Analysis of site effects on induced seismic monitoring: An Oklahoma case study

Induced seismicity from activities such as wastewater injection has raised growing concerns about the structural integrity of surface and subsurface structures, public safety, and environmental risks. Several induced events in Oklahoma have caused structural damage, highlighting the need for proactive seismic hazard assessment and mitigation strategies. A critical component of understanding seismic risk is analyzing site effects—variations in ground motion caused by local geologic and geomorphic conditions such as soil properties, rock type, impedance contrasts, and depth to bedrock. These factors can amplify seismic waves, increasing shaking intensity and prolonging wave duration. Although induced seismicity has become more frequent in Oklahoma due to fluid injection from oil and gas operations, relatively few studies have explored site effects in the region. This study investigates site effects by (1) analyzing waveform distortions from real seismic events recorded by a temporary station and (2) simulating seismic waveforms using 2D finite difference methods. The velocity models used in the simulations were constructed using well logs and geologic data provided by Oklahoma State University and industry partner Bob Springman. Results show consistent P-to-S wave conversions within the P-wave coda and S-wave reverberations originating around 1,200 meters depth, which coincides with the Oologah Limestone—a thick, regionally extensive carbonate unit. Previously, such conversions were attributed to the granite basement, but this study demonstrates they instead originate at the high-impedance contrast boundary between the Oologah Limestone and the overlying Cleveland Sandstone. These conversion phases strongly influence surface recordings and represent a critical site effect that should be incorporated into hazard models. The findings offer a revised interpretation of seismic wave behavior in Oklahoma and suggest that similar wave conversions may occur across other regions where the Oologah Limestone is present. This study emphasizes the importance of integrating detailed geologic data—particularly well log constraints—into velocity models to better simulate wave propagation and improve ground-motion predictions in regions affected by induced seismicity.