SEA-LEVEL RESPONSE TO PRESSURE FORCING IN A BAROTROPIC NUMERICAL-MODEL

A barotropic shallow-water model is used to study the large-scale sea level response to realistic barometric forcing, at periods ranging from 1 day to 1 year. Results are presented from coarse resolution "open" ocean experiments (i.e., no shallow continental shelf regions or marginal seas) with coastal geometries and bottom topography representative of the North Atlantic and Pacific basins. The validity of the inverted barometer (IB) approximation is examined in detail, including nonlocal effects which result from taking into account the constant volume of the ocean. These effects are found to be important at low latitudes, where a considerable part of the sea level variability is related to pressure forcing over higher latitudes. Root-mean-square deviations from an IB response in the range of 1-3 cm are typical, with most of the variance occurring at high frequencies. Basin-averaged estimates yield IB deviations of only a few percent at time scales longer than 1 week increasing to 5%-20% over the range from 1 week to 2 days, but significantly larger departures from isostatic behavior are possible locally. Over these time scales, deviations in the Pacific are generally larger than in the North Atlantic. For periods shorter than 2 days, the IB approximation is not reliable, with the largest departures from equilibrium occurring at frequencies where basinwide resonances are present. Model results agree, in general, with findings of previous studies and are interpreted using simple dynamical ideas. Some of the implications of our results are explored. Preliminary calculations suggest that nonisostatic effects in the ocean may significantly contribute to variations in the earth's rotation at short time scales (1 to 2 weeks or shorter). Issues concerning the need to interpret satellite altimeter estimates of surface height in the presence of pressure-driven variability are also briefly discussed.