Df. Argus et Ga. Lyzenga, SITE VELOCITIES BEFORE AND AFTER THE LOMA-PRIETA AND GULF OF ALASKA EARTHQUAKES DETERMINED FROM VLBI, Geophysical research letters, 21(5), 1994, pp. 333-336
We use geodetic data from VLBI to determine the pre- and postseismic v
elocities of two sites. We then place limits on variations in intersei
smic strain buildup. The 1987 and 1988 Gulf of Alaska earthquakes (eac
h Ms=7.6) broke the Pacific plate interior. During the earthquakes the
Cape Yakataga site moved 78 mm toward southwest. During the 1989 Loma
Prieta earthquake (Ms=7.1) the Fort Ord site moved 48 mm toward north
. Baselines (a) from Fairbanks to Cape Yakataga and (b) from Mojave to
Fort Ord change at nearly the same rate before and after the earthqua
kes. Postseismic transients, which we determine from differences betwe
en post- and preseismic rates, are minor: at Cape Yakataga the transie
nt is 3 +/- 4 mm in a postseismic interval of 23 months, and at Fort O
rd the transient is 6 +/- 5 mm in 21 months. The slip beneath the Loma
Prieta rupture needed to generate the Fort Ord transient is 0.22 +/-
0.19 m, one-tenth the coseismic slip (2 m). We analyze elastic lithosp
here-viscous asthenosphere models to determine that the characteristic
time describing exponential decay in deep fault slip is longer than 6
years. The VLBI measurements are consistent with uniform interseismic
strain buildup. They disagree with fast postseismic rates caused by a
n asthenosphere with very low viscosity.