[MITgcm-support] Basic GRDCHK Questions

[Timothy Andrew Smith]

Hey all,

I have a couple questions regarding the GRDCHK package. For context, I am
resurrecting the bottom_control_5x5 code to get sensitivities to bottom
topography. I am working on a small domain (sNx x sNy = 25 x 30 , nPx x nPy
= 1 x 1, nSx x nSy = 1 x 1). Here goes:

1. Is there a good method for understanding where i,j,k glopos lands in the
domain, or is this more guess work?

2. in STDOUT the two rows with printed "grad-res" information seem to have
conflicting i,j,k positions. See for example:

grad-res -------------------------------
 grad-res  proc    #    i    j    k   bi   bj iobc       fc ref
fc + eps           fc - eps
 grad-res  proc    #    i    j    k   bi   bj iobc      adj grad
fd grad          1 - fd/adj
grad-res -------------------------------
 grad-res     0    1    4   13    1    1    1    1  -7.79340478504E+02
-7.79340463216E+02 -7.79340493800E+02
 grad-res     0    1    1   10    0    1    1    1   1.52921513738E-03
1.52921389827E-03  8.10295748721E-07

why is i different between the two bottom rows? I'm similarly confused
about j, and k.

3. The above printed positions don't intuitively reflect what I put in my
data.grdchk file. If someone could explain what is in STDOUT vs what is
chosen in data.grdchk that would be great. My data.grdchk is shown below:

 &GRDCHK_NML
 grdchk_eps       = 1.d-2,
 iglopos         = 10,
 jglopos         = 10,
 kglopos         = 1,
 nbeg             = 10,
 nstep            = 1,
 nend             = 15,
 grdchkvarindex   = 20,
 &

4. Lastly, how do the nbeg, nstep, and nend variables work? Unfortunately,
none of these variables are described in the documentation or grdchk.h.

Let me know if I should include more information. Many thanks!

Cheers,
Tim

-----------------------------------------------------------------------------
Tim Smith
SIAM | Industry Liaison
PhD Student & Graduate Research Assistant
The Institute for Computational Engineering and Sciences
The University of Texas at Austin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20171115/9bea4348/attachment.html>