Simulation of water use process by film mulched cultivated maize based on improved AquaCrop model and its verification

Abstract: Model simulated crop growth and productivity has been a widely accepted and powerful tool for assessing agricultural production in response to weather, soil, water and nutrients management. A water-driven AquaCrop model recommended by FAO can evaluate the various crops growing across climate, soil, water deficit and irrigation management conditions apart from surface soil mulching process. In this study, AquaCrop model for simulating maize (Zea mays L.) canopy growth, soil water utilization and grain yield formation of mulching maize with plastic film was developed in the northwest semi-arid region of Liaoning province, China (121.70°E, 42.11°N). Based upon invariance of growing degree days (GDD) principle, we modified mean daily air temperature calculation method of AquaCrop model for maize with plastic film mulch according to compensatory effect of cumulative soil temperature to cumulative air temperature, and calibrated this developed model using measured experimental data of growing days, water consumption and grain yield in 2011; the two-year experimental data from 2012 and 2013 were used to validate the developed model for simulating canopy cover (CC), soil water content (SWC); three-year filed experimental data from 2011 to 2013 were used to validate the developed the model for grain yield and water use efficiency under maize rainfed conditions. The modified calculation of air temperature showed that, depending on the linear regression relationship between mean daily air temperature and soil temperature (at 5cm depth) under plastic film mulching and non-mulching (R2＞0.8), the raised soil temperature in the mulched maize field was remarkable before tasseling stage. The compensatory coefficient (Cmaize) of mulching plastic film maize and air increment of cumulative soil temperature to cumulative air temperature can be generated continuously by transparent algorithms, which Cmaize was 1.356 from sowing to emergence, 0.635 from emergence to tasseling stage, and 0 after flowing. Furthermore, mean daily air temperature with the addition of air increment formed a new file (*.tmp) and input to AquaCrop's climate module which can modify by function could be applied to mulching plastic film maize growing. Model coefficient of efficiency (CE), coefficient of determination (R2), the relative root mean square error (RRMSE), prediction error (Pe) and coefficient of residual mass (CRM) were used to test the model performance. The developed AquaCrop model was calibrated for simulating maize growing days, water consumption and grain yield for mulched and no-mulched maize with the prediction error statistics -4%0.88, 0.090.87. Upon validation, the Pe in simulation of water consumption and grain yield under mulched and no-mulched maize was among ±6%. In addition, R2, RRMSE, CE of grain and water use efficiency during 2011 to 2013 were 0.96 and 0.93, 0.042 and 0.06, 0.91 and 0.89, respectively. The developed AquaCrop model predicted maize grain yield with higher accuracy and performed better yield than water use efficiency for mulching plastic film maize, which indicated this improved model were better mechanism and application for simulated maize mulching. The present research implicated that the developed AquaCrop model can be applied to the same to semi-arid region and simulated maize potential, or prescribing yield in response to water and temperature limiting factors under climate change.