Dear Andreas and Martin

I agree with Martin that if layers can be defined numerically these numerical
limits should not appear in the standard_name. The usual way to do it would
be with a coordinate variable (either with a dimension of size 1, or a scalar
coordinate variable) whose bounds specify the limits concerned e.g. a
variable of air_pressure with bounds of 500 and 750 hPa.

This will not work straightforwardly when one of the bounds is numerically
defined and the other is a named surface such as the tropopause. That hasn't
come up before. Still, I think we should find a solution which avoids having
a number in the standard_name.

Best wishes

Jonathan

----- Forwarded message from "Schultz, Martin" <[email protected]> -----

> From: "Schultz, Martin" <[email protected]>
> To: "[email protected]" <[email protected]>
> Date: Fri, 21 Dec 2012 07:55:00 +0000
> Subject: Re: [CF-metadata] standard names for column amounts (atmospheric
>       chemistry)
> 
> Hi Andreas,
> 
> > -----Original Message-----
> > From: CF-metadata [mailto:[email protected]] On Behalf
> > Of Andreas Hilboll
> > Sent: Wednesday, December 19, 2012 2:53 AM
> > To: [email protected]
> > Subject: Re: [CF-metadata] standard names for column amounts
> > (atmospheric
> > chemistry)
> >
> 
>    in addition to what Philip said, let me try to guide you a little more 
> specifically concerning some of your questions. Don't feel frightened about 
> the discussions that are going to come up. Eventually, there will be an 
> agreement, and in the end it's usually a wise decision if you look at it from 
> a little distance later. The following comments may help you to identify the 
> potential areas for discussion - the better your proposal will address these 
> points "in the spirit" of existing CF names, the easier it will be to get the 
> proposal accepted smoothly.
> 
> > * The canonical way to specify tropospheric NO2 columns in
> >   molecules cm^-2 would be to use the standard name
> >   "mole_content_of_nitrogen_dioxide_in_troposphere_layer", or rather
> >   "troposphere_mole_content_of_nitrogen_dioxide",
> >   together with a units attribute of "molecules cm^-2"?
> 
> I agree with Philip here, that "troposphere_mole_content..." seems the more 
> natural choice. Obviously, this can easily be extended to 
> "stratosphere_mole_content..." if you provide a clear definition of what 
> "stratosphere" means (in fact I just saw that "troposphere" is also not 
> defined very well, it only says "in this layer of the atmosphere"!). We 
> should probably change this definition to something more specific, such as 
> "the layer of the atmosphere between the earth surface and the tropopause. 
> Generally, the tropopause will be defined via the reversal of the temperature 
> gradient with altitude as specified by WMO. Other definitions of tropopause 
> are possible, and if one of these is used, it should be explained in a 
> comment attribute." -- In analogy, the "stratosphere" should be defined as 
> the layer between the tropopause and the stratopause.
> 
> 
> >
> > * What can I do for
> > "atmosphere_mole_content_of_nitrogen_dioxide_between_surface_and_XhPa
> > ",
> > where X is 500?
> 
> Here, my suggestion would be to go for the general term "atmosphere_layer" 
> and request either a generic comment attribute or a more specific attribute 
> (something like "layer_extent" ?) where the exact specification of "between 
> ... and ..." should be provided. I don't believe that it would be possible to 
> agree on a small confined set of numbers, and since a standard_name is static 
> and cannot contain variable elements, one would have to define thouands of  
> standard names in order to be able to distinguish any layer between X and Y.
> 
> >
> > * What about
> > "atmosphere_mole_content_of_nitrogen_dioxide_between_XhPa_and_tropopa
> > use",
> > where X is 500?
> 
> Here again, we should try to find a general solution with an additional 
> attribute. In fact this discussion could/should be broader than for 
> atmospheric composition only, because similar problems might be present for 
> soil variables (soil layer between X and Y), and in the ocean as well. In 
> fact you can look into the definition of "frozen_water_content_of_soil_layer" 
> as an example. This is formulated rather vaguely, I think. So, perhaps this 
> could be a starting point for a more general discussion on how to uniformly 
> express "layers" in the CF model.
_______________________________________________
CF-metadata mailing list
[email protected]
http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata

Reply via email to