Integrating Ecology and Economics in Agent-Based Models to Support Agricultural Policy

Agriculture is a complex social-ecological system with numerous interactions between policies, markets, farm management, landscapes, and ecosystems. Agent-based models are a valuable method for studying these agricultural systems, as they are able to simulate the spatiotemporal dynamics and heterogeneous actors inherent to them. This potentially also makes such models helpful tools for scientific policy advice. Here, we conduct a literature review to evaluate how agent-based models represent social-ecological systems in agriculture. We analyze 56 published models using a set of 26 criteria to determine whether socio-economic and ecological aspects of agriculture have been studied. We find that there is a particularly robust tradition of using agent-based models for economic impact analyses of policy changes, and much effort has gone into studying decision-making by land users. However, ecological models of biodiversity and ecosystem services are still often very abstract and ignore important land-use processes, impeding their usefulness for providing policy advice. Research areas that are currently underdeveloped include simulating farm management and landscape dynamics in ecological models, risk management in economic models, and bidirectional human-nature interactions in social-ecological models. Agent-based modeling of agriculture would also profit from using model coupling to represent more of the breadth of social-ecological systems.