Section Thirteen





	GRANTEE: TEXAS A&M UNIVERSITY Center for Tectonophysics Department of Geology and Geophysics College Station, Texas 77843-3115 GRANT: DE-FG05-87ER13711 TITLE: Mechanical Properties and Modeling of Seal-Forming Lithologies PERSONS IN CHARGE: A.K. Kronenberg, J.E. Russell, and N.L. Carter (409-845-0132; Fax 409-845-3002; E-mail a.kronenberg@tamu.edu)



	Objectives: This research addresses the mechanical properties of two weak sedimentary lithologies, shale and rock salt, that deform under gravitational and tectonic loads and examines the roles of these lithologies in the development of structural traps and unconventional oil and gas reservoirs. Project Description: The shapes and physical properties of a wide range of structural traps and barriers to mobile hydrocarbons are governed by the mechanical properties of shale or those of salt and by the loading histories to which they have been subjected. Concurrent experimental and modeling studies are underway to determine the fracture and flow properties of these two lithologies, to determine physically- based constitutive relationships that can be applied to geologically relevant conditions, and to make predictions concerning the development and timing of structural traps relative to known depositional histories. Results: Significant progress has been made during this year towards characterizing the mechanical and transport properties of Wilcox shale saturated with fluids of varying chemistry and towards simulating the evolution of salt structures by modeling the deformation of rock salt and overlying sediments with nonlinear, experimentally constrained rheologies. Our investigations of shale deformation have been extended from strain rates of 10^-8 s^-1 to 10^-3 s^-1 to include time scales that represent drained and effectively undrained conditions as determined from experimen		tally determined hydraulic diffusivities and calculations of internal pore pressures. Having established these time scales, we have determined and are currently refining effective stress laws for permeability and failure strength of shale. We have examined the effects of fluid chemistry on permeability using NaCl, KCl, and CaCl₂ brines and distilled water, and we have further explored the effects of water as a pore fluid and as adsorbed layers at clay surfaces, on mechanical properties. Permeability measured at increasing and decreasing effective pressures suggest substantial changes in pore geometries that are inelastic as effective pressure is increased while recoverable poro-elastic effects are observed as effective pressure is decreased. Our numerical studies of salt structure development have led to insights into the initiation, evolution, and timing of complex salt structures of the northern continental shelf of the Gulf of Mexico, as well as the sensitivities of salt structure development to subsurface conditions, viscosity contrasts between salt bodies and overlying sediments, and degrees of nonlinearity in rheology. These model studies have been accompanied by studies of deformation microstructures of natural salt structures of the Gulf of Mexico using cores recovered from actively deforming allochthonous salt nappes. Equivalent stress levels within these structures, as determined from stress-dependent microstructures, offer important constraints for the models and the rheologies on which our models are based.





		GRANTEE: UNIVERSITY OF TEXAS Department of Geological Sciences Austin, Texas 78712 GRANT: DE-FG05-92ER14278 TITLE: Energy Flux and Hydrogeology of Thermal Anomalies in the Gulf of Mexico Sedimentary Basin - South Texas Example PERSON IN CHARGE: John M. Sharp, Jr. (512-471-5172; Fax 512-471-9425)



		Objectives: The objectives of this study were to (1) evaluate if observed heat flow anomalies in the south Texas portion of the Gulf of Mexico sedimentary basin (if reconfirmed by extensive data analysis) can be accounted for by conduction alone or if convection is a significant process, (2) determine if the present potential field is amenable to convection hypotheses, and (3) develop fluid and heat flux histories compatible with both compiled and newly collected data. Project Description: In order to test hypotheses posed to address the above objectives, we compiled an extensive data base of fluid pressures (>25,000 data points at depths of up to 5 kilometers [over 16,000 feet]), water chemistries (>2000 data points), and formation temperatures (>5500 data points) from a variety of private and public sources; installed a G.I.S. to analyze these data; analyzed selected drill cores and cuttings for their mineralogical, thermal conductivity, radiogenic heat production, and porosity/bulk density data. We culled spurious pressure and temperature data, corrected temperature and pressure data to infer predevelopment conditions, calculated buoyancy gradients, and implemented mathematical and numerical models to simulate the study area's temperatures and pressures and to infer the importance of the various processes.		Results: Analysis of the data reconfirmed the thermal anomalies discerned from the previous, much smaller data bases. New data on fault zone locations centered the anomalies along the deepest-seated Wilcox Faults. The pressure data revealed that petroleum production has created significant depressurization that extends well beyond the immediate petroleum reservoirs. The water chemistry data revealed zones of higher salinity coincident with Wilcox and Frio fault zones. Several areas have significant buoyancy gradients. This creates conditions conducive to free convection that was confirmed using the SUTRA computer model. The thermal property data revealed several interesting trends. First, radiogenic heat production in the Gulf of Mexico Basin is significant and should be accounted for in simulation models, but this production is not responsible for the thermal anomalies. Secondly, sandstone thermal conductivities correlate well with quartz content and porosity; a new empirical equation for Gulf Coast sandstone thermal conductivities was developed. Shale cuttings indicated significant shrinkage so that laboratory measurements of shale thermal conductivity, porosity, and bulk density are suspect. Our models indicate that forced convection (advection) is a significant factor in geological-time, regional-scale heat transport and that both free and forced convection are significant in salinity transport.





