[CF-metadata] Cell bounds associated with coordinate ... (in situ data)

From: Jonathan Gregory <j.m.gregory>
Date: Wed, 18 Nov 2009 18:24:57 +0000

Dear Nan

I didn't suggest that everyone should use the centre-time for their time-
coordinates. I suggested that the centre-time should be the recommended choice
if there was no other clear way of deciding what was appropriate.

> As Benno mentioned, some instruments - not just old current meters, but
> new ones, and most anemometers - still record a single direction and an
> averaged speed.

That is fine, I think, with the existing cell_methods. In that case the
time coordinate variable would have bounds, but the cell_methods are different
for the two variables. The instantaneous direction is a "point" in time, and
the bounds aren't relevant, whereas the averaged speed is a time-mean, and
the coordinates aren't relevant. However it does no harm to have both of them;
they can be useful for graphical purposes, for instance.

Best wishes

Jonathan
Received on Wed Nov 18 2009 - 11:24:57 GMT