Extreme rainfall regime characterization in Sardiniaudusing daily rainfall data

摘要

For the design of hydraulic structures for flood conveyance and discharge, orudprotection of territory against flood is fundamental the knowledge of the ``extremeudrainfall regime'' in the area where the hydraulic structures must be set up.udIndeed the design flood is commonly evaluated as output of rainfall-runoff models thatudreceive as input the quantitative description of a rainfall extreme event with a givenudexceedance probability.udThis dissertation assesses the performance of different statistical approaches inudcharacterizing extreme rainfall in the island of Sardinia (Italy).udAfter a detailed review of the theoretical bases of existing methodologies, we compareudthe results obtained from the use of:uda) a Generalized Extreme value (GEV) distribution model, and a Two componentudExtreme Value (TCEV) distribution model, both applied to yearly maxima of dailyudrainfall, and b) a Generalized Pareto (GP) distribution model applied to rainfalludexcesses above a properly specified threshold. For the latter purpose, we use theudMultiple Threshold Method (MTM) developed by Deidda(2010), which demonstrateudgood performance also in the case of quantized records.udIn order to describe the spatial variation of TCEV, GEV and GP model parameters audregional approach based on homogeneous regions, and two versions of Kriging (audcommonly used geostatistical approach) i.e. ordinary Kriging (OK), and Kriging foruduncertain Data (KUD), are compared.udThe obtained results are very promising, pointing towards the use of: a)a GEVuddistribution model for yearly rainfall maxima, and a KUD model to describe the spatialudvariation of model parameters, and b)a GP model for rainfall excesses and either an OKudor a KUD model for the spatial variation of model parameters. The reason why the OKudand KUD approaches lead to the same results in the GP case, is attributed to theudrobustness of the MTM method.