[MITgcm-support] defining my own xx_gentim2d_weight file

[Daniel Goldberg]

Hi Gael

Many thanks. I will try this again when I am able. It seems as though your
input files were 4-byte instead of 8, even though the
MDSREADFIELD_3D_GL command
at line 176 of ctrl_set_pack_xyz.F specifies RL precision? This is the only
difference I can see with what I have been trying, but I will attempt again
with 4-byte precision input files.

Thanks again
Dan

On Thu, Nov 27, 2014 at 1:09 PM, gael forget <gforget at mit.edu> wrote:

> Hi Dan,
> there is an example @
>
> http://mitgcm.org/viewvc/MITgcm/MITgcm_contrib/gael/verification/global_oce_cs32/results/output_adm.txt?view=markup
> <http://mitgcm.org/viewvc/MITgcm/MITgcm_contrib/gael/verification/global_oce_cs32/results/output_adm.txt>
> See the ‘data.ctrl' and 'control vector configuration’ portion.
> Cheers,
> Gael
>
> On Nov 27, 2014, at 5:41 AM, Daniel Goldberg <dngoldberg at gmail.com> wrote:
>
> Hi Gael
>
> I have asked about the xx_gentim2d_weight and xx_genarr2d_weight files
> before, but that was because I wanted not to define them. Now, however, I
> would like to actually use them to weight the gradient data when packing
> for M1QN3.
>
> I might have misunderstood their purpose before. My understanding is that
> if I use the genarr2d control with size=2, then i would define in data.ctrl:
>
> xx_genarr2d_weight='weightfile1','weightfile2'
>
> and then in the run directory i would provide binary files weightfile1.bin
> and weightfile2.bin. These binary files would be the size of the grid, and
> would contain the weight I want to attach to the corresponding gradient
> entry.
>
> however, while these files are read, I am finding this is not the case --
> and I'm not sure if I understand the point of the xx_genarr2d_weight
> parameter array -- or else I'm not sure I am constructing the binary files
> correctly.
>
> for instance, say my grid is Nx=100,Ny=100. For genarr2d control #1, I
> would like to weight the gradient by 2x for i.le.50, and by 0.5x for
> i.gt.50. My understanding is that I would create a file via
>
> q=zeros(100,100); q(1:50,:) = 2; q(51:100,:)=0.5;
> fid=fopen('weightfile1.bin','w','b'); fwrite(fid,q,'real*8'); fclose(fid);
>
> and then place weightfile1.bin in the run directory. However, by
> inspection of ctrl_set_xy_pack.F, I can see that the weightFld2D array does
> not populate with the values I would expect, and I am not sure why.
>
> Is there an example in verification/ of a weighttype being used, with an
> nonconstant weight field being set via file?
>
> Many thanks
> Dan
>
> --
>
> Daniel Goldberg, PhD
> Lecturer in Glaciology
> School of Geosciences, University of Edinburgh
> Geography Building, Drummond Street, Edinburgh EH8 9XP
>
>
> em: D <dgoldber at mit.edu>an.Goldberg at ed.ac.uk
> web: http://ocean.mit.edu/~dgoldberg
>  _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
>
>
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
>
>


-- 

Daniel Goldberg, PhD
Lecturer in Glaciology
School of Geosciences, University of Edinburgh
Geography Building, Drummond Street, Edinburgh EH8 9XP


em: D <dgoldber at mit.edu>an.Goldberg at ed.ac.uk
web: http://ocean.mit.edu/~dgoldberg
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20141127/00ac41c7/attachment.htm>