Leakage and Increasing Fluid Pressure Detected in Oklahoma's Wastewater Disposal Reservoir

Abstract

The Arbuckle Group is the principal reservoir used for wastewater disposal in Oklahoma. In Osage County—a seismically quiet part of the state—continuous measurements of fluid pressure reveal that pressure in the reservoir is increasing by at least 5 kPa annually and sometimes at a much higher rate. Tidal analysis reveals that fluid level changes lead the local strain tides, with no apparent influence from transient permeability changes; this indicates a response that is inconsistent with flow in a radially extensive, confined reservoir. We investigate whether this is due to vertical flow to the water table, vertical flow within the Arbuckle, or local distortions from fractures. While none of these alternative models can fully explain both the observed tidal phases and amplitude ratios, the observed response to teleseismic waves supports a mechanism related to leakage rather than fracture effects. At this location fluid influx associated with wastewater disposal is offset by migration into surrounding layers, which include the Precambrian basement below. Thus, our findings suggest the need to monitor for changes in the induced seismicity hazard, while pore pressures increase in a leaky disposal reservoir.

Induced Seismicity in the Delaware Basin, Texas

“…With those assumptions there would be the equivalent of~5.1-cm fluid head decline associated with withdrawal, equivalent to an~0.5-kPa hydrostatic pressure change. Even given our conservative assumptions, this estimated pressure change is roughly 2 orders of magnitude smaller than stress changes observed for natural earthquake triggering (e.g., Gomberg & Johnson, 2005), about a factor of 3 smaller than the average stress changes associated with either atmospheric pressure loading or solid earth tides (e.g., Barbour et al, 2019), with only scant evidence of tidetriggered seismicity (e.g., Cochran et al, 2004;Vidale et al, 1998). A similar methodology that produced a larger estimated stress change was previously used to argue that seismicity in the Raton Basin was induced by WD rather than production (Rubinstein et al, 2014).…”

The Induced Mw 5.0 March 2020 West Texas Seismic Sequence

“…As hydrocarbon production is spatiotemporally related to WD activities, it is difficult to isolate the individual contributions of each type of operation. However, previous conservative stress estimates between 2007 and 2017 in the Delaware Basin found production to have resulted in a ∼0.5 kPa hydrostatic pressure change (Skoumal et al., 2020), much smaller than the changes associated with atmospheric pressure loading or solid earth tides measured in deep disposal reservoirs (e.g., Barbour et al., 2019). The only other industrial operation that we are aware of that could explain the seismicity is WD.…”

Elevated Seismic Hazard in Kansas Due to High‐Volume Injections in Oklahoma

“…Although direct observation of subsurface pressure change is limited, this mechanism is supported by numerical simulations (Fan et al, ; Keranen et al, ; Shirzaei et al, ). Monitoring of Arbuckle formation pressure provides strong evidence linking the fluid injection and induced seismicity within the U.S. midcontinent through pore pressure changes (Ansari et al, ; Barbour et al, ; Kroll et al, ; Peterie et al, ). However, the relationship between fluid injection and pore pressure evolution is complex in time and space (Zhai et al, ; Zhai & Shirzaei, ).…”