Detecting changes in water level caused by climate, land cover and dam construction in interconnected river-lake systems
Jie Liang,Yuru Yi,Xiaodong Li,Yujie Yuan,Suhang Yang,Xin Li,Ziqian Zhu,Manqin Lei,Qianfang Meng,Yeqing Zhai
DOI: https://doi.org/10.1016/j.scitotenv.2021.147692
2021-09-20
Abstract:There is a growing recognition of the broader environmental significance of exploring the relative importance of climate change and anthropogenic impacts on hydrologic fluctuations in river-lake systems. In the case of Dongting Lake, the typical river-lake system, we collected the water level from 1990 to 2019, spanning before and after the operation of the Three Gorges Dam (TGD) in 2003. This study was conducted to detect water level fluctuations in Dongting Lake and to quantify the relative influence of climate, land cover and dam construction on water levels. We defined the impact of the dam construction as the three inlets inflow of Yangtze River (In-YR), and four waters inflow of Hunan (In-HN). The Mann-Kendall (M-K) test revealed the trends and change points of water level fluctuations. Structural Equation Model (SEM) was used to detect the direct and indirect effects of these factors on water level and quantify their relative importance. The MIKE21 hydrodynamic model reflected the spatial-temporal variability of water levels under the action of key driver. The results showed that the water level appeared a downward trend during 1990-2019 and the change point appeared in 2003; During 1990-2002, the significant factors were: precipitation (V = 0.469, P = 0.013), evaporation (V = -0.424, P = 0.029), non-agricultural cover (V = -0.334, P = 0.025), and agricultural cover (V = 0.235, P = 0.033); During 2003-2019, the significant factors were: In-YR (V = 0.436, P = 0.007), In-HN (V = 0.431, P = 0.012), and precipitation (V = 0.349, P = 0.045); The In-YR was the key factor affecting the changes of the water level during 1990-2019; Under the influence of In-YR, the most obvious fluctuation of water level was in the flood adjustment period (Jun-Aug) and the impoundment period (Sep-Nov) when the average declined by about 0.50 and 0.67 m, respectively. Our findings provide a new insight into how to better maintain the stability of river-water system water resources under the influence of multiple factors.