	GRANTEE: UNIVERSITY OF TEXAS Bureau of Economic Geology University Station, Box X Austin, Texas 78713-8924 GRANT: DE-FG03-95ER14504 TITLE: A Robust Economic Technique for Crosswell Seismic Profiling PERSONS IN CHARGE: Bob A. Hardage (512-471-1534; Fax 512-471-0140; E-mail hardageb@begv.beg.utexas.edu) and James L. Simmons, Jr. (E-mail simmonsj@begv.beg.utexas.edu)



	Objectives: This project is investigating techniques by which crosswell velocity tomograms can be constructed from seismic wavefields generated by a surface-positioned source and recorded by downhole sensors in two or more in-line receiver wells and is verifying if this type of source-receiver geometry produces travel-time tomograms of sufficient accuracy to be used in reservoir characterization and monitoring. Project Description: Crosswell velocity tomograms can be helpful in reservoir characterization and monitoring, yet crosswell applications cannot be attempted in highly attenuating rocks or between widely separated wells because of the limited energy output of present downhole seismic sources. In concept, crosswell data can be acquired in these no-data situations by using a robust surface-based energy source and then measuring the travel times required for these surface-generated wavefields to travel between downhole sensors positioned at different depths in two in-line receiver wells. If this combination of surface energy source and twin receiver wells allows ray paths to be recorded over an adequate angular aperture and also provides a travel-time accuracy that is adequate for velocity inversion, then crosswell applications can be implemented in reservoirs having almost any well spacing and any type of interwell lithology. The purposes of this research are to determine how to best record this type of crosswell data and then to verify if the data satisfy the numerical requirements for creating travel-time tomograms that can be used in reservoir evaluations.		Results: Crosswell data were recorded between two receiver wells that penetrated low-velocity, weakly consolidated Pleistocene rocks in South Texas. The downhole sensors in each well were a six-level hydrophone array with 10-ft spacings between adjacent hydrophone elements. The surface energy source was a Bolt land air gun stationed at four offset positions that were in-line with the two receiver wells. The downhole data were recorded at vertical increments of 10 ft over a depth aperture of 1500 to 2500 ft in each receiver well. This work was done in an active oil field so that the typical background seismic noise of pumping wells, nearby workover activity, and continuous vehicular traffic associated with a producing reservoir would be superimposed on the downhole data. This environment was desired so a determination could be made if production activity has to be reduced in order to record crosswell data of adequate quality when using this type of source-receiver geometry. It is particularly important to know how these types of cultural noises affect free-hanging hydrophones, because a hydrophone array can be moved to a new downhole recording level quicker than a wall-clamped receiver array can, thus allowing crosswell data to be recorded in less time and at less cost. An eikonal technique has been used to reproduce the P-wave arrival times observed in these data and to construct an interwell P-wave tomogram. The interwell velocity distribution resulting from this eikonal computation agrees with the velocity control that is available at the twin-well site.





