Dear Alison,
Thanks for some very detailed work. My answers are below in blue italics, regards, Martin ________________________________ From: Pamment, Alison (STFC,RAL,RALSP) Sent: 08 June 2018 15:21 To: Juckes, Martin (STFC,RAL,RALSP); cf-metadata@cgd.ucar.edu Subject: RE: Final 17 terms for CMIP6 LS3MIP. Dear Martin, Many thanks for making these proposals. I have comments and questions about some of them. In particular, these proposals have made me think that we can make better use of area types than we currently do for surface properties and fluxes such as albedo (1.2), water evaporation (2.2) and snowpack heat flux (3.1) - I think we should take a common approach to these. I have accepted a few of the names and there are quite a number more that are close to acceptance. As with the change_over_time names we need to be clear which areas are included and excluded for all names and also which phases of water are included. Please have a look at my comments and let me know what you think. > Albedo [2] > > We already have names such as “soil_albedo” (Soil albedo is the albedo of the > soil surface assuming no snow). Two new terms are requested, for canopy and > snow > 1.1 albc Canopy Albedo > canopy_albedo (1) > "Canopy" means the plant or vegetation canopy. Albedo is the ratio of > outgoing to incoming shortwave irradiance. The name and units are fine. Does the canopy albedo assume no snow, or does it include snow covered areas of the canopy? This name is still under discussion. Please see also my related comments on 1.2. I think canopy albedo is different from a masked surface albedo. See, for instance, the energy budget diagram on this page: http://geode.colorado.edu/~small/foresthydro.html . This shows that the incoming radiation is partially reflected by the canopy, the portion that passes through the canopy is partially reflected by the surface, so the total amount reflected is greater than what is reflected by the canopy. > 1.2 albsn Snow and Ice Albedo [1] > Albedo of snow and ice covered surface. > snow_and_ice_albedo > Snow and ice albedo. > -- OR -- > This could be handled with existing name surface_albedo and a new area type > "snow_or_ice". [Note that "landice" appears to exclude ice on lakes, which > may be a significant component of this albedo] I'm beginning to think that we should have just one surface_albedo name which would have several aliases (surface_albedo_assuming_deep_snow, surface_albedo_assuming_no_snow, sea_ice_albedo) and add some more area types. For your proposed quantity we'd need an area type of something like land_snow_or_ice as you suggest; we could add deep_snow and no_snow, and sea_ice already exists. This would be similar to the way in which we deprecated several surface_temperature_where_X names and made them all aliases of surface_temperature. The surface_temperature definition says 'The surface called "surface" means the lower boundary of the atmosphere. The surface temperature is the temperature at the interface, not the bulk temperature of the medium above or below. Unless indicated in the cell_methods attribute, a quantity is assumed to apply to the whole area of each horizontal grid box. Previously, the qualifier where_type was used to specify that the quantity applies only to the part of the grid box of the named type. Names containing the where_type qualifier are deprecated and newly created data should use the cell_methods attribute to indicate the horizontal area to which the quantity applies.' If we take a similar approach with the albedo names we should of course add the same recommendation to use cell_methods to indicate the horizontal area. We would then be able to have the albedo of any surface that is described in the area_type table without the need to add further standard names. Do others think that would be a useful step? If we do go down the area_type route, I think we'd still need canopy_albedo as a separate name because it's clearly not the same as the surface, but that name could also optionally be combined with an area_type of no_snow if only the albedo of snow free canopy where required. Regardless of whether we decide to introduce a new standard name or a new area type for the proposed quantity, we need a clear definition of which ice/snow areas are included and excluded. I wonder if excluding lake_ice from our definition of land_ice is a deliberate omission or an oversight? Can we check this point with the ISMIP6 group? Even if lake_ice is not included in land_ice, we do have an existing area_type lake_ice_or_sea_ice and we could introduce lake_ice as a distinct area_type in its own right. I think we'd also need a definition for land_snow, presumably something like ' "Land snow' means any snow lying on the land surface, land ice or lake ice.' If we use surface_albedo, then the area type for this quantity would need to be land_ice_and_lake_ice_or_land_snow or land_ice_or_land_snow, depending on whether lake_ice should be included in land_ice. If we introduce a new standard name the information would have to go in there: surface_albedo_assuming_land_ice_and_lake_ice_or_land_snow. Both approaches lead to quite a long string for the surface type, but I think that's the only way to be clear about the quantity that is being represented. We've used similar long strings for C4MIP names such as surface_upward_mass_flux_of_nitrous_oxide_expressed_as_nitrogen_out_of_vegetation_and_litter_and_soil. This