Rapid Coseismic Landslide Susceptibility Assessment using Newmark Analysis and Decision-tree Algorithm, and Its Web-based Display System

Taiwan, which is located in an active tectonic area, results frequent earthquakes. The combined effects of high seismicity and steep topographic relief cause the high susceptibility of landslides associated with earthquakes. In this study, we use Newmark analysis and decision tree (DT) algorithm to perform the nowcasting and delivery susceptibility map on website. The strong-motion records of earthquakes with local magnitude greater than 6.0 between 1990 and 2020 are collected. The 175 potential planar failure slopes with similar lithology (sandstone or sandstone interbedded with shale) are selected as the target slopes (TS). We first found the representative station (RS) satisfied the specific thresholds of peak ground acceleration (> 196 gal) and Newmark displacement (> 10 cm), and then hillslopes around the TS associated with the RS with potential failures caused by earthquakes were carefully mapped. The classification labels of failure and non-failure are used for the DT algorithm. The C5.0 and CART methods are adopted to construct the DT model. Finally, the web-based display system for rapid coseismic landslide susceptibility assessment is established to report the distribution of risk slopes with traffic lights for emergency response and disaster mitigation.