[CF-metadata] Radar data and the 'dB' and 'dBZ' units

[Stephens, A (Ag)]

Dear John,

Bryan Lawrence (BADC) and I were discussing the CF Conventions and the issue
of non-compliant units in regard to radar data. Bryan mentioned that you are
dealing with radar data and it is likely that you have been encountering the
same issues as ourselves.

Our radar data providers have use 'dB' and 'dBZ for a number of parameters
such as:

radar_reflectivity_factor       dBZ
signal_to_noise_ratio           dB
linear_depolarization_ratio     dB
differential_reflectivity       dB

These are relative dimensions not included in UDUNITS. One solution is
therefore to include them as units="dimensionless" with an appropriate
comment to explain the relative unit. However, radar people will be confused
if they do not see dB as the unit for such parameters. 

Another solution might be to ask udunits to include these units in their
list.

Have you come across this problem? What do you propose is the best solution?
We have chatted to Jonathan Gregory about it briefly but it would be good to
hear your comments.

Kind regards,

Ag
-------------------------------------------------------------
Ag Stephens                           Ph : +44 (0)1235 446220
Data Scientist,                       Fax: +44 (0)1235 446314
British Atmospheric Data Centre,     
Rutherford Appleton Laboratory,               
Chilton, Didcot, OX11 0QX, U.K. 
Email:   A.Stephens at rl.ac.uk      Web: http://badc.nerc.ac.uk
-------------------------------------------------------------