[CF-metadata] NMAT

From: Jonathan Gregory <j.m.gregory>
Date: Mon, 4 Dec 2006 09:43:17 +0000

Dear Julian

> Night is indeed commonly used. ...
>
> However the definition of nighttime is "specialised" for these purposes.
> It is defined as Sunset+1 Hrs to Sunrise+1 Hrs (i.e on local time to the ob)

Is it sufficient to say "night" and define that precisely for your purposes,
or are there several possible definitions of "night" so that there might be
a need to distinguish them in datasets?

> "difference_from_climatology" sounds absolutely fine. However, how and
> where should the climatology be specified (of course this question would
> be just as valid for the anomaly form)? There are several available
> climatologies and it is important to capture which is used in a standard
> fashion.

If this means recording which climatological dataset was used, would it be
sufficient to put it in a comment attribute, or is it possible that you might
have used several of them and need to distinguish in software among anomalies
wrt different climatologies?

Cheers

Jonathan
Received on Mon Dec 04 2006 - 02:43:17 GMT