[CF-metadata] [Standard name request] property changes over time

From: Paul.Durack at csiro.au <Paul.Durack>
Date: Thu, 14 Apr 2011 01:46:41 -0000

Ok can I request the new standard names:

sea_water_salinity_change
sea_water_temperature_change
sea_water_potential_temperature_change
sea_water_density_change


I'd also note that in the current version of the standard names table there are the following entries (amongst others) which, to me, appear to duplicate meaning without a clear "standard":

change_in_atmosphere_energy_content_due_to_change_in_sigma_coordinate_wrt_surface_pressure
change_in_energy_content_of_atmosphere_layer_due_to_change_in_sigma_coordinate_wrt_surface_pressure
...
change_over_time_in_atmosphere_water_content_due_to_advection
change_over_time_in_surface_snow_amount
...
global_average_sea_level_change
global_average_steric_sea_level_change
global_average_thermosteric_sea_level_change
...
tendency_of_air_density
tendency_of_air_pressure
tendency_of_air_temperature


I personally would prefer to use a "standard" way to express absolute changes, or tendencies wrt time.. My preference for this is the *_change suffix (along with it's appropriate climatology attribute).

Cheers,

P

-----Original Message-----
From: Jonathan Gregory [mailto:j.m.gregory at reading.ac.uk]
Sent: Thursday, April 14, 2011 1:50 AM
To: Durack, Paul (CMAR, Hobart)
Cc: CF-metadata at cgd.ucar.edu
Subject: Re: [CF-metadata] [Standard name request] property changes over time

Dear Paul

You have raised two issues.

* Preference for X_change to change_in_X.

* Suggestion that any X could have _change suffixed without being approved as
a new standard name.

I think the first doesn't have a clear answer. It's a bit arbitrary. I would
say that change_in is probably clearer, although a bit longer, because
concatenating too many nouns can get a bit difficult or ambiguous to understand
in English. But we do have some long noun phrases in the stdname table. I don't
think there's a compelling need to change (sorry) our convention, myself.

We have discussed quite a lot of similar cases to the second point already,
and have so far always decided that even though it is desirable to construct
stdnames consistently, we will always approve them individually and manually
rather than automatically. That is because sometimes there's more than one way
to describe a quantity, so consistency is not unique, or because some
automatically constructed names might actually be physical nonsense, or because
there may be a common unsystematic term we'd prefer to adopt. Therefore
discussion about this issue has been concerned with providing tools to make it
easier to construct proposed standard names consistently. You're only asking
for three names, I think, which is not a big deal really. The standard name
table has over 2000 entries already so 3 more isn't going to cause indigestion.

Best wishes

Jonathan
Received on Wed Apr 13 2011 - 19:46:41 BST