利用Google Earth Engine(GEE)平台,采用基于Sentinel-1双极化水体指数(Sentinel-1 dual-polarized water index,SDWI)的Otsu最佳阈值法对鄱阳湖区进行水体范围提取,并对水体面积进行曼-肯德尔(Mann-Kendall)趋势检验,同时对2015—2022年淹没频率的空间格局变化进行统计分析.结果表明:①基于GEE平台,利用Sentinel-1数据可以实现水体面积高时相频率的动态监测,识别精度最高达到0.93;②2015—2021年鄱阳湖区水体面积整体呈显著上升趋势(Z>0,P<0.01),其中,2018年枯水年水体面积波动较2020年丰水年小;③鄱阳湖区西南及东南部分季节性水体广布,水体空间格局变化最为显著,研究时段内淹没频率为(0,20％]的面积年际变化最大.研究表明,基于GEE平台和Sentinel-1雷达数据可以有效监测鄱阳湖区的水体面积动态变化,能够为区域洪涝灾害风险管理、湿地栖息地保护提供支持.
The Spatiotemporal Pattern Analysis of Water Body Area Change in Poyang Lake Region Based on GEE Platform
Leveraging the Google Earth Engine (GEE) plat-form,the Otsu optimal threshold method based on Sentinel-1 dual-polarized water index (SDWI) was used to extract the extent of water bodies in the Poyang Lake area,and the area of water bodies was analyzed by Mann-Kendall trend test,as well as statistical analysis of the spatial pattern changes of in-undation frequency from 2015 to 2022. The results show that,first,the dynamic monitoring of high temporal phase fre-quency of water body area can be realized by using Sentinel-1 data combined with GEE platform,and the recognition accu-racy reaches up to 0.93. Second,the overall water body area of Poyang Lake area shows a significant increasing trend from 2015 to 2021 (Z>0,P<0. 01),among which the fluctua-tion of water body area in 2018 dry water year is smaller than that in 2020 abundant water year. Third,seasonal water bod-ies are widely spread in southwest and southeast parts of Poy-ang Lake area,and the spatial pattern of water bodies is the most significant change,and the area with inundation frequen-cy of (0,20％]has the largest interannual variation during the study period. The study shows that based on GEE platform and Sentinel-1 radar data can effectively monitor the dynamic changes of water body area in Poyang Lake area,which can provide support for regional flood disaster risk management and wetland habitat protection.
Sentinel-1Google Earth EnginePoyang Lake regionwater areaSentinel-1 dual-polarized water indexspatiotemporal dynamics
刘修钰、朱青、王薇、蔡永立
展开 >
上海交通大学设计学院中国城市治理研究院,上海,200240
自然资源部国土空间生态治理数字工程技术创新中心,上海,200240
Sentinel-1 Google Earth Engine 鄱阳湖区 水体面积 Sentinel-1双极化水体指数 时空变化
NETL
NSTL
万方数据
