Evapotranspiration as the key component of the terrestrial water cycle, an accurate estimates of evapotranspiration are of great importance for water irrigation management. Although many applicable ET models have been developed, these are largely focused on low altitude regions, with little attention to alpine ecosystem. In this paper, we evaluate the performance of 14 evapotranspiration (ET0) models by comparison with large weigh lysimeter measurements. Specifically, we use Bowen ratio-energy balance method, three combination models, seven radiation-based models and three temperature-based models driven with data from June 2017 to December 2018 in a humid alpine meadow, northeastern Qinghai-Tibetan Plateau. The daily actual evapotranspiration was obtained by large weighing lysimeters located in an alpine Kobresia meadow. We found that the performances of the 14 ET0 models, ranked on the basis of their RMSE (root mean square error), decreased in the order: Bowen> Priestley-Taylor> DeBruin-Keijman> 1963 Penman> FAO-24 Penman> FAO-56 Penman-Monteith > IRMAK1> Makkink (1957) > Makkink(1967)>Makkink> IRMAK2 > Hargreaves>Hargreaves1>Hargreaves2. For the combination models, FAO24 Penman yielded the highest correlation, followed by Pen-63 and FAO-56 PM. For radiation-based models, PT and DK obtained the highest correlation, followed by Makkink, Makkink(1967) and Makkink (1957), IRMAK1 and IRMAK2. For temperature-based models, HAR, HAR1 and HAR2 obtained the same correlation. Overall, the Bowen performed best, with RMSEs 0.98, followed by radiation-based models, combination models and temperature-based models. Furthermore, all models tended to underestimate measured ETa during periods of larger evaporative demand (i.e. growing season) and overestimate measured ETa during lower evaporative demand (i.e. non-growing season). Our results could provide a new sight for the accurate assessment of evapotranspiration in an alpine ecosystem.

In recent decades, alpine grassland has been serimously degraded across the Qinghai Tibetan Plateau (QTP), although grazing exclusion has been widely adopted to restore degraded QTP grassland. It remains unknown whether this management approach is effective for all degraded alpine grasslands. In this study, plots with three grazing management treatments (free grazing, FG; reduced grazing, RG; grazing exclusion, GE) and four degradation stages (non-degradation, ND; light degradation, LD; moderate degradation, MD; heavy degradation, HD) were compared. Our results showed that the total aboveground biomass (AGB) and species richness (SR) were reduced while total belowground biomass (BGB) increased with increasing degradation, and the responses of SR, AGB and BGB to grazing management varied with the degree of degradation. The total AGB in the LD, MD and HD stages reduced significantly after 6 years under RG and GE, but there was no significant change of AGB in the ND stage. Meanwhile, SR reduced significantly after 6 years under RG and GE across all degradation stages except for HD. Furthermore, the responses of plant functional groups to grazing management varied. After 6 years under RG and GE, the Gramineae AGB increased significantly across all degradation levels; that of the sedges decreased (except in the MD stage); and that of the forbs increased significantly in LD and HD but decreased significantly in ND. Our result suggested that the light degradation grassland can be restored by reducing grazing, and moderate degradation and heavy degradation grassland can restored by grazing exclusion.