On-demand flood predictions and warnings with high-resolution weather and hydrological models: the case of Vejle, Denmark

Floods are often caused by small-scale weather and hydrological phenomena that require very high-resolution models to adequate resolve and simulate. Unfortunately, high-resolution models are expensive to run continuously in real-time, which makes the case for only running these specialized systems “on-demand” when it is deemed necessary. The aim of this study is to investigate a high-resolution digital twin designed for on-demand use and test it on the case of Vejle, Denmark. The city of Vejle is of special interest because it is prone to frequent floods from long-term winter precipitation, convective cloudburst events in summer as well as storm surges from the sea. On top of this, several fast and slow responding rivers meet inside the city

Here, we present results from a hydrological forecasting setup based around the conceptual HYPE model, which is developed by the Swedish Meteorological and Hydrological Institute. The model is developed with high-resolution soil and land use data, forced with high-resolution meteorological observations, and its predictions are evaluated at several gauges along the Vejle and Grejs rivers. The final aim of the research is to assess the benefits of utilizing subkilometer-scale HARMONIE weather predictions, which especially is expected to improve the resolution of local rainfall fields. The full forecasting chain will be put into operation in the coming year.