Longterm multisource satellite data fusion reveals dynamic expansion of lake water area and storage in a hyperarid basin of China

• 全文链接
• 请求原文

作   者：

Chuanhui Zhang;  Aifeng Lv;  Shaofeng Jia;  Shanshan Qi; 

作者机构：

China;  Key Laboratory of Water Cycle and Related Land Surface Processes;  Beijing 100101;  Institute of Geographic Sciences and Natural Resources Research;  CAS; 

关键词：

Multisource satellite data;  Lakes;  Long-term changes;  Qaidam Basin;  Spatiotemporal variations; 

期刊名称：

Journal of Hydrology

i s s n：

0022-1694

年卷期：

2022 年 610 卷 Pt.2 期

页   码：

1278881-1-127888-14

页   码：

摘   要：

Lakes play an important role in global hydrologically and provide substantial water resources in hyperarid re-gions, and they are becoming even more significant as climate change impacts hydrological processes, especially in these sensitive regions. This study proposes a lake-monitoring framework that uses multisource satellite data to automatically estimate the spatiotemporal variations from 1987 to 2020 of lake area, lake water level, and lake water storage (LWS) in the Qaidam Basin (QB), which is a typical hyperarid basin in northwest China. The global surface water dataset provides long-time-series data for lake areas. Multisource satellite images (Landsat 7, Landsat 8, Sentinel 2) and altimetry data (CryoSat 2, ICESat, ICESat 2, Sentinel 3) are used to build hypso-metric curves (the fit to the curve of lake area vs lake water level) to reconstruct lake water level before the 21st century. In contrast with traditional methods, we propose to verify the construction of a hypsometric curve to select a reasonable curve and thereby improve the accuracy of the extrapolation of lake water level. On this basis, we estimated the variation of LWS from 1987 to 2020 and to use products of the gravity recovery and climate experiment and of land surface model simulations to estimate the variation in total LWS in the QB for comparative analysis. The results show that the total LWS in the QB has increased by 82.63% over the past 33 years. Compared with the mean value from 2004 to 2009, the total LWS in 2020 increased by 4.41 km~3, which is equivalent to 12.53% -19.49% of the increase in terrestrial water storage. This ratio cannot be ignored. The analysis of spatiotemporal variations in lakes shows that large shallow lakes are disappearing, leading to ecological deterioration in some areas. Conversely, there is increased water storage in small deep lakes, which may be beneficial for water conservation in hyperarid basins.