Much has been written on how hydrological modelling can be improved: better data, better models, better objective functions, better benchmarking, better uncertainty quantification, better communication, and so forth. Yet, it remains rare to find definitions of what “better” means in each context, and even more to see such recommendations implemented comprehensively.

This session provides a platform to discuss which gaps remain between current and recommended model application, and how such gaps may be bridged. We welcome contributions under the following themes:

1. The theoretical underpinnings of models, centered around the question:
“How good are our models and the approaches we use to decide if they are, and how far can we stretch their applicability?”

2. The technical side of model application, centered around the question:
“How do we efficiently configure, parametrize and run our models?”

3. The social side of model application, centered around the question:
“Why do we use the tools we do, and how do we communicate their strengths and weaknesses?”

4. The future of model application, centered around our main question:
“Where are we today, and where do we want to be tomorrow?”

Contributions may include comparison of existing models and techniques, and development of more robust alternatives; as well as examples of studies that implement existing recommended practices throughout their modelling chain; and lessons from post-audits comparing past predictions with observed outcomes. We particularly invite abstracts that investigate whether model assumptions and representations are supported by evidence and whether models are fit for their intended purpose, and those that take a step back and assess the current state of hydrological modelling in comparison to what may, by now, be theoretically and practically possible.