Make proxies proportional to fluxes

[coroa]

Description

Convention update

CEDS proxies were designed to be proportional to absolute emissions in each grid cell and we adopted that convention unknowingly.

This PR is changing this design to make uniform proxies that are used for CDR like OAE more intuitive. This means that all CEDS proxies need to be divided by the cell area on import.

This happened in the generation scripts also updated in this PR, but also in the new data folder on OneDrive labelled data_2024_09_16 . Proxy rasters for the code for this PR need to be regenerated or used from the new OneDrive folder. (NB @jkikstra).

OAE Alkalinity Additions

OAE Alkalinity additions are not deployed above 67.5ºN any longer (due to efficiency concerns).

Fix CEDS Aircraft emissions proxy

As reported in https://gmd.copernicus.org/preprints/gmd-2022-250/ and https://github.com/JGCRI/CEDS/issues/45 :

• input4MIPS historical files came with a bias of 1/cos(lat), while
• input4MIPS scenario files had even a 1/cos(lat)**2 bias

These resulted from two mistakes in proxy generation scripts.

We half-fix the scenario files bias to 1/cos(lat) to be consistent with historical gridded emissions.

(@jkikstra @gidden This half-fix in commit ac85fd1 has to be undone, when the CEDS Aircraft emissions proxies will be updated, but hopefully you can find a better format than the current R dumps, which will make the current hacky script obsolete)

Remaining known problem:

• New scenario files have still a 1/cos(lat) bias to be consistent with historical files

[gidden]

Great, thanks @coroa for digging into all the nuances here. Well noted on the CEDS air update. Also for @ssmithClimate to be aware of (Steve - the file changed here is how we process the CEDS proxy data to generate emissions grids, FYI).

[gidden]

@coroa - FYI I rebased and merged so we can keep reference to the specific aircraft commit