Numerical Simulation of Topographically Forced Mesoscale Variability in a Well-Mixed Marine Layer

Abstract

A simple nonlinear numerical model of a well-mixed marine layer is used to study topographically forced mesoscale variability off coastal California. The model is used to simulate a persistent wind maximum observed near Point Conception during northwesterly winds. The model also demonstrates the development of a coastally trapped Kelvin wave and a marine-layer eddy when the large-scale forcing is suddenly reduced. The model is a one-layer, two-dimensional, gridpoint model with idealized coastal topography. The model assumes that potential temperature and wind are constant with height in the layer and that the layer is capped by an inversion. Effects of diabatic heating, water vapor, entrainment, and spatial variations of potential temperature are neglected in order to focus on topographic effects. The model solves for the two horizontal components of the marine-layer wind and the marine-layer height. A comparison of the model results with observations taken near Point Conception during the 1983 OPUS (Organization of Persistent Upwelling Structures) project shows that the model simulates the general features of the observed mesoscale wind maximum. The success is due to the very fine grid size of 3.5 km. The model wind perturbation and along-trajectory acceleration show the effect of the prominent Arguello headland on the marine-layer wind. The northwesterly flow is blocked by the headland on the upwind side, and this causes the marine-layer height to rise there. On the downwind side the northwesterly flow removes mass from the region, and the marine-layer height decreases. This perturbation in the marine-layer height creates a local pressure-gradient force that is responsible for the existence of the wind maximum. The model simulation of the marine-layer height is found to be in agreement with observations in the region. The model also simulates a solitary atmospheric Kelvin wave crest in the marine layer north of the Arguello headland and a marine layer eddy to the south of the headland when the large-scale forcing is sharply reduced. Model simulation of these phenomena supports the hypothesis that they are coastally trapped marine-layer responses to changes in synoptic-scale forcing.