Dear David, John,

thanks for those comments.


Just to re-iterate, the current UDUNITS library does not include "dB", but it 
does include "dBZ" and some other variations which are decibels with a fixed 
specified reference value.


The 2014 UDUNITS discussion which John links to is interesting. It appears to 
have covered the ground we have covered below, and Sean Arms has just 
re-started the discussion there with a link to this thread (thanks Sean).  The 
latest proposal appears to be that a generic `dB` unit could be introduced with 
the reference level be specified in an additional attribute (yet to be defined).


The 2nd link from John is a discussion about the R package wrapper for UDUNITS, 
which does appear to have support for "dB" added, but simply defined in terms 
of the logarithm of a dimensionless number. This looks like a neat programmers 
solution which avoids having to worry about what the physical parameter is.


I've looked into this a bit further, and found that "bel" and "decibel" are 
recognised by ISO and the International Electrotechnical Commission (IEC 
60027-3:2002), along with the "neper", which is the natural logarithm of a 
ratio. A "bel" is either the log10 of the ratio for a "power" quantity or 2 
times log10 of the ratio for a "field" quantity. The IEC standard goes onto say 
that "dB" can be used to define power relative to a reference level, and 
information about reference levels should be attached to the quantity not to 
the unit -- which is in line with the current usage in 4 CF standard names.


There might be some value in trying to follow the ISO/IEC approach.


regards,

Martin


________________________________
From: CF-metadata <[email protected]> on behalf of Moroni, David 
F (398G) <[email protected]>
Sent: 12 November 2018 06:24
To: John Graybeal; Jonathan Gregory
Cc: CF Metadata List
Subject: Re: [CF-metadata] Decibel units in CF standard names


John and others interested,

I was part of the initial request that dated back to 2014. Here’s the original 
GitHub ticket (still open) capturing the correspondence with the UDUNITS team: 
https://github.com/Unidata/UDUNITS-2/issues/33



I was corresponding with “mhidas” and “semmerson”, so I don’t know if these 
people are still on this project, as no further correspondence has taken place 
since July 2015. I made another attempt last year to bring some life to this 
request, but to no avail.



There was another user inquiring about representing dB as a unit, but that 
ticket has since closed and it’s not clear to me from the thread whether there 
was a positive resolution. Here’s that link: 
https://github.com/r-quantities/units/issues/176



Others on here are more than welcome to pursue this further.



Cheers,

David



From: CF-metadata <[email protected]> on behalf of John Graybeal 
<[email protected]>
Date: Sunday, November 11, 2018 at 9:33 PM
To: Jonathan Gregory <[email protected]>
Cc: CF Metadata List <[email protected]>
Subject: Re: [CF-metadata] Decibel units in CF standard names



Just as an aside (or maybe not), the udunits support list has been asked before 
to include dB, and I understood that they had (I think I actually saw it, but 
can’t find written confirmation). So it wouldn’t surprise me if the library 
included dB.



In case it’s useful I pasted in a bit of the old thread below.



john

---------------------------------------
John Graybeal
[email protected]<mailto:[email protected]>





On Nov 4, 2018, at 09:03, Jonathan Gregory 
<[email protected]<mailto:[email protected]>> wrote:



Dear Martin

Your points are good ones and have been raised before. More than once we have
talked about maintaining a CF version of the udunits definition to include dB
and sverdrup, or ask udunits to add them (if they're not there). dB is a dimen-
sionless unit, equivalent to 1. I suggest that dBZ should be changed to dB,
as I don't think we ought to have several of them. I believe that the default
reference levels are mostly conventional and stated in the definitions of the
standard name, as you say. They can be overridden by supplying a size-one or
scalar coordinate variable. You have previously suggested an xml table to
contain more information about the definition of standard names, haven't you?
It seems to me that an arrangement like that would be the right place to store
the default reference levels and scale factor in a machine-readable way.

Best wishes

Jonathan

----- Forwarded message from Martin Juckes - UKRI STFC 
<[email protected]<mailto:[email protected]>> ——











Begin forwarded message:



From: "Moroni, David F (398M)" 
<[email protected]<mailto:[email protected]>>

Subject: Re: [CF-metadata] [cf-satellite] 
normalized_radar_backscatter_coefficient

Date: August 8, 2014 at 3:10:50 PM PDT

To: John Graybeal 
<[email protected]<mailto:[email protected]>>, 
"[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>

Cc: "Weiss, Barry H (398B)" 
<[email protected]<mailto:[email protected]>>, Jonathan 
Gregory <[email protected]<mailto:[email protected]>>, 
"[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>, CF 
Metadata List <[email protected]<mailto:[email protected]>>



Hi John,



Thanks for providing the email link. I've already emailed Unidata and they are 
processing my request to include dB within UDUNITS.



Cheers,

David



From: John Graybeal 
<[email protected]<mailto:[email protected]>>
Date: Friday, August 8, 2014 10:57 AM
To: "[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>
Cc: "Weiss, Barry H (398B)" 
<[email protected]<mailto:[email protected]>>, David F Moroni 
<[email protected]<mailto:[email protected]>>, Jonathan 
Gregory <[email protected]<mailto:[email protected]>>, 
"[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>, CF 
Metadata List <[email protected]<mailto:[email protected]>>
Subject: Re: [CF-metadata] [cf-satellite] 
normalized_radar_backscatter_coefficient



This request has now been made by me of UDUNITS on general principle, but I 
think individuals could also express their desire as a way to move the ball 
forward.



John



On Aug 7, 2014, at 07:57, 
[email protected]<mailto:[email protected]> wrote:



Dear Barry and David:



For udunits support, send an email to 
[email protected]<mailto:[email protected]>.







very respectfully,



randy







________________________________

From: "Weiss, Barry H (398B)" 
<[email protected]<mailto:[email protected]>>
Sent: Thursday, August 07, 2014 10:06 AM
To: "Moroni, David F (398M)" 
<[email protected]<mailto:[email protected]>>, "Jonathan 
Gregory" <[email protected]<mailto:[email protected]>>
Cc: "Niedfeldt, John C (398M-Affiliate)" 
<[email protected]<mailto:[email protected]>>, "CF 
Metadata List" <[email protected]<mailto:[email protected]>>, 
"John Graybeal" 
<[email protected]<mailto:[email protected]>>, 
"[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>
Subject: Re: [cf-satellite] [CF-metadata] 
normalized_radar_backscatter_coefficient



Jonathan,

As the data product engineer for the SMAP project, I second David's
request.

We are attempting to employ CF metadata in our products. This is not a
challenge at level 2 and above where our products provide geophysical
measure, but is a challenge at level 1, where our products provide
instrument measurements.

Please consider inclusion of dB units. That should include dB based on
unit less measure, as well as dB relative to watts and volts.

Thanks and Regards,

Barry


On 8/6/14 7:49 PM, "Moroni, David F (398M)" 
<[email protected]<mailto:[email protected]>>
wrote:

>Hi Jonathan,
>
>Just as follow up from my last email, I noticed an online email exchange
>where you had responded to a request to use units of dB (decibels) even
>though it is not currently in the udunits database:
>http://mailman.cgd.ucar.edu/pipermail/cf-metadata/2013/056572.html
>
>We also agree it would be wise to include dB in the udunits database, and
>we will be applying these units for our scatterometer datasets.
>
>We hope to see this incorporated in the near future.
>
>Thanks again for your considerations.
>
>Cheers,
>David
>




_______________________________________________
CF-metadata mailing list
[email protected]
http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata

Reply via email to