		GRANTEE: UNIVERSITY OF TEXAS AT DALLAS Center for Lithospheric Studies P.O. Box 830688 (FA31) Richardson, Texas 75083-0688 GRANT: DE-FG03-96ER14596 TITLE: Sedimentological and Geophysical Studies of Clastic Reservoir Analogs: Methods, Applications, and Developments of Ground- Penetrating Radar for Determination of Reservoir Geometries in Near-Surface Settings PERSONS IN CHARGE: George A. McMechan (214-883-2419; Fax 214-883-2829; E-mail mcmec@utdallas.edu) and Kristian Soegaard (214-883- 2415; Fax 214-883-2537; E-mail soegaard@utdallas.edu)



		Objectives: Ground-penetrating radar is evaluated for constructing 3-D models of sandstone hydrocarbon reservoirs from near-surface analogs. Constraints include detailed sedimentologic mapping of outcrops and lab and petrophysical analyses of plug and core samples. Project Description: Existing reservoir models are based on only 2-D outcrop studies; their 3-D apsects are inferred from correlation between well data and so are inadequately constrained for reservoir simulations. Field study sites are in the Cretaceous Ferron Sandstone in Utah. Detailed sedimentary facies maps of cliff faces will define the geometry and distribution of reservoir flow units, barriers, and baffles at the outcrop. High resolution 3-D ground-penetrating radar (GPR) images will extend these reservoir characteristics into 3-D, thereby enabling development of realistic 3-D reservoir models. Models will use geometrical information from the mapping and the GPR data, petrophysical data from surface and cliff-face outcrops, lab analyses of outcrop and core samples, and petrography. Results: Several field sites have been identified in the Ferron Sandstone in Central Utah. Previously acquired 3-D GPR data sets are being processed at		ARCO; the results will be incorporated into this project. An extensive new data set has been collected from fluvial deposits at one site at Coyote Basin, covering a 40 m x 16.5 m surface area to a depth of 30 meters. The fluvial sandstone is composed of a channel complex in the upper 12 meters of a section that overlies floodplain mudstone, coal, and crevasse splay deposits. Three 3-D GPR data volumes were acquired at frequencies of 50, 100, and 200 MHz. The surface geology has been mapped and shows fractures and sedimentary features, such as cross beds. A facies map has been created of the adjacent vertical cliff face. Five stratigraphic sections, gamma-ray profiles, and photographs were made at the cliff face, extending through the upper fluvial sandstone; one section extends to a depth of 30 meters, well into the underlying floodplain deposits. Approximately 700 core plugs have been extracted from the cliff face for determination of permeability. Paleo- current data have been obtained at four localities in the vicinity of the GPR survey site. Four 14-meter wells have been cored within the GPR data cube; stratigraphic sections have been logged from the core. Thus, most of the raw data for characterization of this site are now in hand.





	GRANTEE: U.S. DEPARTMENT OF ENERGY CORE AND SAMPLE REPOSITORY c/o RUST Geotech Inc. P. O. Box 14000 Grand Junction, Colorado 81502-5504 GRANT: None (B&R KC040101) TITLE: U.S. Department of Energy, Core and Sample Repository PERSON IN CHARGE: Larry M. Fukui (970-248-6172; Fax 970-248-6040; Internet gjcore@gjpomail.doegjpo.com)



	Objectives/Project Description: The DOE Core and Sample Repository provides the scientific community with ready access to geologic samples and information, ensures proper preservation and storage of samples and data, maintains records of sample requests, and promotes the use of sample inventory and data by qualified investigators.		Results: The DOE Core and Sample Repository has distributed over 4000 samples selected from the more than 13,900 meters (45,000 feet) of drill core to about 90 investigators in the USA and foreign countries. Please note that all of the core will be moved to the University of Utah Earth Science and Resources Institute in Salt Lake City by the end of Fiscal Year 1997.





		GRANTEE: UNIVERSITY OF UTAH Earth Sciences and Resources Institute Department of Civil and Environmental Engineering Salt Lake City, Utah 84112 GRANT: DE-FG02-90ER14133 TITLE: Assessing the Role of Active and Ancient Geothermal Processes in Oil-Reservoir Evolution in the Basin and Range Province PERSON IN CHARGE: Jeffrey B. Hulen (801-581-8794; Fax 801-585-3540; E-mail jhulen@esrilan.esri.utah.edu)



