workers present as an upper their ranges for quartzite (e.g., Gretener, 1981; Blackwell and Steele, 1989). between 17 Wilcox and Frio sample porosities only overlap in the porosity range and 21% and these Frio samples (ID# 69-77) are nonquartzose. Over this limited, overlapping range, Wilcox sandstones always have a higher thermal conductivity than Frio sandstones. Quartz has a much higher thermal conductivity (7.8 W/m/K) than the feldspars and rhyolitic-trachytic rock fragments (2.3-2.6 W/m/K) found in these rocks (Loucks et al., 1986), so this trend is not unexpected. Wilcox and quartzose Frio sandstones (ID# 53-68) do not overlap in porosity space. Thus a direct comparison between the Wilcox and quartzose Frio sandstones is not possible. Muddy/quartzose Frio sandstones (ID# 78-81) have lower thermal conductivities than clean Wilcox sandstones at similar porosities because of the lower conductivity of the clay minerals (< 2 W/m/K; Figure 3.3). The one clean/calcite-rich/quartzose Frio sandstone (ID# 82) has much lower thermal a a lower conductivity than Wilcox samples at similar porosity because calcite has one conductivity (~3.4 W/m/K) than quartz. The sample of muddy Wilcox sandstone (ID# 83) plots within the range of the clean Wilcox sandstones (ID# l­ -52). Weinferthatquartzdominatesheatconductionthroughthissample. A decrease in thermal conductivity with increasing porosity (pore-filling water has a low thermal conductivity of 0.6 W/m/K) is evident for clean, quartzose sandstones (ID# 1-68) (Figure 3.3). Equation 3.2 describes this decrease in thermal conductivity (X) with increasing porosity (Figure 3.3): X = (-0.123 ± 0.005) $ + (5.9 ± 0.3) (3.2) is where X is in W/m/K and porosity ((})) in percent. This empirically derived linear function describes 90% of the variance in the thermal conductivity. Extrapolating the linear trend to zero porosity, the grain-matrix conductivity of the clean, quartzose samples is estimated to be 5.9±0.3 W/m/K (Figure 3.3). The of the data. extrapolation represents less than 10% of the total porosity range Clean Frio sandstones (ID# 53-77), by themselves, do not show a decreasing trend with porosity (Figure 3.3), but they do show a general increase in thermal conductivity with increasing quartz content. This trend is evident for all Wilcox and Frio sandstones, but is not as well defined for the Wilcox samples because Wilcox samples have a relatively large porosity range and a limited range can be visualized in of quartz-content. Thermal conductivity for clean sandstones a plot of thermal conductivity contoured in porosity percent quartz space - (Figure 3.4). This dependence has been noted by Brigaud et al. (1990) and Somerton (1992) for sandstones from other basins but never presented as in Figure 3.4. A few adjacent samples in the core permitted the measurementof thermal conductivity both parallel and perpendicular to stratigraphic bedding. Measurements indicate that the effective thermal conductivity of these water- saturated sandstones is essentially isotropic (Table 3.2). Discussion There are a number of models which predict thermal conductivity of heterogeneous mixtures of solids and fluids (e.g., Maxwell, 1892; Woodside and Messmer, 1961a, 1961b; Hashin and Shtrikman, 1962; Anand et al., 1973; Beck, 1976; Palciauskas, 1986; Vacquier et al., 1988; Zimmerman, 1989; Brigaud and Vasseur, 1989; Brigaud et al., 1990; Bachu, 1991; Somerton, 1992; Mohanty, 1993). We describe several commonly-used models for comparisons with our new data. We refer the reader to Brigaud and Vasseur (1989), Zimmerman (1989), Somerton (1992), and Mohanty (1993) for more detailed descriptions of the different models as applied to sedimentary rocks. Our purpose is to provide a documented data set of thermal conductivities of Wilcox Group and Frio Formation sandstones (important petroleum reservoirs in the study area) and to supply a reliable data set for the future testing of mechanistic thermal conductivity models. a The thermal conductivity of a sedimentary rock is primarily function of: mineralogy, porosity, pore-fluid conductivity, and temperature. Secondarily, it is a function of grain/pore size, shape, and distribution, and pressure (Somerton, 1992). We evaluate only the first two primary effects: mineralogy and porosity. The effects of fluid conductivity, temperature, grain-size, and pressure are less significant because thermal conductivity was measured on samples saturated with one fluid (water), at the same temperature (20 °C / 68 °F) and pressure (3 MPa / are 435 psi), and the rocks primarily very fine to fine-grained sandstones (with are the exception of Frio samples ID# 69-77 which fine to medium-grained sandstones; Table 3.1). Somerton (1992) gives three categories of models which describe the effective thermal conductivity of sedimentary rocks: i) mixing laws for porous mineral aggregates; ii) empirical models in which thermal conductivity is related to easily measured physical parameters and regression to lab data, and iii) theoretical models based on heat transfer in composite materials with simplified geometries. It is preferable to have a theoretical model to describe the physics of heat conduction (Zimmerman, 1989; Somerton, 1992; Mohanty, 1993), but reliable models have not yet been developed. Such models fail and empirical alterations of the equations are needed (Beck, 1976). Empirical models, on the other hand, may not be applicable to other rock types or in other geographic areas. We develop and examine empirical models for thermal conductivity of sandstones from the Gulf of Mexico Basin and test their global applicability by predicting the thermal conductivities of other sandstones (Woodside and Messmer, 1961b, Brigaud and Vasseur, 1989; Somerton, 1992). We restrict our discussion below to include only clean sandstones (< 25% of solids are clay), because the thermal conductivities of clay minerals are not well documented. Mudrock core samples are rarely collected. When samples are available, they are typically desiccated, fractured, and fragile. Reliable measurements on a divided-bar apparatus and extrapolation to in-situ conditions are difficult. Based on precision temperature well-logging, Blackwell and Steele (1989) suggest that the thermal conductivities of mudrocks are 25-50% lower than literature values and do as a not appear to vary function of porosity in the “expected”way. Mudrockscompriseupto80%oftheclasticrocks intheGulfof Mexico Basin. If we want to understand heat transport in this basin, there is still the need to quantify in-situ mudrock thermal properties. We address the thermal conductivities of clean sandstones and expect that muddy sandstones and mudrocks may behave differently. Adequate samples to quantify their properties were not available. Equation 3.2 is a reliable empirical relationship for Wilcox sandstones zero to at from least 20% porosity (Figure 3.3). The linear relationship also appears reasonable for the quartzose Frio sandstones (ID# 53-68), but the relationship is not as convincing because of the limited porosity range (21-30%) of these samples. This Frio subset appears to follow the decreasing linear trend defined by the Wilcox sandstones and would extend the applicability of equation 3.2 to 30% porosity. However, if the data for all clean Frio sandstones are observed separately (Figure 3.3, ignoring the Wilcox data), then a decreasing linear trend is not evident. That this Frio subset should follow the linear trend of the Wilcox sandstones seems reasonable, because their mineral components and volumes are similar (Figure 3.2, Table 3.1). This subset does not overlap with the so on the Wilcox samples in porosity space, it is not possible to test if they are same linearly decreasing trend. Quartzose Frio sandstones with porosity in the 0­ 20% range are still needed to test this hypothesis. We conclude that equation 3.2 is applicable for Wilcox sandstones (ID# 1-52) for porosity from zero to 20% and tentatively conclude that equation 3.2 is applicable for Wilcox and quartzose Frio sandstones (ID# 1-68) for porosity from zero to 30%. Equation 3.2 is a linear function and extrapolation estimates that thermal - This is unrealistic conductivity becomes negative at porosity greater than 48%. the lower limit as must approach the thermal conductivity of water (0.6 W/m/K) porosity approaches 100%. 0.6 W/m/K is reached at porosity equals 43% using equation 3.2. Loucks et al. (1986) show that porosity of clean Wilcox and Frio sandstones (onshore Texas) is typically less than 30% below depths of 300 m in (1000 ft), giving the linear equation wide applicability over a large depth range this area. Estimating grain-matrix conductivity by extrapolating equation 3.2 to zero we porosity, test two mixing models for the grain-matrix conductivity of the Wilcox sandstones (ID# 1-52) which have a limited of quartz content with range an average of 65%. Grain-matrix conductivity is required for most thermal measurements are conductivity models, but actual typically of effective thermal conductivity of rocks with greater than zero porosity. Some assumed model is implicitly required to extrapolate the measurements to grain-matrix conductivity at zero porosity. This is especially true for measurements on drill cuttings using the cell technique (Sass et al., 1971). Our linear extrapolation gives a robust on estimate of grain-matrix conductivity which is based actual measurements of thermal Wilcox effective conductivity over a large porosity range. For sandstones (ID# 1-52), we use 5.9 W/m/K as estimate of grain-matrix an conductivity and 65% as the average quartz content of the solids. The thermal well known conductivity of quartz is relatively (7.8 W/m/K) (Brigaud et al., 1990). If the grain-matrix conductivity can be described by a two-component we can mixing model, estimate the thermal conductivity of the solid component other than quartz. A two-component model is reasonable because these samples and are composed chiefly of quartz, feldspar, and rhyolitic rock fragments estimates of feldspar and rhyolite thermal conductivities (Brigaud et al., 1990; Gretener, 1981) are very close (2.3-2.6 W/m/K). Two simple and commonly-used models for the grain-matrix conductivity are the arithmetic and the geometric mixing models. Somerton (1992) suggests the use of the arithmetic mixing equation of the form = X aQ + b(l,0-Q) (3.3) m to estimate the thermal conductivity of the grain matrix (Xin W/m/K) where Q is m quartz content as a fraction of the solids, and a and b are the thermal conductivities of quartz and the lumped “other” mineral, respectively. With estimates of X (= 5.9), a (= 7.8), and Q (= 0.65), we can rearrange the equation m and solve for b. b equal to 2.4 W/m/K is a reasonable estimate for feldspars and rhyolitic/trachytic rock fragments (Brigaud et al., 1990; Gretener, 1981). Alternatively, Woodside and Messmer (1961b) and Brigaud et al. (1990) suggest using the geometric mixing model: Q ~Q) X =a b(3.4). m Using the same constraints as for arithmetic mixing, equation 3.4 can be solved for b, giving b = 3.5 W/m/K. This is higher than typically reported for feldspars and rhyolite. Consequently, based on laboratory measurements of effective thermal conductivity and petrography, the known thermal conductivity of quartz, and reasonable estimates of the thermal conductivity of the “other” minerals, the arithmetic mixing model to be better for the grain-matrix thermal appears conductivity for this suite of rocks. The estimate for the “other” solids is 2.4 W/m/K. Lumping all other minerals into “other” is viable because quartz appears to dominate heat conduction through these rocks when quartz is greater than 35% of the solids. The arithmetic mixing model is a reasonable physical model with quartz and “other” minerals acting as if they are aligned parallel to the direction of heat transport with the heat transport predominantly being through the quartz. We tested models for calculating the effective thermal conductivity of sandstones with porosity greater than zero by using data for the clean Wilcox and results of the models will Frio sandstones (ID #l-77). For comparative purposes, be shown in plots of measured versus predicted