The November 2002 Denali earthquake in Alaska provided the perfect test, and showed that GPS was up to the task. Two independent groups have shown spectacular results from this earthquake. A group led by Kristine Larson of the University of Colorado showed in Science on May 30, 2003, that displacement records or "GPS seismograms" from sites in Fairbanks and Anchorage clearly agree with the records from nearby seismometers, and also showed the propagation of surface waves across western North America. The surface waves had very large amplitudes, as large as 40 centimeters peak to peak, and were actually too large for broadband seismometers to record on a scale covering a broad swath of western North America. GPS seismograms are "noisier" than those measured by seismometers, but GPS can never go off-scale, so the utility of the technique increases with the size of the earthquake. This year, an independent group led by Yehuda Bock at Scripps Institution of Oceanography used data from a dense GPS array in southern California to track the propagation of the surface waves across that area (Geophysical Research Letters, v.31, L06604). They used innovative filtering techniques to suppress noise in the GPS seismograms, and showed excellent agreement between the GPS displacements and equivalent records derived by integrating standard velocity seismograms.
This new technique will be applied to every moderate to large earthquake in the future as long as there is a GPS site close enough to record the shaking. Far away from an earthquake epiecenter, the GPS seismograms are clearly inferior to those produced by seismometers, if the seismometers stay on scale. But, as shown by the Denali quake, for the largest earthquakes, seismometers can go off-scale if they are located very far away from the earthquake. And close to the earthquake, the shaking is too strong for all but strong motion seismometers to record. Although the use of GPS seismograms is still in its infancy, there are strong indications that this data is complementary to strong motion recordings when used to determine seismic source models. For the next magnitude-9 earthquake, the most important seismic recordings might come from continuous GPS receivers if many seismometers go off-scale.
This new application of GPS will be aided by the deployment over the next five
years of the Plate Boundary Observatory (PBO) component of National Science
Foundation's EarthScope program (Geotimes, April
2002 and March 2004). PBO will consist of 875 new continuous GPS sites across
the seismically and volcanically active regions of western North America, and
each PBO site will record data at 5 Hertz in an internal buffer, ready for download
in case of a significant event. As telemetry capabilities develop over the duration
of EarthScope, an increasing fraction of sites will stream continuous high-rate
data in real time, potentially allowing GPS seismograms to be used in real time
for earthquake location and source studies.
Back to index