A statistical approach to modelling the temporal patterns of ocean storms

Abstract

The world's sandy beaches, already eroding at rates of concern to coastal managers, are facing a further threat based on predictions of global climate change. Although mean sea level rise has long been recognized as a likely consequence of climate change, it has been suggested that there may also be a change in ocean storminess—that is, the intensity and frequency of storm events. While the effects of a change in wave height can be handled with existing analysis methods, the effects of a change in the rate at which storms arrive are less obvious and harder to predict. As a prelude to model studies investigating coastal response to changes in storminess, it is necessary to have a reliable method for generating storm and wave histories that contain waves that not only conform to known distributions of height and period but that also arrive at the proper rate in time. In the current work a statistical model of the arrival rate of storms and the wave heights within those storms has been developed. Storminess can be modelled by describing interstorm period, storm duration, the average wave height, and the temporal distribution of storm significant wave heights using methods that have been applied in rainfall modelling. The model has been verified against existing wave climate parameters of wave exceedence and wave persistence, and it has been shown to reproduce these statistics reliably.