Evaluating robustness of waste load allocation under climate change using multi-objective decision making

Abstract

In recent years, uncertainties in climate change have made the allocation of waste load a severe challenge. This research develops a new model framework integrating a soil and water assessment tool (SWAT) for hydrological simulation of the river basin, Borg many-objective evolutionary algorithm (Borg MOEA) for optimization, and a software for many-objective robust decision making (OpenMORDM) to minimize these uncertainties and identifying the most robust waste load allocation plans resisting the impact of system uncertainties. This framework has been applied to the Golgol River basin, Iran, which is the main source of the Ilam dam water supply. In this study, first, temperature and precipitation were predicted in 2080–2099. Then, The SWAT model was used to simulate runoff and biochemical oxygen demand (BOD). Afterward, by specifying the objective functions, the Borg MOEA produced the most optimal waste load allocation scenarios in the driest month of the future (August), So that the amount of BOD in the dam reservoir remains less than 5 mg/L. In the next step, the robustness of the solutions analyzed using OpenMORDM. Then, the results were analyzed using the patient rule induction method (PRIM) and classification and regression tree (CART) methods to identify the system vulnerability domains and minimizing false positives. Through the analysis, four climate scenarios have been used to explain how the methodological selection of them affects the resulting candidate planning solutions. The results indicated that individual design selections from the water managers or researchers could influence the efficacy of the model framework, which in turn results in failure of the most robust waste-load allocation strategies under future climate change.