Coseismic deformations detectable by satellite gravity missions: A case study of Alaska (1964, 2002) and Hokkaido (2003) earthquakes in the spectral domain

Coseismic deformations observed on the Earth surface or modeled by conventional dislocation theory cannot be compared directly with those observed by gravity satellite missions because of the spatial resolution limit of the missions and the signal attenuation of the gravity field. Coseismic deformations in the spectrum domain should be considered instead. For this purpose the conventional dislocation theory [e.g., Sun and Okubo, 1993] for a spherical Earth model can be used because it is expressed in the form of a spherical harmonic. In this study, analytical expressions of degree variances of the coseismic geoid and gravity changes for shear and tensile sources are derived and calculated for three real earthquakes. Those results are compared with expected errors of the Gravity Recovery and Climate Experiment (GRACE) to elucidate whether or not coseismic geoid and gravity changes are detectable by gravity satellite missions. Behaviors of the degree variances for four independent seismic sources are investigated. Results indicate that both the gravity and geoid changes are near two orders of magnitude larger than the precession of the gravity missions in low harmonic degrees. On the basis of these results, we derived the minimum magnitudes of earthquakes detectable by GRACE. We concluded that coseismic deformations for an earthquake with a seismic magnitude above m = 7.5 are expected to be detected by GRACE.