name is still under discussion. You raise some good points. As far as "landice" goes, I believe that the term comes from the oceanography community and the exclusion of river and lake ice is deliberate, but perhaps confusing now that we need to take these other forms of land-based ice into account. I would prefer a more compact formulation for the additional area types, e.g. "ice_on_land" for ice in glaciers, ice caps, ice sheets & shelves, river and lake ice, and any other ice on a land surface (I remember as a child skating on a large area of ice left behind after flood water had frozen and then subsided to leave a sheet of ice covering a large area of land .... probably not relevant to today's climate models, but it makes sense to define a term which includes all ice surfaces). The combined term could be "ice_and_snow_on_land". > 2. Evaporation and transpiration fluxes [3] Variations on the theme of the > existing water_evaporation_flux term. > 2.1 ec Interception evaporation [kg m-2 s-1] > > water_evaporation_flux_from_canopy_due_to_intersection > “Canopy interception is the precipitation, including snow, that is > intercepted by the canopy of a tree and then evaporates from the leaves”, > expanding on the existing name water_evaporation_flux_from_canopy I assume there's a typo in the proposed name and it should say 'interception'. Thank you for providing a definition for interception. Adding text from our usual definitions I think we would end up with: water_evaporation_flux_from_canopy_due_to_interception (kg m-2 s-1) ' "Water" means water in all phases. Evaporation is the conversion of liquid or solid into vapor. (The conversion of solid alone into vapor is called "sublimation".) In accordance with common usage in geophysical disciplines, "flux" implies per unit area, called "flux density" in physics. The specification of a physical process by the phrase "due_to_" process means that the quantity named is a single term in a sum of terms which together compose the general quantity named by omitting the phrase. "Canopy" means the plant or vegetation canopy. "Canopy interception" is the precipitation, including snow, that is intercepted by the canopy of a tree and then evaporates from the leaves.' Okay? Yes, thanks for the spelling correction. > 2.2 eow Open Water Evaporation [kg m-2 s-1] > > A new term "water_evaporation_flux_from_open_water" would work here, but it > might make more sense to define an area type for open water > and use the existing standard name "water_evaporation_flux". I agree that using the existing name water_evaporation_flux is a good choice, although perhaps we should make that one into an alias of water_evaporation_flux_from_surface which is a more accurate representation of the quantity. We could certainly introduce a new area_type, but we'd need a definition of 'open_water'. Does it mean 'ice_free_water'? We already have ice_free_land and ice_free_sea. Does the open water in this case include the sea or does it mean only land based water? Does it include rivers, lakes, etc.? This name is still under discussion. Good question. The phrase "open water" appears to have different meanings for oceans and freshwater. For freshwater it means areas that are free of vegetation or other obstacles. I'm not sure it is really right for an area type: when used in the context of swimming, "open water" means water some distance from the shore or banks, but for evaporation it just means evaporation from the water/air interface as opposed to evaporation of water which is in contact with a solid boundary. We could, perhaps, use "fresh_free_water_surface" (Areas where there is a free interface between freshwater and the atmosphere without vegetation or other obstructions). > 2.3 et Total Evapotranspiration [kg m-2 s-1] > “Evapotranspiration” is a new concept to the CF standard names, but appears > to be clearly defined. > evapotranspiration_flux (kg m-2 s-1) > “Evapotranspiration refers to the flux of water into the atmosphere from a > combination of transpiration by plants and evaporation from soil and > other land surfaces.” The name and units are fine. Thank you for providing a definition for evapotranspiration. I think the full definition should read as follows: ' "Evapotranspiration" means the flux of water into the atmosphere from a combination of transpiration by plants and evaporation from soil and other land surfaces. Water means water in all phases. Evaporation is the conversion of liquid or solid into vapor. (The conversion of solid alone into vapor is called "sublimation"). Transpiration is the process by which water is carried from the roots of plants and evaporates from the stomata. In accordance with common usage in geophysical disciplines, "flux" implies per unit area, called "flux density" in physics.' Okay? One question: does 'other land surfaces' specifically mean dry land, i.e. does it exclude the open water of proposal 2.2? This name is still under discussion. The open water concept in 2.2 is included: we can rephrase as "other land surfaces and inland water surfaces". Note that the open water concept is more specialised than evaporation from inland water surfaces (though it might be the same in CMIP6). > 3. Heat fluxes [2] > 3.1 hfrs Heat transferred to snowpack by rainfall [W m-2] > A variation of > “temperature_flux_due_to_rainfall_expressed_as_heat_flux_into_sea_water”. > > temperature_flux_due_to_rainfall_expressed_as_heat_flux_into_snow_and_ice (W > m-2) The name itself does follow existing patterns and the units are correct for a heat flux. The existing temperature_flux_due_to_rainfall_expressed_as_heat_flux_into_sea_water name is defined as follows: 'The quantity with standard name temperature_flux_due_to_rainfall_expressed_as_heat_flux_into_sea_water is the heat energy carried by rainfall entering the sea at the sea surface. It is calculated relative to the heat that would be carried by rainfall entering the sea at zero degrees Celsius. It is calculated as the product QrainCpTrain, where Qrain is the mass flux of rainfall entering the sea (kg m-2 s-1), Cp is the specific heat capacity of water and Train is the temperature in degrees Celsius of the rain water entering the sea surface. In accordance with common usage in geophysical disciplines, "flux" implies per unit area, called "flux density" in physics. The specification of a physical process by the phrase due_to_process means that the quantity named is a single term in a sum of terms which together compose the general quantity named by omitting the phrase.' Assuming that the proposed name would be defined similarly, it seems to me that this is another case where we could use area types. The energy being carried by the rainfall is essentially a surface flux - it depends on where the rain originated and the properties of the air it has fallen through, rather than the type of surface it is falling onto. I suggest we could make the existing name into an alias of temperature_flux_due_to_rainfall_expressed_as_heat_flux. We could then use this for both the original and proposed quantities if an appropriate area type is supplied. We already have an area type of sea. Is the proposed quantity like the albedo in proposal 1.2, i.e. does it mean land_ice_and_lake_ice_or_land_snow? This name is still under discussion. I believe that the heat flux here depends on the temperature of the target medium (or, rather, the difference in temperature between the target and source), so a simple masking does not quite capture the full meaning. But it would be interesting to hear other views. > 3.2 hfsbl Energy of sublimation [W m-2] > Variation of: surface_upward_latent_heat_flux The definitions of latent heat > flux state that the latent heat flux includes sublimation heat flux, so it > makes sense to use the same pattern: > surface_upward_sublimation_heat_flux (W m-2) The name and units are fine and the definition can be constructed from existing text: 'The surface called "surface" means the lower boundary of the atmosphere. "Upward" indicates a vector component which is positive when directed upward (negative downward). Sublimation is the conversion of solid into vapor. In accordance with common usage in geophysical disciplines, "flux" implies per unit area, called "flux density" in physics.' This name is accepted for publication in the standard name table and will be added in the June update. > 4. Nudging increments [2] > There are no terms referring to “nudging increments” in the CF convention, > but there are a number referring to “flux correction”. The following help > text is proposed for nudging increments: “A nudging increment refers to an > amount > added to parts of the model system. nudging_increment_in_X refers to > an increment in quantity X over a time period which should be defined in the > bounds of the time coordinate.” These should be encoded with cell_methods > “time: > sum” and a bounds variable to give the time over which nudging is > summed. The convention requires that, with this construction, the time > periods should be contiguous. > > 4.1 nudgincsm Nudging Increment of Water in Soil Moisture [kg m-2] > nudging_increment_in_mass_content_of_water_in_soil Thank you for providing a definition for 'nudging increment'. I think this is very much modelling jargon, although so is 'flux correction' I suppose. I can't think of a better phrase, so don't have any real objection to including this name as it is. The rest of the definition can be constructed from existing text: 'A "nudging increment" refers to an amount added to parts of a model system. The phrase "nudging_increment_in_X" refers to an increment in quantity X over a time period which should be defined in the bounds of the time coordinate. "Content" indicates a quantity per unit area. "Water" means water in all phases. The mass content of water in soil refers to the vertical integral from the surface down to the bottom of the soil model. The "soil content" of a quantity refers to the vertical integral from the surface down to the bottom of the soil model. For the content between specified levels in the soil, standard names including "content_of_soil_layer" are used.' This name is accepted for publication in the standard name table and will be added in the June update. > 4.2 nudgincswe Nudging Increment of Water in Snow [kg m-2] > nudging_increment_in_surface_snow_and_ice_amount The definition would be as follows: 'A "nudging increment" refers to an amount added to parts of a model system. The phrase "nudging_increment_in_X" refers to an increment in quantity X over a time period which should be defined in the bounds of the time coordinate. The surface called "surface" means the lower boundary of the atmosphere. "Amount" means