[CF-metadata] standard names for defined global attribute names

From: Jonathan Gregory <j.m.gregory>
Date: Thu, 29 Mar 2012 16:26:12 +0100

Dear Jim

> after our earlier discussion about auxiliary coordinate variables, this
> sure looks like a source or institution variable shouldn't be an auxiliary
> coordinate variable. It is not remotely related to a coordinate axis is
> it?

It is expected, as in my example and in Mark's original posting, to be used
when there is an ensemble dimension. This is quite common. The ensemble
dimension is an axis that doesn't usually have a numeric coordinate, but it
can and should have auxiliary coordinate variables to describe it.

> this being used as a label coordinate so that you can make plots that show
> values across sources?

For instance, yes. The facility to provide labels in exactly this way is
described in CF sect 6.1. The idea is not new. The proposal is just to
define a couple more standard_names that could be used for such labels.

Best wishes

Jonathan
Received on Thu Mar 29 2012 - 09:26:12 BST