面向鄱阳湖洪涝风险分析的HY-1C/D卫星CZI影像水体面积与水位关系研究

海洋一号C星(HY-1C)和海洋一号D星(HY-1D)搭载的海岸带成像仪(coastal zone imager，CZI)广泛应用于海洋、海岛和海岸带监测，同时也是内陆水体动态监测的重要数据源.该文选择水体范围变化剧烈、旱洪灾害频发的鄱阳湖为研究区，以瑞利校正后的CZI反射率数据为基础，利用归一化水体指数(normalized difference water index，NDWI)与归一化植被指数(normalized difference vegetation index，NDVI)联合的水体范围提取方法，相对精度达到96%，并获取了2019年3月至2020年8月鄱阳湖湖泊淹没面积.在此基础上，结合同步的水文站水位测量数据，建立了适用于鄱阳湖的水面积－水位定量关系模型(R＝0.97).分析结果显示，鄱阳湖9月中旬至1月，水体面积最小，水位最低；2月至6月，水体面积逐渐增大，水位逐渐抬升；7月和8月，水体面积和水位达到年内峰值.洪涝风险区主要集中在鄱阳湖的东南部和西部.该研究对星地联合的洪涝风险分析有一定的借鉴意义.

Relationship between HY-1C/1D CZI-based lake inundation area and water level for flood disaster analysis of Poyang Lake

Coastal zone imager (CZI) onboard HY-1C and HY-1D is widely used in ocean， islands and coastal zones monitoring， as well as an important data source for dynamic inland water monitoring. In this paper， Poyang Lake with a wide range of water bodies and frequent drought and flood disasters was taken as an example and a water area extraction algorithm integrating NDWI (normalized difference water index) and NDVI (normalized difference vegetation index) was established based on CZI reflectance with Rayleigh scattering correction. The relative accuracy reached 96%， and the inundation area of Poyang Lake from March 2019 to August 2020 was obtained. On this basis， combined with the synchronous water level measurement data of hydrological stations， a quantitative relationship model between water area and water level suitable for Poyang Lake (R＝ 0.97) was established. The results showed that the water area and water level of Poyang Lake was the smallest and lowest from mid-September to January， which increased gradually from February to June and reached the annual peak in July and August. Flood risk areas were mainly concentrated in the southeast and west regions. This study has certain reference significance for the flood risk analysis of the satellite-terrestrial union.