mass per unit area.' This seems okay, but I'm wondering if once again this quantity refers to land_ice_and_lake_ice_or_land_snow as in proposals 1.2 and 3.1? If so, then we should probably add something to the definition advising on the use of an area type to describe exactly which areas are affected. This name is still under discussion. Can we use "nudging_increment_in_snow_and_ice_on_land_amount", following my suggestion for the area type above? > 5. River in- and out-flow [2] > water_volume_transport_in_river_channel and > water_volume_transport_into_sea_water_from_rivers exist. The new variables > represent cell averages of river fluxes directed inwards and outwards > respectively. > 5.1 rivi River Inflow Water flux from upstream [m3 s-1] > river_water_volume_transport_into_cell > '"Cell" refers to a model grid-cell. "River water" refers to the water > (liquid and solid) in the fluvial system (stream and floodplain).' I think the suggested name and units look good. I suggest that we add some advice about grid cell bounds to the definition (similar to ice mass and ice area names). ' "Cell" refers to a model grid-cell. The extent of an individual grid cell is defined by the horizontal coordinates and any associated coordinate bounds or by a string valued auxiliary coordinate variable with a standard name of "region". "Water" means water in all phases. "River" refers to water in the fluvial system (stream and floodplain).' Is it correct to say that this one includes "water in all phases", i.e. would it include ice floating on a river? (This question also applies to proposal 1.3 change_over_time_in_river_water_amount in the thread "Standard names for LS3MIP: 8 temporal changes + 1 feature depth". This name is still under discussion. yes, water in all phases. > 5.2 rivo River Discharge [m3 s-1] > river_water_volume_transport_out_of_cell As with proposal 5.1 the name and units look good and I suggest adding advice about cell bounds to the definition: ' "Cell" refers to a model grid-cell. The extent of an individual grid cell is defined by the horizontal coordinates and any associated coordinate bounds or by a string valued auxiliary coordinate variable with a standard name of "region". "Water" means water in all phases. "River" refers to water in the fluvial system (stream and floodplain).' Again, does this one include ice floating on a river? This name is still under discussion. yes, water in all phases, as above. > 6. Roots [1] > > 6.1 rzwc Root zone soil moisture [kg m-2] > This is a variation on mass_content_of_water_in_soil_layer, but with a layer > defined by the presence of roots rather than a coordinate range (cf. > “stratosphere” in atmosphere). > > mass_content_of_water_in_root_zone > “The root zone refers to the soil which surrounds the roots of vegetation.” I think it would be helpful to include 'soil' in the name. We have an existing name root_depth defined as 'Depth is the vertical distance below the surface. The root depth is maximum depth of soil reached by plant roots, from which they can extract moisture.' I suggest we call the proposed quantity mass_content_of_water_in_soil_layer_defined_by_root_depth. This is a bit like existing ocean_mixed_layer names, e.g., ocean_mixed_layer_thickness_defined_by_temperature. The ocean name definitions refer to a coordinate variable whose value contains the temperature, etc., that actually determine the layer thickness. We could do something similar here: ' "Content" indicates a quantity per unit area. The content of a soil layer is the vertical integral of the specified quantity within the layer. The quantity with standard name mass_content_of_water_in_soil_layer_defined_by_root_depth is the vertical integral between the surface and the depth to which plant roots penetrate. A coordinate variable or scalar coordinate variable with standard name root_depth can be used to specify the extent of the layer. "Water" means water in all phases.' What do you think? Yes, this is a very good approach. > 7. Water fluxes [5] > 7.1 qgwr Groundwater recharge from soil layer [kg m-1 s-1] > There are no usage of “groundwater” in existing CF terms, the following has > been suggested in connection with the LS3MIP term dgw: > “Groundwater is the water present beneath Earth's surface in soil pore spaces > and in the fractures > of rock formations.” > water_flux_from_soil_to_groundwater I think there's a typo in the suggested units and it should say kg m-2 s-1. I assume that this quantity must refer to the transfer of liquid water - frozen soil moisture would have to melt before it could permeate to lower levels, wouldn't it? Hence I think this name should be liquid_water_mass_flux_from_soil_to_groundwater (kg m-2 s-1) 'In accordance with common usage in geophysical disciplines, "flux" implies per unit area, called "flux density" in physics. Groundwater is subsurface water below the depth of the water table, including soil moisture and underground aquifers.' I notice the wording of the suggested definition of groundwater differs from the one we have discussed in other LS3MIP proposals, although I think the meaning is the same. For consistency I have used the same wording as other proposals for now. I am aware that there is still some work being done on clarifying the definition of groundwater and we will need to wait for the outcome of that before we can finalise