Optimal irrigation water allocation in Hetao Irrigation District considering decision makers' preference under uncertainties

In this study, a fuzzy multi-objective m(lambda)-measure dependent-chance programming model was proposed to optimize limited irrigation water among three crops and five subareas in Hetao Irrigation District (HID) under uncertainties, with the objectives of ensuring food security and maximizing economic benefits. Three levels of droughts (100%, 85%, and 70% of annual average water diversion) that may happen in HID were set as future possible droughts scenarios. The optimization model considered the detailed physical process of soil water balance, which is necessary for irrigation districts where groundwater recharge plays a major role when available water limited. Moreover, sensitivities of crop yield to water deficit during crop growth periods were involved in model formulation to reflect the difference of crop water demand in different subareas. In optimization model, m(lambda)-measure was introduced into the first objective to quantify optimistic and pessimistic decision attitudes of decision makers, and a series of alternatives were obtained. The obtained results indicate that: (1) the irrigation water of wheat, maize, and sunflower at rapid growth stage and middle growth stage should be preferentially guaranteed due to their high water-deficit sensitivities. The water demand at initial growth stage and late growth stage can be satisfied by soil water and groundwater recharge. (2) maize has the priority of irrigation in almost all subareas while sunflower has to face water deficit when drought occurs. (3) allocating more irrigation water to Yongji and Jiefangzha subareas can help increase food production. (4) the irrigation water allocated to grain crops increases along with the decrease of lambda, representing decision makers' more conservative preference. The optimization model and results in this study provide decision makers abundant water-allocation schemes under drought conditions.