New standard name table for the CF netCDF convention
Under the CF metadata home page http://www.cgd.ucar.edu/cms/eaton/cf-metadata/
we have posted a new version of the CF standard name table. (The standard name
table specifies standardized names for distinct physical quantities, as
described in section 3.3 of the CF convention.) The first draft was originally
posted early in 2001, but few comments were received until 2002. In the new
version, we have made extensive changes to take these comments into account, in
particular to use shorter names where possible and to adopt more usual
phraseology.
In the two years since the draft was published, users of the CF convention may
have begun to use the draft standard names. We have therefore also provided an
equivalence table to the new names for those which have changed. We also
describe the changes and the reasons for them. For simplicity, we would like to
regard the superseded names as obsolete. However, in cases where this would
cause you great inconvenience, we could define the old names to be aliases for
the replacement ones. A mechanism to do this has already been proposed in the
CF convention. Please post to the CF-metadata group stating which superseded
names you would like to be retained.
We propose that all standard names in the table on 31 January 2003 will be
permanently valid. None will be made obsolete after that date. Future changes
will require aliases to be introduced for backwards compatibility.
The new standard name document provides explanatory comments for many standard
names. We have also supplied tables of equivalences to PCMDI variable names and
ECMWF and NCEP GRIB codes. These other parameter lists specify some quantities
which for a complete definition require metadata in addition to the
standard_name, as shown in the equivalence tables. The equivalences for the
NCEP private GRIB codes (>127) have not yet been included, and there are a few
other holes that will be filled. We will shortly propose a standardized method
for distinguishing between an annual and a decadal standard deviation, which is
needed in the PCMDI table and will be more generally useful. It would be very
helpful if those familiar with PCMDI, ECMWF or NCEP conventions could check the
equivalences which have been completed. In addition, we would welcome answers
to the questions appended, which show our ignorance and/or the lack of clarity
in the parameter tables!
According to the CF convention, the primary definition of the standard name
table is given in XML. The XML version will be made available soon.
Thanks to all for comments on the CF conventions. Continued discussion is most
welcome. Happy New Year.
Jonathan Gregory and Brian Eaton
* Is pseudoadabatic potential temperature the same thing as pseudoequivalent
potential temperature?
* What is Montgomery streamfunction (GRIB code 37)?
* What is the distinction between absolute and relative divergence (GRIB codes
42 and 44)?
* Are GRIB codes 15 and 16 *daily* max and min temperature, as we have
assumed? If not daily, then what?
* Is vorticity (ECMWF GRIB code 138) relative or absolute?
* Is log surface pressure (ECMWF GRIB code 152) natural or common log? In what
units is the pressure?
* Is there a quantitative definition of "deep soil" in ECMWF GRIB codes 170,
171, 183 and 184?
* In view of the note to the GRIB table which reads "Net (vertical) fluxes
shall be calculated by subtracting the downward flux values from the upward
flux values" we have assumed that they are positive upward. Is this correct?
* What is ICAO standard atmosphere reference height (NCEP GRIB code 5)?
* What are the parcel and best lifted indexes (NCEP GRIB codes 24 and 77)?
* NCEP GRIB code 89 is "density" of what?
* What are the precise definitions of the radar spectra (GRIB codes 21-23),
wave spectra (28-30), thermocline depths (68-70), primary and second wave
directions and periods (107-110), image data (127), atmospheric tide (128) and
budget values (129)?
* What is the difference between PCMDI variables cl "cloud amount" and clt
"total cloud amount"?
* What is PCMDI variable mrsofcs "surface soil moisture field capacity" a
percentage of?
* Are PCMDI net radiative fluxes upward or downward? We have assumed upward.
* What is the "model top" in PCMDI variable rtmt? In what way is it different
from TOA?
