HYCOM
417
dense that its isopycnic reference density, the generator attempts to move the
bottom interface downward so that the flux of denser water across this
interface increases density. If the layer is denser than its isopycnic reference
density, the generator attempts to move the upper interface upward to
decrease density. In both cases, the generator first calculates the vertical
distance that the interface must be relocated so that volume-weighted density
of the original plus new water in the layer equals the reference density.
Repeated execution of this algorithm at every time step does maintain layer
density very close to its reference value as long as a minimum thickness
does not have to be maintained. To insure that a permanent p-coordinate
domain exists near the surface year round at all model grid points, the
uppermost layers are initialized with reference densities smaller than values
found anywhere in the model domain. The minimum thickness constraint is
not enforced at the bottom in the open ocean, permitting model layers to
collapse to zero thickness there as in MICOM.
The capability ofҏ assigning additional coordinate surfaces to the
HYCOM mixed layer allows the option of implementing sophisticated
vertical mixing turbulence closure schemes [see Halliwell (2004) for a
review]. The full set of vertical mixing options contained in the latest
version of HYCOM (http://hycom.rsmas.miami.edu) includes five primary
vertical mixing submodels, of which three are vertically “continuous”
models and two are predominantly or totally bulk models. The three
continuous models, which govern vertical mixing throughout the water
column, are: K-Profile Parameterization of Large et al. (1994) (KPP), the
level 2.5 turbulence closure of Mellor and Yamada (1982) (MY), and the
Goddard Institute for Space Studies (GISS) level 2 turbulence closure of
Canuto et al. (2001, 2002). The other two are the quasi-bulk dynamical
instability submodel of Price et al. (1986) (PWP) and the bulk Kraus-Turner
(1967) submodel (KT).
The following procedure is used to implement the three continuous
vertical mixing submodels. Velocity components are interpolated to the p
grid points from their native u and v points. The one-dimensional submodels
are then run at each p point to calculate profiles of viscosity coefficients
along with T and S diffusion coefficients on model interfaces. The onedimensional vertical diffusion equation is then solved at each p point to mix
T, S, and tracer variables, which involves the formulation and solution of a
tri-diagonal matrix system using the algorithm provided with the KPP
submodel (Large et al., 1994). To mix momentum components, viscosity
profiles stored on interfaces at p grid points are horizontally interpolated to
interfaces at u and v grid points. Then the vertical diffusion equation is
solved on both sets of points. For more details on the implementation of the
various mixing schemes, the reader is referred to Halliwell (2004).
417
dense that its isopycnic reference density, the generator attempts to move the
bottom interface downward so that the flux of denser water across this
interface increases density. If the layer is denser than its isopycnic reference
density, the generator attempts to move the upper interface upward to
decrease density. In both cases, the generator first calculates the vertical
distance that the interface must be relocated so that volume-weighted density
of the original plus new water in the layer equals the reference density.
Repeated execution of this algorithm at every time step does maintain layer
density very close to its reference value as long as a minimum thickness
does not have to be maintained. To insure that a permanent p-coordinate
domain exists near the surface year round at all model grid points, the
uppermost layers are initialized with reference densities smaller than values
found anywhere in the model domain. The minimum thickness constraint is
not enforced at the bottom in the open ocean, permitting model layers to
collapse to zero thickness there as in MICOM.
The capability ofҏ assigning additional coordinate surfaces to the
HYCOM mixed layer allows the option of implementing sophisticated
vertical mixing turbulence closure schemes [see Halliwell (2004) for a
review]. The full set of vertical mixing options contained in the latest
version of HYCOM (http://hycom.rsmas.miami.edu) includes five primary
vertical mixing submodels, of which three are vertically “continuous”
models and two are predominantly or totally bulk models. The three
continuous models, which govern vertical mixing throughout the water
column, are: K-Profile Parameterization of Large et al. (1994) (KPP), the
level 2.5 turbulence closure of Mellor and Yamada (1982) (MY), and the
Goddard Institute for Space Studies (GISS) level 2 turbulence closure of
Canuto et al. (2001, 2002). The other two are the quasi-bulk dynamical
instability submodel of Price et al. (1986) (PWP) and the bulk Kraus-Turner
(1967) submodel (KT).
The following procedure is used to implement the three continuous
vertical mixing submodels. Velocity components are interpolated to the p
grid points from their native u and v points. The one-dimensional submodels
are then run at each p point to calculate profiles of viscosity coefficients
along with T and S diffusion coefficients on model interfaces. The onedimensional vertical diffusion equation is then solved at each p point to mix
T, S, and tracer variables, which involves the formulation and solution of a
tri-diagonal matrix system using the algorithm provided with the KPP
submodel (Large et al., 1994). To mix momentum components, viscosity
profiles stored on interfaces at p grid points are horizontally interpolated to
interfaces at u and v grid points. Then the vertical diffusion equation is
solved on both sets of points. For more details on the implementation of the
various mixing schemes, the reader is referred to Halliwell (2004).