Modeling malt barley water use and evapotranspiration partitioning in two contrasting rainfall years. Assessing AquaCrop and SIMDualKc models

Two contrasting rainfall barley seasons, dry (2012) and wet (2013), were used to parameterize and assess the performance of the SIMDualKc and AquaCrop models. Field data were obtained from malt barley cropped in a farmer's field in Ribatejo, Portugal. SIMDualKc applies the dual crop coefficient approach for computing and partitioning crop evapotranspiration (ET), while AquaCrop uses an empirical approach to estimate potential crop transpiration (Tc) and soil evaporation (Es) depending upon the canopy cover (CC). The calibration and validation of both models was performed through comparing observed and predicted soil water content (SWC) for both seasons. The goodness-of-fit indicators were very good for SIMDualKc, with low errors of estimate (RMSE < 0.015 cm3 cm−3). AquaCrop was first parameterized for the CC curves of both years using LAI observations. When tested for SWC, indicators have shown less accuracy than SIMDualKc. The analysis focused the partition of ET into Tc and Es by both models through the analysis of the daily basal crop coefficients (Kcb) and evaporation coefficients (Ke), and the actual crop transpiration (Ta) and Es values cumulated to each crop growth stage and the season. Differences between the dry and wet year were evident, also in terms of model behavior when simulating SWC. Differences were also notable relative to the water balance terms, namely Ta and Es. Problems with estimating Kcb and Kewith AquaCrop were identified, which likely cause its less good performance in simulating SWC and impacted yield estimation. Results of assessing both models led to conclude that computation procedures used in AquaCrop for Tc, Ta and Es lead to inaccuracies that make AquaCrop less appropriate to support irrigation scheduling.