Zhang, S.; Tan, S.; Zhou, J.; Sun, Y.; Ding, D.; Li, J. Geological Disaster Susceptibility Evaluation of a Random-Forest-Weighted Deterministic Coefficient Model. Sustainability2023, 15, 12691.
Zhang, S.; Tan, S.; Zhou, J.; Sun, Y.; Ding, D.; Li, J. Geological Disaster Susceptibility Evaluation of a Random-Forest-Weighted Deterministic Coefficient Model. Sustainability 2023, 15, 12691.
Zhang, S.; Tan, S.; Zhou, J.; Sun, Y.; Ding, D.; Li, J. Geological Disaster Susceptibility Evaluation of a Random-Forest-Weighted Deterministic Coefficient Model. Sustainability2023, 15, 12691.
Zhang, S.; Tan, S.; Zhou, J.; Sun, Y.; Ding, D.; Li, J. Geological Disaster Susceptibility Evaluation of a Random-Forest-Weighted Deterministic Coefficient Model. Sustainability 2023, 15, 12691.
Abstract
The assessment outcomes of regional susceptibility to geological disasters can directly indicate the extent and intensity of risks within the study area, thus, providing targeted guidance for disaster management efforts. This study selects eight evaluation indicators, namely elevation, gradient, terrain relief, lithology of strata, normalized difference vegetation index, distance from the fault, distance from road, and distance from the river. The study focuses on Huize County in Yunnan Province as the research area, utilizing the certainty factor (CF) and random forest (RF) models for evaluating the susceptibility to geological disasters. The non-geological disaster points in the study area are determined using the deterministic coefficient prior model, and the deterministic coefficient values for each evaluation factor serve as the classification data for the random forest model. The optimal parameters for the random forest are selected through iterative calculations of bag error in PyCharm, while the weight of the evaluation factor is determined based on the ran-dom forest model with the optimal parameters. The results of geological disaster susceptibility zoning in Huize County are obtained by overlaying the weighted deterministic coefficients of each evaluation factor. The accuracy of the evaluation results is verified using zoning statistics and ROC curves with a test sample of 30% of the points. The results demonstrate the high accuracy of the model in evaluating the susceptibility to geological disasters in Huize County. Compared to the single deterministic coefficient model, this approach offers advantages in terms of reliability and accuracy. The evaluation results can serve as a scientific reference for related work in Huize County.
Keywords
geological hazard; susceptibility; random forests; certainty factor; Huize county
Subject
Environmental and Earth Sciences, Geophysics and Geology
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