		Objectives: The project is structured around investigation of the premise that active and ancient moderate-temperature hydrothermal systems, by various means, have been instrumental in the generation, migration, and entrapment of oil in the Basin and Range province of the western United States. Project Description: The eastern Basin and Range encompasses several shallow and hot (<2 km; up to 130°C) oil fields (for example Blackburn and Grant Canyon/Bacon Flat) that geologically resemble the Carlin-type, Paleozoic sediment-hosted gold deposits occurring in the same region—in particular those of the southern Alligator Ridge mining district about midway between the towns of Elko and Ely, Nevada. We are investigating the distinct possibility that at least some of these gold deposits are the exhumed and oxidized, paleogeothermal analogues of the modern, exploited geothermal oil fields. Our approach is multidisciplinary, involving (1) detailed geologic mapping, (2) logging of drill cuttings and cores, with emphasis on alteration, porosity characteristics, vein mineralization and paragenesis, and hydrocarbon type and distribution, (3) three-dimensional stratigraphic/structural analysis to allow reconstruction of fluid-flow paths used by both thermal waters and hydrocarbons, (4) fluid-inclusion microthermometry, to ascertain the compositions and temperatures of these fluids at different times during the duration of the hydrothermal system, (5) whole-rock and vein-mineral geochemistry, (6) hydrogeochemistry of oil-field vs. regional		waters, and (7) stable-isotopic systematics of thermal waters and vein and alteration minerals. Results: During the past year, we have mapped several small, new open-pit gold mines in the southern Alligator Ridge district, adding new details to the stratigraphic/structural and hydrothermal picture previously established for this area. All the new mines penetrate the same oil-bearing, altered, and mineralized Paleozoic sedimentary sequence encountered in prior excavations. The oil, freely-flowing and in fluid inclusions, occurs within and around both low- and high-grade gold ore bodies. From detailed petrographic and fluid-inclusion work coupled with field relationships, the oil appears to have been introduced in the same hydrothermal system responsible for the precious-metal mineralization. The temperature of mineralization apparently did not exceed 130°C. This surprising finding is confirmed not only by pressure-corrected, fluid-inclusion homogenization temperatures but also by temperature-dependent biomarker transformation preserved by hydrocarbons in oil-rich fluid inclusions. There is a distinct depletion in ¹⁸O/¹⁶O of the oil-bearing ore bodies' wall rocks relative to their unaltered and unmineralized counterparts. Two deep oil wells completed in the immediate area penetrated neither intrusive rocks nor sedimentary rocks hydrothermally altered at high temperatures. One of these wells, just a mile from the ore bodies, bored through the hydrocarbon source rocks, which at this location and depth were shown by Rock-Eval pyrolysis to be near peak oil-gen



	eration capacity. It now appears highly likely that the gold-depositing hydrothermal system was directly responsible here for the generation, migration, and entrapment of oil; however, we still do not know to what extent, if any, the oil actually contributed to the mineralization process. In other words, the formation of this particular fossil oil reservoir may well have been just a beneficial side effect of the ore-forming hydrothermal event. By contrast, at another oil-rich gold deposit, Gold Point near Ely, we have determined that hydrocarbons were crucial to mineralization, most likely providing highly adsorptive substrates for electrum precipitation. In huge Railroad Valley, about 25 km southwest of Gold Point and the site of most of Nevada's oil production, our study of oil-well drill-stem-test temperatures has indicated that the western side of this fault-bounded valley is probably a major regional hydrologic downflow zone, whereas the eastern side, along which the hottest oil fields occur, is a region of localized geo		thermal upflow of the same waters. These geothermal plumes have fostered the maturation of hydrocarbon source rocks as well as the migration and entrapment of the newly generated oils. A pilot study of the regional hydrocarbon-sealing mechanism in this valley has indicated that volcanic-ash-rich beds at the base of the valley-fill sequence have been widely altered to montmorillonite, thereby inhibiting the escape of oils structurally entrapped in underlying Tertiary ignimbrites and brecciated Paleozoic dolomites. At Kyle Hot Springs, near Winnemucca in western Nevada, an active, moderate-temperature geothermal system has generated paraffin-rich heavy crude oils from hypersaline-lacustrine Tertiary source rocks, which are otherwise well below the favorable oil-generation "window." The Kyle system, in a more energetic early phase, also precipitated hydrocarbon- and gold-bearing siliceous sinter, bolstering the geothermal/oil/gold connection we have been documenting in the eastern part of the state.





