Challenging our knowledge on flood frequency

In recent times water-related hazards have captured public attention and led to an increased interest in developing hydrologic forecasting systems, and quantifying their reliability. Despite major investments in flood forecasting and flood protection measures undertaken after recurring widespread inundations (e.g. 2002 and 2013 in Eastern Europe), heavy rainfall events led to severe flooding in Western Europe during July of the past summer, resulting in severe impacts and massive losses, including over two hundred deaths. Extreme precipitation, breaking observed records, is expected to have an increased likelihood under global warming, all the more we should question and scrutinize our knowledge about the frequency and severity of floods. Reliable estimates of these, and a better quantification of their uncertainty are key information for improving our preparedness and developing adaption measures in the future.

For this purpose we explore the potential of running pooled weather reforecasts through the flexible hydrological model framework DECIPHeR (Coxon et al.2019) in Switzerland, modified to include snow, glaciers and the effect of lakes and reservoirs on the river network. For three case studies of different climatic regions (one including regulated lakes), we compare floods generated by the 10 largest precipitation events and precipitation events of low frequency (return period >= 100 years), both extracted from the pooled data, for three accumulation periods (1, 3, 5 days) with the official flood frequency curves and flood frequency curves generated with long continuous simulation using the pooled data (> 1000 years). This analysis will show us possible deficiencies of record-based flood frequency curves, if running selected precipitation events according to a return period already gives us a representative “flood sample”, and what is the gain of running long continuous simulations (e.g. are there overlooked events, summoning hydrological extremes through particular spatio-temporal patterns? What is the effect of different initial conditions?).

Coxon G., Freer J., Lane R., Dunne T., Knoben W.J.M., Howden N.J.K., Quinn N., Wagener T. and Woods R. : DECIPHeR v1: Dynamic fluxEs and ConnectIvity for Predictions of HydRology. Geosci Model Dev.,12,2285-2306, https://doi.org/10.5194/gmd-12-2285-2019, 2019.