Crustal deformation in Kyushu derived from GEONET data

Abstract

An investigation was conducted on the horizontal crustal deformation in Kyushu using the coordinate data from GPS Earth Observation Network (GEONET) for the 4 years from 1998 to 2002, for which the annual variations are corrected. We examined the spatial differences in the displacement rate relative to the Amurian plate by subtracting the velocity of the Amurian plate at each of the Global Positioning System (GPS) stations. Models for interplate coupling in Hyuganada and the dilatational source beneath Sakurajima are obtained. The crustal deformation in eastern Kyushu is well explained by the model of interplate coupling in Hyuganada, which is strong in the north and weak in the south. When the effects of the interplate coupling and the dilatational source are removed, a southward extensional field is recognized in and to the south of the Beppu‐Shimabara graben in western Kyushu. In southern Kyushu, to the south of around latitude 32°N, a southeastward movement is conspicuous. It must be noted that the displacement field is extensional, indicating that southern Kyushu is being dragged at the base by a flow in the mantle, not moving like a raft on the flow. It is probable that the flow in the mantle is driven by the rising of hot materials in the Okinawa Trough and a retreat of the trench where the Philippine Sea plate is subducting. A gap in the magnitude of the eastward displacement rate around latitude 32°N implies that the velocity of the flow in the mantle decreases substantially to the north of the line. The Beppu‐Shimabara graben may be a continuation of the Okinawa Trough, but the significant southwestward displacement rate observed in the eastern part suggests that the deformation in the graben, at least to the east of Aso volcano, is affected by the westward movement of the forearc sliver produced by the oblique subduction of the Philippine Sea plate on the Nankai Trough off Shikoku Island. We think that the strong coupling in the Bungo Channel in our model is probably not real, but might be a result of our assumption that all the westward displacements in northeastern Kyushu, a part of which are actually due to the movement of the forearc sliver, are produced by interplate coupling.