		GRANTEE: UNIVERSITY OF UTAH Department of Geology and Geophysics 717 Browning Building Salt Lake City, Utah 84112 GRANT: DE-FG03-93ER14313 TITLE: High Resolution Imaging of Electrical Conductivity Using Low Frequency Electromagnetic Fields PERSON IN CHARGE: Dr. Alan C. Tripp (801-462-2112 or 801-581-4664; Fax 801-581-7065; E-mail actripp@mines.utah.edu)



		Objectives: The project seeks to determine means of increasing the resolution of low frequency electromagnetic techniques by means of an optimal use of a priori information. Project Description: The research concentrates on extending to inductive sources a theory of focussing DC sources investigated under previous funding. This theory couples a priori information to data weights to give optimal resolution of a perturbation on a prespecified conductivity structure. This approach is viable in cases where something is known about back		ground geo-electric structure, and resolution of perturbations about this structure is very important. In the DC case, optimization variables include the electrode distribution and the source weighting. In extending the theory to the inductive source, optimization variables besides these include frequencies, loop sizes, and source phasing. The theory will be applied to feasibility studies for enhanced oil recovery and energy-related environmental monitoring. Results: Work on extending the focussing theory to the inductive case has only been recently initiated.





	GRANTEE: UTAH STATE UNIVERSITY Logan, Utah 84321 GRANT: DE-FG03-95ER14526 A000 TITLE: Three-Dimensional Hydrogeology of Fault Zones PERSONS IN CHARGE: Kevin Hestir (Utah State University; Department of Mathematics and Statistics) (801-797-2826; Fax 801-797-1822), James P. Evans (Utah State University); Stephen Martel (University of Hawaii, Honolulu, HI 96844), and Jane C. S. Long and Janet Jacobsen (Lawrence Berkeley National Laboratory, Berkeley, California 94720)



	Objectives: The project is designed to examine the three-dimensional hydrogeologic structure of fault zones by field mapping, mechanical modeling, and probabilistic modeling. The results of these efforts will be used in developing inverse techniques for determining fault zone hydrologic behavior from well test data. Project Description: We are examining the three-dimensional permeability structure of faults in crystalline rocks and integrating these results into fluid flow models, developing mechanical models for the nucleation and growth of faults in three dimensions, and using these mechanical models in developing and testing the numerical models of fluid flow. Goals of the project are to (1) Establish the geometry and permeability structure of small faults by geologic mapping and examinations of both hydrothermal mineral deposits and hydrothermally altered rock along the faults. (2) Develop a three-dimensional, numerical-mechanical technique to model fracture growth. The results of the model will be compared to the natural faults we investigate. We also will conduct parameter studies to help predict how different states of stress and different material properties affect the distribution and linkage (i.e., organization) of permeable and impermeable features along faults. (3) Develop stochastic models, based on field data and our mechanical models, for fault development. These models will represent how permeable features		and impermeable features are likely to be organized along faults. This work will form the foundation of a stochastic model for examining the hydrology of faults. (4) Compare the long-term permeability structure of faults, as revealed by geologic evidence, with the short-term permeability structure revealed by well tests. (5) Implement and develop software for visualizing in three-dimensions both field data on fault structure and hydrologic models of fracture flow systems. This research integrates field work with deterministic and stochastic modeling to gain insight into how the three-dimensional permeability structure of a fault develops through time. This work will lead to an increased understanding of fault zones from geologic, geomechanical, and hydrologic points of view and to the development of a methodology for building physically realistic stochastic models for fault zone hydrology. Results: Detailed maps of the fault zone in the Bear Creek region of the Sierra Nevada, California, have been made and visualized in three dimensions. Thin sections have been prepared from nearly all of the rock samples collected last summer in the Bear Creek region. Nearly all the samples collected for the purpose of dating the fault zones show significant amounts of fracture-bound white mica; this indicates that the samples can yield dates for the timing of fracturing and faulting. The thin sections collected for mineralogical and geochemical analyses have also been



