Dear Jim

Is it a trajectory feature using the discrete sampling geometry convention
of section 9? In that case it is required that there should be a nominal
longitude, latitude and time for each observation along the trajectory.
However you could also supply data variables of longitude and latitude with
cell_methods along the same trajectory, and these data variables could be
named by the coordinates attribute of other data variables (serving as
auxiliary coordinate variables), to provide a more precise indication of
location, as in section 9.5.

Yes, the cell_methods would be "cell methods where there are no coordinates"
of section 7.3.4 since you don't have longitude and latitude dimensions. It is
permitted to specify "area:" for a cell method even if there are no horizontal
dimensions specified.

Cheers

Jonathan

----- Forwarded message from Jim Biard <[email protected]> -----

> Jonathan,
> 
> What I mean is, this is not gridded data.  The operations being performed 
> (mean latitude, etc) are over a geographic region that is not described 
> anywhere in the file, which is the region considered to be the extent of the 
> weather system.  So, you are saying that this trajectory case falls under the 
> ?Cell methods when there are no coordinates? case, is that right?
> 
> Jim
> 
> Visit us on
> Facebook      Jim Biard
> Research Scholar
> Cooperative Institute for Climate and Satellites NC
> North Carolina State University
> NOAA's National Climatic Data Center
> 151 Patton Ave, Asheville, NC 28801
> e: [email protected]
> o: +1 828 271 4900
> 
> 
> 
> 
> On Mar 4, 2014, at 9:30 AM, Jonathan Gregory <[email protected]> 
> wrote:
> 
> > Dear Jim
> > 
> >> Even though there are no cells?  That?s the part that I?m feeling confused 
> >> about.
> > 
> > Sorry, I'm not sure why you say there are no cells. "Cell" does not imply 
> > any
> > number of spatial dimensions - is that the problem? It means the portion of
> > the domain, of whatever dimensionality, which is represented by a particular
> > data value. The cell_methods attribute indicates how that data value 
> > represents
> > unrecorded variability of the quantity represented within the cell.
> > 
> > Best wishes
> > 
> > Jonathan
> > 
> >> On Mar 3, 2014, at 1:22 PM, Jonathan Gregory <[email protected]> 
> >> wrote:
> >> 
> >>> Dear Jim
> >>> 
> >>> Yes, I think such an approach is applicable. Any data variable can have
> >>> cell_methods and any data variable could be used as an aux coord variable.
> >>> Although unusual, I don't see a reason not to have data variables of 
> >>> latitude
> >>> and longitude, qualified by cell_methods, if required.
> >>> 
> >>> Best wishes
> >>> 
> >>> Jonathan
> >>> 
> >>> ----- Forwarded message from Jim Biard <[email protected]> -----
> >>> 
> >>>> Subject: Re: [CF-metadata] can a cell method be an attribute of an 
> >>>> auxiliary
> >>>>  coordinate variable ?
> >>>> From: Jim Biard <[email protected]>
> >>>> Date: Fri, 28 Feb 2014 14:01:24 -0500
> >>>> CC: [email protected]
> >>>> To: Jonathan Gregory <[email protected]>
> >>>> X-Mailer: Apple Mail (2.1874)
> >>>> 
> >>>> Jonathan,
> >>>> 
> >>>> So, I?ve got a broader question.  We have a case in hand with trajectory 
> >>>> data for a physically extensive weather system.  Two data variables are 
> >>>> the mean longitude and latitude of the system, which we are treating as 
> >>>> the auxiliary coordinate variables.  Other data variables contain 
> >>>> measurements such as maximum latitude at the mean longitude, etc.  Are 
> >>>> cell_methods applicable in this case?  There is no grid, so the cell 
> >>>> would be the area covered by the system, and the extents of this area 
> >>>> are not described.
> >>> _______________________________________________
> >>> CF-metadata mailing list
> >>> [email protected]
> >>> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
> > _______________________________________________
> > 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

Reply via email to