[MITgcm-support] sea ice diagnostics

[Menemenlis, Dimitris (329C)]

The “secret” codes are defined in
MITgcm/pkg/diagnostics/diagnostics_init_early.F
and diagnostics_main_init.F

As of checkpoint65y (2016/07/29), these codes are:
C     \begin{center}
C       \begin{tabular}[h]{|c|c|}\hline
C         \textbf{Positions}  &  \textbf{Characters}
C         &  \textbf{Meanings} \\\hline
C         parse(1)  &  S  &  scalar \\
C                   &  U  &  vector component in X direction \\
C                   &  V  &  vector component in Y direction \\
C                   &  W  &  vector component in vertical direction \\
C         parse(2)  &  U  &  C-grid U-Point  \\
C                   &  V  &  C-grid V-Point  \\
C                   &  M  &  C-grid Mass Point  \\
C                   &  Z  &  C-grid Corner Point  \\
C         parse(3)  &     &  Used for Level Integrated output: cumulate levels \\
C                   &  r  &  same but cumulate product by model level thickness \\
C                   &  R  &  same but cumulate product by hFac & level thickness \\
C         parse(4)  &  P  &  positive definite  \\
C         parse(5 ) &  C  &  with counter array  \\
C                   &  P  &  post-processed (not filled up) from other diags  \\
C                   &  D  &  disable an array for output  \\
C         parse(6--8) & '123'  &  retired, formerly: 3-digit mate number \\
C         parse(9)  &  U  &  model-level plus 1/2  \\
C                   &  M  &  model-level middle  \\
C                   &  L  &  model-level minus 1/2  \\
C         parse(10) &  0  &  levels = 0  \\
C                   &  1  &  levels = 1  \\
C                   &  R  &  levels = Nr  \\
C                   &  L  &  levels = MAX(Nr,NrPhys)  \\
C                   &  M  &  levels = MAX(Nr,NrPhys) - 1  \\
C                   &  G  &  levels = Ground_level Number \\
C                   &  I  &  levels = sea-Ice_level Number \\
C                   &  X  &  free levels option (need to be set explicitly) \\
C       \end{tabular}
C     \end{center}

> On Sep 13, 2016, at 10:27 PM, Martin Losch <Martin.Losch at awi.de> wrote:
> 
> Hi Gregory,
> 
> the diagnostics package prints an “available_diagnostics.log” into the run directory. That should have some “secret” code that tells you where individual variables are stored (given that that has been properly specified at the “init” step); somewhere (online-documentation) you can find what the secret code means. So expect some of the variables on C-points on U and V points. Usually the diagnostics are not averaged between grid points before output.
> 
> Martin
> 
>> On 13 Sep 2016, at 02:27, Georgy Manucharayan <gmanuch at caltech.edu> wrote:
>> 
>> Thanks Martin,
>> 
>> Yes, this all makes sense now! I have added appropriate sigmas into the code and hopefully now computing a proper stress divergence term. I am still a little uncertain about how to calculate the remaining forces in the momentum equations but I think those can be computed offline.
>> 
>> For an offline calculation of the ice advection terms, tilt, and ocean drag terms from the diagnostic output files do I need to consider that different variables are defined on different grids? or does the diagnostic package output all variables at the center of the grid boxes?
>> 
>> Georgy