PUBLIC INTERFACE ~ PUBLIC ROUTINES ~ NAMELIST

Module ocean_sigma_transport_mod

Contact:  Stephen M. Griffies
Reviewers: 
Change History: WebCVS Log

OVERVIEW

Thickness weighted and density weighted time tendency for tracer from transport within a bottom "sigma" layer. The advective portion of this routine is experimental, and has many problems. It is retained in MOM for exploratory use only. Also note that the advection contributes a lot of instability when running realistic simulations with pressure vertical coordinates. The instability mechanism is unknown.

This module computes the thickness weighted and density weighted time tendency for tracer arising from 1. Laplacian diffusion within a bottom turbulent boundary layer. 2. Upwind advection within this layer. Advection velocities determined by model resolved velocity and parameterized downslope velocity. We use first order upwind tracer advection to ensure positive definite tracer transport in the sigma layer. As the sigma layer is a proxy for a bottom turbulent boundary layer, the added mixing from the first order upwind should be physically acceptable. CAUTION: The advective portion of this algorithm has problems and it retained in MOM only for research purposes. It is NOT supported for general use. The diffusivity used to determine the strength of the diffusion is generally set to be a function of the local horizontal grid spacing. Diffusivity is the sum of an a priori background plus a velocity dependent diffusivity. It is large if there is a a heavier parcel living adjacent within the "sigma layer" above a lighter parcel. It is small otherwise. The advection is set to zero if the density is not downslope favorable. That is, rho_{,x} * H_{,x} < 0 for downslope flow in the x-direction, and likewise in the y-direction. The thickness of the bottom layer can span more than a single bottom grid cell. This feature allows the sigma layer thickness to undulate in time according to the convergence or divergence of mass within the sigma layer.

OTHER MODULES USED

        constants_mod
     diag_manager_mod
              fms_mod
           fms_io_mod
      mpp_domains_mod
              mpp_mod
    ocean_domains_mod
    ocean_density_mod
  ocean_operators_mod
 ocean_parameters_mod
ocean_tracer_util_mod
      ocean_types_mod
       ocean_util_mod
  ocean_workspace_mod

PUBLIC INTERFACE

ocean_sigma_transport_init:
sigma_transport:
ocean_sigma_transport_restart:
ocean_sigma_transport_end:
advect_sigma_upwind:
watermass_diag_init:
watermass_diag:

PUBLIC ROUTINES

  1. ocean_sigma_transport_init

    DESCRIPTION
    Initialize the sigma transport module by registering fields for diagnostic output and performing some numerical checks to see that namelist settings are appropriate.


  2. sigma_transport

    DESCRIPTION
    This subroutine computes the thickness weighted and density weighted time tendency for tracer arising from transport in a bottom turbulent boundary layer. The result is stored in tracer th_tendency. NOTE: In this algorithm, we ideally wish to have advection velocity components on full data domain. Unfortunately, from ocean_advection_velocity_mod, they are only known on the following domains: Adv_vel%uhrho_et: (isd,ied) x (jsc,jed) Adv_vel%vhrho_nt: (isc,ied) x (jsd,jed). So to proceed with the sigma_transport algorithm, we transfer into local arrays and then update. These updates may be avoided (possibly), but at the price of much more logic in the algorithm. We choose to instead do the updates and have less logic. This decision may need to be revisited. CAUTION: The advective portion of this algorithm has fundamental problems. It is retained in MOM only for process physics research purposes. It is NOT recommended for use in general use.


  3. ocean_sigma_transport_restart

    DESCRIPTION
    Write out restart files registered through register_restart_file


  4. ocean_sigma_transport_end

    DESCRIPTION
    Write to restart.


  5. advect_sigma_upwind

    DESCRIPTION
    First order upwind to advect tracers in sigma layer.


  6. watermass_diag_init

    DESCRIPTION
    Initialization of watermass diagnostic output files.


  7. watermass_diag

    DESCRIPTION
    Diagnose effects from sigma transport on the watermass transformation.


