Dear Martin We added the regions to CF a long time ago and they are in use, for example, in CMIP data. As Alison says, CF section 6.1.1 describes them. One reason to need them is that they are *not* precise. Other definitions might relate to the real-world geography precisely, but we need region names that can be applied to models and the real world with various geography. Like with standard names, quantities with the same CF region name are regarded as comparable for the purpose. For many such purposes, precision is not needed and would be an obstacle. We use region names, rather than codes, in order to make them self- describing, as usual with CF metadata. We did not simply point to GCMD because it includes other entities that aren't really geographical regions and which we didn't want to include, and the format of its names wasn't CF-like.
Best wishes Jonathan ----- Forwarded message from "Schultz, Martin" <[email protected]> ----- > Date: Fri, 24 Apr 2015 08:02:25 +0000 > From: "Schultz, Martin" <[email protected]> > To: "[email protected]" <[email protected]> > Subject: Re: [CF-metadata] CF region_names > > Dear all, > > what is this thing about region_names? I only now saw this popping up > and must say that I strongly object to adopt these as a standard. Region > definitions are so manifold and depend very much on individual applications > that I don't see any value in adopting one set of names (here GCMD) and call > them "CF region names". In my view this runs completely against what I > conceive as the CF philosophy which is trying to be broad and encompass many > application areas. > > Not only are the names of the regions under dispute, but even more so any > definition of their boundaries. For example: for a lake area such as > "caspian_sea", do you follow the lake boundaries (and if so in which year?) > or do you mean the catchment area including all rivers running into it, or > are you referring to a lon/lat rectangle? > > I have no problem with a concept that is called "gcmd_regions", for > example, if it is detailed somewhere where these definitions come from and > how the boundaries are defined. Note that there are indeed various other > "prominent" region definitions, for example 6(+1) world regions by WMO, see > https://www.wmo.int/pages/members/index_en.html, or the WMO code table 0161 > specifying oceanic areas. But even then, I don't think that this belongs in > the CF domain, but instead should be dealt with elsewhere. What CF can do in > my opinion is to define suitable attributes how a region (code) and its > underlying definition can be provided. > > Best regards, > > Martin > > > > ------------------------------------------------------------------------------------------------ > ------------------------------------------------------------------------------------------------ > Forschungszentrum Juelich GmbH > 52425 Juelich > Sitz der Gesellschaft: Juelich > Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498 > Vorsitzender des Aufsichtsrats: MinDir Dr. Karl Eugen Huthmacher > Geschaeftsfuehrung: Prof. Dr.-Ing. Wolfgang Marquardt (Vorsitzender), > Karsten Beneke (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt, > Prof. Dr. Sebastian M. Schmidt > ------------------------------------------------------------------------------------------------ > ------------------------------------------------------------------------------------------------ > > _______________________________________________ > CF-metadata mailing list > [email protected] > http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata ----- End forwarded message ----- _______________________________________________ CF-metadata mailing list [email protected] http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
