Simultaneous Occurrences of High Waves and High Storm Surges along the Norwegian Coast

Abstract

The systematic analysis of simultaneous occurrences of large waves and high water levels will help to reduce the future risk imposed by such hazard. The present study aimed to estimate the individual probability distribution and extremes of contributing variables, the combined distribution of their simultaneous occurrences and to demonstrate,in a lesser extent, the degree of dependence and correlation through determination of their correlation coefficient and by plotting probability density contours of numbers of events in some selected regions along the Norwegian coast.The extent of the present study kept limited to the statistical analyses of sequential and ordered data of waves and storm surges, the application of the Direct Joint Probability Method (DJPM) for only tides and surges and the approximate estimation of statistical dependence between waves and storm surges based on deep-water waves from‘hindcast’ and the observed and predicted sea level data.During the Peak over Threshold method while determining the extreme values, a sensitivity analysis between the extrapolated results from the Gumbel and Weibull distribution suggested preference on selecting Gumbel distribution as a user-friendly and reliable distribution type. The extreme sea levels comprised of tides and surges estimated by the DJPM, showed quite satisfactory prediction in a sense that the predicted results were expected from their joint occurrence histograms as there were only a few occasions where high tides and large surges occurred simultaneously. It is highly recommended that further analysis of the DJPM with three contributing variables(waves, tides and surges) should be carried out to predict more accurate extreme flood levels. The examination of simultaneous occurrences of waves and storm surges through dependence and correlation theory showed that thesetwo variables are largely statistically dependent. So, the Joint density approach based on the fitted distribution anddependence among water level, wave height and wave period using Monte Carlo simulation can be recommended to be undertaken as the next step in this process.