LAUSR

Abstract Accurately modeling evapotranspiration (ET) and its components of maize grown for seed production is essential for precision irrigation management. In this study, a dual crop coefficient method and a multisource model based on radiation interception by adjacent crop varieties were used to estimate ET and its components in the arid region of northwest China. The dual crop coefficient method and multisource model were validated using observed ET (ETEC), transpiration (T) of female (Tsf) and male (Tsm) parents and evaporation (Es). Observations were made using the eddy covariance system, sap flow measurements, and micro-lysimeter in 2013 and 2014. Results showed that ET estimated by the dual crop coefficient method was close to ETEC at the midseason stage, and was higher than ETEC both at the initial and the development stage due to the constant value of initial basic crop coefficient and linear interpolation at the development stage. The estimated T of female and male parents was greater than Tsf and Tsm in both years. Soil evaporation estimated by the dual crop coefficient method was greater than measured soil evaporation in the late growth stage. While the ET, T of female and male parents and E predicted by the multisource model were closer to the measurements. Estimated ET was 2% less (2013) and 4% greater (2014) than ETEC, T of male parents was 8% and 3% less than Tsm, T of female plants was 8% and 6% less than Tsm, and E was 6% and 3% less than Es. Thus the multisource model based on radiation interception by neighboring species is suitable for estimating ET and its components of maize grown for seed production in the arid region of northwest China.