Using a role-playing game to inform the development of land-use models for the study of a complex socio-ecological system

We present an integrated methodology composed of a role-playing game on land adjudication from which we extract narrative and spatially explicit drivers of land-use decisions. We show how geographic information systems (GIS), qualitative decision-matrix analyses, a simple rule-based model using multi-criteria evaluations (MCE), and a machine learning-based land-transformation model (LTM) can be used harmoniously to study complex socio-ecological systems. We evaluate how each technique performs in the study of complex socio-ecological systems using a multi-tier framework detailing how each method analyzes the resource system, resource units, governance system, users and interactions and outcomes in the system. We show that each approach enhances our understanding of the land-use decision making process. Each method provides various information on the drivers of land-use decision, some focusing more on spatial components of socio-ecological systems (resource system and resource unit) and other having a strong emphasis on social mechanisms (governance system, users, interactions and outcomes). Furthermore, we shed light into the existence of a flow of information between the various methods enhancing our understanding of land-use drivers. We end with a discussion on methodological tradeoffs between models and the value of our more holistic approach to modeling land-use drivers and decisions.Land-use Role-playing games Geographic information systems Multi-criteria evaluation Machine learning tools Comparison of models