Research On The Impact Of The Three Gorges Project On Spatio-temporal Variations Of Groundwater Environment In The Western Jianghan Plain

Groundwater is a key component of the global/regional water cycles, which sustains the ecological environment, industrial and agricultural production, and human well being. Climate change and various human activities have caused adverse environmental problems, such as groundwater depletion, water quality deterioration, land subsidence, and ecosystem degradation in many regions of the world. Large-scale dam construction made over half (172 out of 292) large river systems fragmented by dams. Numerios studies have confirmed that dam operation alters downstream flow dynamics, while less attention has been giving to the evaluation of dams'effects on downstream groundwater environments. The Three Gorges Project (TGP), the world's largest hydroelectric dam, initiated water impoundment in June 2003. Since then, the Yangtze River's flow downstream from the dam has been regulated under a specific operational rule, which altered the river-aquifer interactions. The Jianghan Plain, immediately downstream from the TGP, is a major farming area, where groundwater is the main source of local drinking and industrial water supply. Therefore, identifying the spatio-temporal changes of the groundwater environment in the Jianghan Plain, and quantitatively evaluating the impact of natural processes and human activities, especially the impoundment of the TGP, on groundwater environment in this area is of great significance for the research regarding global dam's impact on downstream groundwater environment and local groundwater resource management.This study relied on 63,109 measurements of groundwater levels at 46 confined groundwater monitoring wells during 1990-2010, and on groundwater quality data for 18 parameters at 29 sampling wells during 1992-2010 within the Western Jianghan Plain (WJHP). The variation characteristics of areal average precipitation and temperature, outflows of the Gezhouba Dam (GZD), Yangtze River stages, and groundwater levels were first determined, and the TGP's impact on downstream groundwater levels were assessed via coupling seasonal adjustment (STL) with successive ARIMAX dynamic regression models. The spatio-temporal variations of groundwater quality and their main causes were subsequently studied based on trend detection, multivariate statistics, and groundwater chemistry theory. The TGP's potential effect on the groundwater quality was also discussed. This research's main findings and results are as follows:(1) Multi-year variations of the groundwater levels and other areal hydro- meteorological variables were detected by means of descriptive statistics and the seasonal Kendall test. The areal average precipitation remained stable, whereas the temperature exhibited significant increasing trends. The outflows of the Gezhouba Dam (GZD) experienced a reduction of 141.75 m3/s during 1990-2010. The average reductions of the Yangtze River's stages at 4 hydrologic stations (Zhicheng, Shashi, Jianli, and Luoshan) were 0.89 m,0.57 m,0.30 m and 0.70 m, respectively. The average, highest and lowest groundwater levels in the WJHP were all decreased after the impoundment. The annual decline rates of the 35 wells with significant declining trend ranged from -0.87 to -23.53 cm/yr, which exceed in absolute value the range of increasing trend (0.55-13.00 cm/yr) observed at 10 other wells.(2) The TGP's impact on the seasonal variations of the regional groundwater level were analyzed by comparing the seasonal components of the hydro-meteorological variables before and after the impoundment based on the STL decomposition technique. There were minor seasonal changes in precipitation and temperature during the two periods. The seasonal pattern of the GZD outflow changes after the impoundment coincided with the dispatch operation mode of the TGR, whose water releases decrease from June through October (average reduction equals 2.52×103 m3/s), with the exception of September, and increased from December through May by an average equal 1.47×103 m3/s. The altered seasonality prevailed at the 4 river stations within the WJHP. The TGP operation dampened the average annual amplitude of the regional groundwater level by 32.5%. The seasonal components of the regional groundwater level increased during January through April, which remained stable in May and September, and decreased at varying degrees in other months(3) The