search

ESCOMP / CTSM

Community Terrestrial Systems Model (includes the Community Land Model of CESM)
http://www.cesm.ucar.edu/models/cesm2.0/land/
Other
300 stars 306 forks source link

CTSM-FATES v1 compsets #1047

Open wwieder opened 4 years ago

wwieder commented 4 years ago

Create working compsets for land only (I) spin-up, historical & SPP's with and w/out active crops. Also, F2000 and F_AMIP for coupled model output.

Would the compsets also define if we're using FATES-sp, FATES-fixed biogeography, FATES-dynamic biogeography.

This is linked with #1046.

ekluzek commented 4 years ago

This requires #1046 in order to move forward.

ekluzek commented 4 years ago

The F compsets will need to go into CAM. So this will require a CTSM tag, CAM tag, and CESM tag to put them all together.

ekluzek commented 4 years ago

These are some notes from @rosiealice on what she thought was important...

For the moment, I think the important ones to get us going are 1850 2000 Transient SSP -something.

-As you note, FATES-CROP will need more coding work, as will FATES-SP (that's on me).
-I don't think that having CLM4.5 compsets is a big priority? Does anyone think otherwise?

-On the other details, it would be nice to have a version which is the fastest possible version - i.e. with NO glaciers, runoff, crops, etc, and then one with something more like a default configuration. People who want to run FATES with a specific river model are quite rare, I suspect.

-The no-anthro compset is mostly reelevant to the situation where we have crops working, right? Maybe we can leave that until later?

-I guess the other thing is climate forcing data? Doing that rather depends on whether we want to create the whole entire universe of FATES compsets in one go? I don't have a good handle on the implications for e.g. testing, of adding a massive number of compsets, so if people have thoughts on that please chime in...

ekluzek commented 2 years ago

We don't have it as a compset yet, but ctsm5.1.dev056 enables the ability to do FATES-SP mode. And we also have the ability to have fixed- biogeography. So this brings in much of the functionality needed for FATES v1.

ekluzek commented 2 years ago

As part of #1609 we'll be adding a single-point compset for FATES I1PtClm51FatesRs, which gives one of the needed compsets for FATESv1 (although it wasn't mentioned above it's still one of the needed ones).

ekluzek commented 2 years ago

1617 is about adding transient FATES compset aliases.

rosiealice commented 1 year ago

I think we have fates compsets now, and the transient compsets are their own issue.

ekluzek commented 2 weeks ago

Do we want this for cesm3.0?

wwieder commented 2 weeks ago

Yes, for FATES sp and nocomp configurations in I cases.
(note, this was edited, as I think we want to have SP and NOCOMP options FATES cases in CLM6)

ekluzek commented 2 weeks ago

We can do all of these configurations. I'm thinking definition of done here is to have compset aliases and long-names that make it easy to do all of these options. @wwieder and @rosiealice does that sound right to you?

Is fixed-biogeography the same as NOCOMP from the CESM3 compsets for CTSM?

https://docs.google.com/spreadsheets/d/1nJl4lbbwBq60quyaERJ6OvVJxF4kDu4xX7zZenZgyB8

wwieder commented 2 weeks ago

Let's discuss at the SE meeting, but I'm not sure these will necessarily all be possible. I also don't know if any need aliases (which at last check were being reserved for select, scientifically supported model configurations in the CESM3.0 release).