IASPEI - Picture Gallery

IUGG XXIV General Assembly July 2-13, 2007 Perugia, Italy
(S) - IASPEI - International Association of Seismology and Physics of the Earth's
Interior
JSS003 Oral Presentation 1858
Real-time, probabilistic and evolutionary earthquake location for seismic
early warning
Dr. Claudio Satriano
RISSC-Lab AMRA Scarl
Anthony Lomax, Aldo Zollo
An effective early warning system must provide probabilistic estimates of the location and size of a
potentially destructive earthquake within a few seconds after the event is first detected.In this work we
present an evolutionary, real-time location technique, based on a robust, equal differential time (EDT)
misfit function, a very fast, global-search algorithm, and a probabilistic approach for describing the
hypocenter estimation. The location technique, at each time step, relies on the combined information
from triggered arrivals and not-yet-triggered stations. With just one recorded arrival, the hypocentral
location is constrained to be within the Voronoi volume around the first triggering station. The Voronoi
volume is defined by the current time and the travel times to the not-yet-triggered stations; this volume
shrinks as time passes, event if no new arrivals become available. With two or more triggered arrivals,
the location is constrained by the intersection of the Voronoi volume defined by the remaining, not-yettriggered
stations with the EDT surfaces for all pairs of triggered arrivals. (At each point on an EDT
surface the difference in the predicted travel times to two stations equates the difference in the
observed arrival times at the two stations.) As time passes and more triggers become available, the
evolutionary location converges to a standard EDT location.We show several tests of the real-time
location technique using arrivals generated by different automatic picking procedures for strong-motion
data from moderate to large earthquakes worldwide. The results indicate that useful, probabilistic
location estimates, suitable for early-warning applications, can be achieved with very few recording
stations (1-2), while, in comparison, a standard location algorithm usually needs 3 or more stations to
begin constraining the hypocenter.We also present an event binding procedure for the real-time location
technique to detect multiple sources and associate phase arrivals when different events occur close in
time.
Keywords: location, evolutionary, probabilistic