Combined coverage of multiple area_type

[sol1105]

Question

I have a single variable with standard_name = "area_fraction" that represents the combined coverage of both lake_and_inland_sea and river. I'm looking for CF-compliant guidance on how to represent this.

In the following two suggestions:

dimensions:
  lat=100;
  lon=100;
  maxlen=20;
  type=2;
variables:
  float inland_water_fraction(lat,lon);                 <----- Note: has no type dimension.
     inland_water_fraction:standard_name="area_fraction";
     inland_water_fraction:cell_methods="area: mean";
     inland_water_fraction:coordinates="inland_water"; 
  char inland_water(type, maxlen);
    inland_water:standard_name="area_type";
data:
  inland_water="lake_and_inland_sea", "river";

dimensions:
  lat=100;
  lon=100;
  maxlen=20;
variables:
  float inland_water_fraction(lat,lon);
     inland_water_fraction:standard_name="area_fraction";
     inland_water_fraction:cell_methods="area: mean";
     inland_water_fraction:coordinates="inland_water"; 
  char inland_water(maxlen);
    inland_water:standard_name="area_type";
data:
  inland_water="lake_and_inland_sea, river";       <---- Note: A single string with comma-separated values.

[JonathanGregory]

Dear Martin @sol1105

I don't think you can do it with the existing area types. I would say that this is a use-case for a new area type of inland_water, which would be consistent also with the consensus so far reached in vocabularies issue 25.

Best wishes

Jonathan

[sol1105]

Dear Jonathan,

thank you for your quick reply. Will this new area_type be added following cf-convention/vocabularies#25 or does it have to be suggested in a separate issue?

Kind regards,
Martin

[JonathanGregory]

Dear Martin

A new area type of inland_water would be consistent with vocabularies issue 25 but has not explicitly been proposed. It would be good if you could propose it in a separate vocabularies issue, so that it isn't held up by issue 25. Thanks.

Best wishes

Jonathan