[CF-metadata] mixing ratio

Jonathan Gregory j.m.gregory at reading.ac.uk
Thu Nov 6 15:29:02 MST 2008

Dear Philip and Christiane

Thanks for the explanation. I appreciate that the ratio of constituent mass
to dry air mass is a sensible quantity to use in a model; I am not disagreeing
with that at all. The question is what the standard name should be. I was
confused by the proposal and as Philip also says

> >'water vapor in dry air' initially seems to make no sense. 

It makes sense as a quantity to use, but that name is likely to be baffling to 
a non-expert user of the dataset, I would say, and so it doesn't seem like an
ideal choice for a standard name.

This quantity is often called humidity mixing ratio, and that is at present
its standard name. We could leave it as that, for the special case of water
vapour. But what about other constituents? As far as I can see from the
web (but you are the experts) mixing_ratio_of_X_in_air is used variously to
mean X/(air including X) or X/(air not including X) and the air might be dry
or ambient. Since the numbers are only slightly different, maybe this vagueness
is tolerable, or even desirable if you want to compare models that make
different choices but still regard the quantities as comparable. If we want
to be precise, I think we need more explicit terms.

The construction mass_fraction_of_X_in_air seems obviously to mean
X/(air including X and ambient moisture). mass_fraction_of_X_in_dry_air
would be fine for any X except water vapour, and would mean
X/(air including X but not water vapour).

Best wishes


More information about the CF-metadata mailing list