term description of stratiform_rainfall_rate misleading?

[atmodatcode]

Hello,

I noted that the term description of CF standard names that contain the term "rainfall" does not contain a definition what "rainfall" means. Instead, it contains a definition what precipitation is.

For example:

CF Standard name: stratiform_rainfall_rate _alias: large_scale_rainfallrate Term description: Stratiform precipitation, whether liquid or frozen, is precipitation that formed in stratiform cloud.

In comparison CF Standard name: lwe_stratiform_precipitation_rate _alias: lwe_large_scale_precipitationrate Term description: Stratiform precipitation, whether liquid or frozen, is precipitation that formed in stratiform cloud. "Precipitation" in the earth's atmosphere means precipitation of water in all phases. The abbreviation "lwe" means liquid water equivalent. "Precipitation rate" means the depth or thickness of the layer formed by precipitation per unit time.

Wouldn't it make sense to more precisely define what rainfall means? E.g. Term description for stratiform_rainfall_rate: Stratiform rainfall is liquid precipitation that formed in stratiform cloud.

Thanks a lot. Regards Angelika

[JonathanGregory]

Dear Angelika

I agree that the there should be definitions of rain and rainfall. In standard names, we have rainfall when we're thinking of it arriving at or collecting on the surface, and rain when we're thinking of it within the atmosphere. The definition of mass_concentration_of_rain_in_air includes

"Rain" means drops of water falling through the atmosphere that have a diameter greater than 0.5 mm.

Best wishes

Jonathan

[taylor13]

I also agree with the above.

[efisher008]

Hello @atmodatcode,

I am revisiting this issue after the migration of standard name proposals from the cf-conventions/discuss repo to the newly-created cf-conventions/vocabularies repo. This issue is one which has been open and unresolved for a while, since 2022.

I would suggest adding the definition of "rain" as mentioned by @JonathanGregory to the description of any names containing rainfall - a quick search of v85 of the standard names table suggests this is 25 names. This could be an amendment in line with the "harmonization of precipitation type standard names" proposed by @larsbarring in #51.

If everyone is in agreement, I suggest transferring this action over to that issue and concluding this one.

Best wishes, Ellie

[atmodatcode]

Dear Ellie, I think this is a very good solution. Thanks a lot. Kind regards Angelika

[taylor13]

I think this deserves more discussion. It seems to me to be inconsistent or at least confusing that we define rainfall_rate, rainfall_flux, and rainfall_amount as the rain arriving at the surface (and similarly for snowfall), but we have no way to specifically reference rain and snow passing through a particular level in the atmosphere.

For precipitation, on the other hand, apparently precipitation_flux (etc.) can apply at any level in the atmosphere (including the surface). Is there no need to keep track of snow and rain fluxes in the free atmosphere, but we need to know precipitation fluxes there?

If we were starting from scratch, I'd recommend that "snowfall" and "rainfall" be used anywhere in the atmosphere (and at the surface), similar to "precipitation". This means of course that precipitation, rainfall, and snowfall at the surface would need to have the scalar coordinate height=0. Alternatively, we could define both "precipitation_rate" and "surface_precipitation_rate" (and similarly for snowfall and rainfall) and not need a vertical coordinate for variables with standard name "surface_precipitation_rate".

We're not starting from scratch, so I don't know what we should do. In all past phases of CMIP (and perhaps in CMIP7), "precipitation_flux" is taken to mean "at the surface". If the standard name can apply to fluxes at any level in the atmosphere, the CMIP variables have not been properly fully defined in the metadata; we would need to include a scalar vertical coordinate with value height=0.

[JonathanGregory]

Dear Karl

I thought that precipitation_flux means the flux arriving at the surface, like precipitation_amount means the mass per unit area that has been collected at the surface. Have I missed some discussion about precipitation fluxes at other levels? Have standard names been requested for fluxes of precipitation, rain or snow, at any level in the atmosphere?

Best wishes

Jonathan

[taylor13]

Well, that's how I've interpreted it too -- as surface precipitation. But the description for these standard names includes the sentence:

"Precipitation" in the earth's atmosphere means precipitation of water in all phases.

