Module ocean_lapgen_friction_mod

OVERVIEW

This module computes the thickness weighted time tendency for horizontal velocity arising from horizontal Laplacian friction.

OTHER MODULES USED

       constants_mod
    diag_manager_mod
             fms_mod
     mpp_domains_mod
             mpp_mod
   ocean_domains_mod
       ocean_obc_mod
 ocean_operators_mod
ocean_parameters_mod
     ocean_types_mod
      ocean_util_mod
 ocean_workspace_mod

PUBLIC INTERFACE

PUBLIC ROUTINES

1. ocean_lapgen_friction_init

   DESCRIPTION

   Initialize the horizontal Laplacian friction module by registering fields for diagnostic output and performing some numerical checks to see that viscosity is set appropriately.

   
   
   

2. lapgen_friction

   DESCRIPTION

   This routine computes thickness weighted and density weighted time tendency for horizontal velocity arising from horizontal Laplacian friction. The algorithm is derived from a functional approach that ensures kinetic energy is consistenty dissipated for all flow configurations. The triad do-loops are expanded in order to enhance the ability of cache-based machines to keep most of the variables on-cache. Fundamental to the scheme are the rates of horizontal deformation
   horizontal tension = DT = (dy)(u/dy)_x - (dx)(v/dx)_y
   horizontal strain = DS = (dx)(u/dx)_y + (dy)(v/dy)_x
   Units of the tension and strain are sec^-1. Four tensions and four strains are computed for each velocity point,
   corresponding to the four triads surrounding the point.
   The following notation is used to distinguish the triads:
   (0,1)=northwest triad (1,1)=northeast triad,
   (0,0)=southwest triad, (1,0)=southeast triad A triad contributes when at least one of its velocities is not a land point. In order to obtain the correct tension and strain next to boundaries, tension and strain should not be masked with umask.

   
   
   

3. BDX_EU_smag

   DESCRIPTION

   Compute backwards derivative in X of a quantity defined on the east face of a U-cell. Slightly modified version of BDX_EU used in ocean_operators.F90. If input is a(i,j) then output is defined at (i-1/2,j). BDX_EU_smag(a) has dimensions of a*m^-3

   
   
   

   INPUT

   a	field defined on the east face of a U-cell    [real, dimension(isd:ied,jsd:jed)]

   
   

4. BDY_NU_smag

   DESCRIPTION

   Compute backwards derivative in Y of a quantity defined on the north face of a U-cell. Slightly modified version of BDY_EU used in ocean_operators.F90. If input is a(i,j) then output is defined at (i,j-1/2) BDY_NU_smag(a) has dimensions of a*m^-3

   
   
   

   INPUT

   a	field defined on the north face of a U-cell    [real, dimension(isd:ied,jsd:jed)]

   
   

5. anisotropic_ncar

   DESCRIPTION

   Spatially-varying anisotropic viscosity initialization This routine defines NCOM-like spatial distributions of viscosity coefficients F_PARA and F_PERP. Uses NCAR CCSM2.0 algorithm with cm^2/sec --> m^2/sec. written by: Stephen Yeager 3/2000
   modified by: Gokhan Danabasoglu (08/2001)
   port to mom4: Stephen.Griffies (9/2002) update to mom4p1 based on new tunes from NCAR Stephen.Griffies (7/2007) "A_viscosity" = F_PARA = Along = viscosity parallel to flow = max{0.5*visc_vel_scale(z)*A*max[dx,dy],vconst_6} where
   A = 0.425 * cos(pi*y*radian/30) + 0.575 for |y*radian| < 30
   A = 0.15 otherwise Here, A provides a horizontal variation for visc_vel_scale. "B_viscosity" = F_PERP = Across = viscosity perpendicular to flow = max( bu, bv) and
   F_PARA = min(F_PARA, AMAX_CFL),
   F_PERP = min(F_PERP, AMAX_CFL),
   F_PARA = max(F_PARA, F_PERP)
   are enforced In the above equations, bu = vconst_1 * ( 1 + vconst_2 * ( 1 + cos( 2*y + pi ) ) )
   bv = vconst_3 * beta_f * dx^3 * exp( - (vconst_4 * distance)^2 )
   with
   beta_f (x,y) = 2 * omega_earth* cos(ULAT(i,j)) / radius
   distance (x,y,z) = actual distance to "vconst_5" points
   west of the nearest western boundary
   dx (x,y) = DXU(i,j)
   dy (x,y) = DYU(i,j)
   visc_vel_scale (z) = vconst_7 * exp(-zt(k)/visc_vel_scale_length)
   visc_vel_scale_length = e-folding scale ( default = 1500.0e2 cm)
   y (x,y) = ULAT(i,j), latitude of "u/v" grid pts in radians
   In MOM, ULAT(radians) = xu*pi/180 with xu(i,j) the longitude of U grid points in degrees "vconst_#" are input parameters defined in namelist ocean_lapgen_friction_general_nml. "vconst_1", "vconst_6", and "vconst_4" have dimensions of cm^2/s, cm^2/s, and 1/cm, respectively. "vconst_5" is an INTEGER. NOTE: The nearest western boundary computations are done along the model i-grid lines. Therefore, viscosity based on these are only approximate in the high Northern Hemisphere when using generalized coordinates with coordinate pole(s) shifted onto land.

   
   
   

6. lapgen_viscosity_check

   DESCRIPTION

   Subroutine to perform linear stability check for the Laplacian operator given a value for the horizontal biharmonic viscosity.

   
   
   

7. lapgen_reynolds_check

   DESCRIPTION

   Subroutine to compute the LLaplacian grid Reynolds number. Large Reynolds numbers indicate regions where solution may experience some grid noise due to lack of enough horizontal friction.

   
   
   

   INPUT

   u	Horizontal velocity field at time tau    [real, dimension(isd:ied,jsd:jed,nk,2)]

   
   

8. compute_neptune_velocity

   DESCRIPTION

   Compute Neptune velocity. Method follows that used in MOM2 and MOM3 as implemented by Greg Holloway (zounds@ios.bc.ca) and Michael Eby (eby@uvic.ca) Coded in mom4 by Stephen.Griffies Neptune is calculated as an equilibrium streamfunction given by pnep = -f*snep*snep*ht and is applied through friction whereby the solution is damped towards the equilibrium streamfunction rather than being damped towards zero kinetic energy. ht = depth of tracer cells snep = spnep + (senep-spnep)*(0.5 + 0.5*cos(2.0*latitude)) Neptune length scale snep has a value of senep at the equator and smoothly changes to spnep at the poles Reference: Holloway, G., 1992: Representing topographic stress for large scale ocean models, J. Phys. Oceanogr., 22, 1033-1046 Eby and Holloway, 1994: Sensitivity of a large scale ocean model to a parameterization of topographic stress. JPO, vol. 24, pages 2577-2588 March 2012 Stephen.Griffies Algorithm updated to Eby and Holloway (1994)

   
   
   

NAMELIST

REFERENCES

NOTES