Date of Award

Spring 1-1-2017

Document Type


Degree Name

Doctor of Philosophy (PhD)

First Advisor

Michael H. Ritzwoller

Second Advisor

Shijie Zhong

Third Advisor

Anatoli Levshin

Fourth Advisor

Anne F. Sheehan

Fifth Advisor

Michael A. Calkins


Ambient noise tomography has been well developed over the past decade and proven to be effective in studying the crust and upper mantle structure beneath the Earth’s continents. With new seismic array deployments beginning in the oceans, the application of the tomographic methods based on ambient noise observed at ocean bottom seismometers (OBSs) has become an important topic for research. In this thesis, I investigate the application of ambient noise tomography to oceanic bottom seismic data recorded by the Cascadia Initiative experiment across the Juan de Fuca plate. With higher local noise levels recorded by OBSs, I find that traditional data processing procedures used in ambient noise tomography produce measurable Rayleigh wave Green’s functions between deep ocean stations, whereas the shallow water stations are severely contaminated by both tilt noise and compliance noise and require new methods of processing. Because the local noise level varies across the study region, four semi-independent studies are conducted to both utilize the quieter deep-water stations and to address the problem posed by noisy shallow water stations. First, I construct an age-dependent shear wave speed model of the crust and uppermost mantle with 18 deep-water stations near the Juan de Fuca Ridge. The model possess a shallow low shear velocity zone near the ridge and has its sedimentary thickness, lithospheric thickness, and mantle shear wave speeds increase systematically with age. Second, I investigate the locations and mechanisms of microseism generation using ambient noise cross-correlations constructed between 61 OBSs and 42 continental stations near the western US coast and find that the primary and secondary microseisms are generated at different locations and possibly have different physical mechanisms. Third, I show that tilt and compliance noise on the vertical components of the OBSs can be reduced substantially using the horizontal components and the differential pressure gauge records. Removal of these types of noise improves the signal-to-noise ratio of ambient noise cross-correlations significantly at beyond 10 sec period. Lastly, I present a new single-station method to estimate the microseism Rayleigh wave strength and directionality based on the horizontal-to-vertical transfer function. The high spatial and temporal resolution of this method may open up the microseism Rayleigh waves for a wider range of studies.