Dear Markus Thanks for your thoughtful consideration.
*** > > "surface_particle_number_concentration_at_stp_in_aerosol (and other similar > > names). Could this be said more clearly as > > surface_number_concentration_of_aerosol_in_air_at_stp? > > That would be consistent with existing names e.g. > > number_concentration_of_coarse_mode_ambient_aerosol_in_air" > > > > The standard names I proposed use the term "aerosol" according to its > > proper textbook definition, i.e. meaning the system of particles and > > carrier gas. Your wording implies that "aerosol" consists of particles > > only, which is a common, but colloquial jargon use of the term. I respect > > the use of "aerosol" in standard names so far, so I worded the proposed > > names to be backward compatible. > > I agree with you about the textbook definition but it appears that existing > CF names are not consistent. We have some names with a construction > number_concentration_of_X_aerosol_in_air > and some with > atmosphere_number_content_of_aerosol_particles > The difference between concentration and content is that the first is 3D and > the second a vertical integral, so that's not a problem. We could change the > first construction in existing names e.g. > number_concentration_of_ambient_aerosol_in_air to > number_concentration_of_X_aerosol_particles_in_air > i.e. insert "particles". Would that be correct? I suppose that in_air is > needed because aerosol is not necessarily in air (although the word looks > like it should be). It could be any gas. Then for consistency, could you use > number_concentration_of_aerosol_particles_at_stp_in_air > in your new names? > > MF: Yes, that would be an option, and probably also a rather consistent and > correct one! A few questions to this consensus from my side: > a) I realized that I partly used the "_in_air" qualifier in my proposed > standard names and partly not. I don't have any particular preference on > using it or not. In the geophysical context, I can't think of any other > carrier gas for an aerosol than air, so omitting "_in_air" would probably be > ok. On the other hand, we may also keep it in for the sake of precision in > wording. Should I consistently use "in_air" in my proposed names or not? If CF is never going to be applied to aerosols which aren't in air, then I suppose this phrase could be omitted. I am not an expert! Is this a safe assumption? I note that there are 42 standard names which contain both "aerosol" and "in_air". I suppose they ought to be modified if we decide that this phrase is not needed. On the whole I would vote to keep it, because more information is better. *** > b) In my proposed names for atmospheric aerosol optical properties, I > referred only to "aerosol", e.g. > volume_absorption_coefficient_in_air_due_to_dried_aerosol > even though the variable is supposed to refer to the absorption coefficient > of the particle phase only. Should I change these names as well, e.g. to > volume_absorption_coefficient_of_dried_aerosol_particles_at_stp_in_air ? I have to confess to not really understanding enough to answer this. Still, one way to answer it may be to ask, is there any conceivably useful distinction between these two? If not, there's probably no need to change them. *** > > You draw attention to the inclusion of "surface" in the above, but I'm not > > clear why it's there. Is the measurement actually exactly at the ground? If > > not, surface should be omitted, and the height indicated by a numerical > > coordinate, or some other phrase e.g. in_atmosphere_boundary_layer (that > > one already appears in the stdname table). > > > > The term "surface" is used according to the description given in the > > "Guidelines for Construction of CF Standard Names" at > > http://cf-pcmdi.llnl.gov/documents/cf-standard-names/guidelines > > 4-5 m above the ground is not really the surface. The constructions surface_ > and _at_surface mean the same thing; they're chosen according to which seems > to be easier to read in the context concerned. So I'd prefer > > > 2) Include in the definition the requirement that the > > sampling height needs to be included as numerical > > coordinate. > > That's what we do with screen-height temperature, for example. Although it is > often called "surface air temperature", its standard name is air_temperature > and it has a height coordinate (e.g. 1.5 m or 2.0 m). > > MF: Ok, I see the problem. I would in any case swap the "surface_" qualifier > with "_at_surface" to avoid confusion, and requiring a height coordinate > makes sense anyway. I'm still not quite happy with dropping "_at_surface" > altogether. When the standard name is used in a data discovery portal > independently of a file, having this information included in the name will be > a big help to the user. He will see immediately what's in the file without > opening it. You say that "_at_surface" means the exact interface of > atmosphere and ground. However, even for models, this definition will depend > very much on model resolution. "_at_surface" for a GCM will be very different > from "_at_surface" for a microscale boundary layer model. How's that > different from my proposed use of "_at_surface? Could I use "_at_surface" AND > require stating the sampling height as vertical coordinate? I don't think it would be consistent with the guidelines or existing names to do that. The intention is that named surfaces and coordinate values are alternatives. Data discovery is a different purpose from the use metadata provided by CF. I think this sounds a bit unobliging, doesn't it, but I can't think of a good solution. You could presumably provide a long_name or some other attributes to describe the quantity further. *** > I wonder if sizing_relative_humidity could be made more self-explanatory > somehow, and also whether for the other RH you also need a new and more > explicit standard name, in order to make the distinction clearer? > > MF: I pondered this one for a good while. I think a separate humidity > variable for indicating the humidity of a CCN concentration would add > confusion. After all, this "relative_humidity" would be a co-ordinate > variable to the CCN concentration or CCN number size distribution, so it's > clear immediately what is meant. The "sizing_relative_humidity" would be > auxiliary information. I could rename it to "relative_humidity_at > aerosol_particle_size_selection", which would be rather self-explaining. > Would that be ok? I think so, yes. Maybe "for" instead of "at"? Thank you! Best wishes Jonathan _______________________________________________ CF-metadata mailing list [email protected] http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
