Seismology From Space: Source Modeling of Anthropogenic and Tectonic Earthquakes Using Satellite Radar Observations
出版項
Ann Arbor : ProQuest Dissertations & Theses, 2023
說明
178 p
附註
Source: Dissertations Abstracts International, Volume: 84-11, Section: B
Advisor: Tiampo, Kristy F
Thesis (Ph.D.)--University of Colorado at Boulder, 2023
This item must not be sold to any third party vendors
Characterizing an earthquake is essential for strong ground motion estimates and probabilistic seismic hazard analysis. Earthquake source characterization is the estimation of parameters that describe the earthquake physics, generally using seismic data, geodetic data, or a combination of both. An earthquake rupture can be approximated by a point-source model or a finite-fault model. Point-source model parameters commonly include earthquake epicenter, depth, strike, dip, rake, and magnitude. Simple finite-fault models approximate the earthquake as a planar surface with parameters of length, width, depth, strike, dip, slip, and rake. More complex finite-fault models subdivide the fault plane into subfaults of either planar or triangular elements, each featuring their own set of parametersIn this dissertation, I seek to improve earthquake source models of both anthropogenic and tectonic earthquakes using synthetic aperture radar (SAR) data as my primary observation, together with a rapid and robust genetic algorithm for my first study and finite element model inversion scheme for my second and third studies. First, I present a point-source characterization method that utilizes differential SAR (DInSAR) line-of-sight (LOS) data and a genetic algorithm scheme to model the 2016 M5.0 Cushing, Oklahoma earthquake. My study reveals that the 2016 M5.0 Cushing earthquake is 3.2 km in depth, which is shallower than the 4.4 km depth estimated by the USGS and is consistent with the observed MMI VII damage in downtown Cushing.Next, I introduce the 3D finite element modeling (FEM) inversion method I developed to characterize earthquakes as finite-fault sources in a medium featuring complex rheology with an application to the 2014 M6.0 South Napa earthquake. I leverage published DInSAR line-of-sight (LOS) and GNSS 3D displacement data from Polcari et al. [2017] and a finite-fault model from Wei et al. [2015]. I use the published geodetic data to create a fused DInSAR-GNSS 3D displacement dataset and modify the published fault model to improve the fit to the fused data. My results suggest the M6.0 South Napa earthquake earthquake did not rupture the surface and is buried below a shallow layer of sediment at a depth of ≤ 3 km, in agreement with trench studies where no evidence of surface breaching shear dislocations was found [Brooks et al., 2017].Lastly, I employ my FEM inversion method, complex rheology, and fused ascending and descending DInSAR LOS displacements, pixel offset displacements, and 3D GNSS displacements to characterize the 2019 M7.1 Ridgecrest earthquake sequence. I exploit published surface rupture traces and the meshing technique to construct the most complex fault geometry of the earthquake to date. Through the modeling of a curved fault branching from the primary fault, I successfully recover a localized region of subsidence in the fault’s near field. I find that the mainshock features three main regions of large slip (6.9+ m), with depths ranging from 2 to 10 km. These regions of slip are bounded by the mainshock hypocenter and the subsequent aftershocks and appear to be related to spatially varying rheological properties
