Maturity of nearby faults influences seismic hazard from hydraulic fracturing

Abstract

Understanding the causes of human-induced earthquakes is paramount to reducing societal risk. We investigated five cases of seismicity associated with hydraulic fracturing (HF) in Ohio since 2013 that, because of their isolation from other injection activities, provide an ideal setting for studying the relations between high-pressure injection and earthquakes. Our analysis revealed two distinct groups: () deeper earthquakes in the Precambrian basement, with larger magnitudes (M > 2), b-values < 1, and many post-shut-in earthquakes, versus () shallower earthquakes in Paleozoic rocks ∼400 m below HF, with smaller magnitudes (M < 1), b-values > 1.5, and few post-shut-in earthquakes. Based on geologic history, laboratory experiments, and fault modeling, we interpret the deep seismicity as slip on more mature faults in older crystalline rocks and the shallow seismicity as slip on immature faults in younger sedimentary rocks. This suggests that HF inducing deeper seismicity may pose higher seismic hazards. Wells inducing deeper seismicity produced more water than wells with shallow seismicity, indicating more extensive hydrologic connections outside the target formation, consistent with pore pressure diffusion influencing seismicity. However, for both groups, the 2 to 3 h between onset of HF and seismicity is too short for typical fluid pressure diffusion rates across distances of ∼1 km and argues for poroelastic stress transfer also having a primary influence on seismicity.

Earthquakes Induced by Hydraulic Fracturing Are Pervasive in Oklahoma

“…Considering the uncertainties in the velocity structure in this part of the state, we repeated this earthquake location procedure using the eight velocity models that were considered in central Oklahoma by Keranen et al () and selected the model (Table S2) that produced the lowest root‐mean‐square deviation. The maximum likelihood locations of the events are much closer to the well pads than the cataloged locations (Figure ), with most within ~2 km of the active well and many within ~1 km, similar to previously documented HF‐induced seismicity (e.g., Bao & Eaton, ;Kozłowska et al, ). However, the median location errors of ~1.4 km are still quite significant and limit the types of conclusions that can be reached (discussed in Section ).…”

“…However, if the Darold et al () locations were accurate, this finding would be surprising. These distances are significantly farther than previous cases of HF‐induced seismicity in which induced earthquakes occurred within ~2 km of the HF well (e.g., Bao & Eaton, ; Kozłowska et al, ). Additionally, the first earthquake we identified in the Eagleton sequence occurred within 20 min of HF initiation at the well, which would have required very rapid poroelastic stress changes and/or pore fluid pressure diffusion over significant distances.…”

“…These include wells that are stimulated in highly overpressured shale formations (Eaton & Schultz, ), in carbonate formations with high permeabilities (Schultz et al, ), and in proximity to the Precambrian basement (Skoumal et al, ). However, while these geologic parameters may influence the likelihood of an earthquake being induced, the only geologic parameter that has been suggested to influence the magnitudes of HF‐induced seismicity are wells in proximity to the Precambrian basement (Kozłowska et al, ). A similar conclusion linking the basement proximity of WD wells to the seismic moment was found by Hincks et al ().…”

“…Schultz et al (2018) found that~15% of HF wells within the Kaybob region of the Duvernay play (Alberta) were associated with induced seismicity. Kozłowska and Brudzinski (2018) found that between 10% and 33% of HF wells in four~20 × 20-km regions of Ohio had induced earthquakes. The HF occurrence rates calculated in our similar-sized regions in Oklahoma are also comparable (Figure 7b).…”

Hydraulic Fracture Injection Strategy Influences the Probability of Earthquakes in the Eagle Ford Shale Play of South Texas

“…About a quarter of the earthquakes occurred prior to the first cataloged earthquake in March 2017, indicating the usefulness of template matching for extending the seismicity catalog prior to when TexNet recording began in January 2017. The enhanced catalog of 2,823 earthquakes showed several short‐lived, spatiotemporally clustered events (Figure a), characteristic of HF‐induced seismicity (e.g., Kozłowska et al, ; Lei et al, ; Schultz et al, ; Skoumal, Ries, et al, ). These bursts of earthquakes, which we refer to as sequences, began in June 2014 and increased in prevalence during late 2016.…”