Sustainable management of water resources is crucial for humanity. However, traditional methods for achieving this are becoming obsolete. This is because they are underpinned by the assumption that we have a good understanding of how water availability and demand will change in the future. However, based on our current experience with climate change, this is not the case. In fact, rather than having a good understanding of what the future might look like, it is, in fact, deeply uncertain. Consequently, a new paradigm for water resources management is needed; one that accounts for deep uncertainty by embracing scenario thinking. We categorize and summarize different causes of deep uncertainty in water resources management and provide examples of how an emerging paradigm rooted in scenario thinking can deal with these. We hope to stimulate discussion to enable this new paradigm to be developed further and embedded in standard practice.

Machine learning (ML) applications in Earth and environmental sciences (EES) have gained incredible momentum in recent years. However, these ML applications have largely evolved in ‘isolation’ from the mechanistic, process-based modelling (PBM) paradigms, which have historically been the cornerstone of scientific discovery and policy support. In this perspective, we assert that the cultural barriers between the ML and PBM communities limit the potential of ML, and even its ‘hybridization’ with PBM, for EES applications. Fundamental, but often ignored, differences between ML and PBM are discussed as well as their strengths and weaknesses in light of three overarching modelling objectives in EES, (1) nowcasting and prediction, (2) scenario analysis, and (3) diagnostic learning. The paper ponders over a ‘coevolutionary’ approach to model building, shifting away from a borrowing to a co-creation culture, to develop a generation of models that leverage the unique strengths of ML such as scalability to big data and high-dimensional mapping, while remaining faithful to process-based knowledge base and principles of model explainability and interpretability, and therefore, falsifiability.