Seismic structure of the Costa Rican subduction system from active-source onshore-offshore seismic data and imaging plate boundary processes at the Cascadia subduction zone offshore Washington.

摘要

The goal of this thesis is to use seismic methods, either wide-angle refraction or multi-channel seismic (MCS) reflection, to characterize the physical processes occurring at the subduction zones occurring offshore Costa Rica and Cascadia. The first two chapters use wide-angle refraction data to characterize lithospheric structure and velocities, based on the modeling of wide-angle refractions and reflections from the crust, Moho and upper mantle. They also use MCS data to characterize the uppermost structure that wide-angle refraction data alone cannot provide.;The first chapter uses both wide-angle refraction and MCS data to address the hypothesis that bending-related normal faulting, clearly imaged in the MCS data, provides a pathway for seawater to percolate down into the uppermost mantle and serpentinize it. This process causes a reduction in the seismic p-wave velocity in the upper mantle, which can be detected by wide-angle refraction analysis. We found the upper 1-2 km of the mantle has reduced velocities of 7.5 - 7.6 km/s in the area of pervasive normal faulting within the CNS-2 segment, and regular upper mantle velocities of 8.0 - 8.2 km/s in the CNS-1 segment, which lacks pervasive normal faulting. Our results suggest a link between bending-related large-offset normal faults seen in bathymetric and MCS reflection data in subduction trenches and serpentinization of the upper mantle.;The second chapter uses both wide-angle refraction and MCS data like the first chapter but addresses the hypothesis that juvenile continental crust is created at some volcanic arcs. We addressed this hypothesis by creating a lithospheric velocity and structural model for the Central American subduction system through Costa Rica. This model allows us to estimate the seismic velocity, structure, infer bulk composition (from seismic velocities), and estimate a magmatic flux rate for the volcanic arc. We found a total crustal thickness of ~44 km and mid-to-lower-crustal velocities of ~6.5 -7.2 km/s under the active arc. Our modeled lower crustal velocities and densities fit approximately at or within the error bounds for bulk continental crust. Using the crustal structure from our velocity model, we were able to determine a magmatic production rate of ~80 km3/km/Ma for the Costa Rican volcanic arc.;The third chapter uses iterative pre-stack velocity analysis to create pre-stack depth migrated seismic images and velocity models. The PSDM reveal: (1) landward vergence of faults; (2) extensive BSR's; (3) a zone of low acoustic impedance underneath the Pleistocene accretionary prism; (4) a lack of a strong decollement reflection throughout the section; (5) discontinuous reflectivity from the subducting oceanic crust; (6) and a shallow dip of the top of the subducting oceanic crust ~1.5 - 4° beneath the Pleistocene accretionary prism. From the inferred porosity variations from our velocity model we are able to estimate the volume of expelled fluid from the Pleistocene accretionary prism. We estimate that over the ~32 km along the deformation front covered by our seismic lines that ~ 750 +150/-110 km3 of expelled fluid has been released at a rate of ~ 1.1 mm/yr.