Partitioning evapotranspiration into soil evaporation and transpiration using a modified dual crop coefficient model in irrigated maize field with ground-mulching

The accurate partitioning of crop evapotranspiration (ETc) into two components, soil evaporation (E-s) and transpiration (T-r), is needed to better understand terrestrial hydrological cycles and develop precise irrigation scheduling. However, there is no easy way to distinguish between the two. Based on FAO-56 dual crop coefficient (K-c) approach, we developed a modified dual K-c model for better predicting T-r through basal crop coefficient (K-cb) and E-s through evaporation coefficient (K-e). Daily K-cb was dynamically calculated by introducing a canopy cover coefficient that could be simply described as a function of leaf area index or fraction of canopy cover. Also, leaf senescence factor was taken into consideration to modify K-cb when leaf suffers functional senescence. K-e was modified through introducing the fraction of ground-mulching (f(m)) to account for the effect of mulching on E-s. The model was parameterized by measurements in 2009, and validated using independent data for grain and seed maize with and without mulching in 2010 and 2011. The results indicate that the predicted K-c values by the modified model were obviously better than those by the original model. The good agreements were found between the predicted ETc, T-r and E-s using the modified model and the measurements for grain maize in 2010 with f(m) = 0.6, with the slope of linear regression of 0.99 (R-2 = 0.90), 1.01 (R-2 = 0.92) and 0.96 (R-2 = 0.78), respectively. The modified model also well reproduced the values of ETc and E-s for seed maize in 2011, which had lower plant height and leaf area index compared to grain maize, under mulching (f(m) = 0.7) and non-mulching (f(m) = 0) conditions. The slopes of linear regression between predictions and measurements were 0.98 (R-2 = 0.91) and 0.99 (R-2 = 0.92) for ETc, and 0.98 (R-2 = 0.79) and 0.97 (R-2 = 0.80) for E-s under f(m) = 0.7 and 0.0, respectively. These results suggest that the modified dual K-c model can accurately predict ETc, E-s and T-r for different crop types under different mulching, thus could be a useful tool for improving irrigation water management. (c) 2013 Elsevier B.V. All rights reserved.