Earthquakes induced by fluid injection: Implications for secure CO2 storage
Abstract
It is well understood that the injection of fluids into the subsurface can trigger seismic activity. Recently, the US unconventional gas boom has lead to an increase in the volumes of produced water being disposed in geological formations and a concomitant increase in triggered seismic events. This issue is especially pertinent for geologic carbon sequestration, where the injection volumes necessary to store the CO2 emissions from a typical coal-fired power station far exceed the volumes known to have triggered seismic activity. Moreover, unlike water disposal operations, where there is no strong buoyancy drive to return injected fluids to the surface, CO2 sequestration requires a sealing caprock to prevent upward CO2 migration. Induced seismic events may create or reactivate faults and fracture networks, compromising the hydraulic integrity of the caprock. Therefore, induced seismic activity at future CCS sites is of doubly significant, given both the direct seismic hazard and the risk to secure CO2 storage. With this in mind, we re-examine case histories of seismic activity induced by waste water disposal into sedimentary formations with the intention of learning lessons that can be applied to future CCS sites. In particular, we examine the spatial and temporal distributions of events to determine whether there are any rules-of-thumb that might be usefully applied when appraising and monitoring operations. We find that in all cases, at least some seismicity occurs at the depth of the injection interval, but the majority (~80% of events) occur at least 500m below the injection depth. Less than 2% of events occur more than 500m above the shallowest injection interval. This observation must be considered encouraging from a CCS perspective, where seismicity in sealing caprocks will be of greatest concern. However, without a phenomenological explanation for the relative lack of seismicity above injection depths, it cannot be guaranteed that such observations would be repeated at CCS sites. We also examine the lateral distance between induced events and injection wells. The maximum distance between wells and events will define a minimum radius of influence, a distance over which geomechanical appraisal and fault characterization studies must be carried out at future CCS sites. We find that 62% of events occur within 5km, and that 99% of events occur within 19km of injection wells. These case examples highlight the importance of seismic monitoring at future CCS sites. Of the two large-scale CCS sites to deploy microseismic arrays, both have detected induced seismic events. During 6 years of monitoring at Weyburn, ~100 events with magnitudes between -3.0 and -1.0 have been detected, while at In Salah more than 1000 events, with magnitudes as large as 1.0, have been detected during 6 months of monitoring. Combined the case examples from water disposal operations, these operations demonstrate the need for dedicated local seismic monitoring networks to be installed at future CO2 injection sites.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.S33C2438V
- Keywords:
-
- 7230 SEISMOLOGY Seismicity and tectonics;
- 0999 EXPLORATION GEOPHYSICS General or miscellaneous