It is well established that conventional observation systems are not suitable for monitoring atmospheric and hydrological generating phenomena with regard to flash floods. However, a flood preventing approach entails a strong capacity of observing the generating events (such as storms and violent rainfalls) in an accurate way, which was not the case at the time of the project launching. To overcome this lack of information, networking activities were needed at the European level for effectively managing flash flood events at a local scale.
To this end, the project set up an innovative information and observation system across Europe and was structured through three main goals.
Firstly, a data analysing approach collated both primary and detailed data from the major related events over the last decades. Primary data represents the minimal information which is required to qualify a flash flood event, i.e. occurrence data, location, peak discharge and basin characteristics. As to detailed data, they consist of fine resolution rainfall information and information concerning the basin and the river streams. A global network of Hydrometeorological Observatories (HO), all placed in high flash flood potential regions, was involved in the data set collecting initiative. This data stands for a representative sample of the various hydroclimatic regimes (from humid to semi-arid and arid) and of the different settings (e.g. mountainous, urban, rural) across Europe. Once all this data is gathered, it is used for identifying and understanding the generating mechanisms of flash floods and to thereby improve their forecasting.
Secondly, a software tool was developed to allow all the involved stakeholders to consult a full information package from an internet accessible platform. This database includes primary data (such as coordinates of the river section, basin area, peak discharge, precipitation data, etc.), detailed data (e.g. high resolution rainfall, aerial photography) and reports describing in particular damages and impacts on the exposed local communities.
Thirdly, a strategic planning approach provided a set of methods and tools to forecast flash flood events in ungauged basins. To do so, a modelling approach was developed to downscale the understanding of flash flood processes in ungauged basins either (i) by transposing model parameters from gauged catchment areas to ungauged catchment areas or (ii) by establishing new indicators to identify the dominant generation mechanisms within the target catchment (i.e. for a given rainstorm size and antecedent conditions). Besides, a radar-based nowcasting system was designed to support this forecasting action. The system is capable of estimating rainfalls to filter out small-size patterns of the precipitation field, which are known to be unpredictable when performing extrapolation techniques.
This project, which focuses on natural disaster management was coordinated by the University of Padua (Italy). As the project ended in 2010, the technology readiness level is estimated to be 9 on the TRL scale.