Horizontal label branding algorithm

[znichollscr]

@taylor13 is the algorithm for the horizontal label described in https://docs.google.com/document/d/19jzecgymgiiEsTDzaaqeLP6pTvLT-NzCMaq-wu-QoOc/edit?pli=1&tab=t.0 final?

If yes, I think this will affect:

• Omon.msftyrho, Omon.msftyrhompa: currently their horizontal label is "hys", but I think it will be updated to "hm" based on dimensions "rho basin time" (although maybe this is an edge case the algorithm currently misses as I don't think the variables are horizontal means, rather their cell methods of "grid_longitude: sum where sea time: mean" suggests they're a zonal sum?)
• Omon.msftyz, Omon.msftyzmpa: currently their horizontal label is "hys", but I think it will be updated to "hyb" based on dimensions of "gridlatitude olevel basin time", although seems wrong as they're not a zonal mean, rather a zonal sum given cell methods of "grid_longitude: sum where sea time: mean"
• E3hrPt.so2: currently the horizontal label is "hxy", but it should be "hs" (for site) right?
  • I think the paper handles this case well. We need to update the mapper code to handle this (assuming you tell us that the implementation is final)

[znichollscr]

@taylor13 ping on this

[taylor13]

Table F3 in the branded variables document (linked above) shows what branded variable gets assigned to a variable, depending on which dimensions are attached to that variable. (I have made some changes in it since you last saw it for "gridlatitude". )

Omon.msftyrho, Omon.msftyrhompa: I have raised a new variable request 1.2.2 dimension issue 16 because the dimensions omit "gridlatitude", which needs to be included. I have also updated Table F3 so that the correct label can be generated: ht. This is a special type of "transit" in that it spans the basin along a model's in the x-direction at constant y.

Omon.msftyz, Omon.msftyzmpa: All variables that are a function of gridlatitude and basin should be assigned "ht".

E3hrPt.so2: Yes, this is site data with the label "hs". (I think my previous review got this wrong.)

[znichollscr]

super, thank you. Is Table F3 (and the other F Tables) final now?

[taylor13]

I should hope they're final (until we find a mistake). I don't think they've changed except with regard to gridlatitude (which I though was getting removed from the data request) since v1.2.2.