This seems to leave open the level in the atmosphere where the flux/amount of precipitating water is measured. If we think precipitation_flux has only been used to describe surface precipitation, then we could simply modify the description along the lines of:

"Precipitation" is used to describe the water arriving at the earth's surface in any of its phases.

I think in meteorology "precipitation" is also used to describe water that is precipitating in the free atmosphere, and in some models we probably need to keep track of it at all altitudes (so that, for example, re-evaporation of precipitating water could be calculated). If we wanted to record precipitation at 5 km about the surface, what standard name would we use?

I guess the same question could be asked regarding "snow" and "rain". We don't have standard_names to describe falling snow and rain anywhere above the Earth's surface. I guess no one has requested them, but I fear we should have included "surface" somewhere in the current standard names for precipitation, snowfall, and rainfall.

[efisher008]

Dear @taylor13, @JonathanGregory,

Thank you for the further comments. Would you suggest changing the description of precipitation in all names which currently contain this, and also changing this in the CF phrasebank so the "new" version is suggested for future names?

We can also discuss if there needs to be a surface component in @JonathanGregory's definition for rain or rainfall.

Best wishes, Ellie

[larsbarring]

Currently I find 37 standard names containing the term precipitation, and I am not sure whether all of them refer to precipitation arriving onto the ground, or if some may in principle (or in practice) refer to precipitation leaving/arriving a certain layer of the atmosphere.

[larsbarring]

There is at least two standard names containing precipitation that specifically do not referring to precipitation falling onto the ground: atmosphere_mass_content_of_liquid_precipitation (unit kg m-2) and mass_fraction_of_precipitation_in_air (unit 1). And I think that there are several others that make sense in relation to some specific layer ofthe atmosphere, e.g. `tendency_of_air_temperature_due_to_stratiform_cloud_and_precipitation'.

Hence I think that we have to have the full picture and look at the full list of"precip" standard names. In #51 I was trying to do this with focus on incomplete descriptions. But with this more general discussion on what we mean by "precipitation" and the distinction between "rain" and "rainfall" the scope is considerably broader.

[larsbarring]

Below is a list of "precip" standard names. To compile it I search the standard name table with the following terms precipitation rain hail graupel drizzle snow, and then deleted irrelevant ones manually (I might have missed or accidentally deleted one or a few).

