In this study we outline a risk-based decision-making framework for flood management purposes using a Bayesian network methodology. Flood risk assessments are often based on single hazard events. We acknowledge that with a changing climate, flood risk assessments need to be extended to consider several hazards and their possible simultaneous occurrence. Further, we argue that there is a need to include additional drivers of extreme events into a risk assessment in order to obtain a robust description of the occurrence of these events. It is widely accepted that large-scale weather systems influence local climate, and this addition to a risk assessment can ensure a more complete description of different flood inducing events. In our case study we focused on describing how our Bayesian method can assess the probability of concurrent events. We analysed high sea water level and precipitation events and used different threshold combinations of these hazards to assess the daily probability of concurrent events in Aarhus—the second largest city in Denmark. The results were validated though comparison of observed concurrent events in Aarhus. We showed that there was a clear variation in seasonal probabilities of concurrent events and distinguish weather systems with high probability of concurrent events.
Republished in part with permission from Åstrøm, H. L. A., Sunyer, M. A., Madsen, H., Hansen, P. F., Rosbjerg, D. and Arnbjerg-Nielsen, K. (2014). Describing concurrent flood hazards in a risk assessment decision framework using a Bayesian network methodology. In: 13th International Conference on Urban Drainage, 7-12 September, Sarawak, Malaysia.
Note: Journal articles and conference papers (and links where available) are available under open access arrangements where possible. Otherwise please contact your institution’s library, the authors, or publishers to organise full access.