Area label for SIday.siitdconc SImon.siitdconc

[znichollscr]

Apologies if this is a duplicate of some other conversation (I read through open issues and couldn't find it).

The area label in the branding suffix is currently "u", while the cell methods are "area: time: mean where sea_ice (mask=siconc)", which is inconsistent.

Is the area label wrong, or should the branding algorithm be updated to identify cell methods with constraints like (mask=siconc) as undefined area label or something else?

@taylor13 you have probably thought through this use case with the branding algorithm best

@bethdingley could you help tag the right people here please?

@wolfiex FYI

[bethdingley]

The branding labels in DR v1.2.2 were calculated from the metadata in one of Karl's spreadsheets - i.e. for the variables where there is still ongoing discussion, we used the metadata Karl has proposed, rather than the metadata in the DR. So any branded name in the DR v1.2.2 which is marked as 'Provisional - likely accepted' or 'Provisional - discussion required' don't be too concerned if there are inconsistencies between the branded name and the metadata

This might also help to answer #105, #104, https://github.com/CMIP-Data-Request/Harmonised-Public-Consultation/issues/10, and #102

[znichollscr]

Branded variable names marked as provisional in DR v1.2.2 so this is a known issue (will leave it open for my own reference, but this is probably a duplicate of something)

[znichollscr]

Thanks for picking that up and explaining @bethdingley

[znichollscr]

Will be fixed by WCRP-CMIP/Variable-Registry#51