Adaptive planning in climate change condition is a significant challenge for effective management of water resources and agricultural systems. One of the major controversial issues in climate change adaptation studies is deeply uncertain nature of such changes. This study aimed at planning optimal agricultural adaptation measures under deep climate uncertainty in a semi-arid basin in Iran (Tashk–Bakhtegan Basin) using a simulation–optimization (S–O) approach. To fulfil this objective, a modified version of Soil and Water Assessment Tool (SWAT) entitled SWAT-PARS was used to simulate the status of the basin regarding water resources and agricultural sector. By combining the SWAT model with a NSGA-II based robust optimization model, the optimal and robust type and location of robust agricultural management measures under climate uncertainty were determined. For this purpose, deficit irrigation, irrigation efficiency increase, cropping pattern change and cropping date change were considered as effective adaptation measures in the agricultural sector. The robustness of optimal measures was analyzed by a multi objective robust decision making approach using satisficing and regret robustness assessment methods. The robustness analysis of management strategies under climate change indicated different prioritization of strategies by “satisficing” and “regret” robustness assessment indicators. According to the results, irrigation management measures, and changes in the cropping pattern and date have been recommended in most strategies provided by different robustness methods. Most measures recommended have been able to reduce system vulnerability concerning in groundwater’s quantitative changes and calorie of agricultural products, but did not cause any significant change in the system concerning inflow to the lake.

气候变化条件下的适应性规划是有效管理水资源和农业系统的重大挑战。气候变化适应性研究中主要的争议性问题之一是这种变化的性质非常不确定。本研究旨在通过模拟-优化（S-O）方法，在伊朗半干旱盆地（塔什克-巴赫特甘盆地）的深层气候不确定性下，计划最佳的农业适应措施。为了实现这一目标，使用了名为SWAT-PARS的土壤和水评估工具（SWAT）的修改版本来模拟流域有关水资源和农业领域的状况。通过将SWAT模型与基于NSGA-II的鲁棒优化模型相结合，确定了在气候不确定性条件下农业管理措施的最佳和鲁棒类型和位置。为此，将缺水灌溉，提高灌溉效率，改变种植方式和改变种植日期视为农业部门的有效适应措施。通过多目标鲁棒决策方法，使用满意度和后悔鲁棒性评估方法，分析了最优措施的鲁棒性。气候变化下管理策略的稳健性分析表明，通过“满意”和“遗憾”的稳健性评估指标对战略进行了不同的优先排序。根据结果​​，在不同的稳健性方法提供的大多数策略中，建议了灌溉管理措施以及种植方式和日期的变化。