[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[netCDF #UBD-124546]: question on dBz

> Hi Russ,
> 
> Thanks for your response of my question. Now I have two questions.
> 
> 1. I checked ununits2-common.xml file for both 2.2.17 and 2.1.24 version and 
> there is definition of BZ = lg(re (1e-6 m)^3) instead of dbz.

Right.

> If dBZ is not there, Should we use BZ as unit and use scaling factor of 0.1 
> to make it equivalent to dBZ?

That's not necessary, "d" is a prefix understood by udunits, so "dBZ" is 
already understood 
by udunits as 0.1 BZ.  You can verify this by using the "udunits2" interactive 
command-line
utility that comes with the udunits software installation:

  $ udunits2
  You have: dBZ
  You want: BZ
    1 dBZ = 0.1 BZ
    x/BZ = lg((1e+18*(1e-18*(pow(10, 0.1*(x/dBZ))))))

Similarly, you should be able to use any of the SI prefixes
in the udunits2-prefixes.xml file in front of BZ, such as
n, m, c, d, etc., and have it understood.

> 2. thanks for pointing me to standard_name of equivalent_reflectivity_factor. 
> Does this suggests that "radar reflectivity" measurement from Lidar or NAXRAD 
> should use this as standard name attribute?

Since it's in the CF standard name table, if it means what you intend,
as given in its definition there:

  "Equivalent reflectivity factor" is the radar reflectivity factor
  that is calculated from the measured radar return power assuming the
  target is composed of liquid water droplets whose diameter is less
  than one tenth of the radar wavelength, i.e., treating the droplets
  as Rayleigh scatterers. The actual radar reflectivity factor would
  depend on the size distribution and composition of the particles
  within the target volume and these are often unknown.

then I would use it as the value of the "standard_name" attribute for
your variable.  If you mean something else instead, you'll need to
apply for a new name by following the CF process described here:

  http://cfconventions.org/faq.html#stdnames_ask

Also note that "dBZ" is merely the canonical unit for variables that
have the attribute 

  var:standard_name=equivalent_reflectivity_factor

You could use any compatible unit, such as BZ, or even the definition
in terms of SI units that's used in udunits2.

--Russ
> -----Original Message-----
> From: Unidata netCDF Support [mailto:address@hidden]
> Sent: Tuesday, December 09, 2014 1:57 PM
> To: Li, Xiang
> Cc: address@hidden
> Subject: [netCDF #UBD-124546]: question on dBz
> 
> Hi,
> 
> > I am working on constructing metadata for a satellite data set
> > packaged using netCDF format. We want it CF-compliant.
> >
> > One of the variable is reflectivity profile with unit of 'dBz'.
> > However, I could not find proper unit definition in UDUNITS.dat for
> > this variable. Would you like to suggest what we should do to make the
> > netCDF file CF-compliant and also describe the field correctly?
> 
> The original udunits package did not support logarithmic units such as dBZ, 
> but the current udunits (and all versions since 2.0.4) includes such support:
> 
> http://www.unidata.ucar.edu/software/udunits/
> 
> "dBZ" is defined there as equal to "0.1 lg(re (1e-6 m)^3)", so the reference 
> level is included in the definition.  There is a CF standard name for the 
> quantity "equivalent_reflectivity_factor"
> which uses dBZ as its canonical units.
> 
> > Thanks for your help,
> 
> Your welcome!
> 
> > Xiang Li
> 
> --Russ
> 
> Russ Rew                                         UCAR Unidata Program
> address@hidden                      http://www.unidata.ucar.edu
> 
> 
> 
> Ticket Details
> ===================
> Ticket ID: UBD-124546
> Department: Support netCDF
> Priority: Normal
> Status: Closed
> 
> 
Russ Rew                                         UCAR Unidata Program
address@hidden                      http://www.unidata.ucar.edu



Ticket Details
===================
Ticket ID: UBD-124546
Department: Support netCDF
Priority: Normal
Status: Closed