Unit Affiliation: Seismology, Geology and Tectonophysics, Lamont-Doherty Earth Observatory (LDEO)
The project will develop new methods for locating earthquakes in remote areas of the world with greater precision. Current methods of earthquake location, based on classical techniques that are used routinely by agencies such as the United States Geological Survey, the International Seismological Centre, and the International Monitoring System, provide no better than 25-km precision in earthquake location in many areas of the world. This lack of precision limits the ability to attribute individual earthquakes to specific earthquake faults, and thus also the possibility to understand better the tectonic context. The project will develop new algorithms based on inter-event correlations of intermediate-period surface waves, and is expected to lead to a location precision of 5 km or better. The algorithm will be used to address questions of plate-boundary tectonics and seismogenesis. The improvements in earthquake location will lead to new data sets with broad utility for research in plate tectonics, plate-boundary processes and global seismic tomography. The new algorithm is based on cross correlation of surface-wave signals from nearby earthquakes recorded on common stations. Effects of wave propagation between the earthquakes and the stations are canceled out, and the remaining signal is related to the propagation time delay between the stations, and to predictable source effects. The results are improved by correcting for laterally varying dispersion. The approach will make use of all available pairs of Rayleigh and Love waves for a regional cluster of earthquakes to invert for relative locations of all events in the cluster. The cluster will then be anchored to geomorphological features to provide absolute locations of the individual earthquakes.
Comprehensive high-precision relocation of global seismicty