Search

The existing researches for leaf area index (LAI)-based Penman–Monteith evapotranspiration (ET) model (PML) are mostly carried out at yearly scale and the analysis of effects of seasonal change and different underlying surface conditions on model parameters is scarce. This study emphasizes on the influences of seasonal change and diverse land surface conditions on ET by optimizing the sensitive parameters, namely soil evaporation coefficient f and maximum stomatal conductance gsx, with particle swarm algorithm. This analysis is based on the observations of eight flux stations in China. The model performance is reasonable with a best Pearson r of 0.87. The seasonal calibration results indicate parameters change evidently in different seasons and have obvious spatial heterogeneity. The seasonal calibration method has an obvious effect on improving the ET accuracy in spring, which is mostly influenced by regional temperature and relative humidity. This study further demonstrates the need to dynamically adjust model parameters over time with PML model for evapotranspiration simulations, rather than simply setting these parameters to constants depended on subsurface conditions such as land use type. [ABSTRACT FROM AUTHOR]