thermal conductivity (Figures 3.5 and 3.6). We compare 5 methods: i) the linear relationship given as equation 3.2 (Figure 3.5a); ii) a multi-component, geometric mixing model (Figure 3.5b); iii) arithmetic mixing a two-component geometric mixing model (Figure 3.5c); iv) an model a for grain-matrix conductivity, and two-component geometric mixing model thermal conductivity (Figure 3.6a); and v) multiple for the effective a linear regression model using quartz and porosity as the independent variables (Figure 3.6b). In addition, we evaluated models using the geometric and arithmetic mixing models for grain-matrix conductivity and the upper and lower thermal conductivity bounds given by the Maxwell mixing model (Hashin and Shtrikman, 1962). All of these “Maxwell-type” models predicted thermal conductivities that deviated farther from the observations than the results presented below. Using the linear model, 88% of the predicted thermal conductivities of quartzosesandstones arewithin10%oftheobservations(Figure3.5a). Asnoted above, this model is not valid for the sandstones; the large nonquartzose prediction error for the nonquartzose samples is evident in Figure 3.5a. The multi-component, geometric mixing model uses mineral and water abundances and conductivities to calculate the effective conductivity (k ): e n . * n Vr(3.5) e= i=l where is the conductivity of the ith of n components and F. is its fractional volume. This has been shown by Brigaud et al. (1990) to be a robust estimate using large data sets derived from measurements on drill cuttings. The geometric model underpredicts thermal conductivity of the higher conductivity rocks by up to 20% for our samples (Figure 3.5b). With this geometric model, only 47 and 50% of the predicted thermal conductivities are within 10% of the observations for the quartzose and combined (quartzose/nonquartzose) sandstones, This model low respectively. does, however, predict the conductivity well. nonquartzose samples very Figure 3.5 c shows the results of a simpler two-component geometric model (grain-matrix/water) with a grain-matrix conductivity of 5.9 W/m/K. This a at geometric model is good predictor high thermal conductivities because it converges to the given “known” matrix conductivity, but it greatly overpredicts lower conductivity measurements. With this geometric model, only 74 and 65% of the predicted thermal conductivities are within 10% of the observations for the quartzose and combined sandstones, respectively, and the model is a poor predictor at lower conductivities. Figure 3.6 a shows the results of a two-component arithmetic mixing model for the grain matrix conductivity (equation 3.3) and a two-component model geometric mixing (grain-matrix/water, equation 3.5) for the effective thermal conductivity. This model predicts 72 and 66% of the thermal conductivities within 10% of the observations for the quartzose and combined sandstones, respectively. Figure 3.6 b shows the results of using a multiple linear regression as equation with quartz and porosity the independent variables calculated using samples LD# 1-77. The regression equation is: - \ =(4.08±0.24) (0.111±0.006)(|)+(0.026±0.003)Q (3.6) where effective thermal conductivity (A,) is in W/m/K, porosity (()>) is in percent e of bulk rock, and quartz content (Q) is in percent of solids. This model predicts 84 and 79% of the thermal conductivities within 10% of the observations for the quartzose and combined sandstones, respectively. It does not predict well as as the linear model for the quartzose samples but for the combined set of clean sandstones (quartzose and nonquartzose), it is the best fitting model. Several anomalous points are conspicuous on Figure 3.6b. One point is strongly underpredicted; this sample has the highest calcite content for all the samples plotted. Calcite’s thermal conductivity is ~3.4 W/m/K, which is higher than the “other” abundant minerals in these rocks, hence the underprediction. The other set of conspicuous samples consists of some samples from the “B” well, which are overpredicted. These samples are the least indurated of the entire sample set implying that less grain-to-grain contactthem to be conductors than causes poorer predicted by equation 3.6. We tested the global applicability of equations 3.2 and 3.6, developed as on using this suite of rocks, by predicting the thermal conductivity measured another suite of rocks. We selected data from the literature which were described in enough detail so that we could infer that they were from clean sandstones and were measured at similar saturation, pressure, and temperature conditions as our data set (Woodside and Messmer, 1961b; Brigaud and Vasseur, 1990; Somerton, 1992). These are all sandstones. The linear expression based on quartzose porosity only (equation 3.2) is not a good global predictor as evident in Figure 3.7a. The linear expression based on porosity and quartz (equation 3.6) is a much Above better global predictor (Figure 3.7b) for rocks with less than 90% quartz. 90% quartz, thermal conductivity is underpredicted, suggesting that these rocks aon a have stronger dependence quartz content when the rock approaches quartzarenite composition. Based on these analyses, we suggest that equation 3.6 is applicable for sandstones with the following characteristics: quartz content and porosity as outlined in Figures 3.2 and 3.8, clay (0-25%), calcite (0-10%), and well indurated. This is an empirical model and as such does not quantify the behavior of heat transport in sandstones, but this suite of clean sandstones (ID #l-77) is an ideal data set to test mechanistic models of thermal conductivity. In the past, theoreticians were forced to rely on measurements made on drill cuttings where an an there is underlying assumption of accurate mixing model during data reduction, or on the limited data set (5 samples) of Woodside and Messmer (1961b). The keys to our data set being robust are: i) thermal conductivities were measured on a well-calibrated divided-bar which is the preferred apparatus method recommended by the International Heat Flow Commission; ii) thermal conductivities are precise within 2% of the measurements and inaccuracy is less than 4%; iii) petrography is well documented and based on detailed thin-section analyses; and iv) porosities measured using a standard method and verified were with another method. Conclusions 1. Thermal conductivities of Wilcox and Frio sandstones from 2.06 range to 5.03 W/m/K over a porosity range from 2.4 to 29.6 %. 2. At a given porosity, clean Wilcox sandstones have higher conductivities than clean Frio sandstones because of the higher quartz content in Wilcox sandstones. 3. The grain-matrix conductivity of clean Wilcox sandstones is 5.9 ±0.3 4. can be 5. The thermal conductivity of clean Wilcox and Frio sandstones (< 25% W/m/K and is best described using an arithmetic mixing model. Most variance in thermal conductivity of the clean sandstones explained by variations in quartz contentand porosity. of solids are clays, < 20% of the solids are calcite and well-indurated) can be described with an empirically derived linear decrease in conductivity with increasing porosity coupled with a linear increase in conductivity with increasing quartz content (equation 3.6). 6. For quartzose samples (all Wilcox, some Frio) the dependence on quartz can be dropped and thermal conductivity can be adequately described by a linear decrease in thermal conductivity with increasing porosity (equation 3.2). This linear relationship describes 90% of the variance in the thermal conductivity. 7. Equation 3.6 predicts the thermal conductivity of other sandstones that are within the following domain: quartz content and porosity as outlined in Figures 3.2 and 3.8, clay (0-25%), calcite (0-10%), and well indurated. 8. Based on limited data, effective thermal conductivities of Wilcox and Frio sandstones are isotropic. Chapter 4: Radiogenic heat production in sedimentary rocks in the Rio Grande Embayment, Gulf of Mexico Basin Abstract Radiogenic heat production within the sedimentary section of the Gulf of Mexico Basin is a significant source of heat. It should be included in thermal We calculate that it models of this basin (and perhaps other sedimentary basins). may contribute up to 26% of the overall surface heat-flow density for an area in south Texas. Based on measurements of the radioactive-decay rate of a-particles, and bulk potassium concentration, density, we calculate radiogenic-heat production for Stuart City (Lower Cretaceous) limestones, Wilcox (Eocene) sandstones and mudrocks, and Frio (Oligocene) sandstones and mudrocks from south Texas. Heat production rates range from a low of 0.07 ±O.OI pW/m3 in 3 clean Stuart City limestones to 2.21 ±.24 pW/m in Frio mudrocks. Mean heat production rates for Wilcox sandstones, Frio sandstones, Wilcox mudrocks, and 3 Frio mudrocks are 0.88, 1.19, 1.50, 1.72, and pW/m respectively. In general, , the mudrocks produce about 30-40% more heat than stratigraphically equivalent sandstones. Frio rocks produce about 15% more heat than Wilcox rocks unit per volume of clastic rock (sandstone/mudrock). A one-dimensional heat-conduction model indicates that this radiogenic heat has a significant effect on source subsurface If a thermal model is calibrated to observed temperatures. heat temperatures by optimizing basal heat-flow density and ignoring sediment source production, the extrapolated present-day temperature of a deeply buried rock will be overestimated. Introduction maturation and the thermal Hydrocarbon diagenesis are functions of history of the host sediments/sedimentary rocks. Numerical modeling is one of the primary tools used to reconstruct this thermal history (Hermanrud, 1993). While considerable effort has gone into developing sophisticated thermal-history models, little effort has gone into quantifying the fundamental thermal rock- properties that are required as model input. Radiogenic heat production of sedimentary rocks is one property that has generally been neglected. To evaluate the thermal history of a sedimentary basin, and sinks of heat need to be sources considered along with heat transfer via conduction and convection. The source of the heat from the Earth’s interior originates from three major sources: i) primordial energy of planetary accretion, ii) segregation and solidification of the Earth’s core, and iii) radiogenic heat production from the decay of uranium (U), thorium (Th), and potassium (K). Although the relative proportions of these sources are uncertain (Durrance, 1986; Pollack et al., 1993), the radiogenic source (in the crust and lithosphere) is estimated to account for about 80% of the heat lost from the Earth’s surface (Birch, 1954; Turcotte and Schubert, 1982). Radiogenic-heat production can be divided into three components: heat generated i) in the lithosphere, ii) in the crust beneath the basin (i.e. basement), and iii) in the sedimentary rocks within the basin. The lithosphere has a low heat­ 3 production rate (0.01 pW/m for peridotite), but is a significant heat source per unitsurfacebecauseofitslargethickness. Theheatdistributedtothebaseofthe crust is dependent on the dynamics of the underlying lithosphere (including asthenospheric upwelling and rifting of the overlying crust), but the heat-flow 2 density at the top of “stable” continental lithosphere is about 25 ±lO mW/m (Durrance, 1986; Fowler, 1990). The lower crust or basement is a more source heterogeneous of heat; average heat production may range from a low of 13 0.01 jiW/m for an ultramafic basement, to 10 jLiW/m for a granitic basement (Roy and Blackwell, 1968; Rybach, 1988). On a local scale, heat production can approach 600 jiW/m3 for a commercial-grade uranium ore. The lithospheric and crustal radiogenic sources, along with convective heat flow into the base of the thermal lithosphere, contribute to what is termed basement or basal heat-flow density or basement heat flux (the heat flowing into the base of the sedimentary basin through a unit area). Heat is transferred into the basin by conduction and convection via rock movement (faulting, diapirism, etc.) and/or fluid flow. Conductive, basal heat-flow density at the lower