Spatio-temporal patterns of grassland evapotranspiration and water use efficiency in arid areas

Understanding spatio-temporal patterns of grassland evapotranspiration (ET) and water use efficiency (WUE) in arid areas is important for livestock production and ecological conservation. Xinjiang, China, was used as an example in the Biome-BGC model to explore spatio-temporal patterns of grassland ET and WUE from 1979 to 2012 in arid areas. The ET ranked from high to low as follows: among seasons, summer (142.4 mm), spring (49.7 mm), autumn (45.9 mm) and winter (7.7 mm); among regions, the Tianshan Mountains (357.9 mm), northern Xinjiang (221.3 mm) and southern Xinjiang (183.2 mm); among grassland types, mid-mountain meadow (387.7 mm), swamp meadow (358.3 mm), typical grassland (343.9 mm), desert grassland (236.2 mm), alpine meadow (229.7 mm), and saline meadow (154.7 mm). The WUE ranked from high to low as follows: among seasons, summer (0.60 g C kg H₂O^?1), autumn (0.48 g C kg H₂O^?1) and spring (0.43 g C kg H₂O^?1); among regions, northern Xinjiang (0.73 g C kg H₂O^?1), the Tianshan Mountains (0.69 g C kg H₂O^?1) and southern Xinjiang (0.26 g C kg H₂O^?1); among grassland types, mid-mountain meadow (0.86 g C kg H₂O^?1), typical grassland (0.84 g C kg H₂O^?1), swamp meadow (0.77 g C kg H₂O^?1), saline meadow (0.52 g C kg H₂O^?1), alpine grassland (0.37 g C kg H₂O^?1) and desert grassland (0.34 g C kg H₂O^?1). In Xinjiang grasslands, the spatio-temporal ET patterns were more strongly influenced by precipitation than by temperature, whereas most high WUE values occurred when precipitation and temperature were relatively conducive to grass growth.