Progress of Agent-based modeling for water resources management:a review

Agent-based modeling (ABM) has enriched the theory and methods of research in water resources management. Better understanding of the state-of-the-art of ABM and its potential in the field of water resources management can promote institutional development and reform in China's water resources management system. In this paper,while ABM of water resources management (WR-ABM) is defined,its key components—agent decision rules and agent interactions—are identified and their modeling approaches are summarized. Various WR-ABMs applied in basin-scale optimal water allocation,urban household water use and agricultural water management published during the period of 2009—2018 are carefully reviewed. Future research on the use of WR-ABMs that should address the challenges and weakness in the water resources management are discussed and several research directions are recommended herein:① further expanding the use of complex adaptive system theory in the field of water resources management;② coupling ABM and water resources system models that include uncertainties;③ exploring the use of machine learning algorithms in the decision-making modeling;④ improving the methods used in the model parameter calibration,result verification and validation;⑤ using standard documentation protocol,such as the ODD protocol,for the description of models;and ⑥ achieving comprehensive and optimal balance between completeness and simplification in model design.