Slopeland disasters triggered by rainfall are related to geology, topography, and vegetation conditions;however, most warning models are only quantified for rainfall indicators. For example, although the current debris flow
warning model in Taiwan and the sediment-related disaster warning model in Japan employ different critical lines,
indicating the differences in the geology and terrain of the two regions, these models do not comprise a defined set of quantitative assessment methods for expressing the physiographic fragility of each region. In this study, we used historical relative rainfall data from QPESUMS to define a rainfall hazard index (HR) and adopted the frequency and scale of new collapses in each QPESUMS mesh to determine a physiographic fragility index (FP). Then, we integrated these two indicators to obtain the risk level of rainfall-induced hazard (Rh) for each QPESUMS mesh during a typhoon or heavy rainfall. Finally, we considered rainfall events from 2005 through 2017 in Liougui District of Kaohsiung City as an example. The results indicate that the warning model can predict the occurrence and location of new collapses and also display the risk zones of rainfall-induced slopeland disasters through a visualization platform.
Key Words: Disaster, fragility, hazard, landslide, warning system