Stress data inversion to estimate collision rate distribution and its application to the Izu Peninsula, Japan

In convergent plate boundary zones, convergent motion between adjacent plates is accommodated by fault slip along the plate interface (subduction) combined with the compressional deformation of the plates (collision). The collision rate is defined as the ratio of the long-term increase rate of the irrecoverable slip deficit at the plate interface to the full plate convergence rate in order to quantitatively represent this mechanical aspect. The distribution of the collision rate on the plate interface is a major active source that generates a tectonic stress field in the plate interior. We developed a new inversion method based on these concepts to estimate the collision rate distribution from the patterns of the intraplate tectonic stress fields. In the inversion, the analytically obtained tectonic stress data were used and a Bayesian approach was considered. The application of the inversion method to the Izu collision zone in central Japan revealed that a broad belt of high collision rates surrounded the base of the Izu Peninsula. The collision belt was divided into three parts: (1) a central strong collision zone spreading landwards at the base of the Izu Peninsula, where the value of the collision rate becomes a maximum, (2) an eastern moderate collision zone widely covering the Philippine Sea-North American plate boundary along the Sagami trough, and (3) a western moderate collision zone concentrated in a shallower part of the Philippine Sea-Eurasian plate boundary along the Suruga trough. The central collision zone coincided with the colliding front of the Izu-Ogasawara (Bonin) island arc, which indicated that the estimated collision rate distribution described the actual arc collision process. The eastern and western collision zones might imply a higher absolute strength of their corresponding large interplate seismogenic zones.