资源环境科技发展态势分析平台

Water resource management is of critical importance due to its close relationship with nearly every industry, field, and lifeform on this planet. The success of future water management will rely upon having detailed data of current and historic water dynamics. This research leverages Google Earth Engine and uses Landsat 5 imagery in conjunction with bathymetry and Shuttle Radar Topography Mission digital elevation model data to analyze long-term lake dynamics (water surface elevation, surface area, volume, volume change, and frequency) for Lake McConaughy in Nebraska, USA. Water surface elevation was estimated by extracting elevation values from underlying bathymetry and digital elevations models using 5,994 different combinations of water indices, water boundaries, and statistics for 100 time periods spanning 1985-2009. Surface elevation calculations were as accurate as 0.768 m root mean square error (CI95% [0.657, 0.885]). Water volume change calculations found a maximum change of 1.568 km(3) and a minimum total volume of only 23.97% of the maximum reservoir volume. Seasonal and long-term trends were identified, which have major affects regarding regional agriculture, local recreation, and lake water quality. This research fills an existing gap in optical remote sensing-based monitoring of lakes and reservoirs, is more robust and outperforms other commonly used monitoring techniques, increases the number of water bodies available for long-term studies, introduces a scalable framework deployable within Google Earth Engine, and enables data collection of both gauged and ungauged water bodies, which will substantially increase our knowledge and understanding of these critical ecosystems.