Research on field water consumption is critical for optimizing crop growth and policy-making, in which the computer models play an increasingly important role. As a water-driven crop model, the AquaCrop model has been used in a large number of studies since its launch in 2009. However, how the model performs in predicting the ecohydrological process of farmland under film-mulched drip irrigation is still unclear, especially its application on partitioning crop evapotranspiration is very rarely reported. To make up for the above insufficiency, maize experiments were conducted under full mulch drip irrigation with observation instruments of eddy covariance systems, heat balance stem-flow gauges, micro-lysimeters and other tools, during seasons of 2014–2018. The AquaCrop model was first calibrated using measured data in 2014, and subsequently validated with data in 2015–2018. Results indicate that the parameterized model could precisely simulate the canopy cover ( $${R}^{2}$$ = 0.97), biomass ( $${R}^{2}$$ = 0.99) and grain yield (standard deviation was 4.13%), as well as reflect the patterns of daily variation in transpiration and evapotranspiration with satisfactory $${R}^{2}$$ of 0.91 and 0.87, respectively. Nevertheless, the $${R}^{2}$$ values of soil water content and evaporation were not good, ranging between 0.23 and 0.45, and 0.26 and 0.75, respectively. The AquaCrop model adopts canopy cover instead of leaf area index to describe the growing process of crops; this is an important innovation for model extension and application but also may lead to some inaccuracies in water balance simulation. Summarizing, this study shows that the AquaCrop model is appropriate for supporting crop production but not for predicting the soil moisture content and evaporation variation for maize under film-mulching drip irrigation.
Read full abstract