Processes Influencing Storm-Induced Currents in the Irish Sea

Abstract

Although the problem of predicting storm surge elevations has received significant attention, the simulation of currents has suffered because of lack of current observations during surges. Current measurements made during surge conditions are presented here and are used in combination with three-dimensional models to understand processes producing storm currents in the Irish Sea. A coarse-grid (resolution of order 7 km) model of the west coast of Britain together with a fine-grid (of order 1 km) model of the eastern Irish Sea is used to examine the processes, namely, open boundary forcing of the west coast model and wind fields, that produced flows within the eastern Irish Sea during the storm surge of November 1977. Simulations of the surge show that the fine-grid model nested within the west coast model can reproduce observed coastal changes in surge elevation. However, an observed major inflow that was recorded by current meters in the region, prior to a storm surge elevation peak, is not represented, although subsequent inflows and outflows are reproduced. The flow fields in the west coast model giving rise to these currents are analyzed in detail. Also, computations are performed with idealized open boundary forcing and wind fields to understand their role in determining the circulation within the region. An analysis of computed flows shows that outflows from the eastern Irish Sea following major storm events are determined by sea surface elevation gradients in the region and topographic effects. Observed flows under these conditions are reproduced by the model. Inflows, however, are more difficult to compute and depend upon a delicate balance of northern and southern boundary forcing of the west coast model and wind fields over the region. The first observed inflow event, which was not reproduced in the model, was associated with a current from the south. A second inflow event that was reproduced arose from a combination of an inflow from north and south, and a third event was again reproduced in the model due to a current from the north. Without a more comprehensive observational dataset, it was not possible to determine the exact reason why the first inflow was not reproduced.