Identifying data challenges to representing human decision-making in large-scale land-use models

Land-use models are by now an accepted method in scientific research, both to increase our understanding of land-use change processes and to project future land-use trajectories. Many of these models simulate changes as a function of spatial data layers, such as elevation, accessibility and soil type. However, land-use changes are ultimately the result of human decisions. Therefore representing human decision-making processes in models is essential to advance our understanding of land-use change processes as well as our capacity to support policy making. Agent-based models allow human decision-making to be represented explicitly. However, their application is constrained by the availability of data about actors and their decision-making processes. Empirical data can be obtained from case studies, but the geographic extent of these case studies is constrained by time and resources. Therefore we argue that we need new sources of data to support model representation of these processes. In this chapter, we further specify this data demand and discuss potential methods of data acquisition. Data acquisition methods include metastudies, aligning with various ongoing large-scale data collection efforts, dedicated projects and crowdsourcing.