	examined; they indicate episodic precipitation of hydrothermal minerals in the faults. One stage of two-dimensional analytical modeling has nearly been completed that bears on the mechanics of fracturing near the ends of the faults. The results so far mark a distinct improvement in our ability to account for the orientation of fault-related fractures as observed in the field, appear to place constraints on forthcoming three-dimensional mechanical modeling, and also suggest how the most essential aspects of fracturing might be accounted for in a simple but realistic manner in the stochastic modeling. A manuscript describing a two-dimensional stochastic fracture hydrology model is awaiting final com		ments from the coauthors and will be submitted this fall to the Journal of Geophysical Research. Two preliminary three-dimensional stochastic models for fault zones have been constructed. These models incorporate fault zone structures suggested by the mechanical and field results of the project. The first stochastic model will be used with field maps to estimate the likely size and density of fracture components in the mapped fault zone. The second model will be used to study the hydrology of fault zones, including an inverse modeling with pressure data from a similar fault zone at the Finnsjon site in Sweden.





	GRANTEE: VIRGINIA POLYTECHNIC INSTITUTE AND STATE UNIVERSITY Fluids Research Laboratory Department of Geological Sciences Blacksburg, Virginia 24061 GRANT: DE-FG05-89ER14065 TITLE: Geochemical Studies of Fluid Systems PERSON IN CHARGE: R. J. Bodnar (703-231-7455; Fax 703-231-3386; E-mail bubbles@vt.edu)



	Objectives: The objective of this project is to experimentally determine the pressure-volume-temperature-composition (PVTX) relationships of H₂O-CO₂-NaCl mixtures over the complete range of PTX conditions encountered in crustal energy, resource, and waste-related environments. These data are used to develop equations of state to predict the behavior of these fluids in crustal rocks. Project Description: Volumetric (PVT) data provide the fundamental information needed to understand the physical and chemical behavior of fluids in energy, resource, and waste-related environments. Further, these data represent the basis for developing empirical or theoretical equations of state to predict the thermodynamic properties of fluids over crustal PTX conditions. In this study the PVTX properties of H₂O-CO₂-NaCl are being experimentally determined using the synthetic fluid inclusion technique. With this technique, fluids of known composition are trapped as inclusions by healing fractures in quartz at known temperatures and pressures. Phase relations and P-T locations of isochores in the H₂O-CO₂-NaCl system are obtained by observing the temperatures and modes of homogenization of the synthetic fluid inclusions during subsequent heating and cooling experiments in a fluid inclusion stage mounted on a petrographic microscope. Results: The location of the solvus and isochores for an H₂O-CO₂-NaCl solution having a composition		of 20 wt.% NaCl and 10 mol% CO₂ (both relative to H₂O) were determined over the range 350°-580°C and up to 5 kb. These data significantly extend the range of PTX conditions over which data are available for this geologically important fluid system. The results confirm the earlier suggestion that the solvus displays a pressure minimum in P-T space with a very steep slope at temperatures below the minimum and a very shallow slope at temperatures above the minimum. A technique has been developed for determining compositions of synthetic H₂O-CO₂ inclusions, based on laser Raman microprobe (LRM) analysis. With the technique, inclusions are analyzed at 350°C using LRM, and the intensities are related to composition through a simple polynomial equation. The technique is valid for CO₂/(H₂O-CO₂) ranging from 0-1 and is relatively insensitive to inclusion density variations. The Hydrothermal Diamond-Anvil Cell (HDAC) has been developed as a pressurized fluid inclusion stage. The HDAC permits analysis of fluid inclusions that develop high internal pressures during conventional microthermometry. Previously, data could not be obtained from such inclusions because they decrepitated (exploded) before the homogenization temperature was reached as the internal pressure exceeded the tensile strength of the host mineral. This development permits PVT data for H₂O-CO₂-NaCl to be obtained over a much wider range of PTX conditions than possible previously.





	GRANTEE: WASHINGTON UNIVERSITY Department of Earth and Planetary Sciences St. Louis, Missouri 63130 GRANT: DE-FG02-92ER14297 TITLE: Development of an Experimental Database and Theories for Prediction of Thermodynamic Properties of Aqueous Electrolytes and Nonelectrolytes of Geochemical Significance at Supercritical Temperatures and Pressures PERSONS IN CHARGE: Everett L. Shock (314-726-4258; Fax 314-935-7361; E-mail shock@zonvark.wustl.edu) and R. H. Wood (302-831-2941; Fax 302-831-6335; E-mail rwood@brahms.udel.edu)



