Insights into the earthquake fault mechanics from space geodesy

Résumé: Precise measurements of surface strain due to co-, post-, and inter-seismic deformation allow robust inferences about the mechanical behavior of the seismogenic crust and the underlying ductile substrate. The observed coseismic deformation patterns due to large shallow earthquakes are well explained by a classic elastic-brittle model of the upper crust. Analysis of small-scale features of the coseismic displacement field from several major earthquakes, including the 1992 Landers, 1999 Hector Mine (California), and 1999 Izmit (Turkey) earthquakes reveals the presence of kilometer-wide low rigidity zones that presumably manifest highly fractured and damaged rocks around active crustal faults. The inferred reduction in the effective shear modulus within the damage zones varies between 40 and 60%. In the framework of the damage mechanics, such variations in the effective elastic modulus imply that the average crack density within the fault zone exceeds the crack density in the ambient crust by a factor of 5 to 7. Geodetic inferences about the presence of wide low rigidity zones were recently verified using seismic tomography. Preliminary results form the Calico fault zone experiment (collaboration among UCSD, UCLA, and USC) reveal a low velocity structure that coincides with a compliant fault zone inferred from earlier InSAR work. The presence of wide damage zones is also suggested by observations of near-field postseismic deformation. In particular, data from the Hector Mine (California) and Bam (Iran) earthquakes show centimeter-scale subsidence and contraction of kilometer-wide zones centered on the earthquake rupture trace. This deformation can be explained in terms of postseismic healing and microcrack closure, presumably reversing the effects of coseismic dilatancy. On a larger scale, measurements of deformation following large earthquakes are increasingly used to discriminate among several commonly assumed mechanisms of postseismic relaxation, such as visco-elastic flow, poro-elastic rebound, and localized afterslip. Ultimately these measurements may help resolve a long-standing debate about the effective thickness, strength, and rheology of the continental lithosphere. -------------------------------------------------------------------------------- Séminaires généraux de l'Institut de Physique du Globe de Paris --------------------------------------------------------------------------------