boundary of a sedimentary basin 2 is typically estimated to be on the order of 15-150 mW/m Convection of heat . frombasementtobasin-fillisdifficulttoestimatebut besignificant.Within may a sedimentary basin, there is a heterogeneous distribution of radioactive minerals. Sedimentary rocks typically generate heat ranging from < 0.1 fiW/m1 for salt and 3 anhydrite to 5.5 JiW/m for “black shale” mudrocks (Rybach, 1986). In this paper, quantify heat production in south Texas sedimentary we rocks and determine the the significance to overall temperature pattern by calculating the heat produced within the sediments and comparing it to the surface heat-flow density. To simplify the problem and to evaluate the first-order effects of this internal source, we consider one-dimensional, steady-state heat conduction. We do not model either sediment deposition/compaction nor convective transport of heat. The former has an effect on temperatures over long times (Sharp and Domenico, 1976) and the latter has a local effect on temperatures (Bodner and Sharp, 1988; Pfeiffer and Sharp, 1989; McKenna and Sharp, 1996). We evaluate the following: 1. the heat production of several stratigraphic units 2. the quantity of heat generated in the sedimentary sequence, 3. the relationship of heat generated in the sediments to the overall surface heat-flow density, and 4. implications for estimating subsurface temperature. Previous WorkandSignificance as on Though it is recognized important, little data have been published radiogenic-heat production of the sediments within deep basins its or on significancetotheoverallthermalregime(KeenandLewis,1982;Rybach, 1986; Hermanrud, 1993). In a thermal model for the eastern North America divergent margin. Keen and Lewis (1982) calculated that radiogenic-heat production in 2 km (32,800 ft) of sediment contributed about 15% (8 mW/m ) to the total surface 2 heat-flow density of 50-68 mW/m Calculated temperatures increased to 17 . up °C (30.6 °F) at depths from 6 to 10 km (19,700 to 32,800 ft) in models that included heat production relative to those where the effect was ignored; the same basal heat-flow density was used in both models. In that study, heat production 3 varied with lithology and ranged from a low of 0.3-0.6 (xW/m for limestones to a 3 high of 1.4-1.8 pW/m for shales. Allowing for heat production in the sediments increased the calculated rate of hydrocarbon maturation and Keen and Lewis concluded that maturation may have occurred 10-30 million years earlier in the source rocks when radiogenic-heat in the sediments was included rather than neglected. Blackwell and Steele (1989) noted that if similar heat-production values were assumed for the Gulf of Mexico Basin, radiogenic-heat production would contribute from 25-50% of surface heat-flow density, but no heat- production data were then available to test their hypothesis. Hermanrud (1993) suggests that radiogenic heat-production within the sedimentary sections of basins is generally insignificant with a contribution of only a few percent or less to heat- flow density. For the Western Canada Basin (Alberta, Canada), Bachu (1993) estimates that radiogenic heat-production within the sediments from 1­ ranges 22% of the surface heat-flow density. Measurements of radiogenic heat production in Gulf of Mexico sediments are are sparse. The few other observations that have been reported for pelagic 3 muds (4.3, 4.4, and 4.5 pW/m ; Epp et al., 1970; Pollack et al., 1993), and silty 33 mudstones (1.36 pW/m ), and carbonates (0.66 pW/m ) from the deep, central part of the Gulf (Nagihara et al., 1996). No data have been published for the major onshore units of the Gulf of Mexico Basin. Host minerals for the volumetrically important radiogenic heat-producing isotopes in the sediments of south Texas are indicated in Table 4.1. Major sources for uranium and thorium include the heavy minerals apatite, zircon, and monazite, and detrital and authigenic sphene (Milliken, 1988; Milliken and Mack, 1990), noncrystalline forms sorbed onto clay surfaces (Durrance, 1986) as identified by bulk rock analyses of mudrocks (L. S. Land, 1993, personal communication), and organic sources (Durrance, 1986; Bartow and Ledger, 1994). Major potassium sources include detrital illite/smectite, authigenic illite, and feldspars (Hower, 1976; Loucks et al., 1986; Lynch, 1995). area There are no heat-flow density measurements published for the study (Figure 4.1). Smith et al. (1979) estimate the heat-flow density to be 15±3 2 mW/m at a depth of ten kilometers (32,800 ft) in northern Mexico about 100 km (62 mi) to the west of the area. Nagihara et al. (1996) estimate heat-flow density from the lithosphere (at a subsea depth of 11-13 kilometers [36,000-42,700 ft]) is 2 40-47 mW/m in the Sigsbee abyssal plain, about 400 km (249 mi) to the southeast of the south Texas area. Thermal modeling is typically used to predict the timing of hydrocarbon generation. The lack of information on heat production in the sedimentary section poses a problem with accuracy when calibrating these models. Neglecting a heat production in the sedimentary section and calibrating model to observed temperature and surface heat-flow density data may be possible by adjusting the basal heat-flow density and thermal conductivity of the sedimentary section, but an this will not necessarily yield accurate representation of the present-day thermal state of a basin. This is especially significant when using a thermal model to extrapolate temperatures to depths below observations as is often done to determine the present temperature of deeply buried hydrocarbon source rocks. This extrapolated temperature may be the only “data” point in time for subsequent thermal history modeling of the source rock. As shown below, ignoring radiogenic heat production in a model may cause overestimates of With the thermal maturation rate of organic material in present-day temperature. a source rock doubling with every 10 °C (18 °F) rise in temperature (Waples, 1980; Lerche, 1990), the calculated rate of hydrocarbon maturation will be overestimated, resulting in a premature prediction for the onset of hydrocarbon generation. Methodology We calculated the radiogenic-heat production of 100 samples of sandstones, mudrocks, and limestones from south Texas. Rock samples were collected from cores and drill cuttings (Figure 4.1, Table 4.2) from the Core Research Center and Department of Geological Sciences collections at The University of Texas at Austin. Additional lithologic and geochemical datafor the core samples are available in Stanton (1977), Prezbindowski (1981), Fisher (1982), Lynch (1994), and McKenna et al. (1996). Additional geochemical data also available S. for the drill cuttings were (L. Land, 1995, personal were communication). Mudrock samples hand picked from drill cuttings over approximately 30-meter (100-foot) intervals; the depth given in Table 4.3 is for the midpoint of each interval. We measureda-particle decay rates in an alpha scintillation counterusing the method described by Huntley and Wintle (1981), Huntley et al. (1986), and Huntley (1988). Counting statistics are used to calculate the concentrations of U and Th as a function of the count rate (Huntley and Wintle, 1981; Daybreak, 1990). U and Th are differentiated by the number of “slow pairs” detected. A slow pair is defined as two counts occurring within 0.4 seconds. True pairs result = from the decay of 216P0 (half life 0.145 seconds) in the Th decay chain. These so pairs are recorded infrequently and other “pairs” may be recorded by chance, the calculated U and Th concentrations are much less precise than the measured rate of cx-particle emission. Precision is based on counting statistics, such that the longer the samples are counted, the more precise the measurement. Our experiments concur with those of Huntley (1988) that 1000 counts (about 1 day) is an appropriate time limit. Inaccuracy between laboratories and imprecision of count-rate measurements are on the order of 12% for a-particle counts. For assessing accuracy, the isotope dilution method was used at The University of Texas and the neutron activation method was used at XRAL Laboratory [Ontario, Canada]; a-particle counts were calculated based on the U and Th concentrations. Imprecision of U and Th concentrations is on the order of 30% with the error caused chiefly by the small number of slow pairs counted. To be counted, both pairs must be emitted towards the alpha counter, and occasional “pairs” will not actually be pairs, but will record two alpha particles from different nuclei that were emitted in the 0.4 second time window of the counter. Huntley (1988) be convincingly shows that the a-particle counts can used directly to estimate heat production without the intermediate step of calculating the concentrations of uranium and thorium. Therefore, the imprecision and inaccuracy of our measurements are,atmost, 12%ofthereported value. Potassium concentration data were available from ICP spectrometry analyses (L. S. Land, 1995, personal communication) or were measured by the x- ray fluorescence method by XRAL Laboratory (Ontario, Canada). Measurement imprecision and inaccuracy is 5% of the measured value. Bulk density was determined by direct measurements of mass and volume for sandstones and limestones. Mudrock density was established using Dickinson’s (1953) curve for Gulf of Mexico mudrocks because desiccation of mudrocks while in core storage made accurate direct measurements of in-situ bulk volume and porosity impossible. Measurements of matrix volume via gas porosimetry also fail because minute amounts of water in the mudrocks prevents stabilization in the measurement vessel. Bulk density inaccuracy gas-pressure and imprecision is about 5% and 10% of the reported value for sandstone/limestone and mudrock samples, respectively. Heat production is calculated (Rybach, 1986; Huntley, 1988) according to: A=(ot*a+K*b)*p(1) b 3 where Ais the bulk-rock heat-production rate (|iW/m ), ais the alpha-count rate in number of counts per kilosecond per square centimeter (counts/ ks/cm 2), ais the conversion factor from the number of alpha particles emitted to 42 the amount of heat produced [7.288x10‘ (|iW / (kg))/ (counts/ks/cm )], K is solids the potassium concentration in weight percent of solids (%), b is the solids conversion factor from the percentage of potassium to the amount of heat 53 produced (3.48x10 |iW / % ) and pb is the bulk density of the rock (kg/m ). solids Radiogenic-heat-production rate is an isotropic petrophysical property independent of pressure, temperature, and chemical surroundings. Errors propagate in the calculations; resulting inaccuracy and imprecision is less than 13% of the calculated radiogenic heat production values. Results ofHeat ProductionCalculations 3 from a low of 0.07 |TW/m in clean Stuart City Heat production ranges limestones to a high of 2.21 pW/m3 in Frio mudrocks (Table 4.3). Arithmetic 33 means for Wilcox sandstones to 1.72 pW/m for Frio range from 0.88 pW/m mudrocks (Figure 4.2; Table 4.4). Means were not calculated for the limited limestone data. Mudrocks produce 30-40% more heat than stratigraphically equivalent sandstones and Frio rocks produce about 15% more heat than Wilcox rocks when viewed as a bulk unit of clastic rock (consisting of 80% mudrock and 20% sandstone) (Figures 4.2 and 4.3). Frio rocks produce more heat because they contain abundant volcanic rock fragments containing higher concentrations of uranium and thorium. Muddy limestones produce heat similar to the sandstones (Figure 4.2). There may be a small increase in heat production with depth for both Frio and Wilcox mudrocks (Figure 4.3), because of the increase in bulk density, but the trend is uncertain due to the 13% error in heat production and the assumed bulk density function for the mudrocks. No areal trends were identifiable. The heat production that we measured in clean limestones is significantly lower (0.07-0.18) than the “typical” average(0.62) given in Rybach (1986). Heat production is 5-15% and 20% higher in mudrocks and sandstones, respectively, than in the “typical” averages (Rybach, 1986). This exemplifies the importance of site-specific