Detailed Inventory And Micro-zoning Of Landslide Susceptibility Based On Big Date Of Reservoir Bank Landslide

Since the experimental water storage of the Three Gorges Reservoir,the periodical storage and discharge of reservoir has caused the periodic rise and fall of the reservoirs,which has aggravated the resurrection of landslide and the instability of the potential landslide in reservoir area.The typical reservoir-related landslide include: Yunyang Peak package Ridge landslide,Yunyang chicken chop landslide,Wushan Hongyan landslide,etc.The landslide along the bank coast not only threatens the safety of life and property of the residents living on the shore,but also threatens the safety of the vessels passing through the Yangtze River channel.As an important part of the Yangtze River Economic Belt,the Three Gorges Reservoir area is of great social significance to guarantee the safety of the life and property of the residents and the vessels along the bank.This article takes the left bank of the Yunyang section of the Three Gorges reservoir area as the research object,and determines the scope of the research area according to the characteristics of the Three Gorges Reservoir water level fluctuations and the local topography and geological conditions.Based on the remote sensing images and geological maps of the study area,69 landslides were mapped through visual interpretation and field survey.According to the characteristics of landslides in the bank,13 landslide susceptibility factors were selected to establish a landslide susceptibility impact factor index system;based on geographic information systems,logistic regression models were applied to analyze the geospatial big data modeling of the study area.Thus,the zoning evaluation of landslide susceptibility in the study area was conducted and the accuracy of the evaluation results was examined.The paper mainly achieved the following results and conclusions:(1)Establish an evaluation index system.The indicator system chosen in this paper comprehensively considers the characteristics of the complex geological environment and bank landslides in the study area,and selects elevation,slope,aspect,slope position,curvature,micro-geomorphology,cis-converse slope,lithology,distance to road,and NDVI,TWI,distance to river,and dip and escarpment bank were used.A total of 13 landslide susceptibility impact factors were used as indicator systems.(2)Compare the influence of different influence factors on bank landslide.In this paper,13 influence factor data are normalized,and the absolute size of the correlation coefficients of each influence factor obtained by using logistic regression model reflects its influence on the landslide.The effects of 13 factors on landslide are as follows: slope,curvature,slope position,elevation,dip and escarpment bank,cis-converse slope,microtopography,slope direction,distance to the river,lithology,NDVI,TWI,distance to the road(3)Get landslide susceptibility zoning map in the study area.A sample was selected to establish a landslide susceptibility logistic regression model in the study area,and the resulting model was used to simulate and calculate the geospatial data in the study area in the geospatial database.The probability distribution of the landslide in the study area was obtained,and landslides were selected from the highest to the lowest landslide probability.The risk is divided into five levels: extremely high,high,medium,low and extremely low.The statistical evaluation method was used to test the evaluation results.It was found that the evaluation results were basically consistent with the actual landslide distribution.The receiver operating characteristic curve(ROC curve)was used to test the evaluation results.The test results showed that the AUC value of the ROC curve was 0.892.Using landslide examples to verify that the results of the susceptibility partitions are consistent with the actual situation.Explain that the evaluation result has high accuracy.This study makes a quantitative and scientific evaluation on the vulnerability of Reservoir Rock landslide,and explores a method to evaluate the vulnerability of landslide in reservoir area,which can provide accurate and scientific data reference information for landslide prevention and early warning,construction exploration and geological research.