Font Size: a A A

Evaluation Of Landslide Susceptibility In Shifang City Based On GIS And RS Technology

Posted on:2020-02-10Degree:MasterType:Thesis
Country:ChinaCandidate:Y LuoFull Text:PDF
GTID:2370330578958376Subject:Cartography and Geographic Information System
Abstract/Summary:PDF Full Text Request
Shifang City in Sichuan Province is selected as the study area.The geomorphology of this area has transited from Chengdu Plain to the western Sichuan Plateau.The geological structure conditions are complex and landslides and other geological hazards develop in this area,which poses a serious threat to the life and production of residents in this area.Landslide susceptibility assessment can help us to understand the probability of landslide occurrence in the region,provide scientific reference for the prevention and mitigation of landslide disasters in the region,provide data and model reference for the evaluation and prevention of landslide disasters in similar geomorphic transition areas,and improve the evaluation system of landslide susceptibility.Based on ASTERG-DEM 30m resolution data,geological data,meteorological data and human activity data in this area,this paper uses spatial analysis technology to extract and quantitatively analyze the information of many factors,such as topographic and geomorphological characteristics,hydrometeorological characteristics,geological characteristics,human activity characteristics,etc.in Shifang City,and chooses the appropriate model benefit on the basis of logistic regression model and weight of evidence method.The susceptibility of landslide hazards in the study area was assessed by several influencing factors extracted and landslide data interpreted from remote sensing images.The main research achievement of this paper is divided into the following 3 aspects:(1)Landslide factor extraction and sensitivity analysis:Firstly,through the interpretation of historical landslides,the number and distribution of historical landslides in the study area were identified,and then 11 impact factors including topography,hydrology and geology were extracted to determine the impact of all landslides.The attribute value of the factor establishes a complete landslide susceptibility evaluation factor system,and the sensitivity coefficient method(CF)is used to explore the sensitivity degree of each impact factor and each grading.The results show that the elevation,slope,elevation coefficient of variation,formation rock Factors such as sex,fault distance and rainfall have high sensitivity to landslides.(2)Spatial prediction of landslide disaster susceptibility:Based on eight significant factors,logistic regression model and weight of evidence method are constructed to evaluate the landslide susceptibility in the study area,and the spatial zoning and analysis of landslide susceptibility are carried out.The results of logistic regression model landslide susceptibility zoning show that the area of landslide-prone area is 435.96 km~2 distributes in the villages and towns of Huangdi,Luoshui,Shigu and some areas of Shifang State-owned Forest Farm.The terrain is flat or the slope and the topographic fluctuation are great,showing a trend of two-level differentiation.The landslide disasters are basically undeveloped,the risk of landslide is small,and landslide disaster prevention and control is not necessary;the area of the extremely high-risk area of landslide is 84.67 km~2.It mainly distributes in some areas of state-owned forest farms and Hongbai Town in Shifang City.There is also a small area distribution in Yinghua Town.The density of landslides is very high and the number of landslides is large.The topographic,hydrological and lithological characteristics are suitable for the development of landslide disasters.It is the key prevention and control area of landslides.(3)Accuracy verification of landslide susceptibility and selection of model results:The two models are verified by ROC curve and AUC value.The results show that the accuracy of logical regression model 0.832 is slightly better than that of evidence weight method 0.746.The best evaluation method of landslide disaster in the study area is logical regression model and logic.The results of the regression model show that the method has a certain generalization in the landslide susceptibility evaluation in the terrain transition area,and it is a useful exploration for the landslide susceptibility evaluation in the terrain transition area of Plateau plain.
Keywords/Search Tags:Landslide, geographic information system, remote sensing, logistic regression model, susceptibility evaluation
PDF Full Text Request
Related items