Hello Jonathan, Karl,
I originally proposed use of "amount" because that is used in other precipitation terms such as rainfall_amount, graupel_fall_amount, but, on closer consideration, I can see that "amount" is used for terms expressed as mass per unit area, and that would give the wrong meaning here. We don't want the mass per unit area on snow divided by the mass per unit area on the whole grid cell, but rather, as Jonathan has spelled out, the mass on snow divided by the mass on the whole grid cell. Hence, I support Karl's suggestion of using "mass" in the name, rather than "amount". I also agree with other changes suggested and summarised in Karl's email, regards, Martin ________________________________ From: CF-metadata <cf-metadata-boun...@cgd.ucar.edu> on behalf of Karl Taylor <taylo...@llnl.gov> Sent: 01 June 2018 17:45 To: cf-metadata@cgd.ucar.edu Subject: Re: [CF-metadata] Precipitation fractions for LS3MIP Hi Jonathan and Martin, I agree, rainfall and snowfall are better because they are more explicit. I guess precipitation by itself is already clear. "onto" instead of "on" would also be less prone to misinterpretation as would be using "surface_snow" in place of "snow". As for the meaning, I interpreted it as you did in terms of the fraction of falling mass deposited onto the snow-covered portion of the grid cell. I think of snowfall "amount" and "rainfall" amount as answering the question "how much snow (rain) did we get yesterday?". The answer is usually expressed in terms of a depth measure (e.g., "cm"), not mass. Interpreted in this way, it would be confusing to say fraction_of_snowfall_amount_falling_onto_surface_snow because it might suggest that this is the "fraction_of_snowfall_depth_falling_onto_surface_snow, which isn't what we mean. So, I would favor fraction_of_precipitation/rainfall/snowfall_mass_falling_onto_surface_snow best regards, Karl On 6/1/18 12:34 AM, Jonathan Gregory wrote: > Dear Martin and Karl > > I'm not sure I have understood this. Writing out what Martin first said > at greater length, does it mean > > mass of rainfall (in kg, meaning the area- and time-integral rainfall flux > in kg m-2 s-1) which falls onto the snow-covered portion of the gridbox in > a certain time-interval, divided by the mass of rainfall (in the same sense) > which falls onto the entire gridbox in the same time-interval? > > and the same for "precipitation" and "snowfall" instead of "rainfall"? > > If so, I have some suggestions regarding >> fraction_of_precipitation_mass_falling_on_snow >> fraction_of_rain_mass_falling_on_snow >> fraction_of_snow_mass_falling_on_snow > First, I think we could use "rainfall_amount" and "snowfall_amount" as Martin > first said, because these are time-accumulations, aren't they? They are > actually the gridbox precipitation/rainfall/snowfall_amount multiplied by the > area fraction of the gridbox which is covered by snow. I think this would > be preferable because we currently don't have terms with "mass" for > precipitation. On the other hand, you could prefer "mass" if you write it as > I did above and regard it as a mass fraction. In that case I would advocate > "precipitation/rainfall/snowfall_mass" i.e. insert "fall". This > is because "snowmass" could be misunderstood to mean the mass of lying snow - > a sense in which the word is often used. Including "fall" would make these > mass terms look like the existing amount, content, flux and rate terms. > However > I still prefer "amount". > > Second, I think "onto", as Martin said, is better than "on". It makes it > clearer (again) that it's been added. > > Third, instead of onto_snow I would say onto_surface_snow, because we have > consistently used the phrase surface_snow to mean snow lying on the ground, > to be absolutely clear we don't mean snowfall. Since we have to talk about > both snowfall and lying snow in one standard name here, clarity is useful > about which is which. If it means snow and ice, not just snow, then we can say > surface_snow_and_ice, which is a phrase already used. > > Thus I end up with > > fraction_of_precipitation/rainfall/snowfall_amount_falling_onto_surface_snow > and if that's still not quite transparent to parse, we could consider saying > which_falls instead of falling. Neither is yet used in standard names. > > Best wishes > > Jonathan > > ----- Forwarded message from Martin Juckes - UKRI STFC > <martin.juc...