V54C-01
Relationship Between Active Faulting, Strain Transfer, and Relief on the Brittle-Ductile Transition Zone, Coso Range, Eastern California
The Coso Range is located within a right-releasing stepover between the dextral Airport Lake and Owens Valley fault zones along the southeastern margin of the Sierra Nevada microplate. Upper crustal extension in the stepover region is accommodated by the Coso Wash graben at the southeast end of the stepover region, and by a series of left-stepping, north-northeast-striking, right-normal oblique faults in the northern Coso Range at the northwest end of the stepover region. The upper crustal faults overlie a NW-trending zone in the 5 to 10 km depth range beneath the central and northern Coso Range that is characterized by high heat flow and low P-wave and S-wave velocities. The brittle-ductile transition zone (BDT) is arched upward over the low velocity zone as evidenced by shallowing of the base of seismicity to 4-5 km depth beneath the Coso geothermal field. The positive relief on the BDT may guide or control the down-dip geometry of active faults in the step-over region. South of the geothermal field, the Coso Wash fault dips east, toward the east-dipping flank of the low velocity zone and in the direction that the BDT is deepening away from the zone of highest heat flow. At the latitude of the geothermal field, the traces of active normal faults dip west and they likely terminate against or sole into the BDT as it deepens westward toward Rose Valley. The right-normal faults in the northwestern Coso Range graben similarly sole into or terminate against the elevated BDT above the low velocity zone. The low velocity zone may be acting as a strain guide for the transmission of NW dextral shear across the stepover region. Brittle faulting and seismogenic deformation above the shallow BDT contribute to development of permeability in the geothermal reservoir, and provide pathways for upward circulation of hydrothermal fluids.
V54C-02
Coso Hot Springs: A Condensate Fed Geothermal Feature
The Coso Hot Springs are located almost two miles from the Coso geothermal field within the China Lake Naval Air Weapons Station, China Lake, California. The hot springs are about 16 m above the adjacent valley floor and because of their position are not believed to be in good hydrologic connection with the regional groundwater water table. Shortly after the onset of geothermal production in 1987, both water levels and temperatures within the South Pool of Coso Hot Springs increased. Although water levels appeared to be stable by 1989-1990, temperatures continued to rise until about 1993. We postulate that Coso hot springs are a condensate fed geothermal feature associated with vapor flux emanating from the Coso Wash Fault. We developed a suite of generic and site specific numerical models using FEHM to test whether these measured changes in the hot springs might be due to the removal of fluids associated with production from the Coso geothermal field. The idealized models were based on observed conditions including the locations of faults and geologic framework as well as temperature/pressure/production history of the geothermal field. The model results suggest that pressure declines associated with geothermal production led to an expanded steam cap which resulted in an enhanced vapor flux up the Coso Wash Fault zone.
V54C-03
3D Finite Element Analysis of Stress, Second Order Fractures and Fluid-Flow Pathways for the Coso Geothermal Field
The efficiency of geothermal energy production at the Coso Geothermal Field in eastern California is reliant on the knowledge of fluid flow directions associated with fracture networks. We use 3D finite element analysis to simulate the 3D state of stress of the tectonic setting of the Coso Range within the ECSZ. The modeling is based on a multi-scale approach of the ECSZ and the Coso Range releasing bend setting. The modeling results are calibrated against stress magnitudes and orienations and the received 3D state of stress is used for subsequent anaylsis. The tectonically induced differential stresses are used to infer second order fracture likelihood and orientation. The mean stress ditribution is used to derive possible fluid flow vectors. The results show that the Coso Range and adjacent areas are regions of increased likelihood of second order fracture generation and that 2nd order fracures coincide with clusters of regional seismicity. The mean stress distribution indicates that the geothermal field occurs in a favorable location for fluid through-flow locally derived from the north and west, but more regionally from the Sierra Nevada, and that fluid storage may occur at the southern end of the Coso Wash Fault. Predicted second order structures either support or constrain fluid flow. This indicates the importance of fracture networks for fluid migration in tectonically active regions such as the Coso Range.
