Calculation of Bosten Lake Water Storage Based on Multiple Source Remote Sensing Imagery
Weiwei Wang,Fei Zhang,Jingchao Shi,Qi Zhao,Changjiang Liu,Mou Leong Tan,Hsiang-Te Kung,Guang Gao,Gang Li
DOI: https://doi.org/10.1109/tgrs.2023.3338635
IF: 8.2
2023-12-20
IEEE Transactions on Geoscience and Remote Sensing
Abstract:Bosten Lake is a crucial water source in arid northwest China, which has been maintaining the ecological balance of the southern Xinjiang region. It contributes to the sustainable development of the local economy and watershed ecology. Insufficient hydrological data for the basin, however, cause uncertainty in hydrological modeling and makes it difficult to calculate water availability and lake-water storage using conventional methods. Therefore, this article proposed a novel method to retrieve the lake's water depth from Landsat-8 OLI and ICESat-2 satellite data using neural network (NN) model. Specifically, the Rayleigh-corrected top-of-atmosphere (TOA) reflectance ( ) in the 443–2300 nm range was employed as the input to the NN model for the retrieval of water depth. This avoids the requirement to correct the effects of aerosols. In addition, the water depth retrieved by the NN model was compared with a conventional dual-band ratio model (DBRM). To evaluate the accuracy of the corrected ICESat-2 photon data, the in situ water depths were compared with the corrected ICESat-2 photon data (considering water-level variations at different times), and 20 in situ water depth values near the ICESat-2 track were selected for verification. The results showed that 1) the in situ water depth and the corrected ICESat-2 photon data had the coefficient of determination ( ) values of 0.94; 2) NN inversion of bathymetry presented a feasible method of obtaining regional bathymetry, with of 0.87, root mean square error (RMSE) being 1.17 m, and mean absolute error (MAE) being 0.26 m. In contrast, DBRM generated the , RMSE, and MAE values of 0.74, 1.62, and 0.93 m, respectively; and 3) the water storage in Bosten Lake's from April to September in 2022 ranged from to . The findings can provide a strong scientific basis for the sustainable development and the rational allocation regional water resources.
imaging science & photographic technology,remote sensing,engineering, electrical & electronic,geochemistry & geophysics