measurements. Thermal Models The results of seven numerical models of one-dimensional, steady-state, conductive are heat transport for a 10 km (32,800 ft) thick stratigraphic section presented. The first set of models (1,2, and 3) is for a true stratigraphic section through the Frio depocenter in south Texas (Figures 4.1 and 4.4; Table 4.2). The second (4 and 6) and third (5 and 7) sets of models are for idealized stratigraphic sections composed entirely of sandstone or mudrock, respectively. Models 4 and 6 are for an idealized section composed entirely of Wilcox sandstone and models 5 and 7 are for an idealized section composed entirely of Frio mudrock. Details on input parameters and constitutive equations are given in Table 4.5. Porosity is a function of lithology and depth; thermal conductivity is a function of lithology, a porosity, and temperature; heat production is function of lithology (Table 4.5, Figure 4.4). Lithology is based on electric logs and geologic interpretations (Sohl et et al., 1991; Galloway al., 1994). Boundary conditions are specified at the top and specified heat-flow density at the base of the model. temperature We assume the sediment pile is in thermal equilibrium (i.e. no sedimentation- moving boundary effect); this assumption may not be entirely realistic, but this simplification is adequate to quantify the effect of ignoring sediment radiogenic on heat production estimating the present-day temperature of a potential source rock. The only variables modified within a set of models are the amount of radiogenic heat production in the sediment pile and the basal heat-flow density. Model 1 includes sediment heat production and the basal heat-flow density was optimized to a visual best-fit to observed temperature data. Model 2 neglects heat production, and as in model one, the basal heat-flow density was optimized to on suites temperature data. “Observed” temperature data are best estimates based of Kehle (1971) corrected bottom-hole temperatures within 20 km (12.5 miles) of the stratigraphic section. Model 3 uses the basal heat-flow density determinedin model 1, neglects heat production, and is not calibrated to observed temperatures. Models 4 (all sandstone) and 5 (all mudrock) include sediment heat production and models 6 (all sandstone) and 7 (all mudrock) neglect heat production. Models 1 and 2 result in optimized basal heat-flow densities of 35 mW/m2 and 45 mW/m2 respectively. Therefore 10 mW/m2 higher basal heat- a , flow density is needed to match the observed temperatures when sediment heat production is ignored. Resulting temperatures at 10 km (32,800 ft) are 28 °C (50 °F) higher in model 2 than in model I (Figure 4.5a). Therefore, if sediment heat production is ignored and an optimized higher basal heat-flow density is used to match the observed temperatures, the present-day source-rock temperature at km (32,800 ft) is overestimated by 28 °C (50 °F). There is most likely some error associated with the temperature observations. If we assume 10 °C error bars on the observed temperature data and assume that the thermal conductivity is well constrained and again optimize the basal heat-flow density, the resulting temperature difference at a depth of 10 kilometers is still higher in the case with no heat production by about 5 °C. As the error in the temperature observations above 10 °C, than this effect of radiogenic heat production becomes goes indiscernible. Comparing model 1 to model 3, it is evident that sediment heat production accounts for 26% of the surface heat-flow density. we For comparative purposes, also modeled stratigraphic sections consisting entirely of Wilcox sandstone and Frio mudrock (Figure 4.5b). These serve as end-member simulations for temperature-depth profiles with and without inclusion of heat production. Models 4 (sandstone) and 5 (mudrock) include sediment heat production and models 6 (sandstone) and 7 (mudrock) neglect heat production. The basal heat-flow density for models 4,5, 6, and 7 was assumed to 2 be 45 mW/mand was not optimized to any temperature observations. For the idealized sandstone section, resulting temperatures in model 4 (with heat production) at 10 km (32,800 ft) are 16 °C (29 °F) higher than in model 6 (without for 17% of the surface heat production) and sediment heat production accounts heat-flow density. For the idealized mudrock section, resulting temperatures at 10 km (32,800 ft) are 64 °C (115 °F) higher in model 5 (with heat production) than in model 7 (without heat production) and sediment heat production accounts for 27% of the surface heat-flow density. As expected, radiogenic heat in the mudrock section produces a significantly greater effect on temperature and surface heat-flow density than in the sandstone section. Implications basal heat-flow density and thermal conductivity to match Calibrating When observed subsurface temperatures is frequently done in thermal modeling. there are no other thermal indicator data (e.g. vitrinite reflectance, apatite fission track length) to calibrate the model, the fit to present-day temperatures is the only of the thermal model. For the eastern North America measure accuracy of divergent margin, Keen and Lewis (1982) showed that including radiogenic-heat production in modeling resulted in increased temperatures up to 17 °C (30.6 °F) at depths from 6to 10 km (19,700 to 32,800 ft). Our analysis for south Texas indicates increases in calculated temperature over 1.5 times that amount (28 °C [SO °F]) at 10 km (32,800 ft) for a simulation that neglects sediment radiogenic This is heatproduction relativetoonethatconsidersradiogenic heatproduction. the opposite effect of that concluded by Keen and Lewis (1982). The reason for in our calculation (models 1 and 2) the discrepancy is that the two models we use used different basal heat-flow densities that were optimized to match observed Keen and Lewis (1982) used the same basal heat-flow density in temperatures. models that included and neglected heat production. This is equivalent to comparing our models 1 and 3 which show a similar effect that arises due to the increased amount of heat put into the system by heat production. Our purpose is to show the importance of radiogenic heat production in estimating the present­ daytemperatureofasourcerock asthistemperaturemaybetheonlydatapointin time for thermal history modeling of the source rock. Therefore, by ignoring radiogenic heat production and optimizing basal heat-flow density to temperature observations, thermal maturity may be overestimated and the predicted onset of thermal models be earlier than the hydrocarbon generation by subsequent may actual generation. Conclusions Radiogenic-heat production within the sedimentary section of the Gulf of Mexico Basin is a significant source of heat. We agree with Keen and Lewis (1982)thatradiogenicheatproduction insedimentsshouldbeincludedinthermal history modeling. 1. Radiogenic-heat production within the sedimentary section may 2. Heat production is in the mudrocks and lowest in the contribute up to 26% to the overall surface heat flow in south Texas. greatest limestones. Mudrocks produce 30-40% heat than time-equivalent more sandstones. Frio rocks produce about 15% more heat than Wilcox rocks as a bulk unit of clastic rock (sandstone/mudrock), because they have higher concentrations of U and Th. Frio sediments contain more volcanic rock fragments than Wilcox sediments. 3. Radiogenic heat in the sediments has a significant effect on within the sediments. Ignoring sediment heat and temperatures production calibrating a thermal model to temperature observations by optimizing the basal heat-flow density may overestimate the present-day temperature of a potential source rock at 10 km (32,800 ft) by as much as 28 °C. Chapter 5: Summary Results were presented from analyses of the thermal properties of rocks from the Embayment and from the development and analysis of a large geographic-information-system database of subsurface temperatures, fluid and salinities. These data, along with available regional geological and pressures, geophysical interpretations, were used to evaluate the hypothesis that fluids are episodically expulsed from overpressured rocks, discharging vertically along the regional Wilcox and Frio Fault Zones in the Embayment. Results indicate that the hypothesis is viable; it is consistent with the pressure, temperature, and salinity data and with other geologic observations. sediments Fluids are episodically expulsed from extremely overpressured during natural hydrofracturing and discharge vertically along regional fault zones in the Embayment. The basal 9 km (29,500 ft) of the stratigraphic section is extremely overpressured with fluid at 80 to 90% of the overburden pressures close to the minimum pressure needed for the onset of hydraulicpressure, section is fracturing. At least 3 km (9,800 ft) of the extremely overpressured within the greenschist metamorphic facies and possibly part of the amphibolite facies. Small increases in fluid pressure in this extremely overpressured regime, caused by recharge the system from hydrocarbon generation, to diagenetic/metamorphic fluids, or other areas where the limiting pressure of hydraulic fracturing is higher, trigger hydraulic fracturing and episodes of fluid discharge. The Wilcox, Vicksburg, and Frio Fault Zones all fluid discharge are zones, but the Wilcox Fault Zone is a more recent discharge area for fluids from a deeper fluid source. The most recent pulse of fluid discharge was along the Wilcox Fault Zone where the largest thermal anomaly in the Gulf of Mexico Basin occurs coincident with a positive fluid pressure anomaly and salinity inversions. The thermal anomaly is centered on the most basinward and deepest of the Wilcox faults. Heat conduction is not a viable mechanism for producing the anomaly. Hydraulic fracturing in the fault zone creates a relatively permeable vertical pathway that focuses fluid discharge from the extremely overpressured rocks faults. along the deep-seated Heterogeneous salinity distributions along both the Wilcox and Frio Fault Zones represent the cumulative effect of expulsion events in both fault zones. Neither distinct thermal anomalies are nor pressure evident in the Frio Fault Zone, suggesting that fluid expulsion events had a diffuse than in the shallower fluid source, were of smaller magnitude, were more Wilcox Fault Zone, or that similar thermal and pressure anomalies have already dissipated. The deep source of fluids, hydraulic fracture, and episodic fluid expulsion hypothesis is consistent with temperature, fluid pressure, and salinity data. The Rio Grande Embayment may a regional discharge be for the thick zone sedimentary section in the western Gulf of Mexico Basin. The thermal conductivity of 83 Wilcox and Frio sandstones from the Embayment were measured and conductivity was correlated to petrographic variables. Thermal conductivity ranges from 2.06 to 5.73 W/m/K over a porosity range of 2.4 to 29.6 %. For a given porosity, because of a higher quartz content, Wilcox sandstones are more conductive than Frio sandstones. Thermal conductivities of clean (< 25% clay) sandstones can be described by multilinear a function of decreasing thermal conductivity with increasing porosity and increasing thermal conductivity with quartz content. For clean, quartzose (> 35% of the solids) sandstones, the dependence on quartz content can be dropped and thermal conductivities can be predicted with a linear decrease in conductivity with increasing porosity. These sandstones appear isotropic with respect to thermal conductivity. Radiogenic heat production within the sedimentary section of the Gulf of Mexico Basin is significant of heat. It should be included in thermal a source models of this basin (and perhaps other sedimentary basins). Modeling calculations indicate that it contribute to 26% of the overall surface heat- may up an area was flow density for in south Texas. Radiogenic-heat production calculated for Stuart City (Lower Cretaceous) limestones, Wilcox (Eocene) sandstones and mudrocks, and Frio (Oligocene) sandstones and mudrocks from 3 south Texas. Heat production rates range from a low of 0.07 ±O.OI |iW/m in clean Stuart