Wang Jing, Gao Yongnian
School of Earth Sciences and Engineering, Hohai University, Nanjing, Jiangsu 211100, China.
Key Laboratory of Soil and Water Processes in Watershed, College of Geography and Remote Sensing, Hohai University, Nanjing 211100, China.
Sci Total Environ. 2025 Feb 25;966:178662. doi: 10.1016/j.scitotenv.2025.178662. Epub 2025 Feb 4.
Accurate underwater topography and water storage data of lakes and reservoirs are crucial for evaluating their functionality and sustainable water resources management. However, due to the scarcity of in-situ data, acquiring such information for remote lakes and reservoirs remains challenging. In this study, we derived multiple underwater elevation points of 14 lakes and reservoirs in the Tarim River Basin of China using ICESat-2 data, combined with the cubic B-spline fitting method. Subsequently, we utilized multi-temporal Sentinel-2 remote sensing data to map the underwater topography of these 14 lakes and reservoirs by inverting underwater elevation. Monthly time series of water area, levels, and storage were subsequently derived, offering a long-term perspective on the water dynamics of the 14 lakes and reservoirs from 1990 to 2021. The results indicate that, compared with the existing contour data of A.lake (Bosten Lake), the underwater elevation points exhibits high accuracy, with an R of 0.90, a mean-absolute-error (MAE) of 0.72 m, and a root-mean-square error (RMSE) of 1.01 m. Additionally, the underwater topographic map of A.lake (Bosten Lake) demonstrates a high degree of fit, with an R of 0.86, a MAE of 1.27 m, and an RMSE of 1.69 m. The findings of this study suggest that ICESat-2 data, combined with optical remote sensing imagery, are effective for mapping lakes and reservoirs in remote inland regions. Additionally, the storage dynamics of these 14 lakes and reservoirs between 1990 and 2021 reveal a notable interannual increase in water storage within the study area during the 2010-2012 period, whereas changes in the upper limit of reservoir storage were comparatively less significant. This study offers an effective approach for investigating the underwater topography and water storage of remote lakes and reservoirs with limited in-situ data, providing valuable reference information for characterizing these water bodies and managing water resources.
湖泊和水库准确的水下地形及蓄水数据对于评估其功能和可持续水资源管理至关重要。然而，由于原位数据稀缺，获取偏远湖泊和水库的此类信息仍然具有挑战性。在本研究中，我们利用ICESat-2数据并结合三次B样条拟合方法，获取了中国塔里木河流域14个湖泊和水库的多个水下高程点。随后，我们利用多时相哨兵-2遥感数据通过反演水下高程来绘制这14个湖泊和水库的水下地形。随后得出了水域面积、水位和蓄水量的月度时间序列，提供了1990年至2021年这14个湖泊和水库水动力的长期视角。结果表明，与A湖（博斯腾湖）现有的等高线数据相比，水下高程点具有较高的精度，相关系数R为0.90，平均绝对误差（MAE）为0.72米，均方根误差（RMSE）为1.01米。此外，A湖（博斯腾湖）的水下地形图显示出高度的拟合度，相关系数R为0.86，平均绝对误差为1.27米，均方根误差为1.69米。本研究结果表明，ICESat-2数据与光学遥感影像相结合，对于绘制偏远内陆地区的湖泊和水库是有效的。此外，这14个湖泊和水库在1990年至2021年期间的蓄水动态显示，研究区域内蓄水量在2010 - 2012年期间有显著的年际增加，而水库蓄水量上限的变化相对较小。本研究为在原位数据有限的情况下调查偏远湖泊和水库的水下地形及蓄水量提供了一种有效方法，为表征这些水体和管理水资源提供了有价值的参考信息。
