| description abstract | Using a nonlinear ?reduced gravity? model it is shown analytically that a large buoyant midlatitude outflow situated along the northern boundary of a ?-plane ocean produces an unusually broad westward flow. The steady nonlinear outflow consists of a narrow jet (whose width is the familiar midlatitude Rossby radius) and a broad, nearly stagnant region whose width is the equatorial Rossby radius. The steady, inviscid solution reported here is constructed with the aid of the momentum-flux equation. For high-Rossby-number flows (i.e., zero potential vorticity flows), the total outflow's width is ?1.228 times the equatorial Rossby radius. A finite potential vorticity outflow produces a slightly narrower westward flow. The above solution breaks down in the linear limit, and it is expected that a linear outflow would consist of a single flow whose width is on the order of the midlatitude Rossby radius. Numerical simulations are in very good agreement with the above nonlinear solution. The new southward outflow solution reported here complements recently derived solutions for northward outflows, eastward outflows, and westward outflows. In the earlier northward outflow case (i.e., a light water source situated along a southern boundary), a chain of westward-propagating eddies and an eastward current are produced. This corresponds to the situation in the Gulf of Mexico where Loop Current eddies are generated. In the earlier eastward outflow case (i.e., a light water source situated along a western boundary), a steady gyre is generated. This corresponds to the Tsugaru gyre (in the western Pacific) and the Alboran gyre (in the western Mediterranean). In the earlier westward outflow case (i.e., a source situated along an eastern boundary), a chain of westward- propagating eddies and a northward current are produced. (The Southern Hemisphere analog of this case corresponds to the generation of teddies and the Leeuwin Current.) A new solution with the new length scale corresponds to outflows such as the one associated with the surface flow (from the Atlantic to the Caribbean) through the Windward or Mona Passage. At present, there are no observations that can support or reject the scales suggested by the authors. | |
