Serious conflicts over water resources and the determination of new dams locations often happen in undeveloped transboundary basins. The resolution of these conflicts through equitable allocation depends on the uncertain distribution of the surface water. This study is an extension of the multi-objective model presented Roozbahani et al. Ann Oper Res 287: 323–349, (2020), adding a stochastic modeling approach to settle the water conflicts. We consider the uncertainty of the available streamflow through fitting a multivariate distribution on the historical runoff data in the each node of a basin network. Then, several streamflow scenarios are generated according to the fitted distributions, utilizing the NORmal-To-Anything (NORTA) algorithm. The water shares of the stakeholders, besides the location and capacity of required dams, are obtained using a three-step algorithm for each generated runoff scenario. The water allocation policy is to maximize the minimum ratio of a realized profit over the highest possible profit of the stakeholders from the water utilization, while satisfying the environmental water requirements in the entire basins. The results are analyzed using the frequency approach to determine the optimal location and capacity of new dams. Furthermore, the proposed approach is illustrated through a case study of water allocation modeling in the Sefidrud Basin, Iran. The outputs of the approach implementation show that a dam construction in a node (Node 10) has a substantial role in the sustainable water development of the basin as it is proposed by 52 runoff scenario, in which the 35% of them propose 1230 MCM capacity for the dam.
- Mixed-integer linear model
- Dams location-allocation
- Water allocation under hydrological uncertainty
- The sefidrud basin