ABSTRACT This study assessed the DSSAT/CERES‐Maize model's ability to simulate daily and cumulative ET, water productivity (WP) and water use efficiency (WUE) for maize in tropical conditions. Observed ET data were obtained using the Bowen ratio (BREB) method in Piracicaba‐SP, and the soil water balance (SWB) method in Serra Talhada‐PE. Simulations used the FAO‐56 Penman‐Monteith (ET PM ) or Priestley–Taylor (ET PT ) equations for ET, combined with the Ritchie (R‐2) or Suleiman–Ritchie (S–R) methods for soil evaporation. The model was able to simulate daily and cumulative ET (Piracicaba irrigated: RMSE = 1.13–1.75 mm d −1 , d = 0.69–0.89 and bias = 4.92%–25.81%; Piracicaba rainfed: RMSE = 1.70–2.02 mm d −1 , d = 0.46–0.73 and bias = 21.10%–35.37%; Serra Talhada irrigated: RMSE = 0.95–1.42 mm d −1 and d = 0.45–0.75 and bias = −20.74%–3.19%), daily ET by phenological phase (Piracicaba irrigated: RMSE = 1.60–3.18 mm d −1 and d = 0.18–0.78; Piracicaba rainfed: RMSE = 1.59–2.27 mm d −1 and d = 0.14–0.73; Serra Talhada irrigated: RMSE = 0.62–1.77 mm d −1 and d = 0.54–0.87), WP and WUE (Piracicaba irrigated = 1.08–1.85 mm kg −3 and 3.15–4.64 mm kg −3 ; Piracicaba rainfed = 0.79–1.85 mm kg −3 and 2.63–4.64 mm kg −3 ; Serra Talhada irrigated = 0.80–1.12 mm kg −3 and 2.37–3.09 mm kg −3 ). The ET PT method with the R‐2 approach showed better agreement with observed data in Piracicaba‐SP, while the ET PM method combined with S–R performed better for the data from Serra Talhada‐PE.
Dias et al. (Thu,) studied this question.