actual correlations between the hydro-meteorological variables were obtained by applying seasonal adjustment and pre-whitening to the original series before calculating the Cross Correlation Function. The outflow of the GZD is significantly correlated with the precipitation of the previous month and the current month. The correlations between the GZD outflow and the Yangtze River stages at the 4 hydrologic stations decreased along the channel, with the coefficient for the current month being 0.799,0.836,0.797, and 0.637 at the Zhicheng, Shashi, Jianli, and Luoshan stations, respectively. The effect of the precipitation on groundwater levels lags 2-3 months, while the temperature exhibits nonsignificant negative correlation with groundwater levels. The exponential distance-decay patterns were observed in the correlation coefficients between groundwater levels and GZD outflows and between groundwater levels and the Yangtze River stages at the corresponding nearest stations, which indicates that the former correlations were generally larger than the latter ones. Significant relations occurred mainly within 10 km from the river bank.(4) Flow-regulated and undisturbed groundwater scenarios were estimated by coupling STL seasonal adjustment and ARIMAX dynamic regression models to quantify the TGP's impact on trends of the groundwater levels. A comparision of the forecasted undisturbed GZD outflow/Yangtze River stages after the impoundment and the measured GZD outflows revealed that the latter decreased 829.8 m3/s on average, and that the measured Yangtze River stages decreased on average 0.72 m,0.71 m,0.22 m,0.56 m at the Zhicheng, Shashi, Jianli, and Luoshan stations, respectively. The TGP-induced average reduction of groundwater levels ranged between 0 and 0.25 m at various monitoring wells. The influence of the TGP's streamflow regulation decayed exponentially with increasing distance from the Yangtze River bank, and almost vanished beyond 6 km from the river bank. The lowered groundwater level would cause enhanced soil oxidation along the Shimatou-Xiaogang Section in Honghu, which is proven by the increased average Eh values (65.81% increase), decreased Fe2+ concentration (-30.24% reduction), and decreased total reducing substances (-31.00% reduction) during 2004-2008.(5) Multiple statistical methods, trend analysis, and theories of groundwater chemistry were applied to study the spatio-temporal characteristics and main causes of groundwater quality variations. The confined groundwater is predominantly of the HCO3-Ca-Mg type, which is neutral to slightly alkaline. The total dissolved solids (TDS) value varied from 287.86 to 1,380 mg/L, which increased along the groundwater flow direction from northwest to the southeast. Cl-, SO42-, NH4-N, and NO3-N exhibited remarkable spatial variations. The correlation analysis revealed that Ca2+, Mg2+, K+, HCO3-, OD, Fe2+, CO2, and TDS exhibited significant pairwise positive correlations, and thus form a relatively stable cluster, while their correlations decreased in 1998-2003 and 2004-2010. The correlations between pH and Ca2+ and between pH and CO2 became increasingly negative during the latter periods. Cl- and SO42- are exclusively correlated elements in 1992-1997 and 1998-2003. Thereafter, NO3- begin to establish positive correlations with Cl- and SO42-. The seasonal Mann-Kendall test revealed that pH, NO3-N, and Cl" concentrations at 27,26 and 15 wells, respectively, exhibited significant increasing trends during 1992-2010. Meanwhile, the concentration of HCO3-and CO2 at 16 and 15 wells, respectively, exhibited significant decreasing trends. The increase of pH may be attributed to CO2 degassing caused by large-scale groundwater extraction. Increased regional average concentrations of NO3-N, Cl- and SO42- were caused by increased use of fertilizer. Abnormally high values of Cl- and SO42- at some wells were induced by industrial chemicals. The change-point analysis on the average pH values of the 29 wells demonstrated that the groundwater pH experienced an abrupt change at about 2003, which increased?0.43 pH units as a whole. The year 2003 coincided with the start of operation of the TGP. In addition, the similarity of the temporal variations of the regional average of pH, NH4-N, and NO3-N concentrations in groundwater with those in the Yangtze River at the outlet of the Three Gorges Reservoir (TGR) suggests that the variations of these parameters in the WJHP was partly due to water storage by the TGR.