Linearized and nonlinear travel time tomography for upper crustal velocity structure of the western Great Basin

In this dissertation, I use linearized and nonlinear travel time tomographic methods to estimate upper crustal velocity structure in the western Great Basin. The northern and southern fringes of this region have seen quite extensive seismic profiling and 2-dimensional velocity modeling works, while the central part, with the exception of Long Valley caldera, has remained like a gap in terms of upper crustal velocity modeling. I obtain an upper crustal {dollar}Psb{lcub}g{rcub}{dollar} refractor velocity map to bridge this gap of information in the central part of the western Great Basin.; Vis-a-vis the dilemma of the linearized methods' being closely dependent on a priori information and the fully nonlinear methods' being oftentimes computationally intractable, I adopt a compromise between these two using different combinations of both as the problem at hand warrants. I use a fully linearized least squares method of travel time inversion for hypocenters and local three-dimensional velocity model in the Eureka Valley area, a conjugate-gradient based method for mapping the {dollar}Psb{lcub}g{rcub}{dollar} refractor at a regional scale and a fully nonlinear method of optimization by simulated annealing for obtaining a starting velocity model in an area where no prior model exists.; In the first part of this dissertation, I use a tomographic scheme in order to invert about 200,000 {dollar}Psb{lcub}g{rcub}{dollar} travel times for an upper crustal refractor velocity mapping in the western Great Basin. The results feature a north-northwest trending upper crustal low velocity corridor that connects the eastern California shear zone with the Furnace Creek-Fish Lake Valley fault. A positive correlation between {dollar}Psb{lcub}g{rcub}{dollar} velocity and gravity along the dominant structural trend of the western Great Basin (north-northwest) has also been inferred.; The second part of the dissertation proposes an inversion methodology composed of an in-tandem combination of nonlinear simulated annealing optimization and linearized inversion in order to solve the problem of determination of hypo-center locations and a 3-d velocity model of the Eureka Valley area. The velocity model derived adequately describes broad structural features such as a sedimentary basin at the local scale. The hypocenter locations, however, are a little scattered.; Finally, I do elaborate two-dimensional synthetic tests to determine an efficient definition of the critical temperature to be used in the cooling schedule of simulated annealing optimization for determination of source locations and velocities. I define the critical temperature on the triple criteria of change of shape of the error-versus-iteration curve for short runs at different temperatures, average least square error as a function of temperature, and number of accepted models as a function of temperature. I also propose a heuristic combination of the conventional "acceptance" criterion and a "rejection" criterion for probabilistic acceptance of models. Tests for separate velocity model and source location determinations proved promising.