COULOMB STRESS ANALYSIS OF THE SITES HOSTING THE THREE LARGEST RECORDED EARTHQUAKES IN OKLAHOMA HISTORY: PAWNEE, OK AND PRAGUE, OK

Abstract

Induced seismicity has been identified as a major concern in areas of hydrocarbon production, subsurface carbon sequestration, and enhanced geothermal systems, among others. Induced earthquakes are now commonly seen in many places around the world, including Oklahoma. In Oklahoma, the frequency of earthquakes with a Mw 3.0+ surged from just 2 per year in 2009 to 579 in 2014, peaking at 903 in 2015 before decreasing to 194 by 2018 and 16 in 2023. Wastewater injection was widely regarded as a significant contributor to this dramatic rise in seismic activity across the state. Most of the seismic activity has been located in the north-central portion of the state. The two sites of interest in this study are in Pawnee County, OK, and near Prague, OK. In both cases, a Coulomb stress analysis is being conducted. Coulomb modeling is used to determine how slip on faults can impact stress changes on surrounding faults, which can help determine where aftershocks patterns should be, how one earthquake impacts another, and the overall stress transfer patterns in these locations.The first area of interest is the region termed the Pawnee Triple Junction (PTJ) in north-central Oklahoma, which consists of the Sooner Lake Fault (SLF), the Stillwater Fault (SWF), and a newly identified fault zone informally named the Northeast Trending Fault Zone (NEFZ). This region was the site of Oklahoma’s largest recorded earthquake—a Mw 5.8 tremor in 2016—along the SLF and there are two main hypotheses for the production of this event that are of interest here. The first hypothesis is that aseismic slip along the NEFZ, resulting from wastewater injection, triggered the rupture along the SLF. The second hypothesis is that slip along the SLF itself triggered the Mw 5.8 rupture along the SLF. These will both be tested and compared to see if either, both, or neither are probable hypotheses. A second site of interest in this study is a series of faults just north of Prague, OK, the Wilzetta Fault System (WFS) and the Meeker-Prague Fault (MPF). These two faults run parallel to one another for a portion of their lengths and have ruptured two mainshocks in the past 14 years. The first occurred on the MPF on November 6, 2011, and was a Mw 5.7 event. The second occurred on the WFS on February 3, 2024, and was a Mw 5.1 event. The 2011 event, similarly to the SLF mainshock, is attributed to wastewater injection as a byproduct of hydrocarbon collection. The 2024 event has had little research done on it, so this thesis is aiming to increase the understanding of this event. To investigate stress interactions in the vicinity of the SLF and Prague, OK sites, a Coulomb stress analysis was conducted with the goal of assessing the stress transfer related to the two hypotheses discussed above as well as for the three mainshocks in these sites. Additionally, for Prague, OK this same analysis was done to assess how much how much stress the 2011 event transferred to 2024 event’s epicentral region and how these two ruptures differentiate from one another with a focus on aftershock pattern differences. We utilized previously collected seismic data from the Oklahoma Geological Survey to better identify fault geometries and complexity. Software such as ArcGIS Pro, Coulomb Modeling Software 3.4 in MATLAB, and Fault Slip Potential 2.0 were used in the study to better understand the interplay of stress and seismicity in the region. The results of the area around the SLF showed the likelihood of the two foreshock hypotheses that are discussed in this thesis. It helped to quantify that slip on the NEFZ and/or SLF could have contributed to the SLF Mw 5.8 rupture in 2016. This was concluded based on Fault Slip Potential 2.0 results showing how much Coulomb stress change was calculated for the SLF to slip and Coulomb Modeling Software 3.4 results showing how much Coulomb stress was added to the hypocenter of the Mw 5.8 rupture along the SLF from each of these foreshocks. Both of these outputs were shown in the form of MPa of stress and by comparing them it was possible to see if these hypotheses were able to cause the mainshock to occur. If the hypothesized foreshock did not produce the same or more added Coulomb stress then it was determined that it was invalid since it would not have caused enough of a stress increase for the fault to slip. If the hypothesized foreshock produced the same or more added Coulomb stress increase then it was determined the hypothesis was valid since the foreshock generated enough Coulomb stress increase for the fault to slip. Comparing the Coulomb stress values for both foreshock hypothesis allowed for the identification of the more probable hypothesis as well. Whichever one added a higher amount of Coulomb stress during the Coulomb modeling portion would be interpreted as the higher probability foreshock since it caused more stress to accumulate at the hypocenter of the mainshock. Additionally, these results helped to assess the time-dependent aftershock distribution along the SLF based on ArcGIS Pro mapping and Coulomb stress change plots from Coulomb Modeling Software 3.4. For Prague, OK these results helped to determine the interplay between the 2011 and 2024 events and how they produced different aftershocks patterns. These findings offer insights into the stress patterns that caused these aftershocks in the locations they did since the areas of increased Coulomb stress are the areas we expect to see aftershocks. In addition to this, these results also helped to better explain the observed aftershock pattern differences in the two sites and how these two events interplayed to some extent. The results from the modelling experiments of the two study sites provided insight into the questions proposed in this thesis. It was able to test the two SLF foreshock hypotheses, construct Coulomb stress change plots to compare to aftershock data for Pawnee and Prague, OK, as well as compare the Mw 5.7 Prague 2011 earthquake with the Mw 5.1 Prague 2024 earthquake. Each of these aspects of this research helped to contribute to the bigger picture of the propagation of these earthquakes in Oklahoma and the role co-seismic Coulomb stress change plays a role in that propagation.