B41I-0173:
Systematic High-Resolution (30 meter) Inventory of Global Lakes: Pan-Arctic and Beyond

Thursday, 18 December 2014
Yongwei Sheng1, Jida Wang2, Laurence C Smith1, Evan A Lyons1, Gary Te1, Jordan Woods1, Dorian Garibay1 and Benjamin Knox1, (1)University of California Los Angeles, Los Angeles, CA, United States, (2)Kansas State University, Manhattan, KS, United States
Abstract:
[Abstract] Lakes play a crucial role in the global water cycle and balance, are sensitive to global warming, and are vital for human water supply. There clearly is a pressing need to understand temporal and spatial variations in lake water storage globally, especially in the lake-rich Arctic/Sub-Arctic regions. An accurate systematic lake inventory, however, is unavailable at global scales. Owing to its broad spatial coverage and repeat-pass monitoring capability, satellite remote sensing is the only feasible approach to inventory global lake dynamics. Global lake mapping at high resolutions is a rather challenging task. Since lakes are abundant in small-size classes and their seasonality varies from region to region, a huge number of high-resolution satellite images need to be acquired in “appropriate” seasons. The appropriate seasons refer to the period in a typical year when lakes are relatively stable, and are determined spatially using precipitation and temperature datasets. Thousands of cloud-free Landsat images at 30 m resolution have been acquired during lake-stable seasons. Satellite lake mapping succeeds at different levels from place to place and from season to season. A set of highly replicable automated lake mapping methods and tools have been developed to tackle various situations across the entire Earth and to handle such a large volume of satellite data. The current goal is to produce a circa 2000 high-resolution global lake database in a systematic way. Millions of lakes larger than 0.5 ha have been inventoried. The product is currently examined in an intensive quality control and quality assurance process. Over six million lakes in pan-Arctic (45 deg N and above) are captured in the database and have been validated for release.