PNSN Logo

GPS detection of very slow fault slip on the deep part of the Cascadia Subduction Zone
July 8 - 24, 2004

When slip on an earthquake fault occurs, even deep in the earth, there is a small but often detectable shift of the ground at the earth's surface. This can be monitored on a day-to-day basis with high precision Global Positioning System (GPS) instruments. An array of GPS instruments in Washington allows us to see evidence of "silent slip" occurring at depths of 40-65 km under the Puget Lowland


(also available is a smaller scale map showing GPS instruments in the entire Pacific Northwest).

The surface displacements due to slip on the deep subduction zone interface is generally best seen in the east (longitude) component of the GPS time-series. The time series below shows how the computed position of the GPS station changes with time. For example if the dots move upward on the longitude plot, it means that the station is shifting to the east. If the dots move downward, it means that the station is shifting to the west.

The plot below shows east-component GPS data for several stations. The initial deep seismic tremor of July 2004 began July 8, in northern Puget Sound. A vertical line shows July 8 when the motion was first detectable from GPS data.

Two gray lines on the left part of the plot define the time period of the last slip event from February 26 to March 19, 2003. Deep seismic tremor was also reported at that time.

The next figure shows a map view summarizing displacements between July 8th and 20th , 2004. Notice that the sites in nothern Puget Sound are moving in a direction opposite to the usual NE direction.

GPS data from tracking sites located across the Pacific Northwest were processed at UW using GIPSY/OASIS II software, with orbit and clock information provided by the Jet Propulsion Laboratory. Data are initally processed in point position mode before application of ambiguity resolution. Local results are combined with solutions from other sites on North America and are expressed relative to a fixed-North America datum. For details see Dan Johnson (dj@ess.washington.edu) or Tony Qamar (tony@ess.washington.edu)