I believe that most of Jon's comments regarding the overlap between gridspec 
and ugrid were related to the topic of mosaics. This is a very important aspect 
in gridspec and since we will need to store mosaics of unstructured meshes as 
well I had included some initial ideas on this topic in the first ugrid docs. 
We didn't include this part in the official ugrid-0.9 version because of the 
remaining discussion and since we didn't need it yet at the time. The sgrid 
proposal (a staggered structure mesh convention alike ugrid) is a more direct 
alternative to gridspec's supergrid concept.

Good night from Europe,

Bert

Chris Barker <[email protected]>schreef:



On Thu, Dec 10, 2015 at 8:19 AM, Jonathan Gregory 
<[email protected]<mailto:[email protected]>> wrote:
After that proposal was agreed, we had some further discussions concerning the
relationship of gridspec and ugrid and how they could be made more similar.

I remember that conversation, though I never took the time to look at gridspec.

For reference, here is the current UGRID spec:

https://github.com/ugrid-conventions/ugrid-conventions/blob/master/ugrid-conventions.md

I just looked at:

http://www.gfdl.noaa.gov/~vb/gridstd/gridstdse3.html#x5-240003.4

And my first impression is that while it is handling the same information as 
UGRID, it's pretty different, at least with respect to variable names. My 
impression is thus:

gridspec was developed to support a certain set of (structured) grids. The 
unstructured bit was then added to well-match the model and conventions that 
had been established.

UGRID, n the other other hand was built from scratch, based on CF and the 
vocabulary and conventions that unstructured grid (oceanographic) modelers use. 
So it's a more wordy, but actually matched pretty well with what FVCOM, ADCIRC, 
SELFE, etc already put out.

So: I don't think that we in the UGRID community are going to want to turn what 
we'be got on end to match gridspec. I suspect the gridspec community is in the 
same position. Thugh: how much is gridspec used for unstructured grids? If not 
much then maybe just drop that support, and use UGRID.

HOWEVER: I haven't looked closely enough -- maybe the differences really are 
just vocabulary -- so maybe we could unify them if we allow multiple names for 
the same thing. i.e, in a triangular mesh grid, you specify either:


var_name:cf_role = "face_node_connectivity" ;

or

var_name:standard_name = "neighbor_cell_index"

though, honestly, gridspec doesn't look very CF-y to me anyway :-)


-CHB



--

Christopher Barker, Ph.D.
Oceanographer

Emergency Response Division
NOAA/NOS/OR&R            (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

[email protected]<mailto:[email protected]>
DISCLAIMER: This message is intended exclusively for the addressee(s) and may 
contain confidential and privileged information. If you are not the intended 
recipient please notify the sender immediately and destroy this message. 
Unauthorized use, disclosure or copying of this message is strictly prohibited. 
The foundation 'Stichting Deltares', which has its seat at Delft, The 
Netherlands, Commercial Registration Number 41146461, is not liable in any way 
whatsoever for consequences and/or damages resulting from the improper, 
incomplete and untimely dispatch, receipt and/or content of this e-mail.
_______________________________________________
CF-metadata mailing list
[email protected]
http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata

Reply via email to