Passive seismoacoustic imaging from the seafloor to the lithosphere: Methods and applications to New Zealand and Ascension Island

Passive-source seismic methods often rely on the isolation of transient signals such as distant earthquakes from a pervasive background of ambient noise. In marine seismology, discriminating signals from noise is particularly complicated due to the efficient wave propagation characteristics of the ocean and sediments, and oceanographic noise sources including long-period ocean surface gravity waves. Furthermore, high-amplitude structural reverberations near the receiver modulate and obscure teleseismic arrivals targeted by the analyst. In this thesis I further the development of methods to both accommodate the signal-generated noise and utilize the rich ambient noise wavefield in the ocean. I apply these methods to image subsurface structure at scales from the shallow sediments to the lithospheric mantle beneath the South Island of New Zealand and Ascension Island.;I first utilize ambient noise in the form of infragravity waves and Rayleigh waves, which both sense shear structure at depth, in conjunction with reverberations in P-S wave receiver functions to model shallow sediment structure offshore New Zealand using a Markov Chain Monte Carlo algorithm.;I then turn my focus to the theory and application of Rayleigh/Scholte wave noise interferometry. First I investigate the effects of bathymetric variations on microseism-band modal propagation between two hydrophones moored off Ascension Island. I model the range-dependent dispersion observed in the noise correlation functions from Ascension data as the result of double mode-converted Scholte-Moho headwave propagation, and thereby demonstrate the feasibility of probing oceanic crustal and upper mantle structure using moored hydrophone data.;Lastly I apply a combination of ambient noise and teleseismic Rayleigh wave tomography to image the shear structure of the mantle lithosphere beneath the continental collision zone of the South Island of New Zealand. The resulting models include high-wavespeed anomalies potentially associated with subducted Pacific lithosphere and a relict Eocene passive margin, and a low-speed zone that correlates with Cenozoic surface volcanism.