Earth Sciences Division Berkeley Lab
Energy Resources Program
THREE-DIMENSIONAL IMAGING OF GEOTHERMAL RESERVOIRS USING ACTIVE AND PASSIVE METHODS
Annual Report 1999 - 2000
Ernest L. Majer, Roland Gritto, Tom Daley, Ann Kirkpatrick and John Peterson RESEARCH OBJECTIVES
Contact: Ernest L. Majer, 510/486-6709, elmajer@lbl.gov
The overall objective of this research is to develop and apply passive (microearthquake (MEQ)) and active (3-D surface reflection, vertical seismic profiling (VSP) and cross-well) methods to characterize and monitor geothermal reservoirs. The application would be both for exploring and monitoring the performance of the resources.
activities has been possible through microseismic analysis. Future work will be carried on by The Geysers field operators to apply and refine the method. The 3-D seismic reflection work has shown that surface reflection methods show promise, but they must be modified for geothermal environments. The final data set represented a 3-D cube of the subsurface structure in the reservoir. Additionally, the travel times were used to perform tomographic inversions for velocity estimates to support the findings of the surface seismic imaging. The 3-D imaging shows promise for geothermal applications. It is clear that petroleum processing cannot be directly transferred to the geothermal case. However, much of the technology can be modified and adapted for geothermal applications. An example would be work in the oil and gas industry focused at fractured reservoirs. A detailed report has been issued by LBNL on this work (Feighner et. al. 1999).
APPROACH
The overall approach has been to use both passive and active methods to collect data over potential and producing geothermal reservoirs (mainly in The Geysers geothermal field in California, but other areas have also been investigated). The MEQ effort involves using state-of-the-art, high-resolution instrumentation to record three-component digital data at high frequencies (500 Hz) over tight arrays (10 to 15 stations over a 5 x 5 km area). The data are then inverted for location in space and time as well as velocity and attenuation structure. These data are then correlated with production data to infer reservoir parameters (flow paths, lithology and fluid content). Although the approach is still experimental, it may also be possible to map the high-permeability zones common to most geothermal reservoirs, including The Geysers, by detecting any statistical trend in microseismic source mechanics that would indicate that open space associated with irregularities along fracture planes was being created or destroyed during microseismic events. In terms of active work, we have been evaluating 3-D surface reflection methods at the Rye Patch, Nevada, geothermal field. LBNL has been cooperating with The Industrial Corporation (TIC) and Transpacific Geothermal Inc. (TGI) to evaluate and apply modern state-of-the-art seismic imaging methods for geothermal reservoir definition. The overall objective of the work was to determine if modern off-theshelf, commercially-available techniques in 3-D surface seismic profiling could be successfully applied in geothermal environments. If not, could they be modified to derive useful information on reservoir structure. Past efforts using 2-D seismic reflection have proved marginally successful in some cases, but due to extreme heterogeneity in many geothermal areas 2-D seismic have not been cost effective.
RELATED PUBLICATIONS
Daley, T.M., T.V. McEvilly and E.L. Majer, Analysis of P- and S-wave vertical seismic profile data from the Salton Sea Scientific Res., 93, drilling project, J. Geophys. B11,13025-13036, 1988. Feighner, M.A., R. Gritto, T.M. Daley, H. Keers and E.L. Majer, Three-dimensional seismic imaging of the Rye Patch Geothermal Reservoir, Berkeley Lab report LBNL-44119, 1999. Majer, E.L., T.V. McEvilly, F. Eastwood and L. Myer, Fracture detection using P-wave and Swave vertical seismic profiling at the Geysers, Geophysics, 53, 76-84, 1988.
ACCOMPLISHMENTS
In general, the MEQ work has shown that injection at wells resulted in higher levels of microseismicity, but the increased seismicity also corresponded to larger flow rates in the production wells. While there was not a lag time between increased water injection and increased steam production rates, there was up to several months’ lag time between increased water injection and increased seismicity. It appears that there was a threshold of water injection which caused increased seismicity, but once the system was primed, lower injection rates also increased the seismicity (i.e., as injection continued, the lag time between injection and seismicity decreased).
ACKNOWLEDGEMENTS
We are grateful to Bill Teplow for his support in the interpretation of the seismic and tomographic data sets. This work has been supported by the Assistant Secretary for Energy Efficiency and Renewable Energy, Office of Geothermal and Wind Technologies of the U.S. Department of Energy under Contract No. DE-AC03-76SF00098.
SIGNIFICANCE OF FINDINGS
Overall, a clearer understanding of the effects of injection and production 85
HTTP://WWW-ESD.LBL.GOV