2020年度一带一路洪涝冰湖溃决多发区灾害监测产品

针对东亚地区特别是巴基斯坦，构建洪水提取模型，提取巴基斯坦地区2022年夏季的水体淹没区。使用Google Earth Engine平台上的Sentinel-1 SAR GRD数据、SRTM DEM数据提取洪水，使用东南亚土地覆盖分类图排除洪水提取干扰项，再以Google Earth Engine平台上的Landsat 8数据辅助验证。构建冰湖提取模型，提取喜马拉雅地区2020年冰湖区，基于GEE平台，采用Landsat年度合成数据，利用基于全局水体指数与非局部活动轮廓模型的大尺度冰湖轮廓提取方法，得到了长时间序列的高亚洲地区冰湖编目数据。以30米分辨率描述2008年至2017年冰川湖的年度覆盖率。这是第一个跨HMA的冰川湖泊年鉴，它可以提供不同类型冰川湖的详细信息和演化模式。虽然湖泊制图的方法是自动的，但湖泊都经过人工核查，并在需要时进行手动编辑。它可用于冰川-气候-湖泊相互作用、冰川水文模型、冰湖溃决洪水、潜在下游风险和水资源研究。

Focusing on East Asia, particularly Pakistan, a flood extraction model was developed to extract water inundation areas in Pakistan during the summer of 2022. Flood extraction was conducted using Sentinel-1 SAR GRD and SRTM DEM data available on the Google Earth Engine (GEE) platform, with the Southeast Asia land cover classification map employed to eliminate interference factors during flood extraction, and Landsat 8 data on the GEE platform used for supplementary validation. A glacial lake extraction model was also constructed to identify glacial lake areas in the Himalayan region in 2020. Based on the GEE platform, Landsat annual composite data was utilized, combined with a large-scale glacial lake outline extraction method based on the Global Water Index and Non-local Active Contour Model, to generate long-time series glacial lake inventory data for High Mountain Asia (HMA). The annual coverage of glacial lakes from 2008 to 2017 is documented at a 30-meter resolution. This is the first cross-HMA glacial lake yearbook, which can provide detailed information and evolutionary patterns of various types of glacial lakes. While the lake mapping methodology is automated, all lakes have undergone manual inspection and manual editing when required. This dataset can be applied to research on glacier-climate-lake interactions, glacial hydrological models, glacial lake outburst floods (GLOFs), potential downstream risks, and water resources.