How can models support better decisions?

Much of the science conducted at ZMT aims to contribute to better management or policy decisions. However, the path from scientific insight to action is often long and arduous. In this talk, I will explore the role that models and modeling practice can play in this process. Starting from the idea that good decision-making needs to bring together predictions about impacts with people's values — in other words, expectations and evaluations — I will present examples of different objectives models may address. This includes streamlining the identification of shark catch from images with machine learning, modeling a coral reef management case and understanding the role of preference and prediction uncertainty for this decision, conceptual modeling of wastewater and coastal systems as a participatory process, and using agent-based modeling to examine decision dynamics in societies. As we will discuss, these models all have drawbacks, but their capacity for making complex issues explicit is a promising direction for more informed decisions. Thus, I also hope to spark discussion on how modeling could be more extensively employed to tackle the wicked problems we face.