A Data Fusion Approach to Induced Seismic Risk Assessment in Enhanced Geothermal Systems

Geothermal energy, an important source of renewable energy, generates electricity from circulating fluids in the Earth's crust. Enhanced Geothermal Systems (EGS) use high pressure injection of fluids into the subsurface to improve geothermal reservoir quality in regions with poor natural permeability and increase the storage and production capacity of the system. Documented cases of induced seismicity at EGS sites raise concerns about the risk of felt seismic events (>Mw=2) such as in Soultz, France and Basel, Switzerland that caused alarm in the surrounding regions and shut down development in the latter case (Majer et al. 2007). Evaluating induced seismic hazard and risk during EGS project planning is key to mitigating potential property damage and negative public perception of EGS projects.

Seismic hazard assessment uses a variety of data about previous seismic activity, geology, structure, stress state, and poroelastic rock properties of the surrounding formations. This data is used to develop models to predict treatment response and if seismicity will be generated. Integrating a variety of data is difficult due to the uncertainty and assumptions made in the data, processing, and modelling. We develop an assessment system using a combination of unsupervised hybrid k-means clustering (Rogova et al. 2009) and supervised learning to aid processing of a multi-disciplinary data set. Unsupervised clustering finds patterns in each data type, then data clusters are combined and assigned a final grouping using a pairwise combination method within the framework of the Transferable Belief Model, which is used to quantitatively represent the level of support and ignorance for combined cluster assignments. Public data from the FORGE EGS site in Utah, USA is used in testing and development, including regional seismic catalogs, maps of faults, geologic structure, and borehole logs.

Majer EL, Baria R, Stark M, Oates S, Bommer J, Smith B, Asanuma H, 2007, Induced seismicity associated with Enhanced Geothermal Systems. Geothermics; 36:185-222. http://dx.doi.org/10.1016/j.geothermics.2007.03.003.

Rogova, G.L., M. I. Bursik, and S. Hanson-Hedgecock, 2009, Intelligent system for interpreting the pattern of volcanic eruptions. ISIF Journal of Advances in Information Fusion. http://www.isif.org/jaif.htm