Our journey towards predicting extreme floods faced its own set of hurdles, particularly evident in our initial attempt to forecast the estimated 8,000-year flood in the Ahr River. We quickly recognized that the existing German-wide daily hydrological model wasn't suitable  for such extreme events.  To overcome this challenge, we made significant adaptations, transitioning to hourly hydrological modeling and calibration tailored specifically for flood forecasting. Additionally, we enhanced our weather forecasting inputs,  leveraging high-resolution forecasts from the German Weather Service to replace the coarser atmospheric forecasts from the ECMWF.

While our water level predictions didn't perfectly match the estimated flood peak, analysis identified a high (>50%) chance of exceeding 100-year flood levels at the gauge 17 hours beforehand. Yet, this alone wouldn't aid flood managers needing  more detailed impact assessments to safeguard lives and infrastructure. Collaboration with GFZ colleagues  proved instrumental. We integrated near-real-time flood forecasting into our modeling, providing detailed impact forecasts for buildings and infrastructure at 10-meter resolution, along with probabilistic lead-time maps.

Teamwork was pivotal in this endeavour, with my  UFZ colleagues dedicating tireless hours within the first week after the flood to transition to hourly data and recalibrate models.A draft paper was ready by summer 2022, but impact forecasting and visualizations required further effort, highlighting the dedication needed for impactful research. Thanks to all co-authors!

The journey from disaster indication to publication involved addressing reviewer feedback and emphasising the research's ultimate goal of enabling more effective early warnings and decision-making for flood management.  We acknowledge transparent peer review, like Nature Communication's, which we believe is crucial as it exposes both the strengths and weaknesses of a publication, fostering accountability, trust, and integrity within the scientific community  [peer review file]. 

The research lays the groundwork for  advancements like the UN's EW4ALL initiative, focusing on workflow integration, uncertainty communication, and preparedness. By pushing hydrological modeling and forecasting boundaries and fostering collaboration, the research aims to provide actionable insights that can potentially save lives and minimise damage during extreme events.

Link to the paper: [https://www.nature.com/articles/s41467-024-48065-y]