Drought effects on evapotranspiration and energy exchange over a rain-fed maize cropland in the Chinese Loess Plateau

Characterizing dynamics of evapotranspiration (ET) and energy exchange is of particular importance for the water resource use and management in the water-limited agricultural areas under global climate change. Based on eddy covariance method, this study analyzed the seasonal and interannual variations of ET and energy fluxes over a rain-fed spring maize cropland in the Chinese Loess Plateau during 2 years which presented contrasting precipitation (P) totals and patterns in 2019 (drought year) and 2020 (normal year). Results showed obvious intra- and inter-annual variabilities of ET and energy fluxes related with crop phenology and water availability. Annual ET were 339.3 and 386.5 mm yr–1 in 2019 and 2020, respectively, with separate annual ET/P ratios of 0.84 and 0.70. Annual ET was dominated by ET during growing season (May−September), with close proportions of 75.8% and 72.9% for the two years. The partitioning of net radiation (Rn) was dominated by latent heat flux (LE) during growing season, but was primarily driven by sensible heat flux (H) during non-growing season. Annual ratios of LE/Rn and H/Rn were comparable in the drought year (0.46 −0.47), while LE/Rn (0.52) dominated over H/Rn (0.46) in the normal year. Similar seasonal courses were observed in surface conductance (gs), decoupling factor (Ω), and Priestley-Taylor coefficient (α), consistently peaking during the rapid growth period of maize (July−August). The low annual means of Ω (0.20 −0.25) and α (0.44 −0.45) implied the strongly aerodynamical coupling and biophysical control on ET for the study area. Moreover, daily α in this study increased significantly with increasing gs until a threshold around 10 mm s−1. Structural equation modelling revealed that the environmental constraints on ET were mainly affected by Rn, soil water content and vapor pressure deficit (VPD) in the drought year, and prone to be dominated by Rn, leaf area index (LAI) and VPD in the normal year. These results indicated that drought may increase the sensitivity of maize ET to water supply condition, and decrease the sensitivity to changes in available energy, aerodynamic condition and vegetation coverage. This study will provide insights into the effects of drought on energy partitioning and the environmental and biophysical controls on ET in the region.