Time-Dependent Deformation at Brady Hot Springs Geothermal Field (Nevada) Measured With Interferometric Synthetic Aperture Radar and Modeled with Multiple Working Hypotheses of Coupled Behavior

Monday, 14 December 2015: 13:55
306 (Moscone South)
Kurt L Feigl1, Syed Tabrez Ali2, John Akerley3, Elena Baluyut2, Michael A Cardiff4, Nicholas C Davatzes5, William Foxall6, Dante Fratta7, Corné Kreemer8, Robert J Mellors9, Janice Lopeman3, Paul Spielman3 and Herb F Wang7, (1)University of Wisconsin, Madison, WI, United States, (2)University of Wisconsin-Madison, Madison, WI, United States, (3)Ormat Technologies Inc., Reno, NV, United States, (4)University of Wisconsin-Madison, Geoscience, Madison, WI, United States, (5)Temple Univ-Geol, Earth & Env, Philadelphia, PA, United States, (6)Lawrence Berkeley National Laboratory, Berkeley, CA, United States, (7)University of Wisconsin Madison, Madison, WI, United States, (8)University of Nevada Reno, Nevada Bureau of Mines and Geology, Reno, NV, United States, (9)Lawrence Livermore National Laboratory, Livermore, CA, United States
To measure time-dependent deformation at the Brady Hot Springs geothermal field in western Nevada, we analyze interferometric synthetic aperture radar (InSAR) data acquired between 2004 and 2014 by five satellite missions, including: ERS-2, Envisat, ALOS, TerraSAR-X, and TanDEM-X. The resulting maps of deformation show an elliptical subsiding area that is ~4 km by ~1.5 km. Its long axis coincides with the strike of the dominant normal-fault system at Brady. Within this bowl of subsidence, the interference pattern shows several smaller features with length scales of the order of ~1 km. This signature occurs consistently in all of the well-correlated interferometric pairs spanning several months. Results from inverse modeling suggest that the deformation is a result of volumetric contraction in shallow units, no deeper than 600 m, that are probably associated with damaged regions where faults interact via thermal (T), hydrological (H), mechanical (M), and chemical (C) processes. Such damaged zones are expected to extend downward along steeply dipping fault planes, providing high-permeability conduits to the production wells. Using time series analysis, we test the hypothesis that geothermal production drives the observed deformation. We find a good correlation between the observed deformation rate and the rate of production in the shallow wells. We explore first-order models to calculate the time-dependent deformation fields produced by coupled processes, including: thermal contraction of rock (T-M coupling), decline in pore pressure (H-M coupling), and dissolution of minerals over time (H-C-M coupling). These processes are related to the heterogeneity of hydro-geological and material properties at the site. This work is part of a project entitled "Poroelastic Tomography by Adjoint Inverse Modeling of Data from Seismology, Geodesy, and Hydrology” (PoroTomo) http://geoscience.wisc.edu/feigl/porotomo.