City limestones to 2.21 ±.24 pW/nv in Frio mudrocks. Mean heat production rates for Wilcox sandstones, Frio sandstones, Wilcox mudrocks, and 3 Frio mudrocks are 0.88, 1.19, 1.50, 1.72, and |xW/m respectively. In general, , more the mudrocks produce about 30-40% heat than stratigraphically equivalent sandstones. Frio rocks produce about 15% more heat than Wilcox rocks per unit volume of clastic rock (sandstone/mudrock). A one-dimensional heat-conduction source a on model indicates that this radiogenic heat has significant effect subsurface If a thermal model is calibrated to observed temperatures. temperatures by optimizing basal heat-flow density and ignoring sediment heat production, the extrapolated present-day temperature of a deeply buried source rock will be overestimated. Chapter 6: Suggestions forFuture Work The Rio Grande Embayment is one of the most studied areas in the Gulf of Mexico Basin. Extensive information has resulted from scientific studies, hydrocarbon exploration and production, uranium exploration and mining, evaluation of geothermal resources, and the permitting of injection wells. Compiling and analyzing these data for studies of regional fluid and heat flow in the Embayment are difficult due to the widely disseminated and disparate nature of the information. The development of a geographic information system (gis) for this study facilitated the synthesis of some of these data into a coherent structure that can be easily updated and analyzed by a geologist with a beginner’s knowledge of gis. This gis data set provides a framework for further study in the Embayment. As stated in the introduction, the two most significant questions that must be addressed in our way towards a good understanding of fluid and heat flow in the basin are: i) the fluid-volume problem and ii) our limited understanding of the thermal and hydraulic properties of the mudrocks. These are discussed below along with some other suggestions for future work. Where do the large volumes of water come from that are needed to do the pervasive diagenesis and how are the observed extreme fluid overpressures maintained if large volumes of fluids are flushing through the basin? Is there a source of recharge where the basin fill and crust are underlying undergoing prograde metamorphism? Answering these questions will require analyses of waters and associated gases that could identify a deep source of fluids from a metamorphosing crust and/or degassing mantle. These analyses should include identification of isotopes of helium and nitrogen in natural gas (hydrocarbons) and carbon in carbon dioxide (Sassen, 1990; Bredehoeft and Ingebritsen, 1990; et Land, 1991; Torgersen and Clarke, 1991; Torgersen al., 1995). Do fluids recirculate via free convection in the hydropressured and extremely overpressured zones whose positions be variable in and time? Preliminary may space calculations of the importance of buoyancy forces on fluid flow in the Embayment indicate that free convection is probable in the hydropressured regime (Bodner et al., 1985; Simmons et al., 1997), improbable in the overpressured regime, and possible in the extremely overpressured regime. As this is difficult to document with observations, numerical modeling is the preferred method to test the feasibility of free convection (Sharp, 1997). Mudrocks comprise 80-90% of the basin fill but their thermal and hydraulic properties are not well known. The numerical models of free convection noted above also could be used along with salinity observations to estimate the regional permeability of mudrocks (Simmons et al., 1997). This is difficult to document with observations. Since it also is difficult to measure the thermal conductivity of mudrocks in the laboratory (along with the question of whether these measurements can be “scaled up” to the field scale), an in-situ method is more appropriate. The best estimate could be ascertained by a combination of precision temperature and spectral gamma ray logging in a well with available sandstone core. Temperature gradients could be calculated from over a clean sandstone interval identified with the the temperature log gamma-ray log and the rock core. Thermal conductivity could be measured on the sandstone core (or cuttings if no core was available). Radiogenic heat production in the mudrock and sandstone intervals could be calculated from the spectral gamma ray log and measurements / estimates of the bulk density. Heat-flow density in the sandstone interval could be calculated from the temperature gradient and temperature-corrected thermal conductivity. Thermal conductivity of the mudrocks could then be estimated using the temperature gradient through the mudrock interval and the calculated heat flow density corrected for the amount of radiogenic heatproduction in the mudrock. Other problems that could be addressed include: 1. recognition of episodic expulsion and heating events using thermal indicators such as apatite fission tracks (Corrigan, 1990; Cloos and Corrigan, 1989; Lerche, 1990), vitrinite reflectance (Lerche, 1990; Tyler, et al, 1986), analysis of bands of sulfides in salt-dome cap rocks (Hallager et al., 1990), and recognition of multiple episodes of fracture formation in sandstones (Laubach, 1997), 2. further characterization of the thermal conductivity of sandstones (McKenna et al., 1996) by measurements on Frio sandstones that fall outside the petrographic range used in this study (Figures 3.2 and 3.8), 3) incorporation of the data in this study into a three-dimensional geologic model better to constrain the relationships between geology, pressure, temperature, and salinity (McKenna, 1997), numerical of convective heat determine the 4) modeling transport to magnitude and duration of transient fluid pulses that could create the thermal anomaly (McKenna, 1997), 5) acquisition and analysis of pressure, temperature, and salinity data in Mexico to evaluate how these parameters change towards the arches that define the southwestern edge of the Embayment, 6) precision temperature logging in boreholes to examine the long- to remove standing problem of correcting bottom-hole-temperatures the cooling effects of drilling and to get estimates of heat-flow density and the thermal conductivity of mudrocks, and on 7) mapping of fluid pressures the oil/gas field scale along the geologic cross section to determine the relationship of these “local” pressures to the “regional” pressures mapped in this study. Figure 1.1. Regional location map. Figure 2.1. Local loaction map. Figure 2.2. Geologic cross section (see Figure 2.7 for location). Figure 2.3. Depth to basement from Sawyer et al. (1991). Figure 2.4. Bouger gravity anomaly map (SEG, 1982). Figure 2.5. Magnetic anomaly map (Godson, 1982). 1991). (Ewing, provinces diapir Salt 2.6. Figure Figure 2.7. Location Map for Cretaceous margins, fault trends, and section. cross Tics cross-sectionline in10kmincrementsfromthenorthwestend. on are Figure 2.8. Stratigraphic chart with depositional episodes, depocenterlocations, and tectonic events (Galloway et al., 1991). Figure 2.9. Location of Wilcox Rosita and Rockdale Delta Systems (Edwards, 1981). Figure 2.10. Location of Frio Formation depositional systems (Galloway et al., 1982). Figure2.11.Orientationofmaximumhorizontal stress(fromZobacket.al, 1991). Figure 2.12. Depth to top of fluid overpressure. Figure2.13. DepthtothebaseoftheTexasGulfCoastaquifersystems(Ryder, 1988). This was mapped as the shallower of the top of fluid the Paleocene Midway overpressure (irregular pattern in the east) or Group (linear pattern in the west). extreme fluid Figure 2.14. Depth to the topof overpressure. 119 Figure 2.15. Thickness ofthe overpressured regime. Figure 2.16. Fluid/overburden pressure ratio in the extremely overpressured regime, definitions. regime Hydrologic 2.17. Figure Figure2.18. Depthtothetopoftheoilindustry's"hard"overpressure. Figure2.19. Thermalanomalyinmapview(isothermmapforthe3km [9,840 ft] depth-slice). 124 Figure2.20. Gridoverlayforlocationsofdataused intemperature-depth and pressure-depth plots. Figure2.21. Isothermmapforthe4.5km (14,760ft)depth-slice. Figure2.22. Isothermmapforthe4.0km (13,120ft)depth-slice. Figure 2.23. Isotherm map for the 3.5 km (11,480 ft) depth-slice. Figure 2.24. Isotherm map for the 3 km (9,840 ft) depth-slice. Figure 2.25. Isotherm map for the 2.5 km (8,200 ft) depth-slice. 130 Figure 2.26. Isotherm map for the 2.0 km (6,560 ft) depth-slice. 131 Figure 2.27. Isotherm for the 1.5 km (4,920 ft)depth-slice. map 132 Figure 2.28. Isotherm map for the 1.0 km (3,280 ft) depth-slice. 133 Figure2.29. Isothermmapforthe0.5km (1,640ft)depth-slice. 134 Figure2.30. Temperature indegreesCelsiuscontouredongeologiccrosssection (seeFigure 2.7forlocation). 135 Figure 2.31. Temperature-depth plot for data northwest of the Wilcox Fault Zone along the cross section in Figure 2.30 (50-100 km [3l-62 mi] from northwestern side of section). Figure2.32. Temperature-depthplotfordatanorthwestoftheWilcoxFaultZone along the cross section in Figure 2.30 (101-150 km [63-93 mi] from northwestern side of section). Figure 2.33. Temperature-depth plot for data in the Wilcox Fault Zone along the cross section in Figure 2.30 (151-190 km [94-118 mi] from northwestern side of section). 138 Figure2.34. Temperature-depthplotfordatabetweentheWilcoxandVicksburg Fault Zones along the cross section in Figure 2.30 (191-225 km [ll9-140 mi] from northwestern side of section). Figure 2.35. Temperature-depth plot for data between the Wilcox and Vicksburg Fault Zones along the cross section in Figure 2.30 (226-245 km [l4O-152 mi] from northwestern side of section). Figure 2.36. Temperature-depth plot for data in the Vicksburg Fault Zone along the cross section in Figure 2.30 (246-275 km [153-171 mi] from northwestern side of section). Figure 2.37. Temperature-depth plot for data in the Frio Fault Zone along the cross Figure 2.30 (276-340 km [172-211 mi] from section in northwestern side of section). Figure2.38. Temperature-depthplotforcombineddatafromtheWilcox andFrioFaultZonesalongthecross sectioninFigure 2.20. Figure 2.39. Temperature-depth plot for the Wilcox Fault Zone in grid area 1007 in Figure 2.20. Figure 2.40. Temperature-depth plot for the Wilcox Fault Zone in grid area 1107 in Figure 2.20. Figure 2.41. Temperature-depth plot for the Wilcox Fault Zone in grid area 409 in Figure 2.20. Figure 2.42. Temperature-depth plot for the Vicksburg Fault Zone in grid area 514 in Figure 2.20. 147 Figure 2.43. Temperature-depth plot for the Frio Fault Zone in grid area 1012 in Figure 2.20. Figure 2.44. Temperature-depth plot for the Frio Fault Zone in grid area 1311 in Figure 2.20. Figure2.45. Temperature-depthplotfortheFrioFaultZoneingridarea815in Figure 2.20. Figure2.46. Isobarmapforthe3.5km (11,480ft)depth-slice. Figure2.47. Isobarmapforthe3.0km (9,840ft)depth-slice. 152 Figure2.48. Isobarmapforthe2.5km (8,200ft)depth-slice. 153 Figure2.49. Isobarmapforthe2.0km (6,560ft)depth-slice. Figure 2.50. Isobar map for the 1.5 km (4,920 ft) depth-slice. Figure 2.51. Isobar map for the 1.0 km (3,280 ft) depth-slice. 156 Figure2.52.Ratiooffluidpressuretooverburdenpressurecontouredongeologic cross section (see Figure 2.7forlocation). Figure 2.53. Pressure-depth plot for data northwest of the Wilcox Fault Zone along the cross section in Figure 2.52 (40-100 km [25-62 mi] from northwestern side of section). 158 Figure 2.54. Pressure-depth plot for data northwest of the Wilcox Fault Zone along the cross section in Figure 2.52 (101-150 km [63-93 mi] from northwestern side of section). Figure 2.55. Pressure-depth plot for data in the Wilcox Fault Zone along the cross section in Figure 2.52 (151-190 km [94-118 mi] from northwestern side of section). Figure2.56. Pressure-depthplotfordatabetweentheWilcoxandVicksburg Fault Zones along the cross section in Figure 2.52 (191-225 km [ll9-140 mi] from northwestern side of section). Figure 2.57. Pressure-depth plot for data between the Wilcox and Vicksburg Fault Zones along the cross section in Figure 2.52 (226-245 km [l4O-152 mi] from northwestern side of section). Figure 2.58. Pressure-depth plot for data in the Vicksburg Fault Zone along the cross section in Figure 2.52 (246-275 km [153-171 mi] from northwestern side of section). Figure 2.59. Pressure-depth plot for data in the Frio Fault Zone along the cross section in Figure 2.52 (276-340 km [172-211 mi] from northwestern side of section). Figure 2.60. Pressure-depth plot for combineddata from the Wilcox andFrioFaultZonesalongthecross sectioninFigure 2.20. 