[CF-metadata] projections/default cell_methods

From: Jonathan Gregory <jonathan.gregory>
Date: Fri, 07 Jun 2002 09:19:21 +0100

Dear Brian

> I think any new
> attributes for grid mappings should be attributes on variables.
I agree. The description of the rotation should be attached to each data
variable in the file. This is necessary because it is possible the file might
contain data on a mixture of grids, some rotated, some not.

> So here's a fairly complete example of a variable on a rotated pole grid:

The example all looks fine to me - thanks. grid_mapping is better than
projection as the attribute name. I have a preference for grid_latitude and
grid_longitude as standard names instead of rotated_latitude and
rotated_longitude, because the same concept may arise with all kinds of
projections, and we don't really need to invent separate standard names in
each case, I would argue. (I had forgotten about this pitfall; thanks for
remembering it!)

> To move forward on this we need, as you said, to decide on the set of
> keywords for describing the transformations, and decide on the standard
> names to identify the transformed coordinates. This will take a serious
> effort. Maybe this summer...

However, maybe in the particular case of the rotated pole, we could decide
now that what we have just developed is OK. To make that decision now would
help Burkhardt write his data, I imagine. Rotated pole is also the only
mapping that we use in the Hadley Centre models, so it would help us too.
Shall we do that?

I'm glad you agree with sum and point.

Cheers

Jonathan
Received on Fri Jun 07 2002 - 02:19:21 BST