← Back to Sessions

Using Converted Phases to Investigate Induced Seismicity in the Midland Basin, Texas

Description: 

Wastewater injection triggers seismic activities in the Midland basin, west Texas, where the seismic events as large as M5.2 have been recorded. Mitigation requires understanding the role of shallow (<1.5 km subsurface) versus deep (2.3-3.4 km subsurface) injection, and the determination of causal factors is complicated by the uncertainties of hypocentral depths (+/-3 km), which may be associated with the complicated basin geology and inadequate network coverage over time. Here, we use body wave conversions common for shallow earthquakes in sedimentary basins to constrain hypocentral depths. We first generate a series of synthetic seismograms to gain an intuitive understanding of the subsurface characteristics that lead to the converted phases, and investigate how the waveform is influenced by various source parameters and velocity models. Sonic logs show that the Silurian-Devonian+Ellenburger (SD+E) and the San Andres (SA) formations provide significant impedance contrasts between layers to generate converted body waves (sP). These sP phases are potentially identifiable to earthquake analysts with epicentral distance less than 10 km. The clarity of the converted phases depends on radiation pattern, focal depth, and geology along the propagation paths. Changes to quality factors (Qp and Qs) and density did not produce any notable changes to shape and signal-to-noise of the synthetic sP. Synthetic tests indicate that the relative timings of converted phases are directly controlled by earthquake depths, and the amplitude of converted phases is strongly tied to the amplitude of the S-wave. Events located in the basement will yield two main converted phases, one from the SD+E boundary and one from the SA. The velocity contrast at the top of the SD+E yields an identifiable converted phase whose S-sP time could be incorporated into earthquake location algorithms. Converted phases have the greatest clarity (signal-to-noise) at depths greater than 6 km and operationally should be picked on data between 2-10 km event station distances. Lastly, we provide comparisons between the synthetic waveforms and real data.

Session: Induced Earthquakes: Source Characteristics, Mechanisms, Stress Field Modeling and Hazards - II

Type: Oral

Date: 5/1/2024

Presentation Time: 11:15 AM (local time)

Presenting Author: Heather

Student Presenter: No

Invited Presentation: 

Authors