@stfc.ac.uk> ----- > >> Date: Fri, 1 Jun 2018 06:42:45 +0000 >> From: Martin Juckes - UKRI STFC <martin.juc...@stfc.ac.uk> >> To: Karl Taylor <taylo...@llnl.gov>, "cf-metadata@cgd.ucar.edu" >> <cf-metadata@cgd.ucar.edu> >> Subject: Re: [CF-metadata] Precipitation fractions for LS3MIP >> >> Hi Karl, >> >> >> Thanks, that looks better. I was trying to stick close to existing patterns, >> but the outcome was a bit contrived and, I agree, not very satisfactory. >> >> >> Sorry about the confusion, the two terms we need are: >> >> fraction_of_rain_mass_falling_on_snow >> fraction_of_snow_mass_falling_on_snow >> >> >> However, I need to check with LS3MIP whether they intend "snowfall" to >> include snow and ice, as for the CMIP5 variable "prsn", >> >> >> regards, >> >> Martin >> >> >> ________________________________ >> From: CF-metadata <cf-metadata-boun...@cgd.ucar.edu> on behalf of Karl >> Taylor <taylo...@llnl.gov> >> Sent: 31 May 2018 19:11 >> To: cf-metadata@cgd.ucar.edu >> Subject: Re: [CF-metadata] Precipitation fractions for LS3MIP >> >> Hi Martin, >> >> For CMIP6, I think we define precipitation = rainfall + snowfall . >> >> So do you want to collect "precipitation amount" or "rainfall amount" >> (in addition to "snowfall amount")? You first mention "rainfall" below, >> but later propose a name for "precipitation". >> >> I also think it is a bit confusing that in this name "amount" refers to >> the amount accumulated over some time-period (e.g., a day or a month), >> rather than the total "content" that exists at a given point in time. >> >> I might suggest as alternatives: >> >> fraction_of_precipitation_mass_falling_on_snow >> fraction_of_rain_mass_falling_on_snow >> fraction_of_snow_mass_falling_on_snow >> >> best regards, >> Karl >> >> >> On 5/31/18 9:56 AM, Martin Juckes - UKRI STFC wrote: >>> Hello All, >>> >>> >>> There are two LS3MIP variables for fractions of precipitation onto snow: >>> fraction of rainfall on snow and fraction of snowfall on snow. In both >>> cases the fraction is calculated as the mass of precipitation falling on >>> snow divided by the total mass of precipitation in the grid cell. >>> >>> >>> Since we are dealing with mass, the term should start from >>> precipitation_amount and snowfall_amount. >>> >>> >>> We have terms like lwe_thickness_of_precipitation_amount to describe >>> properties of the precipitation amount, and we have >>> precipitation_flux_onto_canopy do describe a flux onto a particular >>> surface, so I propose: >>> >>> fraction_of_precipitation_amount_onto_snow and >>> >>> fraction_of_snowfall_amount_onto_snow, both with units '1', >>> >>> >>> regards, >>> >>> Martin >>> >>> >>> >>> >>> _______________________________________________ >>> CF-metadata mailing list >>> CF-metadata@cgd.ucar.edu >>> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata >> _______________________________________________ >> CF-metadata mailing list >> CF-metadata@cgd.ucar.edu >> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata >> _______________________________________________ >> CF-metadata mailing list >> CF-metadata@cgd.ucar.edu >> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata > ----- End forwarded message ----- > _______________________________________________ > CF-metadata mailing list > CF-metadata@cgd.ucar.edu > http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata _______________________________________________ CF-metadata mailing list CF-metadata@cgd.ucar.edu http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata _______________________________________________ CF-metadata mailing list CF-metadata@cgd.ucar.edu http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
