Dear Martin

I would favour minus_one_times_tendency. This is clumsy but it is very clear,
I think, and as you say it's already in use for other such awkward situations.
I feel that if we replace the word tendency (as in 2) or use a different idea
altogether (as in 3) there will not be such an obvious relationship between the
names of opposite sign i.e. tendency_of_X and minus_one_times_tendency_of_X
and this is a strong advantage, I feel.

I wonder whether we could reduce the clumsiness by saying e.g. minus_tendency,
negated_tendency or negative_tendency, but I'm not sure these are as clear as
minus_one_times_tendency. I think they might all give rise to questions. Of
those, I think minus_tendency would be best.

Other more exotic alternatives could be to invent a new word e.g. antitendency,
or to allow a - sign in the standard name: -tendency_of_X!

Best wishes

Jonathan

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

> Date: Wed, 16 May 2018 11:58:05 +0000
> From: Martin Juckes - UKRI STFC <[email protected]>
> To: "[email protected]" <[email protected]>, Michael Schulz
>       <[email protected]>, Taylor Karl <[email protected]>
> Subject: [CF-metadata] Dry and wet deposition rates
> 
> There are 20 variables in the CMIP6 data request for a variety of dry and wet 
> deposition rates. Many of these variables have been used in CMIP5 and earlier 
> with standard names of the form: 
> tendency_of_atmosphere_mass_content_of_dust_dry_aerosol_particles_due_to_dry_deposition.
> 
> 
> However, there is a conflict in the intended sign convention. The deposition 
> rates should be positive when material is leaving the atmosphere, but the 
> standard names that have been used should be positive when the atmospheric 
> content is increasing. To resolve this we need a set of new standard names. 
> The existing tendency names are:
> 
> tendency_of_atmosphere_mass_content_of_dust_dry_aerosol_particles_due_to_dry_deposition
> tendency_of_atmosphere_mass_content_of_ammonia_due_to_dry_deposition
> tendency_of_atmosphere_mass_content_of_ammonium_dry_aerosol_particles_due_to_dry_deposition
> tendency_of_atmosphere_mass_content_of_particulate_organic_matter_dry_aerosol_particles_due_to_dry_deposition
> tendency_of_atmosphere_mass_content_of_sulfur_dioxide_due_to_dry_deposition
> tendency_of_atmosphere_mass_content_of_sulfate_dry_aerosol_particles_due_to_dry_deposition
> tendency_of_atmosphere_mass_content_of_nitrogen_compounds_expressed_as_nitrogen_due_to_deposition
> tendency_of_atmosphere_mass_content_of_dust_dry_aerosol_particles_due_to_wet_deposition
> tendency_of_atmosphere_mass_content_of_ammonia_due_to_wet_deposition
> tendency_of_atmosphere_mass_content_of_ammonium_dry_aerosol_particles_due_to_wet_deposition
> tendency_of_atmosphere_mass_content_of_noy_expressed_as_nitrogen_due_to_wet_deposition
> tendency_of_atmosphere_mass_content_of_sulfur_dioxide_due_to_wet_deposition
> tendency_of_atmosphere_mass_content_of_sulfate_dry_aerosol_particles_due_to_wet_deposition
> tendency_of_atmosphere_mass_content_of_particulate_organic_matter_dry_aerosol_particles_due_to_wet_deposition
> tendency_of_atmosphere_mass_content_of_ozone_due_to_dry_deposition
> tendency_of_atmosphere_mass_content_of_seasalt_dry_aerosol_particles_due_to_wet_deposition
> tendency_of_atmosphere_mass_content_of_elemental_carbon_dry_aerosol_particles_due_to_wet_deposition
> tendency_of_atmosphere_mass_content_of_seasalt_dry_aerosol_particles_due_to_dry_deposition
> tendency_of_atmosphere_mass_content_of_noy_expressed_as_nitrogen_due_to_dry_deposition
> tendency_of_atmosphere_mass_content_of_elemental_carbon_dry_aerosol_particles_due_to_dry_deposition
> 
> Three options have been raised in the preliminary discussions:
> (1) Prefix each name with "minus_one_times_": this construction is already 
> used for 3 names.
> 
> (2) Replace "tendency_of_" with "depletion_rate_of_": this is a new 
> construction, but is structurally close to what we have and makes the 
> relation between variables very clear;
> 
