Dear David

Thanks for noticing the incorrect standard_name in Example 7.3. It looks like
a defect to me too.

When Sect 7.2 was written, as far as I remember we had only horizontal area
in mind. As you say, the convention doesn't exclude areas which are functions
of non-horizontal axes, but that probably wasn't intended. So I think the
answer is that the situation is ambiguous, and it should be clarified by
explicitly allowing or disallowing non-horizontal areas as an "area" cell
measure. If they are not allowed, but you have use-case where they are
needed, a differently named measure should be proposed. I think that these
would be substantive changes to the convention either way, not defects.

The same applies to the standard_name. Either cell_area could have its
definition generalised, or a new standard_name could be added.

Best wishes

Jonathan

----- Forwarded message from David Hassell <[email protected]> -----

> Date: Wed, 7 Oct 2015 16:37:48 +0100
> From: David Hassell <[email protected]>
> To: [email protected]
> Subject: [CF-metadata] cell measures for non-horizontal planes
> User-Agent: Mutt/1.5.21 (2010-09-15)
> 
> Hello,
> 
> I was wondering if anyone can help clear something up for me: Can an
> "area" cell measure variable span non-horizontal axes? For example,
> could you have an area cell measure for X-Z cells? And if so, what
> could its standard name be? E.g.
> 
>   float cell_area(longitude, height):
>       cell_area:measure = "area";
>       cell_area:units = "m2";
>       cell_area:standard_name = ?????????????
> 
> I know that the standard_name attribute is optional, thus the two part
> question.
> 
> There is a standard name "cell_area" (= the horizontal area of a
> gridcell), but not one for areas of arbitrarily orientated planes.
> 
> The conventions (section 7.2) do not explicitly disallow
> non-horizontal planes (that's three negatives in half a sentence!),
> but if they're allowed then the assumption is, I think, that the plane
> of the area is implied by the variable's dimensions and there is no
> applicable standard name. This is tricky if the cell measure variable
> does have X and Y dimensions, as in example 7.3. This example uses the
> standard name "area" for its cell measure variable, but this name is
> not in the table - I presume that "cell_area" is meant. I'll raise a
> defect ticket for this if required.
> 
> Many thanks and all the best,
> 
> David
> 
> --
> David Hassell
> National Centre for Atmospheric Science (NCAS)
> Department of Meteorology, University of Reading,
> Earley Gate, PO Box 243,
> Reading RG6 6BB, U.K.
> 
> Tel   : +44 118 3785613
> E-mail: [email protected]
> _______________________________________________
> 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