[CF-metadata] units in cf standard names

Pamment, JA (Alison) J.A.Pamment at rl.ac.uk
Tue Oct 10 04:08:15 MDT 2006


Hi Bryan,

I'm glad this issue has been raised.  At the time of preparing the most recent update to the standard name table I had a brief offline discussion with Jonathan and Brian on the subject of udunits. 

It emerged that there are already some units that have been identified as being useful for CF that are not currently part of udunits.
Jonathan wrote:
> sverdrup = 1e6 m3 s-1   (we can't use the usual abbrev Sv though as it
> already has another definition)
> PSU and psu = 1e-3      (practical salinity unit, effectively mass
> fraction)
There is also decibel (dB).

When I was updating the table it was decided to allow the units of dB for four names relating to sound pressure and sound intensity levels.

We discussed possible ways forward and one would be to define a udunits.dat file including extensions used for CF and link to this from the CF website.  However, Brian expressed the view that a better way forward would be to ask Unidata to include the additions as part of the official udunits, and I agree with that.  As units are so closely integrated with standard names I would be happy to pursue this.

I don't want to open the flood gates to huge numbers of requests for additional units, and I can't guarantee that we will be successful in having them included in udunits, but I think this would be a good opportunity to discuss whether there are other additions needed apart from those listed above.

------
Alison Pamment                            Tel: +44 1235 778065
NCAS/British Atmospheric Data Centre      Fax: +44 1235 445858
Rutherford Appleton Laboratory            Email: J.A.Pamment at rl.ac.uk
Chilton, Didcot, OX11 0QX, U.K.

> -----Original Message-----
> From: cf-metadata-bounces at cgd.ucar.edu [mailto:cf-metadata-
> bounces at cgd.ucar.edu] On Behalf Of Bryan Lawrence
> Sent: 10 October 2006 10:17
> To: christiane.textor at aero.jussieu.fr
> Cc: cf-metadata at cgd.ucar.edu
> Subject: [CF-metadata] units in cf standard names
> 
> 
> Clearly with CF1.0 we are stuck with udunits.
> 
> However, it's worth thinking about the future. I wasn't involved when
> the decision to be udunit compatible was made. Can anyone summarise the
> benefits that requiring udunit conformance confer on the CF community?
> 
> Bryan
> 
> 
> 
> 
> On Tue, 2006-10-10 at 09:38 +0200, Christiane Textor wrote:
> > Hi Egil,
> >
> > This is a good suggestion, however units like kgN/m2/s are not allowed
> > in UDunits as far as I know.
> >
> > (It is not necessary to give additional information for mole fractions,
> > it is only needed for masses.)
> >
> > Best regards,
> > Christiane
> >
> >
> > Egil Støren a écrit :
> > > Hi all,
> > >
> > > I am following this discussion somewhat from the sideline, but with a
> > > growing feeling of uneasiness. If this X_as_Y naming method is
> accepted,
> > > then standard names will tend to grow very large. I also think this
> > > method makes the distinction between physical quantities and units
> less
> > > clear. As I see it, "X_as_Y" and "X_as_such" represents the same
> > > physical quantity, but with different units. I strongly feel that the
> > > "as_Y" part belongs to the units attribute. For example:
> > >
> > > standard_name: mass_flux_of_NO2
> > > units: moles N/m2/s
> > >
> > > This will of course need some changes to CF regarding the units
> attribute.
> > >
> > > Best regards,
> > >
> > >    Egil Støren
> > >    met.no, Norway
> > >
> > >
> > > Jonathan Gregory wrote:
> > >> Dear Roy
> > >>
> > >> Thanks. I note that when you say something is expressed in moles of
> > >> nitrogen
> > >> you mean moles of nitrogen atoms, to be precise.
> > >>
> > >>
> > >>> Does the phrase 'expressed_as' do a better job?
> > >>
> > >>
> > >> It does for me, yes. I interpreted X_as_Y as meaning X is what the
> number
> > >> applies to, but it comes in the form of Y, "as" meaning "in form of"
> like
> > >> "water as ice". By X_as_Y you and Christiane mean Y is what the
> number
> > >> applies
> > >> to, taking "as" to mean "expressed as". So it would be better to have
> > >> expressed_as, I would say, or any neater alternative if someone has
> > >> one to
> > >> suggest.
> > >>
> > >> Is it going to be OK to have names like
> > >> mass_flux_of_ammonium_expressed_as_such
> > >> do you think? Is there a better way to say that?
> > >>
> > >> Cheers
> > >>
> > >> Jonathan
> > >> _______________________________________________





More information about the CF-metadata mailing list