[CF-metadata] Proposed cell_methods method

From: Jonathan Gregory <j.m.gregory>
Date: Mon, 2 Jul 2007 11:17:35 +0100

Dear Mike

I agree with Seth:
> So the standard name would be "change_over_time_in_air_pressure", I
> think, with an associated timebnds variable as described in section 7.1.

This rule was introduced following a similar previous example. The argument
for not treating it as a cell method is that this is not a representation of
subgrid variation, which is the primary purpose of cell_methods. It is the
time-extensive counterpart of tendency_of_air_pressure, which is an intensive
quantity in time. Tendency of air pressure is a different quantity from
air pressure, so change in air pressure is likewise a different quantity
from air pressure.

It might be argued that if we follow this guideline, we could end up with
a change_in_ being proposed for every quantity. In that case we would have
to think again! However, I don't think that is likely, because we have not
been asked for a tendency_in_ for every quantity.

Best wishes

Jonathan
Received on Mon Jul 02 2007 - 04:17:35 BST