When CMIP7 SSP scenarios are available add dataset creation and supported datasets in the XML

[ekluzek]

Currently we do NOT have the CMIP7 SSP future scenarios available. When they are we will need to add support for them in dataset creation, create the datasets and add them to support in the XML files.

This is planned to happen post cesm3.0, so with the CMIP7 datasets this will happen later. What we likely do need to do is to have some kind of blending of some CMIP6 SSP scenarios with the new CMIP7 historical data. So that we can at least do some SSP cases to ensure they work. And there is likely some need for at least some provisional SSP datasets.

Definition of Done:

• [ ] Bring in the rawdatasets, new surface datasets when the CMIP7 SSP scenarios are available
• [ ] This will be part of ctsm5.4

[ekluzek]

@wwieder in the CTSM SE meeting we came up with a plan that makes a lot of sense, and is what we think we should do. The "blended" idea isn't great and probably not something would really want to do. But, people running SSP cases -- will want to run with CMIP6 options. It's possible to do this since ctsm5.4 will be compatible with both ctsm5.4 and ctsm5.3 datasets.

So this is what we do:

• drop the "blended" idea
• allow a CMIP6 option in long compset names
• SSP compsets use the CMIP6 option, but everything else the default CMIP7
• So we keep the current CMIP6 landuse.timeseries datasets in place
• We add ctsm5.3 1850 fsurdat files that can be used with them (this duplicates the 1850 files in the XML)
• When CMIP7 datasets are available we remove the CMIP6 option

There's some details that would be good to workout here, but that's roughly what we would do.

[ekluzek]

From this it looks like the CMIP7 SSP datasets are available, so we don't have to do anything temporary.

https://github.com/ESCOMP/CTSM/issues/3245#issuecomment-2968198185

So this is probably resolved with that fact.

[ekluzek]

In meeting today, we had the update that the datasets that @lawrencepj1 could process are harmonizing the CMIP6 SSP scenario's with the new CMIP7 LUH3 historical data.

In general we feel this isn't an important scientifically, and we should just bring in CMIP7 SSP scenario data when available -- probably ctsm6.1.0?

[ekluzek]

Opening back up as a task to do post cesm3.0 when CMIP7 future scenarios are available.