Shallow injection may cause seismicity in the Delaware Basin, Texas

Abstract:

Beginning in 2014, there has been a surge in seismicity in the Delaware Basin, Texas. Many efforts focused on the spatiotemporal correlation with injection/production activities. However, the underlying physical processes that induce the seismicity remain unclear. Here, we compiled regional wastewater injection and oil/gas extraction volumes from 2010 to 2020 at ~ 2500 sites within the basin. Wastewater injection is mostly targeted in a shallow permeable sandstone formation, except for a subset of ~ 10 deep injections located in Culberson. The shallow injected volume is comparable to that of oil/gas extraction from the underlying low-permeability shale, a hydraulic barrier between the shallow sandstone and the deep basement where earthquakes occur (TexNet). Due to the shale high water cut, the actual fluid extraction volume is several times larger than reported oil/gas volumes. To understand the crustal stress evolution due to fluid injection and production, we build a unified poroelastic model that simulates the processes of shallow injection-caused fluid diffusion in sandstone and deep extraction-caused compaction in shale. We also build an elastic loading model to account for the crustal flexure due to regional mass change within the basin. Preliminary results show that extraction-dominated unloading and shale compaction cause elastic and poroelastic stress changes in the basement, both resulting in a negative Coulomb Failure Stress (CFS) distribution with the magnitude of less than few KPa and few bars, respectively for normal receiver faults. Whereas, shallow injections cause positive poroelastic CFS that favors normal faulting. Using a similar Biot coefficient for the sandstone and shale, the injection caused CFS in the basement is of a smaller magnitude than that of shale extraction due to the proximity of larger extraction to the basement. To explain the widespread Delaware seismicity, a high shale rock rigidity and thus low Biot coefficient may be required to reduce poroelastic coupling that could decrease the production induced basement CFS. We suggest that the Delaware seismicity is possibly caused by injection-induced poroelastic stresses if the shale has a low Biot coefficient, although other possibilities cannot be excluded, e.g., fluid pathways that connect shallow and deep formations.