基于卫星观测的洪灾快速监测与评估平台实现

Implementation of Rapid Flood Disaster Monitoring and AssessmentPlatform Based on Satellite Observation

洪灾往往具有突发性强、影响范围广的特点。建立一个对洪灾进行快速准确监测、灾情初步评估和信息传播等功能的Web系统能够在一定程度上减轻洪灾影响,并为灾害应急响应的高效开展提供重要保障。基于WebGIS技术及多源卫星遥感数据搭建了一套洪灾监测与损失评估系统,采用Python环境下Flask作为Web应用框架,利用开源GIS程序包构建服务器,并结合前端Vue.js框架、Leaflet等组件完成洪涝灾害监测系统的搭建,为用户提供洪水淹没范围信息快速查询和灾情评估功能。基于遥感的洪水淹没范围自动提取与发布基于HydroFloods、Wxee等GEE工具包实现。灾情评估功能通过将淹没范围与人口密度等基础地理信息数据使用GDAL库进行空间统计分析实现。将系统应用于2021年8月12日湖北省随州市柳林镇洪灾监测,淹没范围提取结果与官方报道基本相符,损失评估结果比实际统计较低。该系统的实现和进一步完善将实时遥感观测转换为有效洪水淹没信息,有望为洪水灾害应急决策提供重要支撑。

Floods are often characterized by their strong suddenness and wide impact range.Developing a fast and accurate Web-based system for flood monitoring,preliminary disaster assessment and information dissemination can mitigate the impact of floods to a certain extent and provide an important guarantee for the efficient implementation of disaster emergency response.A flood monitoring and loss assessment system is built based on WebGIS technology and multi-source satellite remote sensing data.The system adopts Flask as the Web application framework in Python environment,uses open source GIS package to build the server,and combines the front-end Vue.js framework,Leaflet and other components to complete the construction of the flood disaster monitoring system,providing users with rapid query of flood inundation extent information and disaster assessment functions.The remote sensing-based flood inundation extent is automatically extracted and published based on HydroFloods,Wxee and other GEE toolkits.The disaster assessment function is implemented by using the GDAL library for spatial statistical analysis of inundation extent and population density and other basic geographic information data.Application of the system to the flood monitoring in Liulin Town,Suizhou City,Hubei Province on 12 August 2021 shows that the inundation extent extraction results are generally consistent with the official reports and the damage assessment results are lower than the actual statistics.The implementation and further improvement of the system converts real-time remote sensing observations into effective flood inundation information,which is expected to provide important support for flood disaster emergency decision-making.