A Spin-Up Model as a Diagnostic Tool for Interpretation of Current and Density Measurements on the Continental Shelf of the Pacific Northwest

Abstract

The primary goal of this paper is to prescribe the practical limitations of the two-dimensional, baroclinic, time-dependent model of Hamilton and Rattray (1978) for the Oregon-Washington continental shelf. A unique time series of density sections at approximately 3-day intervals over a 5-week period makes the test possible. The Columbia River effluent was modeled by a surface freshwater flux that allowed the plume to become detached from the coast during periods of northerly winds. Energy levels and phases of fluctuations in longshore currents and sea level could be effectively predicted with a vertically averaged version of the two-dimensional model whose essential longshore dynamics are a balance between local acceleration (multiplied by shelf depth), surface friction and bottom friction. With respect to simulation of the density field, the two-dimensional model was effective in predicting offshore displacement of surface isopycnals and vertical displacement of isopycnals below the surface layer for periods as long as 15 days. Significant non-local effects were observed twice in 30 days. Some improvement in agreement for deeper isopycnals was obtained when barotropic non-local effects were included by driving the model with h?v?/?t, where h is the shelf depth and v? the depth-mean current. This function, which almost exactly reproduces the barotropic longshore current, is based on an observed time-scale separation between wind stress and current.