Multi-Objective Geophysical Inversion for Earth Structure and Earthquake Parameters

Open Access
Chai, Chengping
Graduate Program:
Doctor of Philosophy
Document Type:
Date of Defense:
February 28, 2017
Committee Members:
  • Charles J. Ammon, Dissertation Advisor
  • Charles J. Ammon, Committee Chair
  • Sridhar Anandakrishnan, Committee Member
  • Andrew Nyblade, Committee Member
  • Parisa Shokouhi, Outside Member
  • Derek Elsworth, Committee Member
  • Geophysical Inversion
  • United States
  • Antarctica
  • Sumatra
  • Earth Structure
  • Earthquakes
Earth structure and earthquake parameters are not only fundamental to earthquake studies but also critical to hazard assessment. Estimates of Earth structure and earthquake parameters can help us attacking important scientific questions such as how the surface geology features relates to subsurface structure, what the relationship between the seismicity and subsurface structure changes, what is the crustal structure beneath polar ice, how reliable are our estimates of subsurface structure parameters, and how does earthquake rupture in a complex fault system. For Earth structure imaging, increasing station coverage provide large quantities of seismic observations. These observations are not free of noise and some suffered from scattering noise. We developed techniques to reduce scattering noise by incorporating observations from adjacent stations and to extract signals from deeper structure using better-determined shallow prior information. Multi-objective 3D inversions using the noise-reduced signals and complementary geophysical observations were conducted to produce reliable subsurface images beneath the western United States and the eastern United States. Stochastic inversions in the polar regions estimated both the first-order subsurface structural variations and associated uncertainties. Additionally, we applied a recently developed technique to relocate earthquakes in the off-Sumatra region. Reliably estimated Earth structure and earthquake parameters can advance our understanding of the correlation between seismicity and subsurface structure. In the meantime, earthquake hazard assessment can benefit from these newly calculated models and parameters. We also made efforts on improving the accessibility of our results. In the first chapter, I introduce the scope of the dissertation, challenges and opportunities in related study areas, and summaries of individual chapter. Future directions that may be approached based on studies in this dissertation are summarized in the last chapter. In the second chapter, we use P wave receiver functions from the western U.S. and adjacent regions to construct a receiver function wavefield interpolation scheme that helps to equalize the lateral sampling of the receiver functions and the surface wave dispersion and to greatly simplify the receiver functions. Spatial interpolation and smoothing suppress poorly sampled and difficult to interpret back azimuthal variations and allow the extraction of the first-order features in the receiver function wavefield, including observations from several ray parameter ranges. We combine the interpolated receiver functions with Rayleigh wave dispersion estimates and surface gravity observations to estimate the 3-D shear wave speed beneath the region. Speed variations in the 3-D model correlate strongly with expected geologic variations and illuminate broad-scale features of the western U.S. crust and upper mantle. The model is smooth, self-consistent, and demonstrates the compatibility of the interpolated receiver functions and dispersion observations. In the third chapter, we simultaneously invert smoothed P-wave receiver functions, Rayleigh wave phase and group velocity measurements, and Bouguer gravity observations for the 3D shear wave speed beneath the eastern U.S. and the northern Mississippi Embayment regions. Using period-coverage-broadened (3-250 s) surface-wave observations and spatially smoothed receiver functions, our velocity models are robust, reliable and rich in detail. Our shear-wave velocity models fit all three types of observations well. The resulting velocity model in for the broad region of the eastern U.S. shows thinner crust beneath New England, the east coast and the Mississippi embayment. Relatively thicker crust was found beneath the stable North America craton. Relatively slower upper mantle was imaged beneath New England, the east coast and western Mississippi embayment. We also explored the relationship between the subsurface structure and seismicity in the eastern U.S. We found earthquakes often locate near regions with seismic velocity variations, but not universally. Not all regions of significant subsurface wave speed changes are loci of seismicity. Eastern U.S. thrust faulting weakly correlates with slower upper mantle speed along the coast and within New England and southeastern Canada. In the northern Mississippi Embayment, the region of the large earthquakes in 1811-12, we imaged a relatively fast lower crust and a relatively slower uppermost mantle. Reverberations of teleseismic compressional (P-) waves within a glacier or ice sheet may mask signals associated with crustal structure beneath the ice. In the fourth chapter, we remove the signal associated with the ice from teleseismic P-waves using a wavefield downward continuation and decomposition technique that depends on known ice layer properties such as ice thickness, velocity, and attenuation. We test the method using data from nine stations in Antarctica and one station in Greenland. We deconvolve the downward-continued seismic wave vectors to create P-wave receiver functions that minimize the ice-layer reverberations in order to better measure signals from deeper structures. The subsurface P-wave receiver functions have similar sensitivities to crustal structure as those calculated from stations installed on bedrock. Synthetic experiments indicate subsurface P-wave receiver functions can constrain crustal structure more tightly than surface P-wave receiver functions when ice layer properties are known. We model the subsurface P-wave receiver functions using a Markov chain Monte Carlo inversion and constrain the product of crustal thickness and the column-average crustal-slowness beneath the stations. Our subglacial shear-speed and thickness estimates are consistent with previous investigations at most stations. At station SUMG in south-central Greenland, our results suggest a thicker crust than from previous estimates. In the fifth chapter, we analyze aftershocks of the 2012 Off-Coast of Sumatra Earthquake Sequence. The aftershocks were relocated using surface-wave cross-correlations to improve relative location precision. Unlike observed patterns along mature transform faults in other regions, the relocated aftershocks seldom align along simple linear trends that are compatible with the strikes of the faults as estimated by the GCMT catalog. The relocation of roughly 60 moderate-earthquake epicentroids suggests that the faulting involved in the 2012 earthquake sequence occurred in a region populated with many short fault segments. The inferred complex fault system agrees with a recent bathymetry survey. Statistical analysis and temporal variations of aftershocks show a relatively low number of aftershocks but possibly a relatively slow decay of the aftershocks. The patterns in the aftershocks suggest that the formation of the boundary and eventual localization of deformation between the Indian and Australian plate is a complicated process.