HR: 1340h
AN: G53A-0109    [Abstracts]
TI: Observation and Modeling of Thermoelastic Strain in SCIGN Daily Position Time Series
AU: * Prawirodirdjo, L
EM: linette@gpsmail.ucsd.edu
AF: University of California, San Diego, 9500 Gilman Drive, MC 0225, La Jolla, CA 92093-0225 United States
AU: Ben-Zion, Y
EM: benzion@terra.usc.edu
AF: University of Southern California, USC Earth Sciences, Los Angeles, CA 90089-0740 United States
AU: Bock, Y
EM: ybock@ucsd.edu
AF: University of California, San Diego, 9500 Gilman Drive, MC 0225, La Jolla, CA 92093-0225 United States
AB: The Southern California Integrated GPS Network (SCIGN) is now well established, and daily position time series for its stations clearly show seasonal variations. With our analysis we suggest that these fluctuations are caused by strain in the elastic part of the earth's crust induced by temperature variations at the surface (thermoelastic strain). We compute the predicted crustal thermoelastic strain using the observed local atmospheric temperature record and compare the seasonal variation in thermoelastic strain to the horizontal position time series of regional GPS stations. We use a crustal model that consists of an elastically decoupled (soil) layer over a uniform elastic half-space. Temperature variations at the base of the decoupled layer results in strain in the underlying elastic half-space. The decoupled layer has the effect of delaying, attenuating, and low-pass filtering the source temperature field. For our analysis we consider three regions (Palmdale, Pinon Flat, and 29 Palms), each with a radius of about 50 km. In each region we analyze data from one temperature station and three or four GPS stations. The temperature time series is used to compute the thermolastic strain at each GPS station, based on its relative location in the temperature field. For each region we assume a wavelength for the temperature field that is related to the local topography. The depth of the decoupled layer is inferred from the phase delay between the temperature record and the GPS time series. In order to compare our predicted strains to the GPS position time series, an arbitrary scale factor is applied. The amplitude of the strain variation at each GPS station is related to the relative location of that GPS station in the temperature field. The goodness of fit between model and data is evaluated from the relative amplitudes of the seasonal signals, as well as the appropriateness of the chosen temperature field wavelength and decoupled layer depth. Our analysis shows a good fit between the predicted strains and the GPS time series and suggests that thermoelastic strain is a primary contributor to the seasonal variations observed by the SCIGN network.
DE: 1200 GEODESY AND GRAVITY
DE: 1208 Crustal movements--intraplate (8110)
DE: 1243 Space geodetic surveys
SC: Geodesy [G]
MN: 2004 AGU Fall Meeting