Electronic Theses and Dissertations

Identifier

780

Date

2013

Date of Award

1-10-2013

Document Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Earth Sciences

Committee Chair

Robert Smalley

Committee Member

Charles Langston

Committee Member

Chris Cramer

Committee Member

John Paul Puchakayala

Abstract

Seismic Array Processing of High-Rate GPS DataCombinations of high-rate GPS (HRGPS) and broadband seismograms show great potential for augmenting the spatial coverage of existing seismic networks. GPS time series' use as seismograms is limited by high noise levels. The primary noise source in 1 Hz GPS seismograms is GPS multipath. GPS multipath is highly dependent on local station conditions and incoherent in an array setting, while seismic waves travelling across an array are coherent and have predictable phase spectra. We propose that the effects of GPS multipath in HRGPS seismograms can be reduced by seismic array processing techniques that enchance the coherent and predictable seismic signals. We use f-k beamforming to demonstrate GPS multipath reduction within an array, and phase-match filtering to reduce multipath in 1 Hz GPS seismograms.We apply f-k beamforming to 1 Hz GPS seismograms for the December 26, 2004, Sumatra earthquake for GPS stations in central North America. Love wave phase velocity for central North America is recovered. Results agree well with current models for continental Love phase velocity dispersion between 20 and 300 seconds. GPS multipath is observed to be incoherent within the array, but its presence still introduces error into the calculations. Methods for reducing multipath in the seismograms prior to additional processing are needed.We use phase-match filtering to reduce GPS multipath in 1 Hz GPS seismograms from the Sumatra, Maule, and Tohoku-Oki earthquakes. We demonstrate significant reduction of multipath for both teleseismic and nearfield data. Tests with multipath-only GPS seismograms suggest that phase-match filtering is useful for 1 Hz GPS seismograms with signal-to-noise ratios of 2 or greater.Using networks of HRGPS stations as seismic arrays with array processing, such as beamforming and phase-match filtering, shows potential for the incorporation of HRGPS seismograms into seismic datasets. Other methods more robust than f-k beamforming and phase-match filtering exist and may further improve future results.

Comments

Data is provided by the student.

Library Comment

dissertation or thesis originally submitted to the local University of Memphis Electronic Theses & dissertation (ETD) Repository.

Share

COinS