A method to define a typology for agent-based analysis in regional land-use research

Land use/cover change (LUCC) is often the cumulative result of individual farmer's decisions. To understand and simulate LUCC as the result of local decisions, multi-agent systems models (MAS) have become a popular technique. However, the definition of agents is not often based on real data, ignoring the inherent diversity of farmers and farm characteristics in rural landscapes. The aim of this paper is to describe an empirical method that defines an agent typology and allocates agents into the different agent types for an entire region. This method is illustrated with a case study in the Netherlands, where processes of farm expansion and diversification of farm practices take place. Five different agent types were defined and parameterized in terms of views, farm characteristics and location. Despite its simplicity, this empirical method captures several relations between farmers’ views, farm characteristics and land-use decisions and strategies. This approach is a step forward in multi-agent systems of land use/cover change (MAS/LUCC) to include the diversity of land-use decisions and strategies in regional studies by empirically defining, parameterizing and allocating different agent types.