165 Figure 2.61. Pressure-depth plot for the Wilcox Fault Zone in grid area 1007 in Figure 2.20. Figure 2.62. Pressure-depth plot for the Wilcox Fault Zone in grid area Figure 2.20. 167 Figure 2.63. Pressure-depth plot for the Wilcox Fault Zone in grid area 409 in Figure 2.20. Figure 2.64. Pressure-depth plot for the Vicksburg Fault Zone in grid area 514 in Figure 2.20. Figure 2.65. Pressure-depth plot for the Frio Fault Zone in grid area 1012 in Figure 2.20. 170 Figure 2.66. Pressure-depth plot for the Frio Fault Zone in grid area 1311 in Figure 2.20. 171 Figure 2.67. Pressure-depth plot for the Frio Fault Zone in grid 717 in area Figure 2.20. 172 Figure 2.68. Salinity map for the 3-3.5 km (9,840-11,480 ft) depth interval. Figure 2.69. Salinity map for the 2.5-3.0 km (9,840-11,480 ft) depth interval. Figure2.70. Salinitymapforthe2-2.5km(6,560-8,200ft)depthinterval. Figure 2.71. Salinity map for the 1.5-2 km (4,920-6,560 ft) depth interval. Figure 2.72. Salinity map for the 1-1.5 km (3,280-4,920 ft) depth interval. Figure2.73. Salinitymapforthe0.5-1km(1,640-3,280ft)depthinterval. Figure 2.74. Salinity plotted on cross section (see Figure 2.7 forlocation). Figure 2.75. Chloride-depth plot for data northwest of the Wilcox Fault Zone along the cross section in Figure 2.52 (40-100 km [25-62 mi] from northwestern side of section). Figure 2.76. Chloride-depth plot for data northwest of the Wilcox Fault Zone along the cross section in Figure 2.52 (101-150 km [63-93 mi] from northwestern side of section). Figure2.77. Chloride-depthplotfordataintheWilcoxFaultZonealongthe cross section in Figure 2.52 (151-190 km [94-118 mi] from northwestern side of section). 182 Figure2.78. Chloride-depthplotfordatabetweentheWilcoxandVicksburg Fault Zones along the cross section in Figure 2.52 (191-225 km [ll9-140 mi] from northwestern side of section). Figure 2.79. Chloride-depth plot for data between the Wilcox and Vicksburg Fault Zones along the cross section in Figure 2.52 (226-245 km [l4O-152 mi] from northwestern side of section). Figure 2.80. Chloride-depth plot for data in the Vicksburg Fault Zone along the cross section in Figure 2.52 (246-275 km [153-171 mi] from northwestern side of section). Figure 2.81. Chloride-depth plot for data in the Frio Fault Zone along the cross section in Figure 2.52 (276-340 km [172-211 mi] from northwestern side of section). Figure2.82.Locationsofsilicaknobs(Freeman, 1968)anduraniummineswith an epigenetic origin. Figure3.1. Locationmapforsamples withthermalconductivity measurements. Wilcox-sample wells:B=GettyBums#1(AmericanPetroleumInstitute#: 4212330375; latitude:29.14N; longitude: 97.27 W); G= ShellGarza #1 (4250531261; 27.02 N; 99.98 W); and M= Shell Muzza #2 (4250531386; 27.04N;98.98W).Frio-sample well:C=CitiesServices StateTract49#2 (4235506348; 27.77 N; 97.30 W). Figure 3.2. QFR ternary diagram depicting the amounts of quartz (Q), feldspar (F), and rock fragments (R) in Wilcox and Frio sandstone samples as percent of Q+F+R using the Folk (1968) sandstone classification (solid lines). Shaded area represents domainofequation 3.6. Figure 3.3. Thermal conductivity versus porosity for Wilcox (Wx) and Frio (Fr) sandstones Descriptions ofsamples (referenced by ID) are given in Table 3.2. The lines depict regression equation 3.2 with approximate standard error (Davis, 1987,p. 204) for the clean, quartzose samples (ID# 1-68). cleanWilcoxandFrio sandstonescontouredin Shown for reference are dashed contours Figure 3.4. Thermalconductivity (W/m/K)of quartz-porosity space (solid contours). representingthermalconductivity calculatedusingageometricmixingmodelofquartz, “other”minerals,andwaterwithconductivitiesequal to7.8,2.4,and0.6W/m/K, respectively. Figure3.5. Predictedversusmeasuredthermalconductivityforcleansandstones,a.)linear regression model(equation 3.2), b.) multi-component geometric mixing model (equation 3.5), c.) a two-componentgeometric mixing model(equation 3.5). The solidline represents 1:1 relationship and dashed lines represent 10% from 1:1. Figure 3.6. Predicted versus measured thermal conductivity for clean sandstones; a.)arithmeticmixingmodelforgrain-matrixconductivity (equation3.3)anda 2-component geometric mixingmodelforeffectivethermalconductivity (equation 3.5), b)multiplelinearregressionmodelwithquartzandporosity asthe independent variables(equation 3.6). The solid line represents 1:1 relationship and 10% from 1:1. dashed lines represent Open squares represent data from the B-well (Figure 3.1). Figure3.7. PredictedversusmeasuredthermalconductivityfordatafromWoodsideand Messmer(1961b),BrigaudandVasseur(1989),andSomerton(1992): a.)linear regression model using porosity as the independent variable (equation 3.2), b.) multiple linearregression modelwithquartzandporosity astheindependent variables(equation 3.6). Solidlinerepresents1:1relationshipanddashedlinesrepresent 10%from1:1. Open symbols on b represent samples having solids greater than 90% quartz. Figure3.8. Domain(shaded)ofmultiple-linear regressionmodel(equation3.6)withcalculated thermalconductivity(W/m/K)contouredinporosity-quartzspace. Datapointsarefromthis paper andWoodsideandMessmer(1961b),Brigaud andVasseur(1989), andSomerton (1992). Figure 4.1. Locationsofsamples withradiogenic heatmeasurements and stratigraphic section. See Table 4.2 for posted sample names (eg. LY, LE)anddetailsofsamplelocations. Sample namesubscriptsindicate sample lithology: ss = sandstone, mr = mudrock, Is = limestone. Figure 4.2. Graphical summary of radiogenic-heat production data. Arithmetic mean and standarddeviation(sd)areindicatedexceptforStuartCitylimestones(shown as individualsamples). Figure 4.3. Radiogenic-heat production versus depth Figure4.4. ModelparameterswithdepthcalculatedfromequationsinTable 4.5. a) lithology (stipple pattern = sandstone, no pattern = mudrock, and block pattern = limestone, b) porosity, c) thermalconductivity as a function oflithologyandporosity, d)thermalconductivity as a functionoftemperature, and e) radiogenic heat production. Figure 4.5. Modelresults, a) Results for true stratigraphic section. Squares are temperature data. Model 1 includes heak production; model2 neglects heat production. Basalheat-flowdensity isoptimizedtofitobservationsinmodels 1and2.Model3neglectsheatproduction andbasalheat-flowdensityisnot optimizedtofitobservations, b) Resultsformodelsofidealizedsectionsofall sandstone (4 and 6) and all mudrock (5 and 7). Models 4 and 5 include heat production; models6and7neglectheatproduction. Table2.1. Temperature picked fromtemperature-dpeth plots. Valuespicked from temperature-depth plots for grid areas on Figure 2.20. Temperature °C at depth grid- code 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 km km km km km km km km km km km km 110 70 80 105 130 111 70 90 106 130 155 112 44 80 95 113 155 190 207 50 62 74 95 117 140 208 45 60 72 88 114 130 209 55 74 85 108 130 157 210 43 55 71 85 102 125 150 150 168 211 75 90 110 135 158 212 115 140 213 72 80 109 214 74 90 215 65 80 101 120 307 90 108 128 149 308 85 100 130 153 309 90 112 140 167 310 45 55 60 90 117 150 160 170 190 215 235 311 45 70 90 313 63 85 109 130 314 45 67 85 105 125 145 165 185 185 315 90 103 130 317 80 105 125 404 90 405 72 85 100 112 125 140 165 195 406 85 100 115 132 150 407 105 120 135 150 408 66 88 108 130 409 48 82 108 130 160 185 410 50 75 100 125 135 151 180 Temperature °C at depth grid- code 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 km km km km km km km km km km km km 411 71 90 413 72 85 103 125 145 414 69 85 100 125 145 160 172 190 415 60 70 80 97 120 140 160 416 40 70 85 95 115 150 417 96 115 504 90 109 505 90 101 506 61 82 105 128 172 507 90 106 145 508 112 135 157 509 90 120 150 511 85 104 512 40 85 97 513 85 101 120 140 160 180 514 42 55 63 80 100 120 138 160 177 195 515 49 58 67 75 95 115 144 516 62 80 90 110 135 150 163 175 517 55 65 78 90 105 125 145 168 604 72 105 120 605 58 72 85 100 115 607 46 60 72 85 105 125 148 608 40 60 75 95 115 610 130 155 175 611 69 88 103 612 70 82 95 120 152 155 613 97 120 140 614 72 95 110 130 150 615 55 66 75 89 no 135 155 616 58 68 78 98 no 132 618 105 128 145 153 170 704 35 52 67 80 97 120 706 60 73 90 105 190 Temperature °c at depth grid- code 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 km km km km km km km km km km km km 707 60 73 90 102 115 708 100 117 140 170 200 709 85 105 125 148 170 710 60 78 92 109 711 105 125 712 80 98 115 133 152 713 53 65 77 90 100 115 135 160 714 60 70 82 93 110 132 150 165 185 718 100 804 80 95 110 805 42 55 71 85 100 115 129 806 79 90 103 127 140 150 160 807 38 55 72 90 110 135 153 190 195 808 55 71 85 107 135 167 195 205 215 809 90 112 810 85 120 140 811 69 85 100 115 812 73 103 120 132 145 151 813 50 99 110 125 140 159 814 69 82 97 110 122 140 161 180 815 65 78 90 100 112 125 135 145 816 105 120 818 48 61 75 160 905 58 70 85 145 155 169 180 194 906 73 90 103 907 37 72 90 107 135 165 185 908 60 69 109 140 909 55 69 85 111 185 202 220 911 100 120 912 68 92 105 913 47 60 70 82 95 110 130 145 160 170 914 117 145 915 68 80 95 108 120 203 Temperature °C at depth grid- code 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 km km km km km km km km km km km km 1004 38 55 68 80 100 115 1005 45 60 72 85 103 115 130 1006 67 85 105 122 142 178 1007 42 55 70 80 105 135 150 175 1008 45 70 90 103 120 145 175 1009 70 115 1010 80 103 125 145 1011 80 98 140 1012 45 60 70 80 93 105 128 160 1013 90 108 125 143 160 175 1014 95 105 115 125 140 1103 55 60 95 110 130 145 1104 110 130 160 1105 90 1106 55 70 85 100 130 145 1107 80 105 130 165 170 1108 40 58 75 90 105 130 150 170 195 1110 63 80 95 1111 65 80 95 105 1112 80 95 110 125 142 160 1113 90 110 130 1203 55 67 80 93 1204 75 100 1205 80 95 112 130 1206 58 80 95 115 175 155 200 1207 40 60 80 75 100 130 155 1208 80 125 155 180 1210 75 95 110 130 1211 45 55 68 80 90 105 125 150 170 1212 95 112 130 1213 110 1304 85 1305 68 82 100 112 130 204 Temperature °C at depth grid- code 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 km km km km km km km km km km km km 1306 55 80 110 135 155 170 185 1307 70 88 105 130 159 1308 138 162 1310 78 93 1311 55 66 80 90 110 130 1312 75 90 115 132 150 1404 120 128 1406 60 75 88 95 120 146 172 195 210 222 1407 40 55 69 85 105 130 1409 155 163 200 1411 65 80 93 110 132 1412 91 110 129 150 1506 60 78 95 120 148 165 172 1507 41 55 70 85 100 122 147 170 1508 130 1510 43 65 90 95 1511 80 95 110 125 1606 42 58 72 90 102 120 151 160 163 185 208 1607 102 122 143 1611 110 Table 2.2. Fluid surfaces and fluid/overburdenratios pressures, tops of overpressure pressure picked from pressure-depth plots. Fluid pressure picked at 0.5 km intervals, depth to top of overpressure, overpressure, and extreme overpressure, “hard” and the fluid/overburdenpressure ratio in the extremely overpressured regime. Values picked from pressure-depth plots for grid areas on Figure 2.20. Column headers are: top op = depthtotopofoverpressure; tophardop=depthto topof hardoverpressure;topextop= depthtotopofextremeoverpressure; X,inext op=fluid/overburdenpressureratiointheextremely overpressured regime. grid- top top top X — Pressure code op hard ext MPa op op in at depth depth depth depth ext 0.5 1 1.5 2.0 2.5 3.0 3.5 4.0 4.5 (km) (km) (km) op km km km km km km km km km 111 <2.8 112 1.4 1.8 19 35 113 <1.5 1.6 22 207 1.5 >2.5 17 30 38 208 <2.5 <2.5 38 209 1.1-2.0 2.3 42 210 1.5 2.3 18 30 40 211 1.5-2.0 >2.7 6 12 17 35 213 0.8 1.4 5 12223444 214 <1.2 1.5 215 <1.6 2.3 28 41 304 1.3 >1.7 17 305 >2.2 >2.2 24 306 <1.7 >2.3 28 307 <1.8 2.2 29 41 308 <2.5 2.5 >3.6 39 52 66 309 1.0 1.3-2.4 >3.0 13 42 310 0.8-2.3 0.8-2.3 3.1 0.85 44 56 311 1.0-2.2 1.0-2.2 11 312 >1.3 >1.3 11 313 0.5 2.5 >2.9 6 12 17 29 40 314 1.5 1.9 2.9 0.85 18 33 46 56 68 80 315 <1.9 2.5 27 38 316 <2.1 2.7 3.0 0.85 34 56 71 79 317 <1.7 2.6 402 >1.7 >1.7 16 403 >2.2 >2.2 22 404 1.7 >2.3 11 24 405 <1.7 >2.1 26 206 — grid-top top top X Pressure code hard ext MPa op op op in at depth