A modified MOD16 algorithm to estimate evapotranspiration over alpine meadow on the Tibetan Plateau, China

The long-term change of evapotranspiration (ET) is crucial for managing water resources in areas with extreme climates, such as the Tibetan Plateau (TP). This study proposed a modified algorithm for estimating ET based on the MOD16 algorithm on a global scale over alpine meadow on the TP in China. Wind speed and vegetation height were integrated to estimate aerodynamic resistance, while the temperature and moisture constraints for stomatal conductance were revised based on the technique proposed by Fisher et al, (2008). Moreover, Fisher's method for soil evaporation was adopted to reduce the uncertainty in soil evaporation estimation. Five representative alpine meadow sites on the TP were selected to investigate the performance of the modified algorithm. Comparisons were made between the ET observed using the Eddy Covariance (EC) and estimated using both the original and modified algorithms. The results revealed that the modified algorithm performed better than the original MOD16 algorithm with the coefficient of determination (R-2) increasing from 0.26 to 0.68, and root mean square error (RMSE) decreasing from 1.56 to 0.78 mm d(-1). The modified algorithm performed slightly better with a higher R-2 (0.70) and lower RMSE (0.61 mm d(-1)) for after-precipitation days than for non precipitation days at Suli site. Contrarily, better results were obtained for non-precipitation days than for after precipitation days at Arou, Tanggula, and Hulugou sites, indicating that the modified algorithm may be more suitable for estimating ET for non-precipitation days with higher accuracy than for after-precipitation days, which had large observation errors. The comparisons between the modified algorithm and two mainstream methods suggested that the modified algorithm could produce high accuracy ET over the alpine meadow sites on the TP.