InSAR has the potential to resolve complex deformation occurring both in space and time by utilizing redundant data in an effort to remove error sources. We have developed an inversion scheme for extracting a time-dependent deformation signal from a data set of differential interferograms. This technique improves upon standard statistical techniques, such as stacking, which strive to reduce orbital, atmospheric, and processing errors. By comparing our result with other geodetic measurements, we demonstrate that InSAR can be a powerful too for accurately resolving nonlinear, crustal deformation processes.
We process 120 differential interferograms of the San Francisco Bay Area. The synthetic aperture radar (SAR) scenes were collected by the European Space Agency using the the ERS1 and ERS2 satellites. The interferograms were processed using the Repeat Orbit Interferometry Package developed at JPL/Caltech. We limit our analysis to those interferograms with perpendicular baselines less than 150 meters. Topography was removed using a USGS 30 meter DEM. Using a least squares inversion, we solve for the displacement field during each time step as determined by the date of each individual SAR scene.
We focus our analysis on three regions in the Bay Area: the northern Hayward Fault, the southern Hayward Fault, and the Santa Clara Valley. Along the Hayward fault, we project the change in range onto a vector 35 degrees counterclockwise from north. We observe a rate of 5.6 mm/yr along the northern portion of the fault. This rate agrees well with observations from a nearby creep meter. Due to strong atmospheric errors which plague individual interferograms, the time series has a scatter of up to a cm, although the overall trend is linear in time. A similar comparison is performed along the southern Hayward Fault where an alignment array is used to monitor surface slip. Lienkaemper et al. (1997) observe left-lateral slip following the 1989 Loma Prieta earthquake followed by a 2 cm right-lateral creep event in February of 1996. While the left-lateral response is too subtle for the InSAR data to observe, the inversion clearly resolves the slip event (Figure 12.1).
In the Santa Clara Valley the time series is able to resolve deformation across the Silver Creek fault and uplift in the Santa Clara basin. The Santa Clara Valley is a large sedimentary basin whose stratigraphy forms a series of overlying aquifers and aquitards. We observe a seasonal pattern of uplift and subsidence superimposed over a long term uplift pattern. During the period from 1992 to 2000, the time series images up to 4.5 cm of maximum uplift centered northwest of San Jose. This uplift is associated with the re-charge of the aquifer. A sharp seasonal signal is also observed across the Silver Creek fault (Figure 12.2). This signal appears not to be tectonic. Rather the deformation is associated with differential subsidence across a discontinuity defined by the fault.
Lienkaemper, J. J., J. S. Galehouse, and R. W. Simpson, Creep response of the Hayward Fault to stress changes caused by the Loma Prieta earthquake, Science, 276, 2014-2016, 1997.