USGS visual identity mark and link to main Web site at

 Automated Near-Real-TimeAnalysis 
        of Continuous GPS Data in the San Francisco 
        Bay Area for Deformation Alerts: Jessica Murray
Project Title: Automated Near-Real-TimeAnalysis of Continuous GPS Data in the San Francisco Bay Area for Deformation Alerts
Mendenhall Fellow: Jessica Murray, (650) 329-4864,
Duty Station: Menlo Park, CA
Start Date: December 1, 2003
Education:Ph.D., 2003 (Geophysics), Stanford University
Research Advisors: Wayne Thatcher, (650) 329-4810,; Ross Stein, (650) 329-4840,; Roland B├╝rgmann, (510) 643-9545,; Barbara Romanowicz, (510) 642-1844,

Project Description: Spatially-dense continuously recording Global Positioning System (GPS) networks are becoming more common world-wide. By 2008, 875 new continuous GPS stations will be installed in the United States as part of the Plate Boundary Observatory component of the Earthscope initiative. The data from such networks have in recent years enabled the identification of transient aseismic deformation in several locations. These events are of interest for a variety of reasons. For instance, aseismic slip relieves stress on a fault that might otherwise be released in a large earthquake. Kinematic models of time-varying fault slip can be used to refine theories concerning the underlying mechanisms and conditions (for example fault frictional properties) that give rise to aseismic slip. It is also thought, based on theoretical considerations and some observations, that aseismic slip may precede large earthquakes and thus provide a potentially measurable precursor to such events.

Transient deformation can be recognized in a position time-series for a GPS station as a deviation from a long-term trend. Many geodetic monuments are known to also experience time dependent noise of varying spectral character (e.g., flicker, random walk, or seasonal noise) which will generally obscure subtle deformation signals. Visual inspection of time series is therefore not a viable means for flagging ongoing aseismic deformation even for a small network, let alone one that consists of hundreds of stations. As continuous GPS networks grow, so does the need for an automated way to detect transient deformation signals in near-real-time. This is the primary goal of this Mendenhall project.
Deformation signals exhibit a spatial coherence not present in site-dependent time-varying noise. This feature may be exploited in a number of ways to identify anomalous displacements across a geodetic network. Existing continuous GPS networks are of widely varying station density and spatial extent. Moreover, these networks monitor a range of deformation sources (e.g., subduction zones, plate-bounding strike-slip faults, and areas where blind thrust faults may be numerous). These factors must be considered in the development of any transient deformation detection system. Therefore, one task of this research is to assess which means of analyzing a network's data is most consistently successful in transient detection, and another is to determine which method is most appropriate for a given network based on its station distribution and tectonic setting.

Go back to Previous Profile   Project Profiles   Next Profile Go forward to

Direct inquiries to Rama K. Kotra at
Page Contact Information: Mendenhall Postdoctoral Fellowship Program Web Team
Last modified: 16:08:31 Thu 13 Dec 2012