How Do Faults React to Rapid Stress Changes?
A study of induced earthquakes in geo-reservoirs and implications for aftershock triggering
Earthquakes are commonly caused by slowly accumulating stresses along tectonic plate boundaries such as the San Andreas fault. Earthquakes at large distances from plate boundaries occur much less frequent - unless human activity contributes, for instance, in geothermal and oil fields. This NSF supported project will investigate human induced earthquakes in Nevada, Oklahoma and Texas to understand their underlying causes. Deep fluid injection operations are thought to be particularly problematic, having caused induced earthquakes up to magnitude 5.8 in Oklahoma.
This team, led by Dr. Thomas Goebel – associate professor in the Center for Earthquake Research and Information (CERI), will evaluate the effect of rapid changes in fluid pumping rates on subsequent earthquake activity. Observations from a densely instrumented geothermal field in Nevada suggest that the rate of tiny, induced earthquakes increases when operational activity is abruptly stopped. They will investigate potential mechanisms of this surprising observation using long-term seismic, pressures and ground displacement data and will study induced stresses and seismicity in one specific, well-documented reservoir and then examine whether similar operations also promote induced seismicity in other areas such as Oklahoma.
Rapid stressing rate changes in energy reservoirs are analogous to aftershock triggering stresses and may help improve the understanding of aftershock frequency and magnitude distribution. This project will connect natural and induced earthquakes to produce a more general understanding of earthquake triggering mechanisms. A better documentation of faults driving stresses has profound implications for seismic hazard mitigation and operational aftershock forecasts. The proposed work will identify specific reservoir operations that amplify induced earthquake activity thereby promoting safer subsurface energy production and CO2 sequestration.
The research outcomes from this work will provide scientific guidance for decision makers and stake holders. Informing the public about induced and triggered seismicity is especially important because of the significant induced seismic hazard in the central and eastern United States. The project includes education and outreach activities in El Salvador which produces about 25% of its energy from geothermal reservoirs. The PI will promote earthquake preparedness in areas with induced seismicity and teach classes on induced seismicity and observational seismology in collaboration with the University of El Salvador.
Technical description of the project
Most seismic events are triggered by small natural and induced stress changes. Triggering intensity is thought to be governed by fluid pressure and dynamic stress. Yet, how a fault will react to specific stress perturbations is hard to predict due to the limited understanding of governing processes and stresses at seismogenic depths. Stress and pressure changes are more easily studied in shallow geothermal reservoirs with induced seismicity. This project investigates induced seismicity and earthquake triggering in the Blue Mountain geothermal reservoir in Nevada which experiences repeated seismicity spikes during annual maintenance shut-downs. Previous observations revealed a rapid seismic response to geothermal operations within ∼24 hours and a gradual decay of seismicity within ∼one week. The respective seismic events occur close to reservoir depths and exhibit strong spatial clustering around a high- volume injection well. The link between seismic events and abrupt operational changes has rarely been examined before and provides a particularly interesting target for the study of earthquake triggering. We will test the hypothesis that rapid injection changes lead to increased seismicity which is primarily driven by poro-elastic effects. Blue Mountain is especially well-suited to test this hypothesis because of its long operational history, and geologic records and dense seismic and pressure instrumentation. We will examine whether insights from Blue Mountain can be applied more generally by examining rapid injection rate changes and seismicity at larger scales across hydrocarbon basins in Oklahoma.
This work will connect induced and natural earthquake triggering to produce a more general understanding of underlying mechanisms. A better documentation of how faults react to different driving stresses has profound implications for seismic hazard mitigation and operational aftershock forecasts. The proposed work will provide insight into the coupling between injection rate changes and seismic activity in different geo-reservoirs, as well as stress conditions that promote earthquake triggering. Identifying specific reservoir operations that amplify a seismogenic response is important for safer subsurface energy production and CO2 sequestration.
For more information, contact Goebel at thgoebel@memphis.edu.