Standard name	Units
atmosphere_mass_content_of_graupel	kg m-2
atmosphere_mass_content_of_graupel_and_hail	kg m-2
atmosphere_mass_content_of_hail	kg m-2
atmosphere_mass_content_of_liquid_precipitation	kg m-2
atmosphere_mass_content_of_snow	kg m-2
convective_precipitation_amount	kg m-2
convective_precipitation_flux	kg m-2 s-1
convective_precipitation_rate NOW ALIASED TO lwe_convective_precipitation_rate (#210)	m s-1
convective_rainfall_amount	kg m-2
convective_rainfall_flux	kg m-2 s-1
convective_rainfall_rate	m s-1
convective_snowfall_amount	kg m-2
convective_snowfall_flux	kg m-2 s-1
graupel_and_hail_fall_amount	kg m-2
graupel_and_hail_fall_flux	kg m-2 s-1
graupel_fall_amount	kg m-2
graupel_fall_flux	kg m-2 s-1
hail_fall_amount	kg m-2
hail_fall_flux	kg m-2 s-1
lwe_convective_precipitation_rate	m s-1
lwe_convective_snowfall_rate	m s-1
lwe_precipitation_rate	m s-1
lwe_snowfall_rate	m s-1
lwe_stratiform_precipitation_rate alias: lwe_large_scale_precipitation_rate	m s-1
lwe_stratiform_snowfall_rate alias: lwe_large_scale_snowfall_rate	m s-1
lwe_thickness_of_convective_precipitation_amount	m
lwe_thickness_of_convective_snowfall_amount	m
lwe_thickness_of_precipitation_amount	m
lwe_thickness_of_snowfall_amount	m
lwe_thickness_of_stratiform_precipitation_amount alias: lwe_thickness_of_large_scale_precipitation_amount	m
lwe_thickness_of_stratiform_snowfall_amount alias: lwe_thickness_of_large_scale_snowfall_amount	m
mass_concentration_of_drizzle_in_air	kg m-3
mass_concentration_of_rain_in_air	kg m-3
mass_fraction_of_graupel_and_hail_in_air	1
mass_fraction_of_graupel_in_air	1
mass_fraction_of_hail_in_air	1
mass_fraction_of_liquid_precipitation_in_air alias: mass_fraction_of_rain_and_drizzle_in_air alias: mass_fraction_of_rain_in_air	1
mass_fraction_of_precipitation_in_air	1
mass_fraction_of_rainfall_falling_onto_surface_snow	1
mass_fraction_of_snow_in_air	1
mass_fraction_of_solid_precipitation_falling_onto_surface_snow	1
precipitation_amount	kg m-2
precipitation_flux	kg m-2 s-1
precipitation_flux_containing_17O	kg m-2 s-1
precipitation_flux_containing_18O	kg m-2 s-1
precipitation_flux_containing_single_2H	kg m-2 s-1
precipitation_flux_onto_canopy alias: precipitation_flux_onto_canopy_where_land	kg m-2 s-1
rainfall_amount	kg m-2
rainfall_flux	kg m-2 s-1
rainfall_rate	m s-1
shallow_convective_precipitation_flux	kg m-2 s-1
snow_grain_size	m
snow_transport_across_line_due_to_sea_ice_dynamics	kg s-1
snowfall_amount	kg m-2
snowfall_flux	kg m-2 s-1
solid_precipitation_flux	kg m-2 s-1
solid_precipitation_flux_containing_17O	kg m-2 s-1
solid_precipitation_flux_containing_18O	kg m-2 s-1
solid_precipitation_flux_containing_single_2H	kg m-2 s-1
stratiform_graupel_fall_amount	kg m-2
stratiform_graupel_flux alias: large_scale_graupel_flux	kg m-2 s-1
stratiform_precipitation_amount alias: large_scale_precipitation_amount	kg m-2
stratiform_precipitation_flux alias: large_scale_precipitation_flux	kg m-2 s-1
stratiform_rainfall_amount alias: large_scale_rainfall_amount	kg m-2
stratiform_rainfall_flux alias: large_scale_rainfall_flux	kg m-2 s-1
stratiform_rainfall_rate alias: large_scale_rainfall_rate	m s-1
stratiform_snowfall_amount alias: large_scale_snowfall_amount	kg m-2
stratiform_snowfall_flux alias: large_scale_snowfall_flux	kg m-2 s-1
tendency_of_air_temperature_due_to_stratiform_cloud_and_precipitation	K s-1
tendency_of_air_temperature_due_to_stratiform_cloud_and_precipitation_and_boundary_layer_mixing	K s-1
tendency_of_air_temperature_due_to_stratiform_precipitation alias: tendency_of_air_temperature_due_to_large_scale_precipitation	K s-1
tendency_of_atmosphere_mass_content_of_water_vapor_due_to_sublimation_of_surface_snow	kg m-2 s-1
tendency_of_atmosphere_mass_content_of_water_vapor_due_to_sublimation_of_surface_snow_and_ice alias: surface_snow_and_ice_sublimation_flux	kg m-2 s-1
tendency_of_mass_fraction_of_stratiform_cloud_condensed_water_in_air_due_to_autoconversion_to_rain	s-1
tendency_of_mass_fraction_of_stratiform_cloud_condensed_water_in_air_due_to_autoconversion_to_snow	s-1
tendency_of_mass_fraction_of_stratiform_cloud_ice_in_air_due_to_accretion_to_snow	s-1
tendency_of_mass_fraction_of_stratiform_cloud_ice_in_air_due_to_convective_detrainment	s-1
tendency_of_mass_fraction_of_stratiform_cloud_ice_in_air_due_to_melting_to_rain	s-1
tendency_of_mass_fraction_of_stratiform_cloud_ice_in_air_due_to_riming_from_rain	s-1
tendency_of_mass_fraction_of_stratiform_cloud_liquid_water_in_air_due_to_accretion_to_rain	s-1
tendency_of_mass_fraction_of_stratiform_cloud_liquid_water_in_air_due_to_accretion_to_snow	s-1
tendency_of_mass_fraction_of_stratiform_cloud_liquid_water_in_air_due_to_convective_detrainment	s-1
tendency_of_sea_ice_amount_due_to_conversion_of_snow_to_sea_ice alias: tendency_of_sea_ice_amount_due_to_snow_conversion	kg m-2 s-1
tendency_of_specific_humidity_due_to_stratiform_cloud_and_precipitation	s-1
tendency_of_specific_humidity_due_to_stratiform_cloud_and_precipitation_and_boundary_layer_mixing	s-1
tendency_of_specific_humidity_due_to_stratiform_precipitation alias: tendency_of_specific_humidity_due_to_large_scale_precipitation	s-1
thickness_of_convective_rainfall_amount	m
thickness_of_convective_snowfall_amount	m
thickness_of_rainfall_amount	m
thickness_of_snowfall_amount	m
thickness_of_stratiform_rainfall_amount alias: thickness_of_large_scale_rainfall_amount	m
thickness_of_stratiform_snowfall_amount alias: thickness_of_large_scale_snowfall_amount	m

