Landslide Susceptibility Analysis Based On ArcGis In Rimian Hydropower Station

Alpines and canyons, high intensity rainfall and large areas of reviers are widelyexisted in the southwest of china, which lead to vulnerable ecological environmentand a variety of geological disasters. Landslides are among the most dangerousdisasters.Therefore, Southwest of China is the country suffered the most serioushazards of landslide hazard areas. As the most dangerous and harmful Influencingfactor in the stability of reservoir bank, Landslides have a great effect on the safety ofthe dam site.How to make scientific and accurate landslide susceptibility assessmenthas very important significance on the construction of hydropower stations anddisaster prevention and mitigation.In this paper, twenty-five landslides and accumulation bodies were interpreted byArcGis software combining Rimian hydroelectric power geological survey with thehigh resolution of SPOT5remote sensing images.And then select the slope,aspect,lithology, geological structure, river erosion, the vegetation coverage of landslide asthe susceptibility evaluation factors. Supported by the ArcGis software, the regionwere divided by50×50meters gird cell.Finally, the models of AHP and theinformation method were estiablished to evaluate the landslide susceptibility forRimian hydroelectric power. The paper achieved the following results:1. Using the orographic digital elevation model (DEM) and the high resolution ofSPOT5remote sensing images, the three-dimensional digital characteristic features ofthe Reservoir Area and the landslides were established.Accordingly; we can interpretthe station, the approximate shape and scale of the landslide.It can bring greatconvenience on spot geological survey with the station coordinate of the landslide.Inthis way, the idea using3S technology(RS,GIS,GPS)on region landslide susceptibilityanalysis was formed.2. Using the orographic digital elevation model (DEM) and the high resolution ofSPOT5remote sensing images and the previous research experence, the slope, aspect,lithology, geological structure, river erosion, the vegetation coverage of landslidewere selected as the susceptibility evaluation factors. Landslide distribution and its statistical relationship of the above factors were discussed by introducing the concept“level weights”.3. The study area was divided by50x50meters grid cell with the ArcGISsoftware powerful spatial analysis. While using the AHP and the information method,The models for evaluating the landslide susceptibility were estiablished.By plottingthe cumulative frequency curve to determine the sensitivity level partition boundarythreshold, finally, concluded the landslide susceptibility mapping of the study area.4. Using the test principle of the ROC curve, we can plot the accuracy test curvesof the AHP and the information method, and strike the area under the curve (AUC).Studies have shown that the two methods are reliable and effective, but the results ofthe evaluation of the information method are superior to the AHP method.