Evaluation of Historical Wildfires in Tohoku Region Using Satellite-Based High-Fire-Severity Index

抄録

Crown fires represent an extreme fire behavior that leads to high fire severity, and dryness plays a vital role in this behavior. Due to the lack of fire severity data in Tohoku, high fire severity was estimated using a satellite-based high-fire-severity index (HFSI). HFSI is the ratio of the identified area of high fire severity sensed using the Landsat-differenced normalized difference vegetation index (dNDVI) to the reported total burnt area. Using the HFSI, only six wildfires could be identified as having high fire severity areas from an evaluation of 55 wildfires with burnt areas greater than 0.1 km² reported in Tohoku from 1995 to 2017. The low HFSI values computed for these wildfires implied that fireline intensity was not high for crown fires to occur in Tohoku. Additionally, the soil moisture (SM) content for three layers, the surface, root, and recharge zones simulated using a land surface model (the Simple Biosphere Model including Urban Canopy (SiBUC) model), were used to assess the dryness. The highest HFSI value calculated among all wildfires was that of the largest wildfire that ever occurred in Japan in the period between 1995 and 2017, the 2017 Kamaishi wildfire. The conditions before this fire were among the driest of the six wildfires with HFSI values.