NAMELIST

&ocean_sigma_transport_nml

use_this_module
Must be true to use this module. Default is false.
[logical]
debug_this_module
For debugging.
[logical]
sigma_diffusion_on
For using sigma diffusion. Default is true.
[logical]
sigma_advection_on
For using sigma advection. Default is false.
[logical]
sigma_advection_sgs_only
In many cases, adding the resolved transport to the sigma-advective transport produces a tremendous level of noise at the bottom. The problem is that there are grid-scale features that may cause large jumps in whether the velocity should be added or not, depending on the logic of the scheme. For this reason, it may be prudent to remove the resolved velocity from that contributing to the sigma transport scheme. Note that its removal from sigma transport does not remove the contributions of the resolved velocity from the resolved advective transport arising from ocean_tracer_advect_mod. It simply removes it from the added transport arising in the sigma transport module. Default is sigma_advection_sgs_only=.true.
[logical]
sigma_advection_check
If true, then will only include the resolved advection velocity in the sigma-layer if the direction of transport is downslope favorable for enhancing deep density. IF false, then will include the velocity regardless. This option aims to reduce the large divergences that occur for the case when only include the velocity if it is favorable for deep water getting more dense. Default is sigma_advection_check=.true.
[logical]
thickness_sigma_layer
Initial thickness of the bottom sigma layer.
[real, units: meter]
thickness_sigma_min
Minimum thickness of the bottom sigma layer.
[real, units: meter]
thickness_sigma_max
Maximum thickness of the bottom sigma layer.
[real, units: meter]
sigma_just_in_bottom_cell
For just having sigma layer in the bottom cell, as in mom4p0. This option must be .false. in order to use sigma_advection_on=.true. Default sigma_just_in_bottom_cell=.true.
[logical]
tmask_sigma_on
IF .true. then masks out fluxes passing into the sigma layer, except those associated with sigma transport. Typically set to .false.
[logical]
sigma_diffusivity
Sigma tracer diffusivity for use if not using micom diffusivity.
[real, units: m^2/sec]
sigma_diffusivity_ratio
When flow along sigma surface is stable (i.e., heavy parcels are below lighter parcels) then sigma diffusivity is reduced by sigma_diffusivity_ratio from the case where heavy parcels are above lighter parcels.
[real, units: dimensionless]
tracer_mix_micom
If .true., then the sigma diffusivity is set according to a velocity scale times the grid spacing.
[logical]
vel_micom
Velocity scale that is used for computing the MICOM diffusivity.
[real, units: m/sec]
campingoose_mu
Dissipation rate for the bottom friction. Campin and Goosse suggest campingoose_mu=10^-4
[real, units: inverse seconds]
campingoose_delta
Fraction of a grid cell participating in the overflow process. Campin and Goosse suggest campingoose_delta=1/3.
[real, units: dimensionless]
sigma_umax
Maximum downslope speed allowed in sigma layer. In some cases, the model will be unstable if sigma_umax is too large.
[real, units: m/s]
smooth_sigma_velocity
To smooth the sigma advective transport velocity. Default is smooth_sigma_velocity=.true.
[logical]
smooth_sigma_thickness
To smooth the sigma thickness. This may be needed especially for case with sigma advection, in which case the thickness can become noisy. Default is smooth_sigma_thickness=.true.
[logical]
sigma_velmicom
For smoothing the sigma_thickness, use this as velocity scale to determine the thickness diffusivity. Default is smooth_velmicom = 0.2
[real, units: m/s]

REFERENCES

  1. A. Beckmann and R. Doscher, 1997: A method for improved representation of dense water spreading over topography in geopotential--coordinate models Journal of Physical Oceanography, vol 27, pages 581--59.
  2. R. Doscher and A. Beckmann, 2000: Effects of a bottom boundary layer parameterization in a coarse-resolution model of the North Atlantic Ocean Journal of Atmospheric and Oceanic Technology, vol 17 pages 698--707
  3. Campin and Goosse 1999: Parameterization of density-driven downsloping flow for a coarse-resolution model in z-coordinate", Tellus 51A, pages 412-430.
  4. S.M. Griffies: Elements of MOM (2012)

top