Subsonic type earthquake dislocation moving at approximately × shear wave velocity on interface between half spaces of slightly different elastic constants

Abstract

Bouchon, Bouin, Karabulut, Toksöz, Dietrich and Rosakis reported recently that rupture appeared to propagate eastward in the 1999 Izmit and Düzce earthquakes in Turkey at a transonic velocity greater than the shear wave velocity CS and smaller than the longitudinal velocity CL. The velocity appeared to be close to the Eshelby velocity ( × shear wave velocity), a velocity at which a transonic glide edge dislocation has a subsonic (with respect to CL) type stress‐displacement field with no supersonic (with respect to CS) component. They suggest that the super shear wave velocity rupture may have been aided by a normal stress reduction in the slip process whose origin lies in the differing elastic properties of the crust on either side of the earthquake fault. They suggest that this normal stress reduction will occur if the north side of the fault has the elastically softer rock In this paper it is shown that a transonic glide edge dislocation on an interface separating material of different elastic shear moduli cannot have only a subsonic type stress field (or only a supersonic field). Moreover, the supersonic part of the displacement field is not the usual step function.