Ocean Waves and Coastal Flood Hazards

In the present study, a recently developed novel approach (Bender et al. in J Hydrol 514:123-130, 2014) has been further extended to investigate the changes in the joint probabilities of extreme offshore and nearshore marine variables with time and to assess design the total water level (TWL) at the shoreline under the effects of climate change. The nonstationary generalised extreme value (GEV) distribution has been utilised to model the marginal distribution functions of marine variables (wave characteristics and sea levels), within a 40-year moving window. All parameters of the GEV were tested for statistically significant linear and polynomial trends over time, and best-fitted trends have been detected. Different copula functions were fitted at the 40-year moving windows, to model the dependence structure of extreme offshore significant wave heights and peak spectral periods, and of wave-induced sea levels on the shoreline and nearshore sea levels due to storm surges. The most appropriate bivariate models were then selected. Statistically significant polynomial trends were detected in the dependence parameters of the selected cop-ulas, and time-dependent most likely bivariate events were extracted to be used in the estimation of the TWL at the shoreline. The methods of the present work were implemented in three selected Greek coastal areas in the Aegean Sea. The analysis revealed different variations in the most likely estimates of the offshore wave characteristics and nearshore storm surges in the three study areas, as well as in the time-dependent estimates of TWL at the shoreline. The approach combines nonstationarity and bivariate analysis, blends coastal and offshore marine features and finally provides non-trivial alterations in the response of coastal sea level dynamics to climate change signals, compared to former work on the subject. The methodology produces reasonable estimates of design quantities for coastal structures and boundary conditions for the assessment of flood hazard and risk in coastal areas.

Extreme events of sea level elevation, due to severe weather conditions, pose great threats to low-land coastal areas by extended inundation hazards. The latter take the form of short- to mid-term flooding due to wave- and storm-induced sea level elevation and run-up on the coast. In this paper, the impact of the combined effect of extreme storm surges and extreme wave set-up in nearshore areas is investigated. The framework is set by future and historic climate change scenarios during a period of 150 years (1951–2100) that affect the occurrence frequency and magnitudes of total (surge- and wave-induced) sea level extremes in the eastern Mediterranean focusing on the coastal zones of Greece. Inter-annual and multi-decadal patterns, trends and return levels of storm surge and wave set-up extremes are calculated based on nonstationary bivariate statistical analysis with copula functions of the Generalized Extreme Value distribution. The numerical data of storm surge- and wave-induced sea levels are derived from post-processing of simulation results by GreCSS and SWAN models, respectively, in order to transfer validated numerical data from offshore regions towards the shoreline of selected areas prone to coastal flooding. An increase and a consequent attenuation of storminess and inter-annual extremes of total sea level on the coast is estimated during the 1st and 2nd half of the 21st century, respectively. Different morphological characteristics of regional coastal zones in the Aegean Sea are found to influence variability of sea level extremes.

This paper describes the operational wave forecasting system that has been implemented at the Portuguese Institute for Ocean and Atmosphere (IPMA) along the west coast of the Iberian Peninsula. This system is based on the Simulating WAves Nearshore (SWAN) model, is forced at the boundaries by the Limited-Area Wave (LAW) model forecasts generated by the European Centre for Medium-Range Weather Forecasts (ECMWF) and by the 10m wind fields from the ALADIN atmospheric model. Moreover, the performance of three SWAN whitecapping parameterizations is evaluated. This evaluation is performed for significant wave height, mean wave period and mean wave direction forecasts through comparisons with observations from four buoys over a 6 months period. Finally, a short evaluation of the operational version of SWAN model at IPMA during the recent Christina Storm is presented.