Research On Disaster Risk Assessment Model Of Single Landslide Using GIS And RS

Risk assessment on the instability single landslide in the range of population will provide scientific basis for prevention plan; mitigation programs; deployment of control projects. Geographic information systems and remote sensing technology provide a guarantee for achieving the "Digital Landslide" and "Spatial Information", becoming more accurate identification, location and quantitative means for single landslide investigation.In this paper, obtaining information by remote sensing data; spatial data management and analysis using GIS;"hazard- vulnerability-risk"three models as the theoretical basis for single slide risk assessment, systematic study of the related issues of single landslide risk assessment model using GIS and RS technology according to the process of "principles and methods - extraction and analysis - building and solving - application and implementation". This paper analyzes the principle of single landslide risk assessment and the extraction of hazard index and vulnerability index, building a fuzzy comprehensive evaluation model after data preprocessing, and evaluation the science of models with principal component analysis and cluster tests. Research results are as following:(1) Using DEM, geological maps, precipitation data, high-precision aerial photographs, SPOT5, Beijing-1, Radarsat and other multi-source data, extracting topography, geology, climate, hydrology, vegetation, human activities, population density, engineering facilities and land use information of Donghekou landslides in QingChuan, Woqian landslide in Qingchuan, Wenjiaba landslide in PingWu.(2) After data preprocessing, Building and solving a fuzzy comprehensive evaluation model for single landslide risk assessment. Results indicate that the Sequence of hazard: Donghekou (0.79)> Wenjiaba (0.56)> Woqian (0.49), the Sequence of vulnerability: Donghekou (0.66)> Woqian (0.067)> Wenjiaba ( 0.010), the Sequence of risk: Donghekou (0.52)> Woqian (0.033)> Wenjiaba (0.0056).(3) Principal component analysis shows that the most important single factor in landslide hazard is the topography (0.32), the geological (0.30), hydrological (0.24), and human activities (0.092). Cluster analysis reveals "river network density, population density, and annual rainfall" have the largest similar features, while the "seismic intensity" is the most independent factor.(4) With the steps of hazard index information extraction and analysis, vulnerability index information extraction and analysis, risk assessment and disaster mapping of to achieve a decision support system by AE.