The Newberry Volcano, a DOE Phase 1 FORGE site in Central Oregon, has been stimulated two times using high-pressure fluid injection to study the Enhanced Geothermal Systems (EGS) technology. Several hundred microseismic events were generated during the first stimulation in the fall of 2012. Initial locations show events occurred in two distinct depth ranges. Within the two depth ranges microseismicity does not clearly outline subsurface structures in part because event location uncertainties are large (Foulger and Julian, 2013). We have initially focused on this stimulation to explore the spatial and temporal development of microseismicity, which is key to understanding how subsurface stimulation modifies stress, fractures rock, and increases permeability. We use an application of PageRank (Aguiar and Beroza, 2014), Google’s initial search algorithm, to assess signal-correlation topology for the micro-earthquakes. We then use this information to create signal families and compare these to the spatial and temporal proximity of associated earthquakes. We relocate events within families (identified by PageRank linkage) using the Bayesloc approach (Myers et al., 2007). Preliminary relocations show tight spatial clustering of event families. Event relocations are significant enough in several cases to change cluster affiliation. In every case cluster affiliation determined by PageRank and event relocation agrees. We also find that signal similarity (linkage) at several stations, not just one or two, is needed in order to determine that events are in close proximity to one another suggesting the importance of having good seismic station coverage. We show that indirect linkage of signals using PageRank is a reliable way to increase the number of events that are confidently determined to be similar to one another, which may lead to efficient and effective grouping of earthquakes with similar physical characteristics, such as focal mechanisms and stress drop. We will also apply this analysis to the new stimulation performed in 2014, and compare with clusters found in the initial stimulation. This will allow us to determine whether changes in the state of stress and/or changes in the generation of subsurface fracture networks can be detected using PageRank topology as well as aid in the event relocation to obtain more accurate subsurface structure. Ultimately, automating and applying this method in real time could potentially aid in a more successful geothermal production at any well-instrumented geothermal site. This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. LLNL-ABS-705557.

Press the Back button in your browser, or search again.

Copyright 2017, Stanford Geothermal Program: Readers who download papers from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the original publisher.