| Hokkaido is the second largest island in Japan,located near the Pacific seismic belt.It has frequent earthquakes all year round,and the earthquake geological disaster is severe.Among them,the earthquake landslide is the most serious,which has brought huge property loss and casualties to local residents.Therefore,it has become an urgent task to carry out the evaluation and zoning of earthquake landslide hazard susceptibility.Taking the earthquake landslide in Hokkaido in Japan as an example,this paper analyzes the distribution law of Seismic Landslides in the study area based on ArcGIS software platform,analyzes and selects the evaluation factors according to the specific geological conditions of the research area,and constructs a scientific and reliable index factor system for seismic slope evaluation.Then,the paper uses logistic regression model and the certainty factor model to evaluate the landslide vulnerability.The main results of this paper are as follows:(1)The establishment of landslide database and the analysis of landslide distribution characteristics.According to Google The 0.3m resolution pleiades-1 remote sensing satellite image provided by earth visually interprets 12586 landslides,and makes statistical analysis on the distribution of landslides.It is found that the development of landslides is closely related to the geological environment and topographic factors;the distribution of landslides is along the trend of water system;human engineering activities will more or less affect the development of landslides;seismic activity plays an important role in the development of landslides It plays an important role.(2)The construction of landslide evaluation index system.The grid cell has the characteristics of fast data processing,high precision and wide applicability.Therefore,based on the situation of the study area,the grid unit is the most suitable for landslide evaluation.By collecting the relevant data of the study area and synthesizing the relevant research results of previous scholars,nine index factors,including elevation,slope angle,slope aspect,curvature,TPI,distance to river,distance to road,distance to the epicenter,lithology,are selected as the evaluation factors.Before the evaluation of landslide susceptibility,correlation analysis is carried out on the selected influencing factors,and the relationship between the factors is determined according to the VIF value in the analysis results.If the VIF value is less than 10,it means that there is no multicollinearity among the factors,that is,they meet the requirements of mutual independence.(3)Using the two models,combined with ArcGIS software platform,the susceptibility of Seismic Landslides in Hokkaido is evaluated.According to the model evaluation results:the evaluation results of the two models are in line with the requirements of rationality test,and the ROC curve is used to test the evaluation accuracy.The accuracy of logistic regression model is 88.4%,and the accuracy of certainty coefficient model is 85.4%.According to the evaluation results,the study area is divided into five areas:extremely high risk area,high risk area,medium risk area,low risk area and extremely low risk area.It can be seen that the evaluation results are in good agreement with the actual landslide distribution.(4)The evaluation results show that the logistic regression model is more accurate than the certainty factor model in the evaluation of landslide susceptibility.The research results can provide reference for the study of earthquake landslide susceptibility in Hokkaido and similar regions,and provide some suggestions for the prevention and control of earthquake landslide according to the research results,which has important reference value for the prevention and treatment of earthquake landslide disasters in Hokkaido and similar regions.Figure 41 Table 8 References 87... |
PDF Full Text Request