Scholarlink

<p>Southwest China is one of the regions with the most frequent occurrence of rainfall-induced landslides. Studies on the rainfall–landslide relationship, particularly those on rainfall thresholds, are crucial for landslide prediction and early warning. In this study, four precipitation products (i.e., Tropical Rainfall Measuring Mission (TRMM 3B42), Multisource Weighted Ensemble Precipitation (MSWEP), China Meteorological Forcing Dataset (CMFD), European Centre for Medium-Range Weather Forecasts Reanalysis (ERA5) and historical landslide data from year 2010 to 2018 were utilized. Through landslide sensitivity analysis, effective intensity–duration (EID) curves, and clustering analysis, the characteristics of rainfall-induced landslides were analyzed. Furthermore, different rainfall–landslide response thresholds were established for each precipitation product. A method for simulating the spatiotemporal occurrence of shallow rainfall-induced landslides based on landslide sensitivity and EID curves was proposed and validated through case studies. Results indicated that landslides in Southwest China were characterized by large-scale and numerous points and were densely distributed on complex slopes. Temporally, 80% of landslides occurred during the flood season (April to July), following heavy rainfall and prolonged precipitation, which demonstrated a clustered pattern. The EID curve based on CMFD precipitation data exhibited the optimal fitting performance, indicating the highest sensitivity to landslide response with an <i>R</i><sup>2</sup> value of 0.9529. When rainfall exceeded the EID threshold, the probability of landslide occurrence was high, followed by TRMM and MSWEP, whereas ERA5 showed the poorest performance. Different rainfall thresholds for landslide occurrence can be selected on the basis of meteorological forecasts, and regional landslide early warning can be issued accordingly. This study is expected to provide a theoretical basis for simulating and predicting shallow rainfall-induced landslides, aiming to contribute to landslide early warning and prevention.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Spatiotemporal analysis, simulation, and early warning of landslides based on landslide sensitivity and multisource precipitation products in Southwestern China

  • 期刊论文

  • 发表日期：

    2025年1月7日

查看全文
  • Rui Zhang,
  • Sheng Chen

摘要

Southwest China is one of the regions with the most frequent occurrence of rainfall-induced landslides. Studies on the rainfall–landslide relationship, particularly those on rainfall thresholds, are crucial for landslide prediction and early warning. In this study, four precipitation products (i.e., Tropical Rainfall Measuring Mission (TRMM 3B42), Multisource Weighted Ensemble Precipitation (MSWEP), China Meteorological Forcing Dataset (CMFD), European Centre for Medium-Range Weather Forecasts Reanalysis (ERA5) and historical landslide data from year 2010 to 2018 were utilized. Through landslide sensitivity analysis, effective intensity–duration (EID) curves, and clustering analysis, the characteristics of rainfall-induced landslides were analyzed. Furthermore, different rainfall–landslide response thresholds were established for each precipitation product. A method for simulating the spatiotemporal occurrence of shallow rainfall-induced landslides based on landslide sensitivity and EID curves was proposed and validated through case studies. Results indicated that landslides in Southwest China were characterized by large-scale and numerous points and were densely distributed on complex slopes. Temporally, 80% of landslides occurred during the flood season (April to July), following heavy rainfall and prolonged precipitation, which demonstrated a clustered pattern. The EID curve based on CMFD precipitation data exhibited the optimal fitting performance, indicating the highest sensitivity to landslide response with an R2 value of 0.9529. When rainfall exceeded the EID threshold, the probability of landslide occurrence was high, followed by TRMM and MSWEP, whereas ERA5 showed the poorest performance. Different rainfall thresholds for landslide occurrence can be selected on the basis of meteorological forecasts, and regional landslide early warning can be issued accordingly. This study is expected to provide a theoretical basis for simulating and predicting shallow rainfall-induced landslides, aiming to contribute to landslide early warning and prevention.