More than 10% of permanent grassland coverage has been lost in Germany since the 1990s. Grassland areas, especially if managed extensively, play a key role in providing important ecosystem services. Policy measures have addressed this issue but were largely unsuccessful in halting this trend. The design of future measures could be improved with a better understanding of the influencing factors of farmers’ decisions and their interaction. Agent-based models can deliver this kind of information.
Material and Methods
We have analyzed different recent approaches, advances and challenges of agent-based models designed to examine changes in socio-ecological systems. The feasibility to use this model approach for the support of policy decisions was explored for the case of grassland conversion in Bavaria.
Our analysis shows that agent-based modelling can be a useful decision support tool. Past models have however mostly failed to incorporate sound and theory-based human decision-making modules. Only few models included two-way feedbacks between social and ecological systems. The use of GIS data, taking into account spatial and social interaction, transdisciplinary methods and the design of structural realistic models will improve a models value and impact for policy making.
With the use of an extensive GIS data set on past agricultural land use and agricultural statistics of Bavaria at hand, through coupling with other models to predict changes in delivered ecosystem services and with the incorporation of behavioral data we will be able to advance the modelling of human decision making in land use systems and provide valuable decision support for future policies.