DSSAT-CERES-maize modelling to improve irrigation and nitrogen management practices under Mediterranean conditions

Abstract Maintain high agriculture production levels while reducing environmental impacts is of primordial importance in intensive irrigated areas. Adequate management of water and nitrogen are critical for a sustainable agriculture. The objectives of this work were to (i) calibrate and validate DSSAT model for maize under different nitrogen availability conditions and (ii) assess the effect of best management practices on irrigation water needs and nitrogen losses by leaching using different scenarios in an intensive irrigated area. For model calibration and validation, three field experiments were conducted with a total of 134 plots. Then, the model application was performed in three soil types in the ‘Del Reguero’ watershed (Spain) considering (i) real irrigation applied by farmers and DSSAT automatic irrigation and (ii) a recommended dose of N fertilizer (250 kg N ha −1 ) compared to a traditional dose of 390 kg N ha −1 . Among all plots, the model simulated reasonably well grain yield with a Root Mean Square Error (RMSE) lower than 708 kg ha −1 and high Willmott agreement index (d statistic) (>0.9). Very similar trends were observed for total biomass and total N uptake with a RMSE of 2018 kg ha −1 and 36.6 kg N ha −1 . The prediction of residual nitrate in soil was acceptable with a RMSE of 43 kg N ha −1 .Modeling results showed that adjusted irrigation would reduce (on average for different soil types) the total amount of seasonal irrigation water by 31% and the nitrate leaching by 97% without a significant reduction in grain yield. Regarding to N fertilizer scenarios, results showed that farmers can reduce the N fertilizer currently applied leading to a significant decrease in the N leached between 33 and 53% depending on soil types.