[CF-metadata] standard_names for emissions - part 2: species

From: Jonathan Gregory <j.m.gregory>
Date: Wed, 08 Jun 2011 17:32:34 -0000

Dear Martin and Angelika

Your proposal looks very careful and sensible to me. The new chemical species
and processes will fit the existing patterns. I support the use of
tendency_of_atmosphere_mass_content_of_X

One comment, though. Since all the processes are due_to_emission_from_Y,
where Y is a sector of activity, I would like the raise the question of whether
it would be a good idea to keep Y out of the standard name and store it in
some other fashion, probably a [scalar] coordinate variable? As usual, this is
a question about an essentially arbitrary boundary. We decided not to do it
with chemical species in order to make sure everything makes sense chemically
in standard names, but perhaps in order to limit the number of standard_names
we might factor out this new distinction? An analogy would be with area_type
and region, which we have now generalised in this way (though some area_types
of particular importance do appear in standard names). I can imagine there
could be a practical advantage of being able to contain the emissions of
a particular species from all the sectors in a single data variable, if Y
were made a dimension. Another reason is that the Y are getting some way from
being "geophysical", which is the core purpose of standard_name descriptions.

Best wishes

Jonathan
Received on Wed Jun 08 2011 - 11:32:34 BST