干旱区生态环境知识资源中心

Shallow groundwater is primarily discharged via evapotranspiration (ETg) in arid and semi-arid riparian systems; however, the quantification of ETg remains a challenge in regional water resource assessments of such systems. In this study, the diagnostic indicators of groundwater evapotranspiration processes and the principles of applying the water table fluctuation (WTF) method to estimate ETg based on seasonal groundwater level changes were presented. These techniques were then used to investigate groundwater evapotranspiration processes at two sites dominated by phreatophytes (Tamarix ramosissima and Populus euphratica) within hyper-arid desert environments in northwestern China for the period 2010-2012. The results indicate that steady declines in the water table, which are commonly attributed to groundwater evapotranspiration, occurred at both sites during the growing season. Based on the proposed WTF method, the estimated ETg was 0.63-0.73 mm/d at the Tamarix ramosissima site and 1.89-2.33 mm/d at the Populus euphratica site during the summer months (June-August). Numerical simulations using a one-dimensional root water uptake model indicate that the seasonal variations in ETg at both sites were primarily dependent on the potential evaporation rates. Comparisons with previous studies on plant transpiration at similar sites in this area show that these results are reasonable. It is apparent that the WTF method can provide a simple and relatively inexpensive method of estimating ETg on a large scale in arid/semi-arid regions. However, there are significant uncertainties associated with time-dependent lateral flow rates, which creates a challenge when applying this method. In addition, the selection of calculation periods that show steady declines in the groundwater level can be somewhat subjective. To enhance the performance of the WTF method based on seasonal water table declines, further research on the estimation of lateral flow rates should be performed using an effective network of groundwater monitoring. (C) 2014 Elsevier B.V. All rights reserved.