多源遥感数据支持下区域滑坡灾害风险评价

Regional Landslide Risk Assessment Using Multi-Source Remote Sensing Data

滑坡风险管理是防灾减灾的有效途径之一,而灾害风险评价是风险管理的基础和依据。以三峡库首区为研究区、多源遥感影像为主要数据源,利用立体像对技术及光谱分析等方法快速提取地形地貌、地表覆被、地质及水文条件等滑坡孕灾环境信息,应用随机森林模型分析区域滑坡危险性;采用面向对象方法建立典型承灾体识别规则,快速提取建筑物及交通道路等信息,综合分析滑坡危险性及承灾体信息,以实现区域滑坡灾害风险评价。结果显示:高风险区面积为41km2,约占研究区面积的9%,主要集中于人口聚集的城镇和交通建设用地等经济价值大的地区。其评价结果与野外实地调查情况基本吻合。

Landslide risk management is an effective way for disaster prevention and mitigation, in which risk assessment is essential. The head area of the Three Gorges is selected as the study area. Using multi-source remote sensing data as the major data source, the authors extracted the environment information pregnant with landslides including geomorphological, land cover, geological, and hydrology by stereo image processing and spectral analysis, and analyzed the landslide hazard by using the random forest model. Furthermore, the rule of extraction for typical elements at risk is established through the approach of object-oriented image analysis, and then the building and road information is extracted from the high-resolution satellite images. Finally, a landslide risk map is developed by integrating the elements at risk and landslide hazard datasets. According to the map, 9% of the study area is identified as high-risk zones encompassing 41 km2, and these high-risk areas are located in county and towns covered by transportation networks and other urbanized areas with high density of populations and property. The results of the risk assessment are basically consistent with the field observation and survey.