Talk:Units

From Federation of Earth Science Information Partners
Revision as of 09:47, November 27, 2006 by ERobinson (talk | contribs)

Go back to Start page for Atmospheric Chemistry and Aerosol Names PLEASE DO NOT USE THE NAVIGATION BAR ON THE LEFT HAND SIDE!


UDUNITS have to be used for the standard_names according to the CF convention. This is however not always obvious for chemistry and aerosol variables, where also other units are common. This page provides a discussion on this matter. See also Unidata Supported Units || Unit Converter, UDUNITS

Important issues are marked in RED , please COMMENT!

Agreed items of former discussions on the "Units" page can be found here.


JGregory: Don't mix units with the Compound

However, kgC and kgS aren't SI units. I think the unit has to be kg. This implies that it's the standard name which must somehow indicate that it is the mass of C or mass of S which is being referred to, rather than the mass of the compound e.g. dry_deposition_flux_of_sulfur_as_sulfate_at_surface. --JonathanGregory 16:51, 22 May 2006 (EDT)

......CTextor:

Very good idea, I will change this. --Christiane 17:11, 2 June 2006 (EDT)

CTextor: SI units

Try to use SI units if allowed within udunits (e.g. not hours but seconds) --Christiane 12:11, 2 June 2006 (EDT)

RanjeetSokhi(RS) / CTextor (CT): remarks on version June 12 from RS and answers from CT

  • RS: Table 1 mass_concentration is usually stated in ug/m3 ie micrograms per cubic metre - this is conventional and used in all air quality literature (scientific and policy)
CT: --Christiane 13 June 2006 (EDT)


CTextor (CT) / Martin Schultz (MS) / Sebastian Rast (RS): UDUNITS in atmospheric chemistry: concentration Christiane 3 July 2006 (EDT)

CT: usually the concentration of chemical compound are not given in UDUNITS, see the remark of Ranjeet above for the air quality community, where aerosol mass_concentration is usually stated in ug/m3ich, or the global chemistry community where the tropospheric column of a gas is given in [molec/cm2], and the dry deposition in [g/m2/month]. This should be changed
cm2 -> m2
molec -> mole
g -> kg
We had a discussion (in german) that this would be difficult, because all papers are in the non-udunits, but at the same time it would be good to agree on one standard within GEMS and HTAP.

CTextor (CT) / Martin Schultz (MS) / Sebastian Rast (RS) / Jonathan Gregory (JG): UDUNITS in atmospheric chemistry: time issuesChristiane 3 July 2006 (EDT)

  • MS: (translated from german by ct) The use of non-UDUNITS can be of advantage, if it concern a variable that is integrated over a certain period, e.g. a the dry deposition flux over a month would need to have mole/m2/month. This is not the same as a mean dry dep. which could be given in mole/m2/s.
  • CT: (corrected) I think the cell method of CF should be applied here, question asked to Jonathan Gregory
  • JG 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 default for extensive quantities), and its time-bounds indicate the period of integration. (See discussion in CF 7.3.)
  • CT This would give a file like this for 2 months, Martin, Sebastian, what do you think?

dimensions:

lat = 18 ;
lon = 36 ;
pres = 15 ;
time = UNLIMITED ;//(2 currently

variables:

float drydep_SO2(time,pres,lat,lon) ;
ddep_SO2:standard_name = "surface_dry_deposition_due_to_turbulence_mole_amount_of_sulfur_dioxide" ;
ddep_SO2:units = "mole/m2" ;
ddep_SO2:cell_methods="time: sum (over month)"
float lon(lon) ;
lon:long_name = "longitude" ;
lon:units = "degrees_east" ;
float lat(lat) ;
lat:long_name = "latitude" ;
lat:units = "degrees_north" ;
float pres(pres) ;
pres:long_name = "pressure" ;
pres:units = "hPa" ;
double time(time) ;
time:long_name = "time" ;
time:units = "days since 1990-1-1 0:0:0" ;
time:bounds = "time_bnds";
double time_bnds(time,nv);

data:

time = 1., 2.;
time_bnds = 0.,1.,1.,2.;
  • CT We will however get the same kind of problem with the chemical production/destructions, which are given per months in GEMS-GRG. I guess we should use:;
chemical_net_production_rate_of_mole_fraction_of_ozone 1/s=mole/mole/s
chemical_gross_production_rate_of_mole_fraction_of_ozone 1/s=mole/mole/s
chemical_destruction_rate_of_mole_fraction_of_ozone 1/s=mole/mole/s
and for the integral over a month using the cell method:
chemical_net_production_of_mole_fraction_of_ozone mole/mole
chemical_gross_production_of_mole_fraction_of_ozone mole/mole
chemical_destruction_of_mole_fraction_of_ozone mole/mole
  • JG I agree that the averaging of "amount" variables could be a problem - This problem comes up with other amount variables too, like precipitation. One solution could be to recognise that if you are averaging them, you are maybe treating them as a rate, not an amount. Hence the standard name should be the one of rate, not amount. The unit does not have to be kg m-2 s-1. It could be kg m-2 day-1, for instance. Although udunits allows "month" we don't recommend it because its definition is not a calendar month but a particular (constant) number of seconds - probably not what you want. But the time bounds of the variable should always indicate the meaning period.

With both rates and amounts, climatological time bounds may help as well, with which you can record that it is (for instance) the January mean over a number of years (see CF 7.4). I hope that tools such as nco may be extended to produce the cell_methods attribute to describe this, since CF is becoming quite important; in fact we could request such an extension.

CT But a quantity with a unit of kg m-2 day-1 is a flux rather than a rate which would have kg/day? See also http://wiki.esipfed.org/index.php/CF_Standard_Names_-_Future_Atmospheric_Chemistry_and_Aerosol_Terms#Table_1:_Components_to_construct_full_CF_standard_names and there the chemical_production and chemical_production_rate...

JG "Rate" is not a precisely defined term, I would say. It has the idea of s-1 but doesn't imply what other dimensions. I suppose a surface emission in kg m-2 s-1 might be called either a rate or a flux. Which is more usual in atmos chem?

CT nco is essential for us! As it is not yet existing I think we should define specific variables, however, I do not know how at the moment. Could it be e.g.

surface_dry_deposition_due_to_turbulence_mole_monthly_flux_of_sulfur_dioxide
surface_dry_deposition_due_to_turbulence_mole_amount_of_sulfur_dioxide using the cell method? or what else?

JG No, I don't think we should put a time unit in the standard name. If you are dealing with real calendar months, I think you'll have to use an amount (no s-1). The time bounds will show the months to which it applies. Is this a bad solution for you?