The Flood Directive 60/2007/EC requires the European countries to verify the effectiveness of existing flood defence infrastructure to mitigate flood risks. The current practice establishes that the river flood control structures must respect a basic requirement, usually consisting of a certain safe freeboard under a design peak flow rate corresponding to a certain probability of exceedance. This requirement has some critical issues. It is based on a univariate frequency analysis of only flood peak, and therefore it assumes a perfect correspondence between the probability of occurrence of the hydrological variable and the failure of the floodcontrol structure. The thesis aims to define a methodology to overcome these issues implementing a bivariate hydrological risk analysis for riverrelated flood risk. The methodology is mainly focused on the overtopping failure for river levees. River levees are the most common river floodcontrol structure. They are raised, predominantly earth, structures (also called dykes, digues or embankments). The overtopping failure of a river levee caused several flood disasters such as Elbe flood (2002), New Orleans flood (2005), Emilia Romagna Flood (2017) or Arkansan flood (2019). The proposed procedure is carried out through two steps: (i) the evaluation of hydrological failure related to the overtopping risk for a levee; (ii) the estimation of the probability of occurrence of the hydrological failure introducing the concept of the Bivariate Failure Return Period. The hydrological failure is determined considering the mutual interaction between a bivariate hydrological load of peak discharge (Q) and the volume of the hydrograph (V), the river conveyance, and the levee resistance with respect to overtopping. The bivariate hydrological load considers an approximation of the real bivariate distribution of Q and V, functional to determine the hydrological failure. The shape of the hydrographs is classified concerning the overtopping introducing the Overtopping Hydrograph Shape Index (OHSI). The hydrological failure condition is represented by a curve in QV space containing all the hydrographs causing the initiation of the damage. This curve demonstrates that not only the peak flow but also the volume of the hydrograph are essential variables to characterise the overtopping failure. The risk of overtopping failure is expressed by the probability of occurrence of the hydrological failure within a new interpretation of the return period. Because of the hydrological failure curve is a function in QV space, the return period is estimated in the bivariate framework. Several definitions of the bivariate return period are available, each of which gives a different interpretation of it. This critical issue is overcome introducing the Bivariate Failure Return Period. The Bivariate Failure Return Period assesses the probability of the failure curve of the hydraulic structure generating possible scenarios through a Monte Carlo simulation. Two case studies are presented to demonstrate the applicability of the methodology and the advantages of using it respect to the current practice. In the thesis, the methodology is also applied to the problem of flood damage estimation demonstrating the flexibility and the validity of the proposed procedure. In this case, the hydrological failure consists in equaleuro curve in QV space, which includes all the hydrographs causing the same euro flood damages in a site. The procedure proposed needs QV data at the target site where the risk is to be assessed but most of the sites are ungauged. This issue is overcome by testing the bivariate regional frequency analysis. It is applied to the case study of the entire Tuscany Region (Italy). By the bivariate regional analysis, the QV series can be estimated at ungauged sites, and the uncertainty is reduced at gauged sites.
A methodology for the bivariate hydrological characterization of flood waves for riverrelated flood risks assessment / Matteo Isola.  (2020).
A methodology for the bivariate hydrological characterization of flood waves for riverrelated flood risks assessment
Matteo Isola^{}
2020
Abstract
The Flood Directive 60/2007/EC requires the European countries to verify the effectiveness of existing flood defence infrastructure to mitigate flood risks. The current practice establishes that the river flood control structures must respect a basic requirement, usually consisting of a certain safe freeboard under a design peak flow rate corresponding to a certain probability of exceedance. This requirement has some critical issues. It is based on a univariate frequency analysis of only flood peak, and therefore it assumes a perfect correspondence between the probability of occurrence of the hydrological variable and the failure of the floodcontrol structure. The thesis aims to define a methodology to overcome these issues implementing a bivariate hydrological risk analysis for riverrelated flood risk. The methodology is mainly focused on the overtopping failure for river levees. River levees are the most common river floodcontrol structure. They are raised, predominantly earth, structures (also called dykes, digues or embankments). The overtopping failure of a river levee caused several flood disasters such as Elbe flood (2002), New Orleans flood (2005), Emilia Romagna Flood (2017) or Arkansan flood (2019). The proposed procedure is carried out through two steps: (i) the evaluation of hydrological failure related to the overtopping risk for a levee; (ii) the estimation of the probability of occurrence of the hydrological failure introducing the concept of the Bivariate Failure Return Period. The hydrological failure is determined considering the mutual interaction between a bivariate hydrological load of peak discharge (Q) and the volume of the hydrograph (V), the river conveyance, and the levee resistance with respect to overtopping. The bivariate hydrological load considers an approximation of the real bivariate distribution of Q and V, functional to determine the hydrological failure. The shape of the hydrographs is classified concerning the overtopping introducing the Overtopping Hydrograph Shape Index (OHSI). The hydrological failure condition is represented by a curve in QV space containing all the hydrographs causing the initiation of the damage. This curve demonstrates that not only the peak flow but also the volume of the hydrograph are essential variables to characterise the overtopping failure. The risk of overtopping failure is expressed by the probability of occurrence of the hydrological failure within a new interpretation of the return period. Because of the hydrological failure curve is a function in QV space, the return period is estimated in the bivariate framework. Several definitions of the bivariate return period are available, each of which gives a different interpretation of it. This critical issue is overcome introducing the Bivariate Failure Return Period. The Bivariate Failure Return Period assesses the probability of the failure curve of the hydraulic structure generating possible scenarios through a Monte Carlo simulation. Two case studies are presented to demonstrate the applicability of the methodology and the advantages of using it respect to the current practice. In the thesis, the methodology is also applied to the problem of flood damage estimation demonstrating the flexibility and the validity of the proposed procedure. In this case, the hydrological failure consists in equaleuro curve in QV space, which includes all the hydrographs causing the same euro flood damages in a site. The procedure proposed needs QV data at the target site where the risk is to be assessed but most of the sites are ungauged. This issue is overcome by testing the bivariate regional frequency analysis. It is applied to the case study of the entire Tuscany Region (Italy). By the bivariate regional analysis, the QV series can be estimated at ungauged sites, and the uncertainty is reduced at gauged sites.File  Dimensione  Formato  

Isola_Dissertation.pdf
Accesso chiuso
Descrizione: Tesi di Dottorato
Tipologia:
Tesi di dottorato
Licenza:
DRM non definito
Dimensione
7.44 MB
Formato
Adobe PDF

7.44 MB  Adobe PDF  Visualizza/Apri Richiedi una copia 
I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.