Two types of seismicity accompanying hydraulic fracturing in Harrison County, Ohio - implications for seismic hazard and seismogenic mechanism
Abstract
While induced seismicity in the United States has mainly been attributed to wastewater disposal, Eastern Ohio has provided cases of seismicity induced by both hydraulic fracturing (HF) and wastewater disposal. In this study, we investigate five cases of seismicity associated with HF in Harrison County, OH. Because of their temporal and spatial isolation from other injection activities, this provide an ideal setting for studying the relationships between high pressure injection and earthquakes. Our analysis reveals two distinct groups of seismicity. Deeper earthquakes occur in the Precambrian crystalline basement, reach larger magnitudes (M>2), have lower b-values (<1), and continue for weeks following stimulation shut down. Shallower earthquakes, on the other hand, occur in Paleozoic sedimentary rocks 400 m below HF, are limited to smaller magnitudes (M<1), have higher b-values (>1.5), and lack post-stimulation activity. We seek the physical explanation of observed difference in earthquakes character and hypothesize that the maturity of faults is the main factor determining sequences b-values. Based on published results of laboratory experiments and fault modeling, we interpret the deep seismicity as slip on more mature faults in the older crystalline rocks and the shallow seismicity as slip on immature faults in the younger, lower viscosity sedimentary rocks. This suggests that HF inducing seismicity on deeper, more mature faults poses higher seismic hazards. The analysis of water and gas production data from these wells suggests that wells inducing deeper seismicity produced more water than wells with shallow seismicity. This indicates more extensive hydrologic connections outside the target reservoir, which may explain why gas production drops more quickly for wells with deeper seismicity. Despite these indications that hydraulic pressure fluctuations induce seismicity, we also find only 2-3 hours between onset of stimulation of HF wells and seismicity that is too short for typical fluid pressure diffusion rates across distances of 1 km. We conclude that a combination of pore fluid pressure changes and poroelastic stress changes are responsible for inducing shear slip during HF.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2017
- Bibcode:
- 2017AGUFM.S14A..05K
- Keywords:
-
- 4475 Scaling: spatial and temporal;
- NONLINEAR GEOPHYSICS;
- 7209 Earthquake dynamics;
- SEISMOLOGY;
- 7223 Earthquake interaction;
- forecasting;
- and prediction;
- SEISMOLOGY;
- 8164 Stresses: crust and lithosphere;
- TECTONOPHYSICS