clarification around fracLut

[paolap]

The fracLut() function re-organise the land-use tile in 4 super-categories.
Based on CMIP6 LUMIP documents these should be:

Primary and secondary land psl (Forest, grasslands, and bare ground)
Cropland crp
Pastureland pst (Includes managed pastureland and rangeland)
Urban settlement urb

CABLE land-use tiles are:

1. Evergreen Needleleaf
2. Evergreen Broadleaf
3. Deciduous Needleleaf
4. Deciduous Broadleaf
5. Shrub
6. C3 Grassland
7. C4 Grassland
8. Tundra
9. C3 Cropland
10. C4 Cropland
11. Wetland
12. empty
13. empty
14. Barren
15. Urban
16. Lakes
17. Ice

I have a few questions about the function

• for the first psl category these tiles are used: [1,2,3,4,5,6,7,11,14]
  when excluding woody vegetation only [6,7,11] are used, shouldn't 14 barren soil be included?
• the second categories according to LUMIP should be crp in the code is defined as
  #no pastures
  vout[:,1,:,:]=vout[:,1,:,:]
  which it leaves it full of zeros
• the urban category is defined as tile 15 but if woody vegetation is excluded is set as
  vout[:,3,:,:]=vout[:,3,:,:].
  so again it's left empty
• finally it looks like nwdFracLut in the CMIP6 table has 5 dimensions: longitude latitude landUse time typenwd and not 4 as fracLut, but the function returns a 4D variable in both cases.

[chloemackallah]

To answer your questions:

1. These tiles are what were given to me when coding these up- check with Tilo/Ying-Ping
2. Pastures are not modelled in CABLE, hence they are empty
3. Urban is not modelled in CABLE, hence they are empty
4. Hmm it does look like I missed getting that last dimension encoded correctly (it is in the metadata, but not applied to the variable itself) - not a major issue tbh, it is redundant because it captures the same info as the variable name, and it only appears as a coordinate in this one

[paolap]

I've updated the comment to fix the table. Thanks for the answers, I guess I will have to double-check with them again.