The growth and decay of meanders in an isopycnic channel model initialized with Gulf Stream data
Date of Award
Doctor of Philosophy (Ph.D.)
First Committee Member
William E. Johns, Committee Chair
A 6-layer version of the Miami Isopycnic Coordinate Ocean Model (MICOM) configured in a 1000 x 2000 km periodic channel with 5 km horizontal resolution is used to study the nonlinear instability of the Gulf Stream system. The model is initialized with isopycnal surfaces derived from Halkin and Rossby's (1985) analysis of the mean Gulf Stream structure at 73°W. The model is run for two years in a cyclic spin-down configuration, and two years in a controlled-inlet configuration. Both configurations exclude surface winds, topographic effects, and thermodynamic forcing, in order to study the meander development solely due to the internal dynamics of the flow during initial meander development and the evolution of the meandering spectrum toward a turbulent equilibrium state.Both model configurations spontaneously develop meanders that share several characteristics of Gulf Stream meanders: the model meander phase speeds decrease as the meander wavelengths increase; they develop a pronounced westward axial tilt as they evolve; and warm and cold core rings form that interact strongly with the jet. Meander phase speeds measured from direct injection of known-wavelength EKE into the cyclic spin-down configuration compare favorably with measured Gulf Stream dispersion curves, as do temporal growth rates determined from the basin-averaged EKE and EKE conversions.Basin-average energetics reveal an initial baroclinic release of eddy kinetic energy. The cyclic spin-down configuration's growth of EKE is controlled at finite amplitude by the reverse barotropic conversion of EKE to mean kinetic energy, consistent with the "barotropic governor" mechanism of James and Gray (1986). This asymmetric life cycle is not shared by the controlled-inlet configuration, which displays an initial baroclinic release of EKE, followed by a baroclinic decay of the EKE, with no significant contribution to the initial growth of EKE by the barotropic conversion.Subsequent development of meanders in the cyclic spin-down configuration is influenced by a "closed-loop" energy cycle involving the production of EKE at 300 km by the baroclinic conversion, followed by the shifting of the EKE wavelengths to between 500--1000 km through the enstrophy cascade. The reverse barotropic conversion depletes the long wavelength EKE by converting it to mean kinetic energy. The excess MKE is restored to the pool of available potential energy, where the process repeats. No such energy loop is observed in the controlled-inlet configuration.An oscillation in the EKE of the controlled-inlet configuration is observed, which may be related to a 9-month cycle in the Gulf Stream meandering observed by Lee and Cornillon (1995). A 4.5 month oscillation in the monthly-mean RMS displacement of the model jet is found, corresponding to multiple simultaneous ring-jet interactions. The meridional symmetry in the model jet and boundaries may be forcing both warm-core and cold-core rings in the model to exhibit "warm-core-like" mean lifetimes of 4.5 months instead of the significant differences in lifetimes for the two ring types as has been observed in the Gulf Stream.Examination of the oscillation in the EKE shows that the period is extending over the course of the 2-year experiment, suggesting that the controlled-inlet configuration has not reached statistical equilibrium after two years. No similar oscillation is observed in the cyclic spin-down configuration.
Spindler, Todd Donald, "The growth and decay of meanders in an isopycnic channel model initialized with Gulf Stream data" (2001). Dissertations from ProQuest. 1776.