> (3) Replace "tendency_of_atmosphere_mass_content_of_" with 
> "surface_downward_mass_flux_of_": this uses an existing construction already 
> used in 8 standard names.
> 
> What do others on the list think?
> 
> regards,
> Martin
> 
> 
> 
> 
> ________________________________
> From: Juckes, Martin (STFC,RAL,RALSP)
> Sent: 16 May 2018 08:14
> To: Michael Schulz; Taylor Karl
> Cc: [email protected]; [email protected]; Bodas-Salcedo, 
> Alejandro; Pamment, Alison (STFC,RAL,RALSP)
> Subject: Re: Conflicting sign conventions in deposition rates
> 
> 
> Hello Michael, Karl,
> 
> 
> OK, it sounds as though we should keep the variables as they are, i.e. 
> positive for deposition from the atmosphere, and fix the standard names.
> 
> 
> I don't think a positive attribute will solve this problem. It may be useful 
> to add "positive=down", but it still leaves an inconsistency with the 
> standard names which definitely should be used with positive "up". Standard 
> names for variables with an opposite sign convention would not be synonyms: 
> there are many examples of pairs of names which differ only in the sign 
> convention, e.g.
> 
> surface_upwelling_shortwave_flux_in_air -- 
> minus_one_times_surface_upwelling_shortwave_flux_in_air  (+ 2 other pairs 
> like this);
> 
> surface_downward_heat_flux_in_air -- surface_upward_heat_flux_in_air (+ 5 
> other pairs like this).
> 
> 
> We can construct standard names for minus the tendency of quantities in 
> several ways, but it may be better  to have that discussion on the CF 
> discussions email list .. so I'll make a proposal there, also raising Karl's 
> suggestion (depletion_rate_of_ ....),
> 
> 
> regards,
> 
> Martin
> 
> 
> 
> 
> 
> 
> ________________________________
> From: Michael Schulz <[email protected]>
> Sent: 15 May 2018 19:46
> To: Juckes, Martin (STFC,RAL,RALSP)
> Cc: Taylor Karl; [email protected]; [email protected]; 
> Bodas-Salcedo, Alejandro; Pamment, Alison (STFC,RAL,RALSP)
> Subject: Re: Conflicting sign conventions in deposition rates
> 
> Hi,
> 
> true - tendencies have not carefully chosen with a sign in mind.
> 
> There is very long tradition that these variables are “positive". So that 
> should not be changed.
> 
> The standard names are also used since quite some time.
> 
> Adding a positive attribute would be my preferred solution.
> 
> Otherwise “surface_downward_mass_flux_of_” is fine for me as well. But that 
> would introduce a synonym in the system, right?
> 
> best wishes
> Michael
> 
> 
> 
> 
> 
> > On 15 May 2018, at 20:13, Martin Juckes - UKRI STFC 
> > <[email protected]> wrote:
> >
> > Hello All,
> >
> >
> > Larry has pointed out a conflict between variable long names and standard 
> > names for a set of deposition rate variables, many of which were in the 
> > CMIP5 aero table, e.g. wetso4, "Wet Deposition Rate of SO4" with standard 
> > name 
> > tendency_of_atmosphere_mass_content_of_sulfate_expressed_as_sulfur_dry_aerosol_due_to_wet_deposition
> >
> >
> > A deposition rate would normally be positive when material is leaving the 
> > atmosphere, making the tendency of atmosphere mass content negative.
> >
> >
> > I will try to check the sign convention adopted by people submitting data 
> > for CMIP5, but I suspect that we should follow the sign convention implied 
> > by the long name .. but this would require new standard names for these 
> > variables. A full list of the variables is here: 
> > https://github.com/cmip6dr/CMIP6_DataRequest_VariableDefinitions/issues/328
> >
> >
> > For new standard names, we could either use the "minus_one_times_" 
> > construction, or replace "tendency_of_atmosphere_mass_content_of_" with 
> > "surface_downward_mass_flux_of_".
> >
> >
> > What do others think?
> >
> >
> >
> > regards,
> >
> > Martin
> >
> >
> 
> _______________________________________________
> CF-metadata mailing list
> [email protected]
> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata

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

Reply via email to