[CF-metadata] CF standard_names for sea water pressure and sound intensity

[Jonathan Gregory]

Dear Mike

> 	sea_water_pressure (canonical units dbar)

Any dimensionally equivalent unit could be used to specify the canonical
units, so dbar is OK.

> Regarding ADCP measured acoustic echo intensity, my current data
> files have this as a count.  There is an algorithm to convert this
> to a measure of acoustic_echo_intensity in decibels (dB).  Is there a 
> need to distinguish this from acoustic_echo_intensity in other media
> by perhaps calling it sea_water_acoustic_echo_intensity?  

Let's call it acoustic_echo_intensity_in_sea_water like other quantities
measured in the ocean.

> dB is a non-SI unit and not in udunits (we have a lot of these in 
> oceanography).

dB is a dimensionless unit for the logarithm of a ratio. The denominator of
the ratio is some standard value. I think it ought to be acceptable to define
a quantity like this, if we state the reference value in the definition of the
standard name, perhaps making it a standard name parameter (an idea which has
been agreed on this email list but not written in the convention yet) with a
default value. What is your reference intensity? The AMS glossary says that
for acoustics it is 1e-10 microwatt per cm2, but perhaps ADCP is different.

It is another problem whether we can use dB as a unit. I think we ought to be
able to, myself. As you say, this isn't the only non-udunit we'd like to use.
PSU is another, also dimensionless, and in CF we would probably prefer Sv to
be sverdrup rather than sievert as udunits has it. This is a question we must
address to Unidata. One way would be to produce a CF version of udunits.dat,
the file which defines the udunits.

Thanks. Cheers

Jonathan