Current global-scale research on environmental change has a strong focus on the physical processes that underlie changes in the earth system. Although global environmental change is driven by individual and collective human decisions, most earth system and integrated assessment models lack a proper representation of variation in human decision making. The importance of variation in local context has caused much social-science research on the underlying driving factors and decision making structures to focus on local case studies. Consequently, important insights from social-science have been ignored in global-scale assessment models.The proposed research, which focuses on land system change as one of the dominant processes of global environmental change, contributes to a new generation of integrated global assessment models. These models will explicitly account for the (spatial) variation in decision making to support the design of earth system governance.An improved understanding of the factors that drive decision making will be obtained through a novel approach for the meta-analysis of existing case studies worldwide. Supplementary empirical evidence will be collected by analyzing a number of transects and disentangling the global and local factors that influence land change decisions. Generalized, global-scale multi-agent modelling systems that represent variation in decision making will be developed. While the current approaches are driven mainly by changes in consumption patterns and demography, we propose an alternative approach that accounts for the full range of ecosystem service demands and explicitly addresses the spatial relationship between demand and supply of those services that influence decision making. Incorporating the findings into existing integrated assessment models will be supported by a trans-disciplinary approach to identify the requirements of the science-policy process in earth system governance.