2-D numerical simulations on temperature fields associated with subduction of the Philippine Sea plate in Southwest Japan

Megathrust earthquakes and deep low frequency tremors have occurred along the Nankai Trough where the Philippine Sea (PHS) plate is subducting beneath southwest Japan. These earthquakes that occurred on the plate boundary are considered to be related to thermal structure. Therefore, estimate of thermal state on the plate boundary can give constraints on generation mechanism of thrust-type earthquakes. In this study, we constructed a 2-D thermal convection model associated with the PHS plate subduction beneath southwest Japan to estimate thermal state, and evaluated reliability of the calculated temperature distribution by comparing the calculated heat flow with observed ones such as Hi-net heat flow data, borehole, heat probe, and BSR. Especially, Hi-net heat flow data that have spatial high resolutions on land enabled us to estimate thermal state on plate boundary at depth ranging from 20 to 50km. We took account of spatio-temporal change of the age of the PHS plate, change of the plate motion direction of the PHS plate during the last seven million years, and shape of the upper surface of the PHS plate. We first calculated temperature fields and heat flow associated with only subduction of the PHS plate, passing through central and east Shikoku and the Kii Peninsula. The results show that the calculated heat flows fit well with the observed data for the profiles near the trough axis. However, the results were not consistent with the observed heat flow data onward land area. Therefore, we need to consider more effective heat transport mechanism other than subduction process. Here, we newly developed a model incorporating frictional heating on the plate boundary and thermal effects of erosion and sedimentation, and calculated temperature and heat flow. Incorporating frictional heating on the plate boundary in subduction model, the calculated temperature on the plate boundary increased on the plate boundary. As a result, heat flow values increased values from trough axis to the region above the corner of the mantle wedge. We found that the lower the value of pore pressure ratio is, the larger the value of frictional heating on the plate boundary is. We showed that thermal effects of erosion and sedimentation in the Quaternary can explain spatial distribution of observed heat flow on land area. Taking account of both frictional heating on the plate boundary and thermal effects of erosion and sedimentation during the Quaternary, the calculated heat flow is consistent with the observation from trough axis to land area.