Landscapes are complex adaptive spatial systems driven by biophysical and socioeconomic processes that are shaped by the cumulative behavior of interacting, but independent decision-makers. Agent-based modeling has been identified as an effective paradigm for the study of such systems because of its ability to represent behavior and interaction at the individual level. These interactions are inherently associated with cause-effect relations in the transition of land use systems through time. Elucidating specific cause and effect relations from model results, however, poses a significant challenge to the research community because of complicated model structures and the complex processes that drive system dynamics. In this article, we call for provenance-based agent-based models of land use dynamics to support the capture and tracking of complex cause-effect relations. We illustrate the importance of explicitly considering provenance in agent-based modeling through the development of a spatially explicit agent-based land use simulation framework. This framework is designed to simulate complex land use dynamics in southwest Montana, USA. Land managers are modeled as geospatial agents who interact with their spatially explicit environments. These agents exchange opinions about land use controls through social networks in an attempt to form consensus on land use policies. The heterogeneous opinion behavior of landowner agents and complicated social structural relationships produce complex dynamics. We conduct experiments to demonstrate the need to explicitly consider provenance in agent-based models of land use systems and its ability to promote our understanding of complex adaptive land use systems.
- Agent-based modeling
- Land use modeling
ASJC Scopus subject areas
- Geography, Planning and Development
- Earth-Surface Processes
- Management, Monitoring, Policy and Law