Estimating rainfed spring maize evapotranspiration using modified dual crop coefficient approach based on leaf area index

Abstract: Evapotranspiration (ET) is vital to energy and water balance in agriculture, and accurate estimation and partition of ET is needed for crop water productivity improvement. In this study, a modified approach of FAO-56 dual coefficients of crop was developed for estimating and partitioning maize ET. Daily basal crop coefficient was dynamically calculated by introducing a canopy cover coef?cient which could be simply described as a function of leaf area index (LAI). Daily evaporation coefficient was also calculated by modified canopy cover that was computed by LAI. A ?eld experiment was conducted from May to September during 2012 and 2013 in a rainfed farmland on the Loess Plateau of North China, and daily spring maize ET and soil evaporation were measured by eddy covariance system and micro-lysimeters to validate the modified approach of dual coefficients of crop. The results indicated that the LAI and the maximum plant height in 2012 were slightly larger than those in 2013, with the LAI ranging from 0 to 4.52 m2/m2 in 2012 and from 0 to 3.97 m2/m2 in 2013, the maximum plant height of 2.98 m in 2012 and 2.97 m in 2013, respectively. Due to large frequency of wetting by rainfall and small canopy cover in initial stage, daily evaporation coefficient fluctuated significantly; then daily evaporation coefficient decreased with the increase of canopy cover in development and mid stages, and increased in late stage. In contrast to daily evaporation coefficient, the change trend of daily basal crop coefficient was relatively small in initial stage; then daily basal crop coefficient increased with the increase of LAI in development and mid stages, and decreased in late stage. Simulated maize ET values in initial, development, mid, late and whole growing season were 47.0, 81.1, 172.7, 64.6 and 365.3 mm in 2012, and 25.1, 78.5, 211.5, 70.5 and 385.6 mm in 2013, respectively, which were in good agreement with the measured ET, with the coefficient of determination (R2) of 0.824, root mean square error (RMSE) of 0.561 mm/d, coefficient of model efficiency (Ens) of 0.817, and average absolute error (AAE) of 0.449 mm/d in 2012, R2 of 0.870, RMSE of 0.381 mm/d, Ens of 0.871, and AAE of 0.332 mm/d in 2013, respectively. The good agreements were found between the simulated soil evaporation using the modified approach of dual coefficients of crop and the measured one by micro-lysimeters; the simulated soil evaporation was 1.71, 1.25, 0.61, 0.78 and 0.98 mm/d in initial, development, mid, late and whole growing season, which accounted for 79.86%, 59.95%, 20.20%, 28.88% and 38.12% of ET in 2012, and 0.75, 0.96, 0.87, 0.77 and 0.86 mm/d in initial, development, mid, late and whole growing season, which accounted for 66.96%, 51.34%, 54.04%, 29.62% and 33.59% of ET in 2013, respectively. The results suggested that the modified approach of dual coefficients of crop could estimate and partition rainfed maize ET accurately on the Loess Plateau of North China, and thus could be a useful method for agricultural water management. The current study doesn't take the effects of maize planting density and direction of maize sowing-line on ET into consideration, and lacks the measured transpiration data, which may bring some uncertainties, and therefore, a further study considering these factors is needed to improve the performances of the approach of dual coefficients of crop.