GLOBMAP SWF: a global annual surface water cover frequency dataset during 2000–2020

[摘要] The extent of surface water has been changingsignificantly due to climatic change and human activities. However, it ischallenging to capture the interannual changes of inland water bodies due totheir high seasonal variation and abrupt change. In this paper, a globalannual surface water cover frequency dataset (GLOBMAP SWF) was generatedfrom the MODIS land surface reflectance products during 2000–2020 todescribe the seasonal and interannual dynamics of surface water. Surfacewater cover frequency (SWF) was proposed as the percentage of the timeperiod when a pixel is covered by water in a year. Instead of determinationof the water directly, the SWF was estimated indirectly by identifying landobservations among annual clear-sky observations to reduce the influence ofclouds and variability of water bodies and surface backgroundcharacteristics, which helps to improve the applicability of the algorithmfor different regions across the globe. The generated dataset shows betterperformances for frozen water, saline lakes, bright surfaces and regions with frequent cloud cover compared with the two high-frequency surface water datasets derivedfrom MODIS data, and it captures more intermittent surface water but mayunderestimate small water bodies when compared with two high-resolutiondatasets derived from Landsat data. Compared with the high-resolution SWFmaps extracted from Sentinel-1 data in eight regions that cover lakes, riversand wetlands, the R 2 reaches 0.46 to 0.97, RMSE ranges from 7.24 %to 22.62 %, and MAE is between 2.07 % and 7.15 %. In 2020, the area ofglobal maximum surface water extent is 3.38×10 6  km 2 , of which thepermanent surface water accounts for approximately 54 % ( 1.83×10 6  km 2 ), and the other 46 % is intermittent surface water ( 1.55×10 6  km 2 ). The area of global maximum and permanent surface water has beenshrinking since 2001, with a change rate of −7577 and −4315  km 2  yr −1 ( p <0.05 ), respectively, while the intermittent surfacewater with the SWF above 50 % has been expanding (1368 km 2  yr −1 , p <0.01 ). This dataset can be used to analyze the interannualvariation and change trend of highly dynamic inland waters extent withconsideration of its seasonal variation. The GLOBMAP SWF data are availableat https://doi.org/10.5281/zenodo.6462883 (Liu and Liu, 2022).

[学科分类] 眼科学