Talk:CF Standard Names - Discussed Atmospheric Chemistry and Aerosol Terms
From Federation of Earth Science Information Partners
Important issues are marked in RED , please COMMENT!
Vincent-Henri PEUCH - replies by Christiane Textor (CT) and Jonathan Gregory (JG)
- Thank you for this work! A few minor comments/suggestions/questions :
- 1) "nitrogen_monooxide" -> "nitrogen monoxide"?
- (CT) both names are allowed by IUPAC, and I chose the monooxid-version to be consistent with dioxid - but I change it.
- 2) there is always the problem that "NOy" has no fully agreed definition in the literature... It is perhaps unwise to use it in the name? Could we use "total_nitrogen_oxides" instead?
- (CT) good idea, I changed it.
- We have also to chose :
- - if only the species with the name listed go in the sum
- - or if it is up to the modeller to select all nitrogen oxides in his/her chemical scheme.
- (CT) would go for the second option. Included in the comment/explanation.
- The explanation is "includes all nitrogen oxides included in the model, to be specified in by the modeller in the long_name attribute if possible ???" I doubt however, that this is CF conform, as long_names are not part of CF...
- 3) the description of the "troposphere_content*" variables is not enough detailed because it is indeed verticaly integrated, but up to the tropopause only. We can specify in the explanation "up to the tropopause level", but we probably also have to specify the tropopause definition to be used (2PVU,380K ?) as the value is quite sensitive to the specific criterion used (for species with strong vertical gradients at the tropopause like ozone). A drawback of specifying is that any other type of hypotheses (other "tropopause" definition : 150 ppb of ozone, 100 hPa,...) or other ways of computation (specific tracer in the model) would then no longer fit with the name. I don't know the solution...
- (CT)Very good point! I explain my ideas below. The 150ppb O3 isosurface is a good measure for atmospheric chemsitry problems and has been used in ACCENT/PHOTOCOMP. Changed to up_to_chemical_tropopause_content as in e.g. "up_to_chemical_tropopause_content_of_ozone_in_air". See also below.
- 4) we could add, for ozone at least, "total_atmosphere_content_of_*_in_air" (in Dobson units for ozone, mol/m2 for others if needed).
- (CT) Dobson units are not possible within the concept of CF which is based on UDUNITS, see also the discussion on units
- 5) the variable "mole_fraction_of_ozone_from_stratosphere_in troposphere" is a modeler's concept, with no chance of being measured. The way it is implemented in a model has an impact on the actual values, due to non linearities etc... I would not be in favor of including it as a standard variable. What do you think?
- (CT) I ask this question to the GEMS-GRG and HTAP communities
- (JG) Remark by Vincent-Henri Peuch: "The variable mole_fraction_of_ozone_from_stratosphere_in troposphere is a modeler's concept." Yes, that is true, but if modellers want to store this quantity, and compare it among models, then it should be given a standard name. There are many quantities like this, which aren't really observable. To begin with CF was designed for models, rather than the real world!
- (CT) At the GEMS GRG meeting in Toulouse we agreed on including this variable in the diagnostics. mole_fraction_of_ozone_from_stratosphere_in troposphere includes the question of what would be the tropopause, which is discussed below.
- 6) add "mole_fraction_of_lead_in_air". Lead is the radioactive daughter of Radon, with wet scavenging as principal sink. There are some observations and it is useful to evaluate models, as decided in GRG+VAL.
- (CT) done
Christiane Textor: Tropopause definition comments by Jonathan Gregory (JG)
- Several tropopause definitions exist:
- - chemical tropopause (150 ppb O3 isosurface)
- - lapse rate tropopause (the lowest level at which the lapse rate decreases to 2 °C/km or less,
- provided that the average lapse rate between this level and all higher levels within 2 km does not exceed 2 °C/km. WMO definition of Tropopause)
- - potential vorticity (PVU2 (at the 2 PVU surface) or PVU1.5 (at the 1.5 PVU surface))
- - potential temperature surface
- For chemical purposes it seems appropriate to define up_to_chemical_tropopause with the explanation from the surface up to chemical tropopause (150 ppb O3 isosurface).
- (JT) You have a number of names of the form up_to_chemical_tropopause_content_of_X_in_air. This order is rather unnatural, I'd say. Also in_air probably isn't right, as here you mean a large-scale quantity. I know that we discussed whether "atmosphere" goes at the start or the end, and I remarked it usually was at the start, but a complete phrase is more awkward at the start. Would you consider X_content_of_chemical_troposphere or X_content_of_atmosphere_below_chemical_tropopause?
- (CT) I have put in_air because satellites only see the fraction in the gas phase, so atmosphere is not correct. below_chemical_tropopause sounds good. This would lead to X_mole_content_below_chemical_tropopause_in_air or X_mole_content_in_air_below_chemical_tropopause. What do you think?
- (JG) If you need to distinguish different definitions of the tropopause, this could be done by defining different standard names. This issue is rather like the definition of the ocean mixed layer, for which we have several standard names:
- If there are particular numbers which are needed for the definition (like your 150 ppb O3) they could be specified as standard name parameters (but we still haven't agreed the mechanism for this!).
- (CT) This would then give X_mole_content_in_air_below_tropopause_defined_by_150ppv_O3_iso_surface.