[larsbarring]

Further, in the list above I just saw the two standard names mass_concentration_of_drizzle_in_air and mass_concentration_of_rain_in_air (both having units kg m-3). This indicates that the termrain is not only used in relation to rainfallto distinguish between "in the air" and "onto the surface" (which might be ground or canopy), but also to distinguish between different drop size distributions and formation processes.

[larsbarring]

After those more general comments I would like to come back to the specific comments by Karl @taylor13; here:

If we were starting from scratch, I'd recommend that "snowfall" and "rainfall" be used anywhere in the atmosphere (and at the surface), similar to "precipitation". This means of course that precipitation, rainfall, and snowfall at the surface would need to have the scalar coordinate height=0.

and here

I think in meteorology "precipitation" is also used to describe water that is precipitating in the free atmosphere, and in some models we probably need to keep track of it at all altitudes (so that, for example, re-evaporation of precipitating water could be calculated). If we wanted to record precipitation at 5 km about the surface, what standard name would we use?

I guess the same question could be asked regarding "snow" and "rain". We don't have standard_names to describe falling snow and rain anywhere above the Earth's surface. I guess no one has requested them, but I fear we should have included "surface" somewhere in the current standard names for precipitation, snowfall, and rainfall.

What if we amend the description to the standard names where this is relevant to recommend inclusion of a height coordinate, which might scalar, and if not present it defaults to 0to represent the surface?

[JonathanGregory]

Thanks for the helpful list, Lars. As far as I can see, "fall" appears with snow, rain, hail and graupel for all standard names which I believe to indicate the flux arriving on the ground, and "fall" doesn't appear for quantities that relate to constituents of the atmosphere. This is already a good consistency.

There's one exception: I think stratiform_graupel_flux should be changed to stratiform_graupel_fall_flux. We already have graupel_fall_flux.

For precipitation, we haven't done the same. There's no phrase "precipitation fall" in English, because "precipitation" has the idea of falling anyway. Wiktionary says it means

(meteorology) Any or all of the forms of water particles, whether liquid or solid, that fall from the upper atmosphere (e.g., rain, hail, snow or sleet). It is a major class of hydrometeor, but it is distinguished from cloud, fog, dew, rime, frost, etc., in that it must fall. It is distinguished from cloud and virga in that it must reach the ground. [My bold text.]

(Also, for amusement, Wiktionary says that it comes from Latin praecipito "cast down, throw headlong" from praeceps "head foremost", from prae- "before" + caput "head".)

In the standard names listed by Lars, precipitation means the flux arriving on the ground, or the hydrometeors in the air, or the process of formation of the hydrometeors (in due_to_[ ...]precipitation). I feel that it doesn't matter than we don't have the distinction for precipitation, except for the possibility which Karl asked about: If we wanted to record precipitation above the surface, what standard name would we use? We haven't been asked for this yet, but it's possible. It's also possible the existing names for precipitation flux, as well as the *fall fluxes, might be misinterpreted as meaning the flux in the air (rather than arriving on the ground).

I suggest that we clarify the descriptions of the precipitation flux/amount/rate standard names, and those of the *fall quantities, to say they refer to stuff arriving at the ground.