this name. This name is still under discussion. It is a typo in the units, which should be kg m-2 s-1. > 7.2 sblnosn Sublimation of the snow free area [kg m-2 s-1] > > This is a variation of the CMIP5 variable "sbl" which used the existing term > surface_snow_and_ice_sublimation_flux. Here, we just drop "and_ice" > to get > a term referring to the ice only. > surface_ice_sublimation_flux Looking at existing names I see we also have surface_snow_sublimation_amount. Rather than talking about sublimation amount in one name and sublimation flux in others I think it would make sense to take a more uniform approach. I think the primary purpose of the existing surface_snow_and_ice_sublimation_flux name is to describe changes at the surface, rather than a water vapour flux into the atmosphere. Hence I suggest introducing aliases: surface_snow_sublimation_amount -> tendency_of_surface_snow_amount_due_to_sublimation surface_snow_and_ice_sublimation_flux -> tendency_of_surface_snow_and_ice_amount_due_to_sublimation and calling the proposed quantity: tendency_of_ice_amount_due_to_sublimation. None of these names specify that the snow and ice are located on land, so as with the albedo and evaporation names we probably need to use them in conjunction with an area type. Do you agree with this approach? Yes, this is good. > 7.3 snmsl Water flowing out of snowpack [kg m-2 s-1] > This is a variation of snm [surface_snow_melt_flux], but considering only the > component of the flux into soil: > > surface_snow_melt_flux_into_soil For this name I suggest a different form: liquid_water_mass_flux_into_soil_due_to_surface_snow_melt (kg m-2 s-1) There are a number of reasons for suggesting this. The existing surface_snow_melt_flux name isn't very clear in my opinion. It doesn't say where the flux is going (or even that it is a mass flux, although that is mentioned in the definition). It might even be better to change the existing name to something like tendency_of_surface_snow_amount_due_to_melting. The other reason I am suggesting a different form for the proposed quantity is because of the use of the phrase 'surface_snow' in standard names to mean snow lying on the surface. I think there's some potential for confusing surface_snow with a surface_flux (i.e. flux in or out of the atmosphere) when that clearly isn't intended. The definition would be as follows: 'In accordance with common usage in geophysical disciplines, "flux" implies per unit area, called "flux density" in physics. The specification of a physical process by the phrase "due_to_" process means that the quantity named is a single term in a sum of terms which together compose the general quantity named by omitting the phrase. The phrase "surface_snow" means snow lying on the surface.' Okay? The sentence explaining that surface_snow means lying snow is new, but I think we should add it to all existing surface snow names. Do others agree? This name is still under discussion. Yes, this looks good to me. Longer, but the additional clarity justifies this. > 7.4 snwc Snow intercepted by the vegetation [kg m-2] > > There is an existing term “canopy_water_amount”, hence: > canopy_snow_amount I think the name and units are fine. The definition would be: ' "Amount" means mass per unit area. "Canopy" means the plant or vegetation canopy. The phrase "canopy_snow" means snow lying on the canopy.' This one is a straight forward addition to the standard name table. It is accepted for publication and will be added in the June update. 7.5 sw Surface Water Storage (excluding snow) [kg m-2] The existing term surface_water_amount refers to “.. the amount on the ground, excluding that on the plant or vegetation canopy”, and "water" is assumed to refer to all phases. There is a term “land_based_water_amount” under discussion for PMIP which refers to “This quantity is often known as Terrestrial Water Storage. It includes surface water (water in rivers, wetlands, lakes, snow, vegetation and reservoirs) and subsurface water (soil moisture, groundwater)”. Assuming that this term is intended to be a variation on the first, surface_water_amount, and the amount includes only liquid phase water (excluding snow and ice): surface_liquid_water_amount I note that the PMIP name you refer to is currently listed as land_water_amount. Given that this variable is related and is clearly limited to surface water on land (i.e. it excludes surface sea water) I suggest: land_surface_liquid_water_amount (kg m-2) 'The surface called "surface" means the lower boundary of the atmosphere. "Amount" means mass per unit area. The quantity with standard name land_surface_liquid_water_amount includes water in rivers, wetlands, lakes, snow, vegetation and reservoirs.' I am aware that there is still some work being done on clarifying the definition of land water and we will need to wait for the outcome of that before we can finalise this name. This name is still under discussion. OK, I'll send a reminder on the groundwater etc discussion next week. Best wishes, Alison ------ Alison Pamment Tel: +44 1235 778065 NCAS/Centre for Environmental Data Archival Email: alison.pamm...@stfc.ac.uk STFC Rutherford Appleton Laboratory R25, 2.22 Harwell Oxford, Didcot, OX11 0QX, U.K. _______________________________________________ CF-metadata mailing list CF-metadata@cgd.ucar.edu http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