	Objectives: The objectives of this research are to combine new experimental measurements on heat capacities and volumes of key compounds with theoretical equations of state to generate predictions of thermodynamic data that allow quantitative models of geochemical processes at high temperatures and pressures. Project Description: This project is part of an ongoing collaboration between Prof. Everett Shock of Washington University and Prof. Robert Wood of the University of Delaware in which we are (1) making substantial improvements in theoretical equations of state for aqueous nonelectrolytes and electrolytes, based largely on data collected during previous DOE-funded research, (2) pursuing novel applications of these equations of state to the study of high temperature/pressure geochemical processes, involving aqueous fluids, and (3) building predictive methods for organic solutes, using functional group contributions that allow estimates of thermodynamic properties at high temperatures and pressures. The experimental work is conducted at the University of Delaware. Geochemical applications of the data and development of predictive methods is done at Washington University. Efforts to improve the equations of state for nonelectrolytes and the functional group contributions are shared between the two labs,		because this task in particular requires close collaboration between the two principal investigators. Results: Progress in developing new equations of state has been presented at national meetings (ACS and GSA) by Mitch Schulte, the graduate student who is conducting this research as part of his PhD thesis. The new equations are considerably more accurate than the revised Helgeson-Kirkham-Flowers (HKF) equations for nonelectrolytes, and the results are being prepared for publication. The equations and estimation procedures have been tested against data recently obtained by Schulte in the lab of Prof. Vladimir Majer at the Université Blaise Pascal in Clermont-Ferrand, France. As a consequence, new methods for making predictions for organic solutes are also near completion. Dr. Jan Amend has been hired as a post-doc to pursue theoretical developments closer to the critical point of H₂O and to expand on the functional group estimation procedure. Usefulness of the revised-HKF equations for inorganic ions, electrolytes, and complexes has expanded. Revised data and parameters for aqueous ions and hydroxide complexes are in press at Geochimica et Cosmochimica Acta (Shock et al., 1996). Exploration continues on methods to use the revised-HKF approach at pressures >5 kb.





	GRANTEE: UNIVERSITY OF WASHINGTON Geophysics Program Seattle, Washington 98195 GRANT: DE-FG06-92ER14231 TITLE: Two- and Three-Dimensional Magnetotelluric and Controlled Source Electromagnetic Inversion PERSONS IN CHARGE: John Booker and Martyn Unsworth (206-543-4980; E-mail unsworth@geophys.washington.edu)



	Objectives: The objective is to develop efficient techniques for high resolution imaging of multidimensional electrical structure of the Earth's subsurface. Project Description: Because the electromagnetic inverse problem for natural sources is generally multidimensional, most imaging algorithms saturate available computer power before the complete data set can be addressed. We have developed an algorithm called the rapid relaxation inversion (RRI) to directly invert large multidimensional magnetotelluric (MT) data sets. This method is orders of magnitude faster than competing methods. The key to its efficiency is that an improvement in the structure directly beneath an MT site can be calculated from a one-dimensional (1-D) inverse problem if one has information about the multidimensional fields as a function of depth beneath the site. The capability of RRI for two-dimensional (2-D) structure is being enhanced, and it is being extended to fully three-dimensional (3-D) structure. In addition, related holographic electromagnetic imaging techniques are being investigated and RRI is being extended to handle data generated by controlled source techniques. Results: In the past year, RRI has been extended to invert data generated by controlled sources. This has been achieved by interfacing the RRI algorithm		and a finite element algorithm that model the EM fields of the transmitter in 3-D. The inversion has been successfully tested on synthetic and field data. In each case the inversion was able to generate a subsurface resistivity model that reproduced the data. The inversion of synthetic data produced a model containing the important features present in the model used to generate the data. In the case of the field data, the model agreed well with independent estimates of subsurface structure, such as well-log information. The approximations used in the original RRI appear to work well for the controlled source electromagnetic signals. Progress has also been made on two algorithms used to obtain accurate MT data from field data. In the first, an alternative formulation of the Groom-Bailey tensor decomposition method is used to remove the effect of shallow structure and to determine the regional 2-D strike direction. In the second algorithm, rho-plus, data is evaluated for consistency with a 1-D subsurface resistivity structure, and the apparent resistivity and phase are checked for mutual consistency. These algorithms have been found to be very useful in evaluating a number of MT data sets and in providing responses that can be analyzed with a 2-D inversion.