Socio-ecological systems include diverse resources of complexity involving complicated interactions, feedback, dynamic behavior patterns, heterogeneity in agents’ characteristics and behaviors, as well as system uncertainties. Agent-based modeling is introduced as a tool to simulate different levels of stakeholders and interactions, including individual, group, and institutional agents in complex socio-ecological systems. In the present study, an agent-based model was developed to explore sustainable solutions in uncertain conditions for groundwater restoration of a critical aquifer in the Lake Urmia Basin, Iran. Different projects for aquifer restoration such as wells monitoring, license adjustment, and modern irrigation development were implemented in the developed framework. The results showed that it is possible to alleviate the negative balance of the aquifer in case of necessary coordination in the projects implementation. So that during the simulation period, while increasing the total annual net income of the cultivation by €1.3 million (about 7%), the groundwater level drawdown can be compensated by about 5 m (about 50% of the decline over the past 30 years). Results of sensitivity analysis on the risk-taking behavior of farmer agents showed that it has a more significant impact on the total income than groundwater level. In the risk-seeking society conditions, a 10% increase of total annual net income is expected in the early years of simulation. The framework is used to learn about the influences of agents’ objectives, characteristics, and behaviors on the system variability. Systematic analysis in such a framework provides a better understanding of system structures to support decision-making under uncertainty in a participatory process.

社会生态系统包括各种复杂的资源，包括复杂的交互，反馈，动态行为模式，主体特征和行为的异质性以及系统的不确定性。引入基于主体的建模作为一种工具来模拟不同级别的利益相关者和互动，包括复杂的社会生态系统中的个人，团体和机构主体。在本研究中，开发了一种基于代理的模型，以探索在不确定条件下用于伊朗乌尔米亚湖盆地关键含水层地下水恢复的可持续解决方案。在已开发的框架中实施了不同的含水层修复项目，例如井监测，许可证调整和现代灌溉开发。结果表明，在项目实施中进行必要协调的情况下，可以减轻含水层的负平衡。因此，在模拟期间，在使耕种总年净收入增加130万欧元（约7％）的同时，地下水位下降可补偿约5 m（过去30年下降的约50％） ）。对农民工冒险行为的敏感性分析结果表明，它对总收入的影响比地下水水位的影响更大。在寻求风险的社会环境中，模拟的初期预计年总净收入将增长10％。该框架用于了解代理的目标，特征和行为对系统可变性的影响。