Search for a command to run...
The first crop chosen to parameterize and test the new FAO AquaCrop model is maize ( Zea mays L.). Working mainly with data sets from 6 yr of maize field experiments at Davis, CA, plus another 4 yr of Davis maize canopy data, a set of conservative (nearly constant) parameters of AquaCrop, presumably applicable to widely different conditions and not specific to a given crop cultivar, was evaluated by test simulations, and used to simulate the 6 yr of Davis data. The treatment variable was irrigation—withholding water after planting continuously, only up to tasseling, from tasseling onward, or intermittently, and with full irrigation (FI) as the control. From year to year, plant density (7–11.9 plants m −2 ), planting date (14 May−15 June), cultivar (a total of four), and atmospheric evaporative demand varied. The conservative parameters included: canopy growth and canopy decline coefficient (CDC); crop coefficient for transpiration (Tr) at full canopy; normalized water productivity for biomass (WP∗); soil water depletion thresholds for the inhibition leaf growth and of stomatal conductance, and for the acceleration of canopy senescence; reference harvest index (HI o ); and coefficients for adjusting harvest index (HI) in relation to inhibition of leaf growth and of stomatal conductance. With all 19 parameters held constant, AquaCrop simulated the final aboveground biomass within 10% of the measured value for at least 8 of the 13 treatments (6 yr of experiments) and also the grain yield for at least five of the cases. In at least four of the cases, the simulated results were within 5% of the measured for biomass as well as for grain yield. The largest deviation between the simulated and measured values was 22% for biomass, and 24% for grain yield. Importantly, the simulated pattern of canopy progression and biomass accumulation over time were close to those measured, with Willmott's index of agreement ( d ) for 11 of the 13 cases being ≥0.98 for canopy cover (CC), and ≥0.97 for biomass. Accelerated senescence of canopy due to water stress, however, proved to be difficult to simulate accurately; of the six cases, the index of agreement for the worst one was 0.957 for canopy and 0.915 for biomass. Possible reasons for the discrepancies between the simulated and measured results include simplifications in the model and inaccuracies in measurements. The usefulness of AquaCrop with well‐calibrated conservative parameters in assessing water use efficiency (WUE) of a crops under different conditions and in devising strategies to improve WUE is discussed.