Michael Dietze, Thomas Hoffmann, Rainer Bell, Lothar Schrott, Niels Hovius

• About 10% of Europe's surface area is prone to rapid flooding of rivers confined in valleys. The devastating potential of such floods is exacerbated by the deficits of existing gauging networks, including low station densities and recording frequencies, and lack of information beyond stage height. Here, we use seismic data of the July 2021 Ahrtal flood, Germany, to extract information to complement sparse hydrometric data, and to reconstruct the rapid evolution of this fatal event. We show that a seismic station can deliver essential flood metrics such as magnitude, propagation velocity and debris transport rate. These seismic products provide high resolution insight to the non-linear flood behavior. We argue that an approach combining distributed low-cost seismometers with existing seismic stations, can provide important real time data on future catastrophic floods and associated hazards in upland catchments, offering precious response time also in currently ungauged landscapes.