|
April 24 2006
IGPP geophysicists Kris Walker & Peter Shearer develop new "back-projection" method for rapid determination of earthquake location
Figure 1: Image of an animation showing the details of
the Mw 8.6, 28 March 2005 Sumatra rupture. The rupture started at the
hypocenter (white star). The black star indicates the Harvard Centroid
Moment Tensor location. This image suggests the greatest amount of energy
was released to the northeast and southeast of the hypocenter. The linked
animation is one of the primary pieces of evidence that suggests the
rupture did not unilaterally propagate to the south as first thought,
but instead propagated to the north and south at 3.1 ± 0.2 km/s.
Click the image to view the complete animation or click here.
IGPP geophysicists Kris Walker and Peter Shearer have developed an improved,
automated method to determine the properties and characteristics of fault
rupture after a large earthquake. The method is based on an old technique
called reverse-time migration or "back-projection". Their technique
works by analyzing the P-waves recorded on the global seismic network,
and back-projecting them to the source region. This permits the
direct imaging of the rupture dynamics within 20-30 minutes following
the onset of the earthquake, and from these images the characteristic
source properties. These source properties are then used by agencies
like the USGS to predict which areas near the epicenter experienced the
greatest surface shaking. These predictions are then used by regional
hazard response coordinators. Walker and Shearer are working with the
USGS to implement this system as part of the Prompt Assessment of Global
Earthquakes for Response (PAGER) system. They plan to have a prototype
algorithm working there by the end of 2006.
Figure 2: Screenshot of a movie showing a *hypothetical* earthquake and how the back projection
technique works on a computer. The surface shaking is recorded with seismometers at stations in different locations
(yellow triangles), then projected backwards in time until all the energy meets, which is the location of the earthquake.
Click the image to view the complete animation or click here.
A large earthquake is actually comprised of many smaller earthquakes along a long fault patch that ruptures,
relieving stresses in the Earth. The primary benefit of the new technique is that it highlights the
entire rupture, which provides useful information when trying to predict how much surface damage or
shaking occurred. It is also useful when trying to determine the potential for a tsunami after
a particular oceanic earthquake.
Figure 3: Screenshot of a quicktime movie of a years worth of earthquakes (1995) and their locations, which
outline Earth's plate boundaries. Most of these earthquakes are so small and far away that we usually don't feel
them. It's only the rare very large earthquakes that cause widespread death and destruction, which this method will
be useful for. Click the image to view the complete animation or click here.
Further information:
|
