Free Surface Effects on the Near-Inertial Ocean Current Response to a Hurricane

View More View Less
  • 1 Division of Meteorology & Physical Oceanography, Rosenstiel School of Marine & Atmospheric Science, University of Miami, Miami, Florida
  • | 2 Atmospheric Physics Branch, Naval Research Laboratory, Washington, D.C
  • | 3 Department of Meteorology, U.S. Naval Postgraduate School, Monterey, California
© Get Permissions
Full access

Abstract

During the passage of hurricane Frederic in 1979, four ocean current meter arrays in water depths of 100–950 m detected both a baroclinic and a depth-independent response in the near-inertial frequency band. Although the oceanic response was predominately baroclinic, the hurricane excited a depth-independent component of 5–11 cm s−1.

The origin and role of the depth-independent component of velocity is investigated using a linear analytical model and numerical simulations from a 17-level primitive equation model with a free surface. Both models are forced with an idealized wind stress pattern based on the observed storm parameters in hurricane Frederic. In an analytical model, the Green's function (K0) is convolved with the wind stress curl to predict a sea surface depression of approximately 20 cm from the equilibrium position. The near-inertial velocities are simulated by convolving the slope of the sea surface depression with a second Green's function. The barotropic current velocities rotate inertially with periods shifted above the local inertial period by 1%–2% and the maximum amplitude of 11 cm s−1 is displaced to the right of the track at x = 2Rmax (radius of maximum winds).

The free surface depression simulated by the primitive-equation model is also about 18–20 cm. The primitive equation model simulations indicate that the vertical mean pressure gradient excites 10–11 cm s−1 depth-averaged currents at x = 3Rmax. The net divergence and convergence of the horizontal velocities induces vertical deflections of the sea surface. The spatial pattern of the barotropic amplitudes simulated by the numerical and analytical models differ by less than 2 cm s−1 in the region 0 < x < 4Rmax, which suggests that the barotropic response to the passage of a moving hurricane is governed by linear processes.

Abstract

During the passage of hurricane Frederic in 1979, four ocean current meter arrays in water depths of 100–950 m detected both a baroclinic and a depth-independent response in the near-inertial frequency band. Although the oceanic response was predominately baroclinic, the hurricane excited a depth-independent component of 5–11 cm s−1.

The origin and role of the depth-independent component of velocity is investigated using a linear analytical model and numerical simulations from a 17-level primitive equation model with a free surface. Both models are forced with an idealized wind stress pattern based on the observed storm parameters in hurricane Frederic. In an analytical model, the Green's function (K0) is convolved with the wind stress curl to predict a sea surface depression of approximately 20 cm from the equilibrium position. The near-inertial velocities are simulated by convolving the slope of the sea surface depression with a second Green's function. The barotropic current velocities rotate inertially with periods shifted above the local inertial period by 1%–2% and the maximum amplitude of 11 cm s−1 is displaced to the right of the track at x = 2Rmax (radius of maximum winds).

The free surface depression simulated by the primitive-equation model is also about 18–20 cm. The primitive equation model simulations indicate that the vertical mean pressure gradient excites 10–11 cm s−1 depth-averaged currents at x = 3Rmax. The net divergence and convergence of the horizontal velocities induces vertical deflections of the sea surface. The spatial pattern of the barotropic amplitudes simulated by the numerical and analytical models differ by less than 2 cm s−1 in the region 0 < x < 4Rmax, which suggests that the barotropic response to the passage of a moving hurricane is governed by linear processes.

Save