[MITgcm-support] rdmnc.m

[David Wang]

Hey Ed,

Thanks a lot. rdnctiles.m works pretty well for me without failures. To 
be a little bit more greedy, I'm expecting more attributes provided, 
such as units, long_name, description for each variable (rdmnc.m does 
retrieve them from mnc output). It will be helpful when a global netcdf 
file needs to be written out (and that's actually what I'd like to do).
anyway, so far so good.

Thanks, D.W.

Ed Hill wrote:

>On Sat, 2005-11-26 at 22:38 -0500, David Wang wrote:
>  
>
>>I've got trouble loading multi-tile two-dimensional mnc diagnostics
>>(i.e., surface quantities like ETAN, TAUX) output into matlab. rdmnc.m
>>works just fine when I read in one single iteration, but failed when I
>>tried to load all the iterations available. What puzzles me is that
>>rdmnc.m works well for 3-dimensional diagnostics (THETA, UVEL etc) in
>>both cases. Did I get anything wrong? I'm attaching my
>>data.diagnostics and data.mnc namelists for reference. The code I use
>>is checkpoint57w_post. matlab is R14 (SP3) on linux FC4.
>>    
>>
>
>Hi David,
>
>I've also had trouble with rdmnc().  So much so that I decided it was
>hopeless mess and, rather than wasting time trying to fix it, wrote a
>much more capable replacement called "rdnctiles":
>
>  MITgcm/utils/matlab/gmt/rdnctiles.m
>
>which you can find in our CVS repository at:
>
>  http://mitgcm.org/cgi-bin/viewcvs.cgi/MITgcm/utils/matlab/gmt/
>
>So please give rdnctiles a try and see if it does a better job on your
>2D arrays.  If it doesn't, then please send an email and I'll iron out
>any wrinkles you find.
>
>Ed
>
>  
>


-- 
Daiwei (David) Wang    Graduate Student
Lamont-Doherty Earth Observatory of Columbia University
103B Oceanography
61 Route 9W, Palisades, NY 10964-8000
Voice: (845) 365-8694