I've just seen Lars's most recent comment. I'm not sure it was there when I began on this! Yes, in addition to clarifying that they mean the flux falling on the ground, we could indeed say that any of those flux quantities are also allowed to refer to a flux above the ground, if the data variable has a vertical coordinate. That is a neat solution, which avoids defining new names or changing the existing ones. Thanks.

[larsbarring]

And further, when reading @JonathanGregory comment

There's one exception: I think stratiform_graupel_flux should be changed to stratiform_graupel_fall_flux. We already have graupel_fall_flux.

I was reminded of another thought: The term fall brings a a clear notion of downward direction. But similar to many other flux standard name we should consider that in some high resolution cloud models the flux of various hydrometeors may be both downwards and upwards, i.e. where the sign matters. While I do not think there is any immediate need for this, we should factor this in when considering how to how to harmonize precip standard names and descriptions. By doing this we will avoid future complications.

[JonathanGregory]

Given Lars's comment just before this one, maybe we should consider an alternative possibility of introducing new standard names for fluxes above the surface. We haven't yet been asked for them, but having them (or one or two, as examples) might help with avoiding confusion about the existing names.

For example, we could have net_downward_mass_flux_of_X_in_air, where X could be precipitation, rain, snow, etc. We might also need downward_velocity_of_X_in_air. I think in_air may be useful, although they obviously can't be in any other medium, to make clear they are not at the surface.

[larsbarring]

Taking the conversation in #59 into account, here is an attempt to draw a "term tree" of different hydrometeors, i.e. the "X" Jonathan's suggested standard names for fluxes:

root	sub1	sub2
precipitation
	liquid water
		rain
		drizzle
	frozen water
		snow
		graupel
		hail

Are there any other hydrometeors that that I have missed?

[taylor13]

I would prefer to reserve the present fluxes involving precip, rain, snow, etc. to apply only at the ground (without the option of specifying a vertical coordinate indicating a different level). (I note that someone might have misinterpreted these rates in the past and applied these standard names to falling precipitation in the free atmosphere, but we can hope this isn't the case.)

I would also support definition of new standard names, as suggested in https://github.com/cf-convention/vocabularies/issues/67#issuecomment-2286304083 above, if we want to record precip, rain, snow, etc. at levels above the surface.

If others agree, I think the only thing we have to do now is to modify the description of many of the precip-related variables by replacing such sentences as

"Precipitation" in the earth's atmosphere means precipitation of water in all phases.

with something like

"Precipitation", as used in this standard name, describes the water arriving at the earth's surface in any of its phases.

A sentence like this would be appropriate for most, but not all, of the standard names listed above in https://github.com/cf-convention/vocabularies/issues/67#issuecomment-2286019247.

[larsbarring]

Yes, I agree that the suggested new standard names for precip fluxes in the atmosphere are good. I believe the suggested pattern can be combined with all terms in the "term tree"?

With respect to the term precipitation we cannot currently say that it always refers to water arriving at the ground/canopy/snow surface, because we have mass_fraction_of_liquid_precipitation_in_air, atmosphere_mass_content_of_liquid_precipitation, and `mass_fraction_of_precipitation_in_air. For the first two we may consider aliasing them by replacing "liquid_precipitation" by "liquid_water". For the last one it might be more difficult it would because replacing "precipitation" by "liquid_and_frozen_water". Or, we could could keep "precipitation" in these three standard names the relevant alternative of the two sentences that Karl suggests.

[JonathanGregory]

I would prefer Lars's second suggestion, to keep the precipitation standard names as they are, and insert different sentences into the descriptions, according to the different meanings of the word. Karl's sentence includes "as used in this standard name", and applies to precipitation arriving at the ground. With the same phrase, "as used in this standard name", we would need some text for precipitation in the air, and precipitation as a process. I don't think we can replace precipitation with (some phase of) water, because not all water in the atmosphere is precipitating.

[github-actions[bot]]

This issue has had no activity in the last 30 days. Accordingly:

• If you proposed this issue or have contributed to the discussion, please reply to any outstanding concerns.
• If there has been little or no discussion, please comment on this issue, to assist with reaching a decision.
• If the proposal seems to have come to a consensus, please wait for the moderators to take the next steps towards acceptance.

Standard name moderators are also reminded to review @feggleton @japamment @efisher008