Background and aim

Models are used to test our understanding of hydrological systems, to estimate hydrological variables in places without observations, and to make predictions of the near or distant future. To ensure meaningful usage, we need to understand how well a model represents the system of interest, i.e. to evaluate the model. Model evaluation should ideally be diagnostic, showing why the model is wrong and how it could be improved (Gupta et al., 2008).

This project investigates the representation of the short and long-term water balance in the hydrological model RoGeR (Runoff Generation Research; Schwemmle et al., 2023), in particular how the model partitions the incoming precipitation into evaporation and streamflow. You will first investigate the spatial variability of the long-term water balance for different catchments in Baden-Württemberg. You will then also investigate the temporal variability within these catchments by examining annual water balances. By performing the same analysis with both modelled and observational data, you will be able to investigate the functioning of the model and how well it represents the real world. For example, if the model produces too much streamflow in comparison to observations, one can investigate whether this is related to uncertainties in the data or to improper process representations in the model.

Data and methods

The analysis uses model output data from the hydrological model RoGeR and observational data from several catchments in Baden-Württemberg. Averaged RoGeR inputs (e.g. precipitation), parameters (e.g. soil type), and outputs (e.g. streamflow) are available for each catchment, as well as observed streamflow (long-term and yearly averages). The project can be divided into three parts. First, you will create a brief overview of the controls on the water balance at different time scales (long-term and yearly) which will serve as a conceptual basis of the project. Second, you will use observations from several catchments to explore hydrological behavior in real systems. Third, you will use the same methods to explore the behavior of RoGeR, check if modelled and observed data align, and explore the reasons if they do not.

Challenges

The project mostly involves data analysis and geospatial analysis and therefore you should have appropriate coding and GIS skills (e.g. Python or R, QGIS could be benefitial). You should also be comfortable with the management of large datasets. In addition, a good understanding of the catchment water balance and the hydrological modelling process are desirable.

Supervision

Sebastian Gnann (University of Freiburg) and Markus Weiler (University of Freiburg)

Contact

Sebastian Gnann

sebastian.gnann@hydrologie.uni-freiburg.de

Tel. +49 (0)761 / 203-9283

Language

English or German

References

Gupta, H. V., Wagener, T., & Liu, Y. (2008). Reconciling theory with observations: elements of a diagnostic approach to model evaluation. Hydrological Processes: An International Journal, 22(18), 3802-3813.

Schwemmle, R., Leistert, H., Andreas, S., & Weiler, M. (2023). RoGeR v3. 0.3–a process-based hydrological toolbox model in Python. Geoscientific Model Development Discussions, 2023, 1-18.

Berghuijs, W. R., Gnann, S. J., & Woods, R. A. (2020). Unanswered questions on the Budyko framework. Journal of Hydrology, 265, 164-177.

Sankarasubramanian, A., Vogel, R. M., & Limbrunner, J. F. (2001). Climate elasticity of streamflow in the United States. Water Resources Research, 37(6), 1771-1781.