/[MITgcm]/manual/s_phys_pkgs/text/gridalt.tex

Diff of /manual/s_phys_pkgs/text/gridalt.tex

Parent Directory | Revision Log | View Revision Graph Revision Graph | View Patch Patch

-revision 1.6 by molod,
Tue Aug  2 21:35:40 2005 UTC
+revision 1.8 by edhill,
Wed Apr  5 02:27:33 2006 UTC
 Line 6
  \subsubsection {Introduction}
- The gridalt package is designed to allow different components of the MITgcm to
+ The gridalt package is designed to allow different components of MITgcm to
  be run using horizontal and/or vertical grids which are different from the main
  model grid. The gridalt routines handle the definition of the all the various
  alternative grid(s) and the mappings between them and the MITgcm grid.
 Line 44 
 the mapping between the two vertical coo
  \vspace*{-0.4in}
  \begin{center}
  \includegraphics[height=2.4in]{part6/vertical.eps}
- \caption{Vertical discretization for the MITgcm (dark grey lines) and for the
+ \caption{Vertical discretization for MITgcm (dark grey lines) and for the
  atmospheric physics (light grey lines). In this implementation, all MITgcm level
  interfaces must coincide with atmospheric physics level interfaces.}
  \end{center}
 Line 116 
 Where $\gamma=1$ here.
  In order to minimize the correction terms for the state variables on the alternative,
  higher resolution grid, the vertical interpolation scheme must be constructed so that
  a dynamics-to-physics interpolation can be exactly reversed with a physics-to-dynamics mapping.
- The simple scheme employed to achieve this is:
+ The simple scheme employed to achieve this is:\\
  Coarse to fine:\
  For all physics layers l in dynamics layer L, $ T_{phys}(l) = \{T_{dyn}(L)\} = T_{dyn}(L) $.
  Fine to coarse:\
- For all physics layers l in dynamics layer L, $T_{dyn}(L) = [T_{phys}(l)] = \int{T_{phys} dp } $.
+ For all physics layers l in dynamics layer L, $T_{dyn}(L) = [T_{phys}(l)] = \int{T_{phys} dp } $.\\
  Where $\{\}$ is defined as the dynamics-to-physics operator and $[ ]$ is the physics-to-dynamics operator, $T$ stands for any state variable, and the subscripts $phys$ and $dyn$ stand for variables on
  the physics and dynamics grids, respectively.
  \subsubsection {Key subroutines, parameters and files }
+ \noindent
+ One of the central elements of the gridalt package is the routine which
+ is called from subroutine gridalt\_initialise to define the grid to be
+ used for the high end physics calculations. Routine make\_phys\_grid
+ passes back the parameters which define the grid, ultimately stored
+ in the common block gridalt\_mapping.
+ \begin{verbatim}
+        subroutine make_phys_grid(drF,hfacC,im1,im2,jm1,jm2,Nr,
+      . Nsx,Nsy,i1,i2,j1,j2,bi,bj,Nrphys,Lbot,dpphys,numlevphys,nlperdyn)
+ c***********************************************************************
+ c Purpose: Define the grid that the will be used to run the high-end
+ c          atmospheric physics.
+ c
+ c Algorithm: Fit additional levels of some (~) known thickness in
+ c          between existing levels of the grid used for the dynamics
+ c
+ c Need:    Information about the dynamics grid vertical spacing
+ c
+ c Input:   drF         - delta r (p*) edge-to-edge
+ c          hfacC       - fraction of grid box above topography
+ c          im1, im2    - beginning and ending i - dimensions
+ c          jm1, jm2    - beginning and ending j - dimensions
+ c          Nr          - number of levels in dynamics grid
+ c          Nsx,Nsy     - number of processes in x and y direction
+ c          i1, i2      - beginning and ending i - index to fill
+ c          j1, j2      - beginning and ending j - index to fill
+ c          bi, bj      - x-dir and y-dir index of process
+ c          Nrphys      - number of levels in physics grid
+ c
+ c Output:  dpphys      - delta r (p*) edge-to-edge of physics grid
+ c          numlevphys  - number of levels used in the physics
+ c          nlperdyn    - physics level number atop each dynamics layer
+ c