V54C-04
ANALYSIS OF 3D MAGNETOTELLURIC MEASUREMENTS OVER THE COSO GEOTHERMAL FIELD
We have carried out an investigation of the Coso Geothermal field utilizing a dense grid of magnetotelluric (MT) stations plus a single line of contiguous bipole array profiling over the east flank of the field. Motivation for this study is that electrical resistivity/conductivity mapping can contribute to better understanding of enhanced geothermal systems (EGS) by imaging the geometry, bounds and controlling structures in existing production, and by monitoring changes in the underground resistivity properties in the vicinity of injection due to fracture porosity enhancement. Initial analysis of the Coso MT data was carried out using 2D MT imaging technology to construct a starting 3D resistivity model from a series of 2D resistivity images obtained using the inline electric field measurements (Zxy impedance elements) along different measurement transects. This model was then refined through a 3D inversion process. The 3D resisitivity model clearly showed the controlling geological structures influencing well production at Coso and shows correlations with mapped surface features such as faults and regional geoelectric strike. We have also correlated the model with an acoustic and shear velocity model of the field to show that the near-vertical high conductivity (low resistivity) structure on the eastern flank of the producing field is also a zone of increase acoustic velocity and increased Vp/Vs ratio.
V54C-05
Evolution of vertical permeability in Coso Geothermal Well 58A-10
Static temperature (T) profiles in geothermal systems record the relative role of conductive and advective heat transport. Under near steady-state conditions these profiles can reveal the long-term average permeability (k) along the T profile to an order of magnitude by solving for the advective component. Well 58A-10 in the undisturbed margin of the Coso Geothermal Field, CA, offers the opportunity to quantify vertical variations in k that provide a critical insight into the life cycle of geothermal systems. This well has three distinct zones of T gradient: a shallow zone from 400-1100 m depth of 100 ° C/km, a near-isothermal intermediate zone from 1100-2900 m, and a basal zone >2900 m of 100 ° C/km. Simple analytical solutions for 1-D heat transport and convective instability can be fit to the observed temperature profiles in the upper two zones. From these solutions the shallow zone requires k<10-17 m2 whereas the intermediate zone requires k>10-13.5 m2, indicating a >103 k difference between these intervals despite similar host rocks and fracture populations. The difference in 1-D, vertical k between the upper two zones appears to arise from the minerals that "heal" fractures as revealed by surface mapping, cuttings and core analysis, and geophysical logs. Initial brittle fracture and frictional slip in low porosity crystalline rocks causes dilation owing to surface roughness along fracture walls, brecciation, and micro-cracking. Yet active precipitation and alteration in geothermal areas implies rapid healing that requires fracture generation or reactivation to maintain k. Fault rocks enriched in neoformed clays minimize dilation during slip and can reduce k to lower than 10-19 m2. Reduced frictional strength in clay-rich faults also promotes their reactivation over the generation of new faults. Since the stability of clays is restricted to relatively low temperature, clay-rich faults are confined to shallow depths. Thus, a fault core enriched in clays at shallow depths becomes a persistent barrier to cross-fault flow that effectively reduces vertical k. Conversely precipitation of calcite or silica dominates healing at greater depth. These minerals retain dilatant behavior during slip that regenerates k as revealed by crack-seal textures. Thus k remains high enough to support convection and isothermal T profiles.
V54C-06
Seismic Characterisation of Hydraulic Stimulation Tests at the Coso Geothermal Area, California
We studied microearthquakes before, during and after fluid injection tests at the Coso geothermal area, California, to map the fractures formed, determine the mode and sense of failure, and characterize the stress cycle associated with injection. Our approach is based on joint interpretation of high-resolution relative earthquake relocations and full earthquake moment tensors. We developed advanced computer programs for this work, which include combining waveform cross-correlation with relative relocation methods, and rigorously assessing the confidence regions for moment tensors derived using linear-programming methods. Using a high-quality permanent network of three-component digital borehole seismometers operated by the US Navy, supplemented by 14 portable three-component digital instruments installed at the surface, we analysed several months of data spanning injection experiments in well 34A-9 in 2004, well 34-9RD2 in 2005, and pre-injection earthquakes near well 46A-19RD. In the case of injection into well 34A-9, the co-injection earthquakes were more numerous, smaller, more explosive and had more horizontal motion, compared with the background earthquakes. Injection modulated the stress orientation in the activated volume for at least two months after injection ceased. In the case of well 34-9RD2, the injection produced spatially coherent seismicity different from the scattered background activity that occurs continuously there. The relocated hypocenters reveal the dimensions and orientation of a well-defined planar structure, 700 m long and 600 m high in the depth range 0.8 to 1.4 km below sea level, striking N 20 degrees E and dipping 75 degrees to the WNW. The moment tensors show that it represented a mode I (opening) crack. The seismicity rate and stress state in the neighborhood of the bottom of the well did not return to its background state for at least two months following the injection. http://cosomeq.wr.usgs.gov/