depth depth depth ext 0.5 1 1.5 2.0 2.5 3.0 3.5 4.0 4.5 kmkm km kmkmkmkm kmkm (km) (km) (km) op 407 <2.4 <2.4 42 54 408 <2.8 <2.8 54 409 <1.9 3.1 3.2 0.85 42 50 68 411 1.8 >2.1 30 413 1.9 2.4 2.6 0.85 18 24 42 414 1.4 2.4 3.1 0.90 18 26 42 60 72 84 415 <1.9 2.7 3.1 0.90 416 2.4 2.6-3.0 3.3 0.85 22 32 66 417 2.3 >2.9 504 2.5 >2.6 22 27 506 1.9 >2.7 1626 40 507 2.2-2.5 2.6 3.1 0.85 38 56 67 508 <3 <3.0 51 68 509 <3.1 <3.1 510 <2.9 <2.9 511 2.1-2.3 >2.3 22 512 1.7 >2.1 11 17 26 513 1.7 2.7 2.6 0.90 27 40 514 2.3-2.6 3.5 0.90 16 22 46 72 82 90 515 2.7 2.8 28 55 516 2.8-3.2 3.2-3.6 517 >2.7 >2.7 28 604 <2.2 >3.2 606 <2.1 >2.7 36 607 1.9-2.3 2.6 11 23 38 48 608 2.0 >3.3 11 17 23 36 43 610 <2.9 <2.9 3.0 0.90 58 611 1.9-2.2 >2.2 11 18 612 2.3 >2.5 24 613 2.4 2.5 22 38 614 2.4 >3.1 22 34 44 615 2.3 2.9-3.5 3.4 0.85 32 66 616 2.9 3.2 3.4 0.85 24 29 38 617 2.9 >3.2 >3.2 11 16 22 38 619 >2.9 >2.9 28 706 0.7-2.2 707 2.3-2.9 2.9 22 50 708 11 709 <2.6 >2.6 710 1.5-2.2 2.4-3.7 2.5-3.7 0.85 11 711 1.8 >2.5 18 28 39 grid- top top top X — Pressure code op hard ext MPa op op in at depth depth depth depth ext 0.5 1 1.5 2.0 2.5 3.0 3.5 4.0 4.5 (km) (km) (km) op km km km km km km km km km 712 <1.9 2.7 26 52 713 >3.0 >3.0 11 17 20 29 34 714 2.7 3.2-3.7 3.9 0.85 40 716 >2.7 >2.7 18 22 28 717 >3.0 >3.0 37 718 >2.3 >2.3 22 719 >2.4 >2.4 801 >2.0 >2.0 22 804 >1.7 >1.7 16 805 >1.9 >1.9 16 807 2.0 2.8 2.9 0.85 6 11 16 22 34 56 66 76 85 808 1.7-2.1 >3.0 11 16 809 1.8 2.0-3.0 3.4 0.85 54 69 85 810 2.0 2.2 2.3 0.85 11 17 24 46 50 64 76 811 1.8 2.3 2.9 0.80 17 28 44 54 62 812 <2.2 2.7 2.8 0.85 38 57 70 813 3.0 3.4 3.6-4.2 0.90 22 27 40 60 814 >3.0 >3.0 28 815 >3.0 >3.0 22 32 816 >2.8 >2.8 22 28 903 >2.5 >2.5 22 905 >3.3 >3.3 33 907 1.7 2.5 11 16 32 40 54 908 2.9 3.5 0.80 29 52 64 80 909 1.4-2.0 2.1 11 30 910 2.0 2.2 11 16 24 911 1.8 2.4 2.5 0.85 26 48 58 912 >2.5 >2.5 22 913 2.7-3.7 2.7-3.7 2.7-3.7 0.90 6 11 16 22 27 84 1005 1.9-2.9 >3.0 16 1007 1.9 2.3 3.5 0.85 6 29 41 52 64 1008 1.5 >2.0 18 29 1009 1.8 2.1 11 31 1010 2.0 2.4 >3.1 11 22 42 54 1011 2.5 >2.6 16 23 32 1012 2.7 2.7 3.1 0.85 16 22 28 56 1013 >2.8 >2.8 >2.8 11 16 28 1106 2.5 >2.6 22 32 1107 2.4 2.6 2.9 0.80 22 32 54 62 1108 1.7 >2.0 12 17 28 1109 1.8 >1.9 16 grid- top top top X — Pressure code op hard ext MPa op op in at depth depth depth depth ext 0.5 1 1.5 2.0 2.5 3.0 3.5 4.0 4.5 (km) (km) (km) op km km km km km km km km km 1110 2.0 2.4 >2.9 23 38 1111 2.4 2.9 3.2 0.80 11 21 34 46 64 1112 <2.6 2.8 3.2 0.85 54 68 79 1113 2.9 2.9 >3.0 27 1203 1204 1205 1206 >4.3 >4.3 >4.3 11 22 1207 <2.4 3.1 3.6 0.85 34 44 1208 1.4 >1.9 11 15 1209 >1.8 >1.8 10 17 1210 2.2 >3.0 32 38 1211 2.8 3.0 >4.0 16 22 44 64 77 1212 1.9-2.6 2.8 3.2 0.85 52 65 1213 <2.5 2.7 3.0 0.85 58 1306 >4.2 >4.2 >4.2 1307 1.6-2.3 2.7 >3.5 11 16 36 50 60 1308 >1.3 >1.3 11 1309 >1.7 >1.7 17 1311 2.4 3.0 3.3 0.85 22 30 44 64 1312 <2.3 2.7 2.8 0.85 30 56 62 75 1409 >1.6 >1.6 1412 2.7 2.9 >3.1 26 as in number; quartz clasts burrows burrows burrows burrows burrows burrows burrows burrows burrows size perpendicular porosity (qtz+cmt); clay minor minor minor minor minor minor minor minor minor (<|)J; grain sample minor description massive massive, massive massive, massive massive, massive, massive, massive, massive, massive, massive massive, massive massive massive W/m/K; section unique meaning cement (mic); (unid); grain size phi 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 in thin qtz + = (>,) in 1 %s 0.6 0.0 1.1 1.1 0.6 1.1 1.6 0.6 0.0 0.6 0.5 0.6 1.2 3.0 1.2 0.5 ID with plus mica minerals unid Ti %s 0.0 0.5 0.0 0.0 0.6 0.0 0.5 0.0 0.0 1.1 0.0 0.0 0.0 0.0 0.6 0.0 units; measured grains (ana): conductivity fsp Og %s 0.0 0.0 0.6 0.5 0.0 0.0 0.0 0.6 0.0 0.6 0.0 0.6 0.6 0.0 0.0 0.0 measurement unidentified and as quartz pyr %s 0.0 0.0 0.0 0.5 0.0 0.6 0.5 1.1 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.0 analcime thermal porosity of sid/ ank %s 1.7 0.5 0.6 2.2 l.l 1.1 0.0 l.l 0.6 1.7 2.2 0.0 0.6 1.8 1.2 0.0 cmt); (Ti); abbreviations cmt); clay cla %s 2.2 0.6 0.0 0.0 0.6 0.5 0.6 2.9 0.0 0.0 0.6 1.2 0.0 0.0 1.1(%b); (qtz Tio2 1.1 orientation effective (cal mic %s 2.3 0.0 1.7 3.8 2.2 1.7 3.8 2.2 0.0 0.0 1.6 1.8 1.2 1.2 1.8 0.0 rock cement 52) ana %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Column (pyr); n cement bulk = meters; measured structures. qtz cal cmt %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 data. in pyrite solids; clay %s 22.0 15.9 144 13.5 11.7 13.4 16.5 15.0 16.2 11.1 13.7 12.3 16.9 15.5 13.9 11.8 of (%s); calcite depth percent rf 20% %s 6.8 5.5 10.0 3.8 9.5 8.4 9.9 9.4 10.4 6.7 8.8 7.6 6.4 11.9 12.0 10.8 < in petrographic bedding; component clay; (sid/ank); sedimentary of to) 3.1); fsp %s calcite 11.3 7.7 7.2 7.6 6.1 6.7 8.2 7.2 9.2 9.4 8.8 8.2 7.0 9.5 11.4 9.1 sv ( (rf); the cement and of qtz cmt %s 25%, 54.2 67.6 63.9 67.0 68.2 66.5 58.2 62.2 60.7 68.9 64.3 68.4 65.1 57.1 57.2 66.7 Figure parallel vessel solid + in of qtz cmt %s clay 3.4 12.6 7.2 11.9 8.9 7.3 2.2 3.3 8.7 7.2 9.9 6.4 4.7 0.0 1.8 14.5 fragments porosity, meaning description < qtz %s 50.8 54.9 56.7 55.1 59.2 59.2 56.0 58.9 52.0 61.7 54.4 62.0 60.5 57.1 55.4 52.2 (shown saturation percent rock siderite/ankerite 35%, a ID and in and .(ts) %b (quartz 11.5 9.0 10.0 7.5 10.5 10.5 9.0 10.0 13.5 10.0 9.0 14.5 14.0 16.0 17.0 7.0 a (fsp); conductivity, II > Well in (qtz) cla); (sv) %b 19.2 16.1 13.8 15.5 17.7 17.1 16.3 16.1 16.9 16.1 16.0 18.1 18.2 20.4 17.7 12.4 = bedding units; 2 X ID to measured grains feldspar (clay phi W/ m/K SANDSTONES 3.61 4.43 4.06 4.27 3.98 4.01 3.83 3.87 3.95 3.90 3.86 3.63 3.74 3.24 3.94 4.50 Thermal + or1 +++++ +++ ++++++++ 3.1. depth (m) 4613.1 4613.8 4615.0 4615.9 4616.2 4617. 4618.3 4619.5 4620.2 4621.7 4622.6 4623.8 4624.7 4626.6 4629.6 4755.5 WILCOX 1 2 MMMMMMMMMMMMMMMM Table 5 1 1234 6789 10 11 12 13 14 15 16 ID Clean burrows burrows burrows burrows minor minor minor minor laminated laminated laminated description massive massive massive, massive, massive massive massive, massive, vaguely vaguely massive massive massive massive massive massive massive massive massive massive massive massive massive massive massive massive massive vaguely grain size phi 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 unid %s I.l 2.0 0.0 0.0 0.6 0.5 0.5 1.5 0.0 0.5 0.6 1.2 0.6 I.l 0.5 0.0 0.0 0.6 0.6 0.5 0.5 0.5 0.0 1.0 1.0 0.0 0.0 1.5 Ti %s 0.5 0.0 0.5 0.0 1.8 0.5 0.0 0.5 0.5 1.0 1.1 0.6 0.6 0.6 0.0 0.5 1.0 2.2 I.l 0.5 0.0 0.5 0.5 0.5 0.5 0.5 0.5 0.5 fsp Og %s 0.5 1.0 0.0 0.0 0.0 0.5 0.0 0.5 1.5 1.5 0,0 0.0 0.0 0.6 1.0 0.5 1.0 0.0 0.0 0.5 1.0 1.5 1.0 1.0 0.0 1.0 0.0 0.0 pyr %s 1.6 0.0 1.0 0.0 0.6 0.0 2.2 0.0 1.0 1.0 1.1 0.6 0.6 1.1 0.0 0.0 0.5 0.6 0.0 0.5 0.5 0.5 0.5 0.0 1.0 0.0 0.0 2.0 sid/ ank %s 1.1 0.5 1.0 0.0 0.0 19.0 0.5 0.0 7.1 4.5 3.4 0.6 0.6 1.1 1.0 0.5 3.5 0.6 1.7 1.6 2.1 2.0 3.6 0.5 1.6 0.0 0.0 1.5 clay cla %s 1.1 0.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 0.0 0.0 1.1 1.1 0.0 0.0 0.5 1.7 1.1 0.0 0.0 0.5 0.0 0.5 0.5 0.0 0,5 1.5 mic %s 1.1 0.5 0.0 1.0 0.6 0.0 0.5 2.0 1.0 1.0 2.2 0.6 1.1 1.1 0.0 0.5 1.5 1.1 1.7 1.0 1.6 0.5 2.0 1.0 1.6 0.0 0.0 2.5 ana %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 cal cmt %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 clay %s 7.1 5.6 9.5 7.0 12.9 5.3 16.2 7.1 2.0 3.0 14.0 ill 15.9 13.6 8.7 16.5 4.0 8.4 9.7 10.5 8.3 5.1 9.1 11.0 15.0 7.1 5.0 14.0 rf %s 14.2 13.3 15.0 16.6 14.7 12.7 9.7 12.8 11.7 11.1 12.3 12.3 10.8 15.8 16.3 16.0 11.5 12.3 13.7 9.4 10.4 14.8 14.2 11.5 18.7 16.8 13.5 12.5 fsp %s 7.7 9.2 7.0 6.0 7.1 7.4 5.9 8.7 6.1 6.6 6.7 7.6 6.8 5.6 6.6 4.0 6.5 6.1 8.0 9.4 13.5 9.2 11.7 11.0 7.3 9.7 7.5 10.5 qtz + cmt %s 63.9 67.9 65.0 69.3 61.8 54.0 64.3 66.8 68.9 69.2 58.7 65.5 61.9 58.2 65.8 61.5 70.0 66.5 62.3 66.0 62.0 64.8 57.4 61.8 52.8 64.8 73.0 53.5 qtz cmt %s 12.0 21.9 18.5 25.1 3.5 7.9 9.2 20.9 20.9 20.7 4.5 7.6 1.7 1.7 19.4 11.0 22.5 15.1 9.7 16.2 14.1 17.9 13.2 13.1 5.7 20.4 22.0 5.5 qtz %s 51.9 45.9 46.5 44.2 58.2 46.0 55.1 45.9 48.0 48.5 54.2 57.9 60.2 56.5 46.4 50.5 47.5 51.4 52.6 49.7 47.9 46.9 44.2 48.7 47.2 44.4 51.0 48.0 (ts) %b 8.5 2.0 0.0 0.5 15.0 5.5 7.5 2.0 2.0 1.0 10.5 14.5 12.0 11.5 2.0 0.0 0.0 10.5 12.5 4.5 4.0 2.0 1.5 4.5 3.5 2.0 0.0 0.0 (sv) %b 11.3 6.8 8.2 5.7 16.7 11.5 12.3 9.1 3.9 3.9 15.1 15.6 19.0 17.8 8.2 4.1 2.4 12.5 13.3 9.3 9.7 7.1 6.9 11.9 11.4 8.9 5.6 6.3 X w/ m/K 4.88 5.11 4.72 4.72 3.36 4.12 4.65 5.13 5.71 5.29 4.48 4.34 3.59 3.81 5.39 5.22 5.73 4.80 4.70 5.10 4.91 4.95 4.94 4.58 4.51 5.25 5.03 4.73 + or 1 + + + + + + + + + + + + + + + + + + + + + + + + + + + + depth (m) 4756.1 4757.0 4758.2 4758.5 4761.6 4762.5 4763.7 4769.2 4902.4 4903.0 4910.9 4911.5 4913.4 4915.2 4918.9 4919.8 4920.7 4922.5 4924.3 4926.8 4928.3 4931.4 4933.8 4940.5 4941.7 4942.9 4953.0 4949.6 ID 2 M M M M M M M M G G G G G G G G G G G G G G G G G G G G 1 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 burrows burrows description vaguely laminated vaguely laminated vaguely laminated vaguely laminated vaguely laminated vaguely laminated vaguely laminated vaguely laminated laminated massive well laminated well laminated vaguely laminated laminated, minor laminated, minor laminated laminated massive laminated laminated vaguely laminated laminated laminated massive massive vaguely laminated grain size phi 2.8 3.2 3.1 3.1 2.3 3.1 3.1 1.5 3.1 3.0 3.1 3.1 3.3 2.8 2.8 3.0 3.1 3.2 3.4 3.2 3.1 3.0 3.1 3.1 2.2 2.6 unid %s 0.0 0.0 3.9 3.9 0.0 1.7 1.7 0.9 1.4 1.4 0.0 0.0 0.0 1.0 1.0 0.0 1.8 0.7 0.7 1.1 0.5 0.0 0.6 0.5 0.0 0.0 Ti %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 fsp Og %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.4 0.7 0.0 0.0 0.0 0.6 0.6 0.0 0.0 0.0 0.0 0.0 0.6 0.0 0.6 0.0 1.2 0.0 pyr %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 sid/ ank %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 mic clay cla %s %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 = 16) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ID depth + X . * qtz qtz qtz + fsp rf clay cal ana I 2 (m) or w/ (sv) (ts) cmt cmt cmt i m/K %b %b %s %s %s %s %s %s %s %s 45* B 2845.9 + 4.18 14.5 6.8 63.5 12.6 76.1 4.2 16.6 0.0 3.1 0.0 46 B 2907.5 1 4.02 13.2 10.0 77.8 5.6 83.3 4.2 9.7 2.8 0.0 0.0 47* B 2913.6 + 4.12 12.2 14.7 65.7 7.0 72.7 5.0 17.2 1.2 0.0 0.0 48 B 2913.6 1 4.60 14.1 14.7 65.7 7.0 72.7 5.0 17.2 1.2 0.0 0.0 49 B 2918.2 1 3.77 12.0 14.2 63,6 9.1 72.7 4.5 22.7 0.0 0.0 0.0 50* B 2929.1 + 3.94 12.3 9.9 67.3 1.7 68.9 12.7 13.4 3.3 0.0 0.0 51 B 2929.1 1 4.31 11.8 9.9 67.3 1.7 68.9 12.7 13.4 3.3 0.0 0.0 52 B 2932.8 1 4.37 15.4 9.8 69.3 5.0 74.3 14.9 8.0 2.0 0.0 0.0 Clean, quartzose FRIO SANDSTONES (quartz > 35%, clay < 25%, calcite < 20% of solids; n 53 C 2609.1 + 2.56 26.1 26.0 56.8 2.0 58.8 13.5 17.6 3.4 4.1 0.0 54 C 2609.4 + 2.57 25.5 29.5 57.4 3.5 61.0 9.2 22.7 2.8 2.1 0.0 55 C 2661.5 + 3.12 21.5 25.5 51.0 2.7 53.7 7.4 18.8 16.1 3.4 0.7 56* C 2661.5 1 2.68 24.3 25.5 51.0 2.7 53.7 7.4 18.8 16.1 3.4 0.7 57 C 2663.3 + 2.59 26.0 22.5 43.2 3.9 47.1 11.0 22.6 11.6 1.3 6.5 58 C 2736.2 + 2.59 21.9 17.5 46.1 10.9 57.0 13.3 18.2 9.7 0.1 0.1 59 C 2736.2 1 2.79 22.5 17.5 46.1 10.9 57.0 13.3 18.2 9.7 0.1 0.1 60 C 2742.3 1 2.97 25.9 17.5 50.9 1.8 52.7 8.5 14.5 23.6 0.6 0.0 61 C 2742.9 + 3.14 24.3 16.0 39.3 1.2 40.5 11.9 18.5 10.7 16.7 0.0 62 C 2745.0 + 2.78 27.7 29.5 48.2 12.1 60.3 10.6 18.4 9.9 0.0 0.0 63 C 2747.8 + 2.59 29.6 24.5 50.3 7.9 58.3 15.2 15.9 9.9 0.0 0.0 64 C 2777.9 + 2.74 25.1 25.0 53.3 5.3 58.7 15.3 14.0 10.7 0.3 0.0 65 C 2779.8 + 2.78 25.6 15.1 47.1 11.2 58.3 16.5 19.4 4.1 0.6 0.0 66 C 2783.4 + 2.90 26.2 20.5 52.2 6.9 59.1 11.9 19.5 9.4 0.0 0.0 67 C 2787.1 + 2.58 28.0 14.0 44.8 14.0 58.7 12.8 15.1 10.5 0.0 1.7 68 C 2787.4 + 2.56 23.4 15.0 48.2 2.9 51.2 15.9 18.8 13.5 0.0 0.1 Clean FRIO SANDSTONES (quartz < 35%, clay < 25%, calcite < 20% of solids; n = 9) 69 C 2449.1 + 2.82 20.1 16.4 29.9 3.0 32.9 11,4 32.3 9.7 12.6 0.0 70 C 2450.6 + 2.40 19.1 18.5 28.8 0.6 29.4 13.5 38.0 3.7 15.3 0.0 212 description well laminated well laminated well laminated massive laminated massive massive burrowed burrowed burrowed burrowed laminated massive, burrowed grain size phi 2.2 2.2 2.0 1.8 2.0 2.3 2.3 3.4 3.4 3.1 3.0 3.0 3.3 unid %s 0.5 0.5 1.3 0.0 2.9 II 1.1 0.9 0.9 0.0 1.7 0.0 1.6 Ti %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 fsp Og %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.5 0.0 pyr %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 sid/ ank %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1) 0.0 1.1 = n mic clay cla %s %s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 = 4) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 solids; 0.0 0.0 1.6 1.1 n of ) rf clay cal ana cmt %s %s %s %s 60.9 6.9 7.5 3.4 60.9 6.9 7.5 3.4 48.4 5.2 5.2 0.6 44.4 1.8 14.6 7.6 38.0 2.4 7.8 5.4 37.2 9.3 6.6 0.0 37.2 9.3 6.6 0.0 calcite < 20% of solids; 13.8 36.2 0.5 0.1 13.8 36.2 0.5 0.1 11.9 26.1 0.2 0.0 9.8 25.2 0.1 0.0 < 25%, calcite > 20% 15.8 3.1 32.2 3.5 20% of solids; n = 1 5.4 24.5 0.0 0.0 fsp %s 3.4 3.4 13.5 10.5 11.4 14.2 14.2 > 25%, 10.2 10.2 10. 