Retrieval study of lake water depth by using multi-spectral remote sensing in Bangong Co Lake

This paper reviewed studies on remote sensing of water depth retrieval. Four water depth retrieval models(single-band, double-ratio-band, multi-band, and BP network models) were evaluated using TM image and water data from Bangong Co Lake, which is located in China's Tibet Autonomous Region and Indian Kashmir. Tested by independent data, comparison of these four models demonstrates that BP network model performed better than the multi-band model, with the single-band model performing the worst. To sum up, this study demonstrates that first, BP network model performed better than the traditional model; second, precise atmospheric correction and radiation study, affected by different water level sand sediment, could improve the precision of water depth retrieval.