Case Study in Detection of Transient Crustal Deformation
Abstract
Detecting anomalies in GPS daily coordinate time series is a matter of considerable importance and concern in geodetic research, especially with the proliferation of GPS stations in regions of active tectonic deformation. With funding from a NASA MEaSUREs project, the Scripps Orbit and Permanent Array Center and Jet Propulsion Laboratory generate time series for over 1800 global and regional GPS stations, with the majority of stations in Western North America. The time series are analyzed with the now standard fitting of slopes (velocities), offsets, periodic (annual and semiannual terms) terms, and (exponential or logarithmic) postseismic parameters. For the purposes of quality control and detection of interesting signals in the time series, we have developed a few simple algorithms to scan the time series every week to detect instances when the model does not represent the data well. Our algorithms look for the following: (1) outliers, (2) “signals”, and (3) residual linear trends, each of which signifies the need for further modeling/inspection of the data. The flagged stations are then posted on the GPS Explorer data portal where we look for spatial clustering to gain insight into any un-modeled geophysical or other physical effects. In a separate effort, we have been investigating a strain transient centered on the Obsidian Buttes fault just south of the Salton Sea detected from analysis of a long-history of survey-mode GPS measurements taken over the last 25+ years. Using interpolated site velocities over a 0.01 degree grid for field survey data collected before and after 2001 followed by 2-D strain analysis, suggests accelerated slip on the Obsidian Buttes fault between 2001 and 2009, equivalent to a magnitude 5.3 earthquake. The associated strain transient has caused an accelerated fault parallel slip rate deficit of 5-7 mm/yr along the southern San Andreas fault near Bombay Beach, corresponding to a Coulomb stress increase since 2001 of ~0.5 MPa centered along the 2009 Bombay Beach seismic swarm. The ongoing strain transient is loading up the transtensional faults in the Salton Sea and the southern San Andreas fault, presumably reducing the time to the expected rupture. We verified that we have observed a real transient by examining the few nearby continuous GPS time series and seeing that their motion agrees well with the motion predicted by our fault model. Although the simple algorithms that we use routinely to detect anomalies in daily position time series do a good job of flagging clusters of stations with anthropogenic, postseismic, and volcanic effects, they did not flag the few stations affected by the strain transient on the Obsidian Buttes fault. We are now investigating what if any changes can be made to our operational approach to detect such transients.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2010
- Bibcode:
- 2010AGUFM.G23A0813C
- Keywords:
-
- 1207 GEODESY AND GRAVITY / Transient deformation;
- 1209 GEODESY AND GRAVITY / Tectonic deformation;
- 1294 GEODESY AND GRAVITY / Instruments and techniques;
- 1295 GEODESY AND GRAVITY / Integrations of techniques