Feedback on the meaning of: sea_ice_area_fraction_of_sea_area_fraction

[ss421]

I wonder if it might be possible to provide more description on the: sea_ice_area_fraction_of_sea_area_fraction variable. The reason I’m asking is that it has generated quite a few questions. First let me provide the description of the variables used by the Met Office:

1. The fraction of the wet part of the cell that contains ice. Some cells are part land and part water (along the coast for example). This variable is used in our atmospheric DA and we include all water bodies so lakes are included for example.
2. The fraction of the cell that contains ice. This is used by ocean DA and from what I understand, the model cells do not contain a land part.

Does anyone know how sea_ice_area_fraction_of_sea_area_fraction fits with these descriptions?

[mkavulich]

@ss421 The current use of this variable in the GFS and its various physics schemes (e.g. GFS_radiation_surface.F90) is the ice area fraction for water points. If a point represents part land and part water, this variable should be used for the fraction of that water part that is ice (the GFS schemes also do not distinguish between lake and ocean points in this regard).

Do you have a proposal for wording that might make this variable's description less ambiguous?

[ss421]

One suggestion (provided by Simon Good at the Met Office) and modified slightly is:

ice_area_fraction_of_water_area_within_cell and ice_area_fraction_of_cell_area.

In addition, maybe we could add more description to the definition in this case?

[mkavulich]

@ss421 Sorry it's taken so long to get back to you. I assume from the context that you're proposing sea_ice_area_fraction_of_sea_area_fraction be renamed to ice_area_fraction_of_water_area_within_cell and adding a new variable ice_area_fraction_of_cell_area? This seems logical to me, but we'd need to be clear about wording, as (at least from the GFS context) we're specifically referring to "floating" ice here and not ice on land. To make it clear we could include a long name with further description, I'd suggest the following or something similar:

      <standard_name name="ice_area_fraction_of_water_area_within_cell"
                     long_name="Fraction of water area within a cell that is covered by ice">
         <type kind="kind_phys" units="frac">real</type>
      </standard_name>
      <standard_name name="ice_area_fraction_of_cell_area"
                     long_name="Fraction of cell area covered by ice">
         <type kind="kind_phys" units="frac">real</type>
      </standard_name>

[ss421]

@mkavulich - thanks.

Yes, I agree with what you have added here in both cases.

The proposed ice_area_fraction_of_water_area_within_cell would be used with our Atmospheric DA.

The ice_area_fraction_of_cell_area would be used with our Ocean DA so tagging @twsearle @s-good to check they are happy with the description.

There is also a land version that is documented in #59 as: land_ice_area_fraction, added by @SamanthaPullen. To fit with the pattern proposed here, perhaps this would be: ice_area_fraction_of_land_area_within_cell with something like:

      <standard_name name="ice_area_fraction_of_land_area_within_cell"
                     long_name="Fraction of land area within a cell that is covered by ice">
         <type kind="kind_phys" units="frac">real</type>
      </standard_name>

@SamanthaPullen - do you agree with the description?

[SamanthaPullen]

@ss421 @mkavulich I think there is a subtle difference with the land_ice_area_fraction name I have proposed (#59 for the Issue, #60 for the PR) as this is intended to represent the area of the entire cell that is covered by land ice (as defined in CF Standard names), regardless of whether that grid cell is entirely land or a mix of land or water. ice_area_fraction_of_land_area_within_cell would be an appropriate (and consistent with the sea ice fraction proposals in this issue) name to use if we want to define the land ice area fraction of just the land portion of a cell (i.e. normalised by the land area fraction in the cell).

Do you think I should append my land_ice_area_fraction proposed name with '_of_cell_area', i.e. land_ice_area_fraction_of_cell_area? I chose land_ice_area_fraction because it already existed in the CF standard names

[mkavulich]

@SamanthaPullen thanks for bringing this up. Reading through the different descriptions in the CF conventions, it seems like the CF standards for sea ice area fraction are a bit vague and possibly inconsistent, but land ice area is more concrete.

Here is the definition of sea_ice_area_fraction (emphasis mine):

• "Area fraction" is the fraction of a grid cell's horizontal area that has some characteristic of interest. It is evaluated as the area of interest divided by the grid cell area. It may be expressed as a fraction, a percentage, or any other dimensionless representation of a fraction. Sea ice area fraction is area of the sea surface occupied by sea ice. It is also called "sea ice concentration". "Sea ice" means all ice floating in the sea which has formed from freezing sea water, rather than by other processes such as calving of land ice to form icebergs.

While the first and second sentences imply that it is a fraction of the whole grid cell's horizontal area, the fourth sentence implies that it should be considered the fraction of sea area only.

Contrast that with the definition of land_ice_area_fraction:

• "Area fraction" is the fraction of a grid cell's horizontal area that has some characteristic of interest. It is evaluated as the area of interest divided by the grid cell area. It may be expressed as a fraction, a percentage, or any other dimensionless representation of a fraction. "Land ice" means glaciers, ice-caps and ice-sheets resting on bedrock and also includes ice-shelves.

This seems to unambiguously refer to the fraction of the total grid cell area, so it wouldn't be a problem.

All that said, there's nothing to say we can't adopt a name within this convention we are discussing here to make sure it is unambiguous and clear. If I had to make a decision, I'd lean towards adopting land_ice_area_fraction_of_cell_area over land_ice_area_fraction. Would that be acceptable with you @SamanthaPullen?

[s-good]

Would it make sense to include sea in the name if we are specifically taking about sea ice? E.g. so we would have the option of land_ice_area_fraction_of_cell_area, sea_ice_area_fraction_of_cell_area or ice_area_fraction_of_cell_area for fraction of land, sea and all ice respectively.

[SamanthaPullen]

@mkavulich I am happy with land_ice_area_fraction_of_cell_area if @svahl991 is also happy from the JEDI point of view. I can update my PR #60 accordingly