V54C-07
Use of U and Th Decay-Series Disequilibrium to Characterize Geothermal Systems: An Example from the Coso Geothermal System
Uranium and thorium decay series isotopes were measured in fluids and solids in the Coso geothermal system to assess the utility and constrain the limitations of the radioisotopic approach to the investigation of rock-water interaction. Fluid radioisotope measurements indicate substantial kilometer-scale variability in chemistry. Between 1988 and 1990, radium isotope activity ratios indicate temporal variability, which is exhibited by apparent mixing relationships observed as a function of time for single wells. Activity ratios of Ra-224/Ra-226 and Ra- 228/Ra-226, and the processes that contribute and remove these radionuclide to and from the fluids, constrain residence times of fluids and may help constrain fluid velocities in the geothermal system. Activity ratios of Ra- 224/Ra-226 > ten were measured. In groundwater and geothermal systems ratios of Ra-224/Ra-226 > ten are limited to zones of thermal upwelling or very young (days to weeks) waters in mountainous areas. Rn-222 results indicate that radon is also an effective tracer for steam velocities within the geothermal system. Analysis of carbon dioxide and Rn-222 data indicates that the residence time of steam (time since separation from the liquid) is short (probably less than four days). Estimates of fluid velocities derived from Rn-222 and radium isotopic measurements are within an order of magnitude of velocities derived from a fluorescein tracer test. Both Rn-222 and Ra-224 activities are higher in single-phase fluids in the northwest as compared to the southeast, indicating a higher rock-surface-area/water-volume ratio in the northwest. Thus, measurements of short-lived radioisotopes and gaseous phase constituents can constrain processes and characteristics of geothermal systems that are usually difficult to constrain (e.g., surface area/volume, residence times). The NRC staff views expressed herein are preliminary and do not constitute a final judgment or determination of the matters addressed or of the acceptability of a license application for a geologic repository at Yucca Mountain.
V54C-08
Isotopic Constraints on the Origin and Evolution of Geothermal Fluids, Long Valley, CA
Successful management of geothermal resources requires hydrologic models that define and predict fluid flow in fracture-dominated systems. Such models are necessary to assess the resource potential, the impact of fluid production on the reservoir and the impact that new wells will have on existing production. We present new data on variations in O, D, and Sr isotopes in thermal waters of the Long Valley (California, USA) geothermal system and use them to assess fluid flow and water-rock interaction in the system. Geothermal well water samples collected June 2005 to June 2007 have temperatures of 41-190°C, δ18O from -16.5 to -13.5‰, and δD from -123 to -111‰. The low values suggest recharge from the west and north rims of the caldera, consistent with a general west-to-east decrease in temperature. Both δ18O and δD are displaced from the local meteoric water line and are positively correlated with Cl- concentrations for all thermal water samples hotter than 50°C. The Sr isotope ratios in the currently producing part of the field are clustered near 0.708, with a small west-to-east gradient of decreasing values from 0.7080 to 0.7078 over a distance of a few kilometers. These values are higher than those of the primary reservoir rocks (0.7060-0.7065), so the Sr isotopic ratio of the hot fluids must be set in another rock type. Granites from the northern and western rims of the caldera have 87Sr/86Sr = 0.7078-0.7100 and are possible sources of hot (200°C+) water entering the Long Valley geothermal system. The O isotope values of the fluids are also far out of isotopic equilibrium with host rocks. The small gradient in Sr isotope ratios, associated with minimal shift in O isotopes in this part of the system, probably reflects water-rock reaction, and indicates that flowing waters are contained in widely spaced (10m or more) fractures. Observed variations in O and D isotopes outside of the producing part of the field are inconsistent with a water rock reaction model, rather it appears another process such as mixing, as suggested in previous studies, or boiling exerts primary control on the water isotopes. The inferred large fracture spacing suggests that the thermal response time and O isotope response time of the system are of order 10-100 years, and therefore that changes to the system hydrology could be reflected in fluid O isotope values. Based on comparison of our 2006 isotope data and earlier data (Goff et al., 1991), there is little evidence of any change to the system hydrology over 15 years.