Revealing the spatio-temporal variability of evapotranspiration and its components based on an improved Shuttleworth-Wallace model in the Yellow River Basin

Identifying the spatio-temporal variations of evapotranspiration (ET) from its components (soil evaporation and plant transpiration) can greatly improve our understanding of water-cycle and biogeochemical processes. However, partitioning evapotranspiration into evaporation (E) and transpiration (T) at regional scale with high accuracy still remains a challenge. This study has aimed to reveal the spatio-temporal variations of evapotranspiration and its components by using an improved Shuttleworth-Wallace (SWH) model to partition ET in the Yellow River Basin during 1981-2010. The environmental factors affecting the spatial and temporal variations of evapotranspiration and its components were also assessed. Results showed that the mean annual ET, T and E in the Yellow River Basin were 372.18 mm, 179.64 mm, and 192.54 mm, respectively, over the last 30 years. The spatial pattern of mean annual ET and T displayed a decreasing trend from southeast to northwest in the Yellow River Basin, and the temporal variation showed a significant increasing trend with rates of 1.72 mm yr(-1) and 1.54 mm yr(-1), respectively. It meant that T accounted for the variations of ET, while E showed no significant changes in recent decades. Moreover, the normalized differential vegetation index (NDVI) and temperature were identified as the main factors controlling the variations of ET and T in the Yellow River Basin. Among them, the area with NDVI as the dominant factor for ET and T could reach 63.82% and 78.47% of the whole basin respectively. However, the variations of E were affected by complex factors, and evaporation in the western alpine region was mainly controlled by temperature. Our findings are expected to not only have implications for developing sustainable policies of water management and ecological restoration in this region, but also provide valuable insight in methodology of ET partitioning in regional or global scale.