DFG Funds New Project on Hybrid Hydrological Models in Anneli Guthke’s Junior Research Group

July 31, 2025

The German Research Foundation (DFG) has approved funding for a new research project within Dr. Anneli Guthke’s junior research group at SimTech. The two-year project, “Diagnostic Evaluation of Hybrid Hydrological Models in the UNITE Framework (UNI-BENCH)”, is a continuation of the group’s successful previous work on model diagnostics.

The project builds on the previous DFG-funded UNITE project (GU1926/1-1), which established a formal evaluation framework based on information theory for comparing physics-based, data-driven, and hybrid hydrological models. UNITE established a set of methods for estimating information-theoretic quantities and using them to better understand the role of the data-driven and physics-based components in hybrid models, UNI-BENCH now aims to expand this work toward three core dimensions of model evaluation: performance, interpretability, and process consistency.

Hybrid models, which combine physically-based and machine learning components, promise both predictive power and process insight – but often fail to deliver trustworthy, interpretable results. UNI-BENCH addresses this by developing new diagnostics, including entropy- and mutual information-based metrics, to understand the internal functioning of hybrid models. The project aims to map model behavior onto physical processes to ensure that models are “right for the right reasons.” Tools developed will be integrated into the existing open-source UNITE toolbox.

The research includes controlled synthetic experiments and real-world case studies, such as the well-instrumented Attert catchment in Luxembourg and the large-sample CAMELS-DE dataset. The overarching goal is to establish a universal benchmarking routine applicable across model types and case studies.

Dr. Guthke’s junior research group, established 2021 within SimTech’s Early Career Program, focuses on model diagnostics and uncertainty quantification in environmental modeling, with a particular emphasis on hydrology. The group combines expertise in geoscience, machine learning, and information theory.

Dr. Guthke acknowledged the strong interdisciplinary environment within SimTech that helped shape the ideas of this continuation proposal. Speaking about the successful grant application, Dr. Guthke said: “I’m deeply grateful for the many inspiring discussions across SimTech project networks about model-data integration. We need to critically rethink the ways in which we infuse knowledge into our (hybrid) models, and I’m looking forward to contributing to SimTech’s research with this new project.”

Applications for this 2-year postdoc project are welcome! Please reach out to Dr. Anneli Guthke.

To the top of the page