Understanding the spatial-temporal dynamics of evapotranspiration in relation to climate change and human activities is crucial for the sustainability of water resources and ecosystem security, especially in regions strongly influenced by human impact. In this study, a process-based evapotranspiration (ET) model in conjunction with the Global Land Surface Satellite (GLASS) LAI dataset was used to characterize the spatial-temporal pattern of evapotranspiration from 1982 to 2016 over the Gan River basin (GRB), the largest sub-basin of the Poyang Lake catchment, China. The results showed that the actual annual ET (ETa) weakly increased with an annual trend of 0.88 mm year−2 from 1982 to 2016 over the GRB, along with a slight decline in annual potential ET (ETp). On an ecosystem scale; however, only the evergreen broadleaved forest and cropland presented a positive ETa trend, while the rest of the ecosystems demonstrated negative trends of ETa. Both correlation analysis and sensitivity analysis revealed a close relationship between ETa inter-annual variability and energy availability. Attribution analysis illustrated that contributions of climate change and vegetation greening on the ETa trend were −0.48 mm year−2 and 1.36 mm year−2, respectively. Climate change had a negative impact on the ETa trend over the GRB. However, the negative effects have been offset by the positive effects of vegetation greening, which mainly resulted from the large-scale revegetation in forestland and agricultural practices in cropland. It is concluded that large-scale afforestation and agricultural management were the main drivers of the long-term evolution of water consumption over the GRB. This study can improve our understanding of the interactive effects of climate change and human activities on the long-term evolution of water cycles.
Application of complementary relationship (CR) approaches using only routine meteorological data is a very convenient method of estimating actual evapotranspiration (ETa). Reanalysis datasets and remote sensing data provide good tools to overcome the difficulties in obtaining observation data. This study of the Hotan River Basin (HRB) in northwest China serves as a prime example for estimation of ETa during 2006–2014 by using the modified generalized CR. Based on comparison and analysis, the ETpmax defined by Szilagyi et al. (2017) was adopted. The estimated ETa rates were verified using a regional water balance method at annual time scales because of the limited available data. The calibration parameter σ was calibrated based on the elevation and underlying surface types. The mean annual ETa ranged from 2.3 mm to 800 mm during 2006–2014. ETa rates in the plains regions were higher than those in the mountainous regions. Most of ETa was concentrated in the months of May to September. A water deficit occurred in the middle and lower regions, while a water surplus occurred in the upper regions. This study not only provided a new concept for calibration, but also a potential solution for different underlying surfaces and time scales.