Despite an increasing number of people exposed to flood risks in Europe, flood risk perception remains low and effective flood risk management policies are rarely implemented. It becomes increasingly important to understand how local governments can design effective flood risk management policies to address flood risks. In this article, we study whether high flood exposure and flood risk perception correlate with the demand for a specific design of flood risk management policies. We take the ideal case of Switzerland and analyze flood risk management portfolios in 18 flood-prone municipalities along the Aare River. We introduce a novel combination of risk analysis and public policy data: we analyze correlations between recorded flood exposure data and survey data on flood risk perception and policy preferences for selected flood risk management measures. Our results indicate that local governments with high flood risk perception tend to prefer non-structural measures, such as spatial planning and ecological river restoration, to infrastructure measures. In contrast, flood exposure is neither linked to flood risk perception nor to policy preferences. We conclude that flood risk perception is key: it can decisively affect local governments’ preferences to implement specific diversified policy portfolios including more preventive or integrated flood risk management measures. These findings imply that local governments in flood-prone areas should invest in raising their population’s awareness capacity of flood risks and keep it high during periods without flooding.

Surface water floods (SWFs) that lead to household losses are mainly localized phenomena. Research on describing the associated precipitation characteristics has previously been based on case studies and on the derivation of local rainfall thresholds, but no approaches have yet been presented on the national scale. Here, we propose a new way to overcome this scaling problem. We linked a gridded precipitation dataset based on both rainfall gauges and radar data with geolocated insurance claims for all of Switzerland. We show that the absolute thresholds vary markedly over complex terrain, and we thus propose basing early warning systems for predicting damage-relevant SWF events on local quantiles of maximum intensity and the total sum of event precipitation. A threshold model based on these two parameters is able to classify rainfall events potentially leading to damage-relevant SWF events over large areas of complex terrain, including high mountains and lowland areas, and a variety of geological conditions. Our approach is an important step towards the development of impact-based early warning systems. Weather warning agencies or insurance companies can build upon these findings to find workarounds for issuing user-targeted warnings at national scale or for nowcasting purposes.

Exploring the impacts of extreme weather events has gained increased attention in recent years after repeated record-breaking events, such as unprecedented river flood events in central Europe in summer 2021. After the flood event, risk management institutions, such as for example, fire brigades, civil protection units, or natural hazard experts asked if it was possible to predict the impacts of such an unprecedented event ex-ante and if similar events could occur elsewhere or if climatic changes can even worsen extreme flood events. Here, we present an approach to explore storylines of physically plausible extreme river flood events in a warming climate. The simulated flood events are based on extreme precipitation events selected from re-forecast archives. River discharge, flood processes, and their impacts on people and infrastructure are simulated for nine storylines under five global warming levels for the alpine and pre-alpine headwater catchments of the main rivers in northern Switzerland. The precipitation intensity was increased linearly with global warming according to the Clausius-Clapeyron relation. The coupled hydrological-hydraulic simulations show that the increase in peak discharge is higher than the increase in precipitation. The flood impacts increase non-linearly with higher global warming levels. The results show that record-breaking, high-impact river floods are possible under the current atmospheric conditions, and climate change substantially aggravates flood impacts. The simulations show a broad range of potential outcomes of flood impact storylines depending on the simulated scenario and local conditions. This leads to the conclusion that storylines of extreme flood events are a valuable tool to explore, describe, and communicate extreme events. Still, the inherent challenge is the communication of the representativeness of a particular storyline and, hence, the practical consequences that should or should not be derived. We, therefore, suggest considering storylines of events with a range of magnitudes and different spatiotemporal precipitation patterns to comprehend the possible range of outcomes and select appropriate storylines out of the set for risk communication.

Beside the flood hazard analysis, a comprehensive flood risk assessment requires the analysis of the exposure of values at risk and their vulnerability. Currently, the main focus of such analysis is on losses on building structure. However, loss on household contents accounts for up to 30% of the total losses on buildings due to floods. Based on insurance claim records, we developed and (cross-)validated two functions. The models based on linear regressions estimate the monetary loss and the degree of loss of household contents by the monetary and degree of loss for building structure, respectively. The main focus herein is to develop functions which provide robustness in prediction and transferability to other regions. Both models generate appropriate results with a comparative advantage of the relative over the absolute loss model. Our results indicate that the ratio of household content to building structure loss is decreasing relatively in regions with comparatively high losses or degrees of loss. A detailed examination of the model residuals, shows that the Box-Cox transformation works well to accurately fit a standard regression model to general right-skewed loss data as the transformed data meet the assumptions of a regression model.