Flood probability mapping by means of conceptual modeling

Flood forecasting systems which provide real time flood maps are a valuable source of information for water managers and emergency units in case of flood events. However, the uncertainty involved in the flood forecasting and its influence on the inundation results and flood maps is most often not taken into account. This paper presents a method which allows producing probabilistic flood maps. These flood maps show the probability of flooding for areas in the vicinity of the modelled river. They can be updated in real time based on rainfall forecasts and related updates in the flood forecast. The approach was tested for the Dender river in Belgium for the flood of November 2010. It is shown that the probabilistic flood maps can provide accurate predictions of the flood probability. The method was validated based on historical data on the flood extent.