Comparison of the Extreme Surge Estimation with the Generalized Extreme Value Distribution, Using the Maximum of Positive Storm Surge and Skew Surge in Two Port Areas with Different Hydrodynamic and Tidal Characteristics
Jonathan F. Dato, M Nica M.E. Fiore, Enrique E. D’Onofrio, Fernando A. Oreiro and Walter H. Grismeyer
Extreme value statistics applied to positive storm surges (PSS) is a necessary tool for coastal management adaptation and decision- making. Generally, when extreme values of PSS are calculated, the maximum height of PSS or the skew surge (SK) for each event is used. In this paper, a comparison of the results obtained by applying the generalized extreme value (GEV) distribution to series of maximum PSS and SK (GEV-MPSS and GEV-MSK, respectively), observed in two port areas with different hydrodynamic and tidal characteristics, is performed for the period from 1956 to 2012. The sites chosen are Buenos Aires and Mar del Plata. The first one is located at the Río de la Plata Estuary, with PSS that tripled the height of the astronomical tide. The second is located on the Atlantic coast of Buenos Aires Province, with PSS that often reach the amplitude of the astronomical tide. The GEV distribution parameters are calculated by the maximum likelihood method, and estimated heights are obtained using MPSS and MSK for return periods up to 500 years. Results show that estimated heights calculated using GEV-MPSS or GEV-MSK differ from one to another for both locations. GEV-MPSS appears to overestimate the meteorological effect in some events, while GEV-MSK appears to underestimate this effect on others. It is also shown that both methodologies are not applicable to all events at the same location.
An alternative to obtain a more representative dataset of meteorological action would be using both methodologies, selecting the MSK or the MPSS that identify the maximum wind effect for each event.
Keywords: Storm Surge, Skew Surge, Return Period, GEV Model, Tide, Argentina