Inference of Geothermal Reservoir Properties from Micro-Seismic Events with Ensemble Kalman Filter

Successful energy recovery from enhanced geothermal systems (EGS) relies on a good understanding of the reservoir response to hydraulic stimulation, which in turn requires integration of monitoring data into models describing the underlying flow, geomechanics, and thermal processes. Micro-seismic or micro-earthquake (MEQ) events that are believed to result from the pore pressure, temperature, and in-situ stress variations can be used to gain a better understanding of the processes that take place during stimulation of geothermal reservoirs. Integration of fluid-induced seismicity clouds into prior models of rock hydraulic and geomechanical property distributions presents a challenging inverse problem because of the nonlinearity and complexity of the governing equations describing the coupled processes involved and the complications associated with the interpretation of the MEQ observations. The clouds of micro-seismic data provide noisy monitoring data that carry information about the spatial distribution and density of micro-seismic events. In addition, subsurface flow and geomechanical property distributions are only known with significant uncertainty. Therefore, a stochastic reservoir characterization framework is more appropriate for inferring these properties. The objective of this work is to develop a probabilistic ensemble-based framework for characterization of hydraulic and geomechanical property distributions in geothermal reservoirs to reproduce the observed production and MEQ data and to quantify their respective uncertainties. In particular, we examine the suitability of the ensemble Kalman filter (EnKF) for application to geothermal reservoir characterization. EnKF is widely studied in the literature as a promising method for subsurface characterization from dynamic measurements. We use a coupled flow and geomechanics simulation as a forward model and discuss an approach for modeling and quantification of miscro-seismic data for application with EnKF. We present several numerical experiments to illustrate the performance and suitability of EnKF for identification of hydraulic and geomechanical properties from dynamic flow and microseismic monitoring data.