+ c NOTES: 1) Pressure levs are built up from bottom, using p0, ps and dp:
+ c              p(i,j,k)=p(i,j,k-1) + dp(k)*ps(i,j)/p0(i,j)
+ c        2) Output dp's are aligned to fit EXACTLY between existing
+ c           levels of the dynamics vertical grid
+ c        3) IMPORTANT! This routine assumes the levels are numbered
+ c           from the bottom up, ie, level 1 is the surface.
+ c           IT WILL NOT WORK OTHERWISE!!!
+ c        4) This routine does NOT work for surface pressures less
+ c           (ie, above in the atmosphere) than about 350 mb
+ c***********************************************************************
+ \end{verbatim}
+ \noindent In the case of the grid used to compute the atmospheric physical
+ forcing (fizhi package), the locations of the grid points move in time with
+ the MITgcm $p^*$ coordinate, and subroutine gridalt\_update is called during
+ the run to update the locations of the grid points:
+ \begin{verbatim}
+        subroutine gridalt_update(myThid)
+ c***********************************************************************
+ c Purpose: Update the pressure thicknesses of the layers of the
+ c          alternative vertical grid (used now for atmospheric physics).
+ c
+ c Calculate: dpphys    - new delta r (p*) edge-to-edge of physics grid
+ c                        using dpphys0 (initial value) and rstarfacC
+ c***********************************************************************
+ \end{verbatim}
+ \noindent The gridalt package also supplies utility routines which perform
+ the mappings from one grid to the other. These routines are called from the
+ code which computes the fields on the alternative (fizhi) grid.
+ \begin{verbatim}
+       subroutine dyn2phys(qdyn,pedyn,im1,im2,jm1,jm2,lmdyn,Nsx,Nsy,
+      . idim1,idim2,jdim1,jdim2,bi,bj,windphy,pephy,Lbot,lmphy,nlperdyn,
+      . flg,qphy)
+ C***********************************************************************
+ C Purpose:
+ C   To interpolate an arbitrary quantity from the 'dynamics' eta (pstar)
+ C               grid to the higher resolution physics grid
+ C Algorithm:
+ C   Routine works one layer (edge to edge pressure) at a time.
+ C   Dynamics -> Physics retains the dynamics layer mean value,
+ C   weights the field either with the profile of the physics grid
+ C   wind speed (for U and V fields), or uniformly (T and Q)
+ C
+ C Input:
+ C   qdyn..... [im,jm,lmdyn] Arbitrary Quantity on Input Grid
+ C   pedyn.... [im,jm,lmdyn+1] Pressures at bottom edges of input levels
+ C   im1,2 ... Limits for Longitude Dimension of Input
+ C   jm1,2 ... Limits for Latitude  Dimension of Input
+ C   lmdyn.... Vertical  Dimension of Input
+ C   Nsx...... Number of processes in x-direction
+ C   Nsy...... Number of processes in y-direction
+ C   idim1,2.. Beginning and ending i-values to calculate
+ C   jdim1,2.. Beginning and ending j-values to calculate
+ C   bi....... Index of process number in x-direction
+ C   bj....... Index of process number in x-direction
+ C   windphy.. [im,jm,lmphy] Magnitude of the wind on the output levels
+ C   pephy.... [im,jm,lmphy+1] Pressures at bottom edges of output levels
+ C   lmphy.... Vertical  Dimension of Output
+ C   nlperdyn. [im,jm,lmdyn] Highest Physics level in each dynamics level
+ C   flg...... Flag to indicate field type (0 for T or Q, 1 for U or V)
+ C
+ C Output:
+ C   qphy..... [im,jm,lmphy] Quantity at output grid (physics grid)
+ C
+ C Notes:
+ C   1) This algorithm assumes that the output (physics) grid levels
+ C      fit exactly into the input (dynamics) grid levels
+ C***********************************************************************
+ \end{verbatim}
+ \noindent And similarly, gridalt contains subroutine phys2dyn.
  \subsubsection {Dos and donts}
  \subsubsection {Gridalt Reference}

 Legend:



Removed from v.1.6
 


changed lines


 
Added in v.1.8
 Legend:



Removed from v.1.6
 


changed lines


 
Added in v.1.8
-Removed from v.1.6
+Added in v.1.8

	ViewVC Help
Powered by ViewVC 1.1.22