Coseismic and postseismic velocity changes measured by repeating earthquakes

Publication Status is "Submitted" Or "In Press: 
LDEO Publication: 
Publication Type: 
Year of Publication: 
Journal Title: 
Journal of Geophysical Research-Solid Earth
Journal Date: 
Oct 7
Place Published: 
Tertiary Title: 
Section / Start page: 
ISBN Number: 
ISSN Number: 
Short Title: 
Accession Number: 
LDEO Publication Number: 
Call Number: 

Repeating earthquakes that rupture approximately the same fault patch and have nearly identical waveforms are a useful tool for measuring temporal changes in wave propagation in the Earth's crust. Since source and path effects are common to all earthquakes in a repeating earthquake sequence (multiplet), differences in their waveforms can be attributed to changes in the characteristics of the medium. We have identified over 20 multiplets containing between 5 and 40 repeating events in the aftershock zones of the 1989 Loma Prieta and 1984 Morgan Hill, California, earthquakes. Postmain shock events reveal delays of phases in the early S wave coda of as much as 0.2 s relative to premain shock events. The delay amounts to a path-averaged coseismic velocity decrease of about 1.5% for P waves and 3.5% for S waves. Since most of the multiplets are aftershocks and follow Omori's law, we have excellent temporal sampling in the immediate postmain shock period. We find that the amplitude of the velocity decrease decays logarithmically in time following the main shock. In some cases it returns to the premain shock values, while in others it does not. Similar results are obtained for the Morgan Hill main shock. Because the fractional change in S wave velocity is greater than the fractional change in P wave velocity, it suggests that the opening or connection of fluid-filled fractures is the underlying cause. The magnitude of the velocity change implies that low effective pressures are present in the source region of the velocity change. Our results suggest that the changes are predominantly near the stations and shallow, but we cannot exclude the possibility that changes occur at greater depth as well. If the variations are shallow, we may be detecting the lingering effects of nonlinearity during main shock strong ground motion. If the variations are deep, it suggests that pore pressures at seismogenic depths are high, which would likely play a key role in the earthquake process.


861PYTimes Cited:11Cited References Count:52

Doi 10.1029/2004jb003011