Data-based, synthesis driven: Setting the agenda for computational ecology

Computational thinking is the integration of algorithms, software, and data, to solve general questions in a field. Computation ecology has the potential to transform the way ecologists think about the integration of data and models. As the practice is gaining prominence as a way to conduct ecological research, it is important to reflect on what its agenda could be, and how it fits within the broader landscape of ecological research. In this contribution, we suggest areas in which empirical ecologists, modellers, and the emerging community of computational ecologists could engage in a constructive dialogue to build on one another's expertise; specifically, about the need to make predictions from models actionable, about the best standards to represent ecological data, and about the proper ways to credit data collection and data reuse. We discuss how training can be amended to improve computational literacy.