[CF-metadata] New UDUnits units for information: "byte" and "octet"

Jonathan Gregory j.m.gregory at reading.ac.uk
Wed Jan 7 08:50:31 MST 2015

Dear Maarten

> >A mole is also a udunit, so mol m-3 and m-3 are different units, and quantities
> >with those canonical units also have to have different standard_names. The
> >standard_name indicates whether the quantity refers to number concentration or
> >molar concentration.
> 1e15 molecules/cm2 NO2 is a valid column amount for NO2, but at the
> same time I'm trying to convince my colleagues that this quantity
> should be expressed in mol/m2 (specifically 16.6 micromol/m2 for
> said column amount). These express the same quantity and should
> _not_ use different standard names, not in the case of molecules. In
> the end the number density should become extinct for molecular
> species, switching completely to mol based units. Having different
> standard names will only serve to hinder the transition.

Ah, I see (perhaps! :-) Is the point that "mole" means a particular number of
molecules, so there should *not* be a distinction between [molecules] m-3 and
mol m-3? Yes, that's a good argument, but in CF we are following SI, and mol
is an SI unit, so a mole of NO2 is dimensionally different from 6e23 molecules
of NO2. If it has a different unit, we have to give it a different standard
name. udunits follows SI as well; you can't convert mol into a plain number.

Best wishes and thanks for the explanation


More information about the CF-metadata mailing list