DYNAMIC-PROGRAMMING FOR IMPROVING IRRIGATION SCHEDULING STRATEGIES OF MAIZE

Efficient and profitable irrigation scheduling strategies are needed, particularly where water available for irrigation is limited. The purpose of the study was to identify optimal irrigation thresholds for several maize irrigation strategies in the southeastern USA. Using dynamic programming (DP), decisions among six possible thresholds were made for each of five maize growth stages. Net returns were maximized subject to a constraint on the total amount of irrigation. Net returns and irrigation amounts were determined for each growth stage using the growth simulator CERES-Maize. Net returns were computed by stage using the futures price for grain and constant ratio of grain yield to biomass. Net returns and water consumption were also computed for fixed threshold irrigation strategies. Dynamic irrigation strategies with differing thresholds for each growth stage produced higher average net returns and required less irrigation water than the best fixed threshold irrigation strategies. Further, irrigation amounts with the dynamic strategies were held below the maximum irrigation level. Irrigation amounts for the fixed strategies often exceeded the allowed maximum.