Abstract. The distance to the surface rupture zone has been commonly regarded as an important influencing factor in the evaluation of earthquake-triggered landslides susceptibility. However, the obvious surface rupture zones usually do not occur in some buried-fault earthquakes cases, which mean lacking of the information about the distance to the surface rupture. In this study, a new influencing factor named coseismic ground deformation was added to remedy this shortcoming. The Mid-Niigata prefecture earthquake was regareded as the study case. In order to select a more suitable model for generating the landslides susceptibility map, three commonly used models named Logistic Regression (LR), Artificial Neural Networks (ANN) and Support Vector Machines (SVM) were also conducted to assess the landslides susceptibility. The performances of these three models were evaluated with the receiver operating characteristic (ROC) curve. The calculated results showed the ANN model has the highest AUC (area under the curve) value of 0.82. As the earthquake triggered more landslides in the epicenter area, which makes it more prone to landslides in further earthquakes, the landslides susceptibility in the epicenter area was also further evaluated.
Received: 27 Feb 2020 – Discussion started: 16 Mar 2020
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
- Printer-friendly version
- Supplement