Comparison of several surface resistance models for estimating crop evapotranspiration over the entire growing season in arid regions

Abstract

How to improve the reliability and accuracy of the single-layer Penman–Monteith (PM) model for estimating crop evapotranspiration (ET) over the entire growing season in arid regions, is still of great challenge for hydrologists. In the study, we developed a coupled surface resistance model (CO) after taking the combined restriction effect of vegetation and soil layers on ET into account. The CO model was compared with the modified Shuttleworth–Wallace model (MSW), and the traditional Jarvis, Katerji and Perrier, Irmak and Mutiibwa, Stannard, Leuning, Shuttleworth and Gurney, Massman, Garcia-Santos, Ortega-Farias, and Todorovic canopy resistance models over the partial and dense canopy stages. Maize and vineyard ET measured by the eddy covariance method during 2007–2013 were used to examine the model performance. Results indicate that the PM equation combined with the coupled surface resistance model yielded the lowest root mean square error against the other methods during all the years under either partial or dense canopy stages. Especially, the PM–CO method also performed superiorly against the dual-layer MSW model during the partial maize canopy period. After considering the meteorological, physiological and soil controls on surface resistance simultaneously, the coupled surface resistance model improved the accuracy significantly against the traditional canopy resistance models, and enhanced the reliability of the PM model for estimating partial canopy ET. Thus the coupled surface resistance equation integrated with PM model is recommended to estimate crop ET for the entire growth stages in arid regions.