Quantitative Detection of Ground Surface Changes Due to Slope Failure Using ALOS-2/PALSAR-2 Data

Slope failure is a process of landform change triggered by geophysical and meteorological factors, and its causes include heavy or prolonged rainfall, earthquakes, and other disasters. Data from the Synthetic Aperture Radar (SAR) satellite ALOS-2 can identify slope failure immediately after a disaster because it can observe a large area of the ground surface at any time, independent of the weather. In general, the commonly used method for slope failure detection is visual interpretation using postdisaster optical aerial image data or satellite data in Japan, which is time consuming and laborious. Therefore, we introduced the concept of extracting ground surface change areas based on slope unit for quantitative purposes. Furthermore, the division of slope unit is improved. The method's feasibility was verified by detecting the heavy rain disaster in North Kyushu, Japan, in 2017. The results show that the improved slope unit has a higher recall of ground surface changes than the previous method. Under the proposal of this study, the location of slope failure can be detected with high accuracy immediately after a disaster.