Anomalous Surface Deformation Associated With the December 22, 2003, San Simeon Earthquake – From RADARSAT-1 Interferometry
By Charles Wicks, U.S. Geological Survey (USGS)
On December 22, 2003, a magnitude (Mw) 6.5 earthquake struck the central California coast in a sparsely populated area NE of San Simeon, California (insert, Figure 1). In the city of Paso Robles (population ~28,000), about 39-km ESE of the epicenter, the two deaths caused by the earthquake occurred in the collapse of a building. The maximum building damage sustained during the earthquake was in city center. To gain insight into the cause of the damage focused on Paso Robles, I used RADARSAT-1 images, made available through the Alaska Satellite Facility (ASF), to study earthquake-related surface deformation in the area.
Satellite InSAR has proven to be very useful in the study of large-scale deformation features, tens of kilometers in extent, but it is also useful in studying small, kilometer scale deformation features. The interferogram in Figure 1 shows the range change between two ascending ST2 mode RADARSAT-1 images with 48 days of separation. The interferogram is unfortunately mostly incoherent in the area of the earthquake epicenter. However, the interferogram does show two areas of apparent subsidence related to the earthquake. One area in Templeton, California, about 8-km south of Paso Robles, coincides with the highest measurement of peak ground acceleration where the measured subsidence is most likely the result of an earthquake induced compaction event. Here, I will concentrate on the other, more interesting area of subsidence in Paso Robles.
The area of subsidence in Paso Robles (Figure 1) is bounded on the NE by a steep NW trending gradient that corresponds with (and parallels) the trend of four new hot springs that formed immediately after the San Simeon earthquake. The steep deformation gradient also corresponds to the area of maximum damage and to the location of the two earthquake-related deaths. Wang, et al. (Geophys. Res. Lett., 2004), showed that streamflow increased in the Salinas River at Paso Robles within minutes after the earthquake. Wang, et al., theorized that the increased streamflow in the Salinas River and formation of the new hot springs, resulting from the earthquake, induced breach of a ~100-m deep seal above a pressurized hydrothermal reservoir. In their scenario fluid from the reservoir exited through the breach and immediately pressurized the overlying fracture zone with fluid that subsequently flowed into the Salinas River. Stream flow measurements indicate that the majority of the hydrothermal fluid was expulsed in the first few hours after the earthquake, followed by a much lower rate of extended fluid expulsion. The area of subsidence in Figure 1 most likely represents the extent of the hydrothermal reservoir that served as a source for the increased streamflow.
ERS-2 images acquired 35 days apart after the earthquake (December 31, 2003 to February 2, 2004) also show the subsidence area, but with a peak amplitude of about 15 mm of subsidence. This observation agrees with the lower fluid expulsion rate inferred from the streamflow measurements shown by Wang, et al. In contrast, the area of subsidence in Templeton does not show signs of subsidence in this postseismic interferogram consistent with a compaction origin.
It is probably not a coincidence that the earthquake damage was at a maximum in structures located on the steep NE flank of the subsidence feature where the fluid saturated hydrothermal reservoir apparently truncates sharply. We are investigating further the nature of this subsidence and the sharp boundary with inteferograms from other scenes and numerical modeling of the deformation field.
Click here to download a copy of the newsletter featuring this article