[MITgcm-support] MITgcm online tutorial problems

Fri Jun 7 11:49:05 EDT 2013

Jonny, if you will be using high_res_cube or quarter_degree_global and things are missing,
unclear, or not working, please complain to me and Hong and we will try to fix.

Dimitris Menemenlis

On Jun 7, 2013, at 11:00 AM, Jonny Williams wrote:

Thanks Martin

that suggested checkout seems to have worked so I'll give it a go!

Jonny

On 7 June 2013 15:55, Martin Losch <Martin.Losch at awi.de<mailto:Martin.Losch at awi.de>> wrote:
I wouldn't check-out the entire contrib-area. I usually have a look here first:

http://mitgcm.org/viewvc/MITgcm/MITgcm_contrib/

to see, what's available and then do something like
cvs co MITgcm_contrib/high_res_cube
(works for me)

Martin

On Jun 7, 2013, at 4:13 PM, Jonny Williams <Jonny.Williams at bristol.ac.uk<mailto:Jonny.Williams at bristol.ac.uk>> wrote:

> Hi Martin
>
> Thank you for that. I am running testreport now and it seems to be working OK.
>
> I have checked out MITcgm_contrib using "cvs co -P MITgcm_contrib" but it does not contain high_res_cube or quarter_degree_global. Am I checking out the wrong version?
>
> Many thanks for you help again!
>
> Jonny
>
>
>
> On 7 June 2013 10:58, Martin Losch <Martin.Losch at awi.de<mailto:Martin.Losch at awi.de>> wrote:
> Hi Jonny,
>
> many of the verification experiments use common files that need to be linked into the run-directory (this is done to reduce the size of the download). Consequently, they fail, if you haven't done that. The easiest way of linking these files is to run "testreport" on them:
> cd MITgcm/verification
> ./testreport
> will compile and run all verification experiments
> ./testreport -t 'exp2 tutorial_global_oce_latlon'
> will compile and run only the specified experiments (exp2 and tutorial_global_oce_latlon in this case).
>
> Did the output files appear as they should?
>
> MITgcm_contrib/high_res_cube
> MITgcm_contrib/quarter_degree_global
> use fairly big grids, but you'll have to get the initialization/forcing data (or create it yourself), I think you can search the email archive for downloading instructions for the forcing data.
>
> Martin
>
> On Jun 7, 2013, at 11:33 AM, Jonny Williams <Jonny.Williams at bristol.ac.uk<mailto:Jonny.Williams at bristol.ac.uk>> wrote:
>
> > Hi again
> >
> > As you suggested, running tutorial_global_oce_latlon was successful although most of the other tutorials failed with various errors concerning missing files.
> >
> > Fundamentally I want to try running a high resolution (approx 1/3 degree or even higher) global or regional model to get an idea of run times etc so if anyone could point me in the direction of something like that then that would be much appreciated.
> >
> > Apparently there used to be some examples in the MITgcm_contrib CVS repository but they are not longer there...
> >
> > Many thanks
> >
> > Jonny
> >
> >
> > On 6 June 2013 10:38, Martin Losch <Martin.Losch at awi.de<mailto:Martin.Losch at awi.de>> wrote:
> > Jonny,
> >
> > as far as I can see exp2 does not use netcdf (there is no useMNC=.true., in input/data.pkg nor any "mnc"-line in code/packages.conf). I think it makes more sense to start with the actual "tutorials" (I know, exp2 unfortunately is the first example in section 3 of the manual). These "tutorial experiments" start with "tutorial_", e.g. "tutorial_global_oce_latlon" is similar to exp2. It's also described in the manual <http://mitgcm.org/public/r2_manual/latest/online_documents/node129.html>
> >
> > The output files should all be in the "working directory" where you run the model (i.e. the mitgcmuv-executable writes to "./" by default)
> >
> > Martin
> > On Jun 4, 2013, at 4:16 PM, Jonny Williams <Jonny.Williams at bristol.ac.uk<mailto:Jonny.Williams at bristol.ac.uk>> wrote:
> >
> > > Hi there
> > >
> > > I am brand new to running the MITgcm but am an experienced climate modeller.
> > >
> > > I am trying to run the MITgcm online tutorials, specifically the experiment called "exp2" in the verification folder in the MITgcm directory tree as successfully downloaded using CVS.
> > >
> > > I am able to compile the job OK (I think) but when I come to run the job and examine any output I get nothing at all, although my "output.txt" (as described here) seems to be pretty much the same as what is expected from the example in the results/output.txt file.
> > >
> > > I believe that I have my NetCDF libraries set up correctly.
> > >
> > > On this webpage, various output streams are detailed but none of them are being produced. Notably, there is no information given as to where to expect these results to be put. In addition to this on the same webpage, the files code/packages.conf and input/data.pkg files are discussed but no information is given as to how to use them.
> > >
> > > Is it possible to give me any further tips on how to proceed?
> > >
> > > Many thanks for your help.
> > >
> > > Jonny Williams
> > >
> > > --
> > > Dr Jonny Williams
> > > School of Geographical Sciences
> > > University of Bristol
> > > University Road
> > > BS8 1SS
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20130607/aa2d2a37/attachment.htm>