1 10.4 35%, clay 6.6 calcite < 9.8 qtz qtz + cmt cmt %s %s 0.1 17.4 0.1 17.4 1.9 25.8 0.6 21.1 0.1 32.0 0.5 31.7 0.5 31.7 35%, clay 3.1 38.3 3.1 38.3 3.6 51.7 1.8 52.1 (quartz > 1.0 38.3 clay > 25%, 3.8 54.3 > X . . qtz w/ (sv) (ts) m/K %b %b %s 2.06 19.9 13.0 17.2 2,09 20.7 13.0 17.2 2.11 20.1 22.5 23.9 2.18 21.5 14.5 20.5 2.50 23.0 17.0 31.9 2.42 19.6 8.5 31.1 2.42 18.8 8.5 31.1 FRIO SANDSTONES ( quartz 2.60 17.9 2.0 35.2 2.82 17.9 2.0 35.2 3.09 15.7 15.8 48.1 2.88 21.6 18.5 50.3 quartzose FRIO SANDSTONE 3.01 2.6 2.0 37.3 SANDSTONE ( quartz > 35%, 3.91 15.3 8.0 50.5 + or 1 + 1 + + + + 1 + 1 + 1 , + + ID depth 1 2 (m) 71 C 2451.5 72 C 2451.5 73 C 2452.7 74 C 2453.3 75 C 2453.9 76 C 2455.5 77 C 2455.5 Muddy, quartzose 78 C 2737.4 79 C 2737.4 80 C 2767.3 81 C 2777.3 Calcite-rich, clean 82 C 2781.6 Muddy WILCOX 83 M 4613.5 Table 3.2. Thermal conductivity measured parallel and perpendicular to stratigraphic bedding. ID parallel perpendicular (W/m/K) (W/m/K) (W/m/K) 47/48 4.60 4.12 1.12 50/51 4.31 3.94 1.09 55/56 2.68 3.12 0.86 58/59 2.79 2.59 1.08 71/72 2.09 2.06 1.01 76/77 2.42 2.42 1.00 78/79 2.82 2.60 1.08 Table 3.3. Mineral thermal conductivities (Horai ,1971; Brigaud et al., 1990). Mineral/ X rock (W/m/K) 7.8 quartz feldspar 2.3 rhyolite 2.5 clay 1.8 calcite 3.4 analcime 1.3 mica 2.3 siderite/ 3.0 ankerite pyrite 19.2 rutile (Ti02) 5.1 unidentified 3.0 Table 4.1. Abundances of naturally occurring radioactive elements in non-ore minerals identified in the Gulf of Mexico Basin. Average concentrations are from van Schmus (1989) and K. Milliken (1994, personal communication). Mineral K U Th ModeofOccurrence wt% wt% wt% Adularia 14.0 local cement detrital - Apatite * * Biotite 8-9 detrital Glauconite 4.6-6.2 detrital detrital Hornblende *** - - Illite/Smectite 0.5-7 detrital/ authigenic 14.0 detrital Microcline - Monazite ** 2-20 detrital Muscovite 9.8 detrital Orthoclase 14.0 detrital Sanidine 14.0 detrital - Sphene * * detrital / authigenic Sylvite 52.4 evaporitic Zircon detrital ** ** - ***0.5 3wt% ** 0.1-0.5 wt % * 0.001-0.1 wt % Table 4.2. Details on sample and stratigraphic section locations. ID Operator Well API# Latitude Longitude Wilcox BU Getty Bums #1 4212330375 29.14 N 97.27 W KO Seaboard Oil Kolodziejcyk #1 4225500897 28.91 N 98.02 W LY Shell Leyendecker #1 4250531222 27.03 N 98.98 W LE Shell H. W. Lehman # 4242704875 26.38 N 98.85 W 1 MU Shell G. P. Muzza #2 4250531386 27.04 N 98.98 W Frio CS Cities State Tract 49 #2 4235506348; 27.77 N 97.30 W Services MI Atlantic Mustang Island 4235506592 27.74 N 97.15 W Richfield Deep Unit #1 StuartCity SC Tenneco Schulz #1 4229700141 28.54 N 98.15 W Stratigraphic Section Shell P. Canales #117 4224903809 27.38 N 98.08 W Table4.3.Radiogenic heatproductiondataforsedimentaryrocksfromsouthTexas. Column headings are: well ID = well identificationas in Figure 4.1; ID # = unique sample identificationnumber; depth in meters; U = uranium concentrationinparts permillionby weight (ppm);Th=thorium concentrationinppm; Uerror=errorassociatedwithuranium concentration(ppm);Therror= errorassociatedwiththorium concentration(ppm);a= measureda-countrate incountsperkilosecond 2 error =error associatedwith per square centimeter (cts/ks/cm ); a 2 determinationofmeasureda-countrate(cts/ks/cm );K=potassium = concentrationinweightpercent(wt%);p bulkdensityinkilogramsper 3 cubic meter (kg/m ; ID #s 1-70 estimatedfrom Dickinson [1953]); A = heat-production rate in microWatts per cubic meter (pW/ m 3); error = error associated with heat production rate (pW/ m 3). ID depthUThUTh ot aK P A A well # error error error error m cts/ks/ cts/ks/ wt % kg/m* pW/in pW/m' ppm ppm ppm ppm 22 cm cm Wilcox mudrocks LY 1 4356.5 3.16 11.89 0.88 3.02 0.823 0.019 2.04 2450 1.64 0.18 LY 2 4373.0 2.35 6.50 0.59 2.04 0.528 0.015 0.57 2450 0.99 0.11 LY 3 4373.6 1.89 11.89 0.79 2.71 0.665 0.017 0.57 2450 1.24 0.14 LY 4 4381.8 3.58 8.68 1.00 3.42 0.760 0.023 0.57 2450 1.41 0.16 LY 5 4382.7 2.59 9.54 0.50 1.72 0.667 0.011 0.57 2450 1.24 0.14 LY 6 4389.4 1.87 4.43 0.45 1.54 0.394 0.012 1.90 2450 0.86 0.09 MU 7 4598.7 3.18 9.51 0.40 1.40 0.739 0.009 2.43 2452 1.53 0.16 MU 8 4702.5 3.92 6.73 0.71 2.44 0.730 0.017 1.28 2461 1.42 0.16 MU 9 4709.9 3.27 11.27 0.70 2.41 0.813 0.015 2.37 2461 1.66 0.18 MU 10 4715.3 3.12 11.18 0.59 2.04 0.792 0.012 2.38 2462 1.63 0.18 MU 11 4716.5 3.71 9.59 0.83 2.82 0.808 0.018 2.38 2462 1.65 0.18 MU 12 4728.7 3.36 11.57 0.60 2.07 0.835 0.012 2.81 2463 1.74 0.19 PE 13 4705.4 3.29 13.61 0.79 2.71 0.902 0.016 2.11 2461 1.80 0.20 PE 14 4613.1 4.26 11.60 0.95 3.24 0.948 0.019 2.09 2453 1.87 0.20 PE 15 4466.8 4.49 10.37 1.25 4.26 0.932 0.026 2.01 2450 1.84 0.20 PE 16 4162.0 4.00 8.80 1.22 4.16 0.815 0.026 2.12 2441 1.63 0.18 PE 17 3997.5 3.60 10.20 1.12 3.85 0.817 0.025 2.08 2428 1.62 0.18 PE 18 3832.9 3.33 11.44 1.15 3.95 0.828 0.025 2.12 2425 1.64 0.18 PE 19 3659.1 4.01 9.43 1.09 3.73 0.840 0.023 2.13 2425 1.66 0.18 ID depthUThUTha aK P A A error well # error error error 33 m ppm ppm ppm ppm cts/ks/ cts/ks/ wt % kg/m jiW/m |iW/m 2 2 cm cm PE 20 3329.9 3.29 11.37 0.90 3.08 0.819 0.019 2.01 2398 1.60 0.17 PE 21 3165.3 3.30 8.93 0.80 2.75 0.732 0.018 1.84 2385 1.43 0.16 PE 22 3014.5 3.56 9.25 0.83 2.87 0.776 0.018 1.88 2375 1.50 0.16 PE 23 2804.2 3.49 8.34 0.93 3.20 0.735 0.021 1.93 2375 1.43 0.16 PE 24 2648.7 3.93 7.30 0.74 2.52 0.721 0.017 1.86 2367 1.40 0.15 PE 25 2470.4 2.33 12.61 0.94 3.22 0.746 0.020 1.88 2353 1.43 0.16 PE 26 2281.4 3.78 8.39 1.11 3.77 0.772 0.025 2.11 2337 1.49 0.16 PE 27 2107.7 3.79 7.97 1.08 3.71 0.759 0.025 1.88 2323 1.44 0.16 PE 28 1946.1 3.78 10.07 0.87 2.98 0.834 0.019 1.88 2310 1.56 0.17 PE 29 1790.7 3.00 11.13 0.69 2.36 0.774 0.015 1.92 2297 1.45 0.16 PE 30 1641.3 1.98 13.63 1.10 3.78 0.739 0.023 1.86 2285 1.38 0.15 PE 31 1461.5 4.62 7.12 1.21 4.12 0.830 0.027 1.76 2270 1.51 0.17 PE 32 1292.4 4.21 8.16 1.15 3.91 0.818 0.026 2.18 2256 1.52 0.17 PE 33 935.7 3.56 8.37 0.99 3.35 0.745 0.022 2.13 2204 1.36 0.15 PE 34 742.2 4.04 9.27 0.97 3.33 0.836 0.021 2.23 2172 1.49 0.16 LE 35 4327.6 3.41 8.48 0.42 1.47 0.730 0.010 2.03 2450 1.48 0.16 LE 36 4144.8 3.37 9.38 0.79 2.69 0.758 0.017 1.91 2440 1.51 0.17 LE 37 3874.6 3.83 7.30 0.45 1.57 0.739 0.010 1.88 2425 1.47 0.16 LE 38 3701.0 3.23 7.64 0.70 2.42 0.677 0.016 1.96 2425 1.36 0.15 LE 39 3534.6 3.64 8.45 0.89 3.05 0.757 0.020 1.83 2415 1.49 0.16 LE 40 3353.3 3.06 9.85 0.79 2.72 0.736 0.017 1.82 2400 1.44 0.16 LE 41 3169.3 2.29 13.46 1.13 3.88 0.772 0.024 2.02 2385 1.51 0.17 LE 42 3004.7 2.71 10.01 0.81 2.79 0.700 0.018 1.64 2375 1.35 0.15 LE 43 2822.8 3.93 9.10 1.16 3.97 0.817 0.026 1.83 2375 1.57 0.17 LE 44 2627.5 2.96 9.84 0.55 1.88 0.724 0.012 1.72 2366 1.39 0.15 LE 45 2274.1 2.57 12.37 0.08 0.33 0.766 0.005 1.96 2337 1.46 0.16 LE 46 1719.1 2.72 9.86 0.52 1.82 0.694 0.012 1.38 2291 1.27 0.14 LE 47 1544.3 3.56 10.02 1.06 3.64 0.805 0.023 1.55 2277 1.46 0.16 LE 48 5181.6 3.71 8.27 0.63 2.16 0.761 0.014 1.83 2475 1.53 0.17 LE 49 5012.6 3.38 10.30 0.84 2.89 0.792 0.018 1.83 2475 1.59 0.17 LE 50 4842.5 3.98 12.61 0.98 3.37 0.950 0.020 1.83 2472 1.87 0.20 LE 51 4691.3 3.65 12.72 0.75 2.57 0.914 0.015 1.83 2460 1.79 0.20 LE 52 4503.9 3.15 12.68 1.25 4.29 0.852 0.026 1.83 2450 1.68 0.19 218 ID depthUThUTha aK P A A well # error error error error m ppm ppm ppm ppm cts/ks/ cts/ks/ wt % kg/m3 3 fiW/m (iW/m 2 2 cm cm LE 53 2099.6 3.70 12.92 0.95 3.24 0.926 0.019 1.96 2322 1.73 0.19 Frio mudrocks MI 54 3861.8 3.35 13.06 0.46 1.62 0.888 0.010 2.09 2425 1.75 0.19 MI 55 4076.7 3.38 12.69 0.88 3.03 0.879 0.018 2.18 2434 1.74 0.19 MI 56 4255.0 4.10 12.50 0.80 2.75 0.961 0.016 2.26 2449 1.91 0.21 MI 57 4433.3 4.24 11.22 0.85 2.92 0.932 0.017 2.27 2450 1.86 0.20 MI 58 4642.1 3.85 12.67 0.59 2.04 0.936 0.012 2.34 2456 1.88 0.20 MI 59 4826.5 3.72 11.93 0.91 3.12 0.894 0.019 2.21 2471 1.80 0.20 MI 60 5009.4 4.48 10.63 0.98 3.37 0.939 0.020 2.40 2475 1.90 0.21 MI 61 5192.3 3.70 12.89 0.79 2.69 0.925 0.015 2.46 2475 1.88 0.20 CS 62 2663.3 3.52 10.64 0.79 2.70 0.821 0.017 1.48 2370 1.54 0.17 CS 63 2442.7 5.47 4.45 1.17 3.99 0.841 0.027 2.28 2350 1.63 0.18 CS 64 2443.0 4.02 7.19 0.85 2.90 0.759 0.019 2.38 2350 1.50 0.16 CS 65 2443.3 4.08 7.78 0.44 1.53 0.788 0.010 2.31 2350 1.54 0.17 CS 66 2443.6 2.55 13.82 1.24 4.24 0.817 0.026 2.17 2350 1.58 0.17 CS 67 2444.5 3.54 7.28 0.41 1.42 0.702 0.010 2.20 2350 1.38 0.15 CS 68 2598.1 4.47 9.95 1.29 4.37 0.916 0.027 2.22 2350 1.75 0.19 CS 69 2597.8 3.89 10.02 1.35 4.60 0.846 0.029 2.17 2363 1.64 0.18 CS 70 2782.5 6.08 10.94 0.70 2.43 1.151 0.014 2.56 2378 2.21 0.24 Wilcox sandstones BU 71 2929.1 2.71 5.63 0.40 1.36 0.541 0.010 1.10 2300 0.99 0.07 BU 72 2918.2 2.53 5.49 0.55 1.88 0.512 0.014 1.11 2300 0.95 0.06 BU 73 2929.1 2.91 6.05 0.62 2.14 0.580 0.015 0.96 2280 1.04 0.07 BU 74 2845.9 1.17 2.69 0.25 0.86 0.244 0.008 0.52 2210 0.43 0.03 BU 75 2907.6 4.21 2.62 0.62 2.11 0.617 0.016 0.56 2250 1.06 0.07 BU 76 2913.6 1.75 4.25 0.25 0.91 0.367 0.007 0.59 2240 0.65 0.04 KO 77 1592.0 1.53 3.30 0.29 1.03 0.309 0.009 0.90 2540 0.65 0.04 KO 78 1587.1 3.77 4.81 0.33 1.15 0.642 0.008 1.36 2470 1.27 0.08 Frio sandstones CS 79 2747.8 2.90 10.63 0.79 2.71 0.754 0.017 1.30 1795 1.07 0.07 CS 80 2449.1 1.47 4.18 0.32 1.12 0.334 0.009 1.12 2067 0.58 0.04 ID depthUThuTha aK P A A well # error error error error m ppm ppm ppm ppm cts/ks/ cts/ks/ wt % kg/m 3 pW/m3 pW/m3 22 cm cm CS 81 2451.5 4.16 7.82 0.87 2.97 0.800 0.019 2.33 1971 1.31 0.09 cs 82 2452.7 3.36 10.33 1.10 3.78 0.790 0.024 2.03 1977 1.28 0.09 CS 83 2453.3 4.91 9.24 1.31 4.47 0.944 0.028 2.35 1981 1.52 0.10 cs 84 2609.1 3.68 5.30 0.49 1.68 0.648 0.012 2.00 1819 0.99 0.06 cs 85 2609.4 2.92 7.38 0.66 2.28 0.631 0.015 1.83 1861 0.97 0.07 cs 86 2661.5 3.27 12.58 0.62 2.14 0.861 0.013 1.35 1908 1.29 0.08 cs 87 2736.2 3.86 3.97 0.68 2.34 0.623 0.017 1.40 1986 1.00 0.07 cs 88 2737.4 4.78 7.46 0.91 3.13 0.863 0.020 1.88 2028 1.41 0.09 cs 89 2742.3 3.37 10.10 0.54 1.89 0.784 0.012 1.17 1894 1.16 0.08 cs 90 2767.9 4.05 9.81 1.22 4.16 0.859 0.026 1.51 2122 1.44 0.10 cs 91 2745.0 6.89 6.44 1.39 5.16 1.089 0.031 1.20 1832 1.53 0.11 cs 92 2779.9 2.51 7.85 0.40 1.40 0.595 0.010 1.39 1906 0.92 0.06 cs 93 2777.9 4.14 11.84 0.66 2.26 0.942 0.014 1.81 1895 1.42 0.09 cs 94 2783.4 2.57 13.48 0.83 2.82 0.797 0.016 1.30 1903 1.19 0.08 Stuart City limestones SC 95 4094.1 1.67 5.24 0.33 1.14 0.397 0.008 0.58 2640 0.82 0.06 sc 96 4096.5 1.87 7.00 0.56 1.91 0.486 0.014 0.90 2660 1.03 0.07 sc 97 4101.1 0.73 0.06 0.04 0.18 0.094 0.004 0.00 2640 0.18 0.01 sc 98 4117.5 0.35 0.00 0.01 0.00 0.044 0.003 0.00 2550 0.08 0.01 sc 99 4144.7 0.28 0.00 0.01 0.00 0.035 0.003 0.00 2560 0.07 0.01 sc 100 4226.4 0.40 0.00 0.01 0.00 0.052 0.003 0.00 2490 0.09 0.01 Table4.4.Summarystatisticsformeasurementsofradiogenicheatproduction. Column headings are: ID # = unique sample identificationnumber; n = numberof samples; Umean=meanuraniumcontentinpartspermillion; Thmean =meanthoriumcontent; Th/Umean= meanofthoriumto uraniumratio; K A =of mean=meanpotassiumcontentinweightpercent; range range radiogenic heatproductioninmicroWattspercubicmeter(pW/m3); A mean=meanradiogenic heatproduction. U ThTh/U K A A Rock type ID# n mean mean mean mean range mean wt%3 3 ppm ppm (iW/m|iW/m Frio 54-70 18 4.03 10.57 2.78 2.23 1.23-2.21 1.72 mudrocks Wilcox 1-53 52 3.37 9.91 3.10 1.86 0.86-1.87 1.50 mudrocks Frio 79-94 16 3.68 8.65 2.58 1.62 0.58-1.53 1.19 sandstones Wilcox 71-78 8 2.40 4.29 1.98 0.89 0.43-1.27 0.88 sandstones Stuart City 95-100 6 0.88 2.05 1.16 0.25 0.07-1.03 — limestones Table 4.5a. Model equations. reference Equations porosity (function of depth) Sclater and Christie (1980) bz (j)=