AN INNOVATIVE APPROACH FOR HAZARD ASSESSMENT AT DIKED RIVERS
Corrado Lucarelli, Walter Gostner
Abstract
Corrado Lucarelli, Walter Gostner
Abstract
Two major scenarios can be differentiated for hazard assessment: either the dikes are overtopped or piping through the dike occurs. In both cases the dikes are eventually destroyed causing the inundation of the floodplains. Whereas the first case is a straightforward one to evaluate, for the second case differentiated considerations have to be made. In order to assess the probability of flooding due to piping and subsequent failure of dikes an innovative approach has been developed. In fact, the probability of piping and dike failure lies between 0% and 100%. In order to calculate the probability of dike failure, the semi-probabilistic safety concept of the Eurocodes have been used (i.e. limit state equation of reliability analysis), by taking into consideration geotechnical and other characteristics of the dikes, such as dike geometry, presence of a counter dike, friction angle and cohesion of the soil types forming the dike, cover of the dike crest, protection of dike toe, presence of diaphragm walls or other sealing systems, etc. The establishment of a best estimate fragility curve, by taking into consideration also the variability of parameters with the help of a Monte-Carlo analysis, served to correlate water depth in the river with failure probability of the dike. Since flow depth in the river can be related to a certain discharge, the failure probability of the dike can be correlated to the return period of the flood event occurring in the river (Fig. 1). With the help of event trees the probability of a dike failure was then combined with the probability of the appurtenant flood event (return period) occurring in the stream. From these calculations it was possible to define the probability for flooding of the affected areas. In order to define the hydrographs inundating the floodplain a historical analysis of the length of the breaches was conducted. The length and the duration of formation of the breaches served to calculate the hydrographs flowing through the breaches. These hydrographs were used as input for numerical 2d-modelling of the floodplains. The 2d-modelling provides maximum water depths and flow velocities, allowing to determine the intensity of the inundation for each of the cells forming the calculation grid. 2d-modelling was carried out for each potential dike failure along the considered river reach.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Two major scenarios can be differentiated for hazard assessment: either the dikes are overtopped or piping through the dike occurs. In both cases the dikes are eventually destroyed causing the inundation of the floodplains. Whereas the first case is a straightforward one to evaluate, for the second case differentiated considerations have to be made. In order to assess the probability of flooding due to piping and subsequent failure of dikes an innovative approach has been developed. In fact, the probability of piping and dike failure lies between 0% and 100%. In order to calculate the probability of dike failure, the semi-probabilistic safety concept of the Eurocodes have been used (i.e. limit state equation of reliability analysis), by taking into consideration geotechnical and other characteristics of the dikes, such as dike geometry, presence of a counter dike, friction angle and cohesion of the soil types forming the dike, cover of the dike crest, protection of dike toe, presence of diaphragm walls or other sealing systems, etc. The establishment of a best estimate fragility curve, by taking into consideration also the variability of parameters with the help of a Monte-Carlo analysis, served to correlate water depth in the river with failure probability of the dike. Since flow depth in the river can be related to a certain discharge, the failure probability of the dike can be correlated to the return period of the flood event occurring in the river (Fig. 1). With the help of event trees the probability of a dike failure was then combined with the probability of the appurtenant flood event (return period) occurring in the stream. From these calculations it was possible to define the probability for flooding of the affected areas. In order to define the hydrographs inundating the floodplain a historical analysis of the length of the breaches was conducted. The length and the duration of formation of the breaches served to calculate the hydrographs flowing through the breaches. These hydrographs were used as input for numerical 2d-modelling of the floodplains. The 2d-modelling provides maximum water depths and flow velocities, allowing to determine the intensity of the inundation for each of the cells forming the calculation grid. 2d-modelling was carried out for each potential dike failure along the considered river reach.
Key concepts: Dike, Geology, Geotechnical engineering, Piping, Levee, Engineering, Environmental engineering, Geochemistry