Alaska Lake and Pond Occurrence Dataset

The Alaska Lake and Pond Occurrence Dataset (ALPOD) is a spatially explicit map of seasonal fluctuations in lake and pond extent across Alaska. ALPOD is constructed from Sentinel-2 imagery and its lake occurrence raster has a 10 m resolution, with each pixel recording the percentage of observations classified as open water during the 2016-2021 ice-free seasons at near-daily temporal resolution. The number of water bodies depends on the occurrence threshold, but a conservative estimate is that ALPOD maps over 800,000 lakes and ponds larger than 0.001 square km. The data format includes lake occurrence rasters tiled by UTM zone and latitude, as well as a vector product defined based on a 25% occurrence threshold. ALPOD was created using a U-Net lake identification model and manual inspection to produce a maximum possible lake extent mask. This mask serves as the region of interest for an adaptive NDWI threshold water classification algorithm written in Google Earth Engine, which was used to classify open water within the maximum lake mask in every available cloud and ice free Sentinel-2 image during the study period. The resulting dataset is suitable for investigations of individual water bodies as well as lake and pond patterns across Alaska.

Geotiffs included here are 10 m resolution lake occurrence rasters with each pixel recording the percentage of observations classified as open water during the 2016-2021 ice-free seasons. They are tiled by UTM zone above 60 degrees north. Regions below 60 degrees north are separated into two geotiffs.

阿拉斯加湖泊与池塘出现数据集（Alaska Lake and Pond Occurrence Dataset, ALPOD）是一套覆盖阿拉斯加全域、具备空间显性属性的湖泊与池塘面积季节变化图谱。ALPOD基于哨兵2号（Sentinel-2）影像构建，其湖泊出现栅格产品分辨率为10米，每个像素记录了2016-2021年无冰季内，以近每日时间分辨率获取的观测中被分类为开阔水域的占比。水体的数量取决于出现阈值，保守估计ALPOD可绘制超过80万个面积大于0.001平方千米的湖泊与池塘。该数据集的数据格式包含按UTM分区与纬度分幅的湖泊出现栅格产品，以及基于25%出现阈值定义的矢量产品。ALPOD通过U-Net湖泊识别模型与人工目视解译构建，以生成最大可能的湖泊范围掩膜。该掩膜被用作自适应归一化差异水体指数（Normalized Difference Water Index, NDWI）阈值水体分类算法的感兴趣区域，该算法基于谷歌地球引擎（Google Earth Engine）编写，用于在研究时段内，对所有可用的无云无冰哨兵2号影像中，最大湖泊掩膜范围内的开阔水域进行分类。最终生成的数据集可用于单个水体的研究，以及阿拉斯加全域的湖泊与池塘格局分析。本次提供的GeoTIFF数据为10米分辨率的湖泊出现栅格产品，每个像素记录了2016-2021年无冰季内，被分类为开阔水域的观测占比。这些栅格按北纬60度以上区域的UTM分区进行分幅，北纬60度以下区域则分为两份GeoTIFF数据。