Concurrent with the development of unconventional oil and gas production, seismicity in the central United States has dramatically increased. Previous studies in many locations suggest that the earthquakes reactivate pre-existing faults, as a result of changing subsurface stresses due to injection of fluids from wastewater disposal wells. This sudden rise in seismicity rate motivates an assessment and possible mitigation of seismic hazard due to the proximity of induced seismicity to metropolitan areas like the Dallas-Fort Worth, Texas.

Successful assessment and mitigation of earthquake hazards requires estimates of ground motion parameters representing path attenuation, site effects and source characteristics constrained by observed ground shaking. The generalized inversion technique (GIT) is a well-known spectral method that separates the three contributions based on a database composed of accelerometric recordings from many small to moderate sized earthquakes.

This dissertation aims to better understand seismic hazards in the Fort Worth Basin (FWB) and source physics of shallow-depth earthquakes that have been shown to be induced by regional wastefluid injection. The GIT is used to separate the three contributions (path, site, and source) from the observed ground motions to yield seismic hazard assessment and source discrimination of induced earthquakes. Since there are no hard-rock recording sites within the FWB, a new method is developed for the application of GIT under this circumstance, called a site correction method. Here, the efficacy of modified GIT is validated and compared against similar estimates using the empirical Green’s functions technique, which isolate source property using co-located large and small events.

In these studies, we find that the GIT derived seismic attenuation suggests the presence of a mid-crustal boundary and partially fluid-saturated material. The GIT site amplification functions document maximum amplification as a high as a factor of 5, slightly larger than the amplification of 3 estimated at the resonant frequency of the fundamental wavelength corresponding to 30 m depth. The site amplification (3 or 5) may be consistent with the thick sequences of sediments in the basin. At the resonant frequency, the GIT amplification is validated against horizontal-to-vertical ratio site functions and synthetic site responses from VS30 data and correlates with geologic conditions. Average stress drop estimates from FWB earthquakes are ~5 MPa, similar to stress drops from tectonic intraplate earthquakes that range from 1 to 10 MPa. All of these results lead to enhanced earthquake hazards for induced earthquakes in the basin structure.

The Dallas-Fort Worth Airport sequence, which occurred shortly after the initiation of injection on a fault close to the well, shows a lower mean stress drop (~1 MPa) than other FWB earthquake sequences (~5 MPa). The Airport stress drops increase with distance from the injector within the first 1.5 km, suggesting a possible fluid effect on the induced earthquake rupture predicted by direct triggering of earthquake via rapid pore pressure diffusion. This effect suggests that stress drop variations with respect to short distances from an injector could be used to discriminant between injection-induced earthquakes and tectonic events.

Degree Date

Spring 5-15-2021

Document Type


Degree Name



Earth Science


Brian Stump

Second Advisor

Heather DeShon

Subject Area

Earth, Atmospheric and Marine Sciences

Number of Pages




Creative Commons License

Creative Commons Attribution-Noncommercial 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License