Model simulations of the Gulf of Maine response to storm forcing

Abstract

The storm response of the Gulf of Maine/Georges Bank region was investigated using the Dartmouth three‐dimensional, nonlinear, finite element coastal ocean circulation model called QUODDY, with uniform density and quadratic bottom stress. A suite of model hindcast experiments were conducted for the 4–8 February 1987 period that featured a strong nor'easter. The model was forced at the open ocean boundaries by the M2 semidiurnal tidal sea level and in the interior by realistic surface wind stresses and, in some cases, atmospheric pressure. The reference model run, with just tidal and wind stress forcing and a time‐space constant bottom drag coefficient Cd of 0.005, captured 65% of the observed detided, low‐pass filtered or “subtidal” sea level variance of a representative array of coastal and bottom pressure stations in the Gulf. However, there remained large unexplained model/observed sea level differences, particularly during the storm. Model sensitivity testing to a realistic range of constant Cd values did not improve model results significantly. However, results of a model run in which atmospheric pressure forcing, derived through linear interpolations of regional measurements, was added to wind stress forcing captured 82% of the observed subtidal sea level variance.