Jonathan Gregory - replies by Christiane Textor (CT)
- The IUPAC issue is not settled yet, perhaps? For example, I think that acetyl and formaldehyde aren't IUPAC names. I may be wrong!
- (CT) formaldehyde is accepted by IUPAG ("In contrast to such systematic names, there are traditional names, semisystematic or trivial, which are widely used for a core group of common compounds. Examples are "acetic acid", "benzene", "cholesterol", "styrene", "formaldehyde", "water", "iron". Many of these names are also part of general nonscientific language and are thus not confined to use within the science of chemistry. They are useful, and in many cases indispensable (consider the alternative systematic name for cholesterol, for example). Little is to be gained, and certainly much to be lost, by replacing such names. Therefore, where they meet the requirements of utility and precision, and can be expected to continue to be widely used by chemists and others, they are retained and, for the most part, preferred in this Guide."). For PAN, I am not sure, I did not find it in the red book, but in some other web articles peroxyacetyl nitrate is given a the IUPAC name. Until somebody knows for sure I assume now that it is a IUPAC name.
- I am sorry that I have not noticed before about some quantities being kg and others mol. I think these must be distinguished by the standard name; even though you make certain choices for your needs, other people might like different choices, and so we may eventually need names for both mass fluxes and mole fluxes for any species. In many existing names, quantities like flux, production and amount are in kg unless specifically stated e.g. rainfall_flux (kg m-2 s-1), gross_primary_productivity_of_carbon (kg m-2 s-1), surface_snow_amount (kg m-2). Perhaps you would not like that default? If not, I think we should use the phrases mass|mole_content, mass|mole_flux, mass|mole_amount, mass|mole_production, mass|mole_destruction wherever you need them.
- (CT) I agree that it is clearer to add the kind of flux concerned. I Changed this and added mass/mole/number everywhere.
- I think it's a good idea to avoid the word "total" if we can, as it is not always obvious what aspect is being totalled! total_nitrogen_oxides is probably clear enough, but maybe all_nitrogen_oxides might be better?
- (CT) But what does 'all' include? anyway, I changed it.
- You also have total_aerosol, which in fact you define as dry aerosol. Could this name be simply dry_mass_concentration_of_aerosol? I suppose that "aerosol" alone ought to be taken to refer to all aerosol species.
- (CT) changed.
- The standard_name guidelines propose the construction surface_X, and there are many existing names like that, rather than X_at_surface. Hence e.g. surface_wet_deposition_flux_of_sulfur_dioxide instead of wet_deposition_flux_of_sulfur_dioxide_at_surface. Is that all right? Similarly, the guidelines and many existing names have toa_X rather than X_at_top_of_atmosphere. (The surface which are single words come at the beginning, and the surfaces which are multi-word phrases come after "at".)
- (CT) Changed!
- I would suggest that due_to_turbulence comes after of_X, because there could be a quantity dry_deposition_flux_of_X, of which dry_deposition_flux_of_X_due_to_turbulence is a part.
- (CT) It is :dry_deposition_mole_flux_of_X=dry_deposition_mole_flux_of_X_due_to_turbulence+dry_deposition_mole_flux_of_X_due_to_sedimentation, so I would rather like to leave the due_to_turbulence and due_to_sedimentation close to deposition, so I would like to change it to dry_deposition_mole_flux_of_X=dry_deposition_due_to_turbulence_mole_flux_of_X+dry_deposition_due_to_sedimentation_mole_flux_of_X, ok?
- flash_frequency_due_to_lightning. I suggest lightning_flash_frequency; the flashes specifically belong to lightning, whereas due_to suggests that flashes are caused by a number of processes of which lightning is only one.
- (CT) Changed.
- grid_box_area. Do you have a data variable containing area? If the only purpose of this is to provide weights, you can use cell_measures to supply the area variable (CF 7.2). However if you do have them as data variables, I agree you need a standard name. I'd suggest surface_area. I think this is really the quantity, isn't it. It is analogous to sea_ice_area (m2), for example. These are extensive quantities in space; they implicitly depend on the size of the grid box.
- (CT) I reread the CF documentation yesterday and realized the possibility of using the cell methods for the grid information, sorry for that. We do not need an additional standard_name.
- atmosphere_reference_pressure_at_surface. Is this a data variable? It sounds like a model-dependent constant. Since we have surface_air_pressure as a standard name, I'd suggest reference_surface_air_pressure for this.
- (CT) It is a model constant, e.g. 10^5, indicating which pressure was exactly used in the models. I changed the name according to your suggestion.
- In your construction rules, I think there should be a _ in in_cloud_water and in_rain_water, like there is in in_sea_water.
- (CT) Changed.
- cell_methods question. In other situations like this we have taken the view that the rate (X s-1) and the integral (X) are different quantities, with different standard names. They are related in a simple way, of course, but it's not quite as simple as a statistical relationship like mean or standard deviation, for which cell_methods are intended. Thus, for instance, we have different names for precipitation_flux (kg m-2 s-1) and precipitation_amount (kg m-2). In your case, we would define a standard name for surface_dry_deposition_mole_amount_of_sulfur_dioxide_due_to_turbulence (mol m-2). Does that make sense? Its cell_methods would be sum, as you say (the the default for extensive quantities), and its time-bounds indicate the period of integration. (See discussion in CF 7.3.)