F2000 initial conditions

[wwieder]

The CAM group, mainly @cecilehannay, is setting up an F2000 compset. I assumed we'd use the same land configuration as in in the 091 run that Gustavo started with updates to datm.streams for the year 2000? We also need to provide initial conditions.

Maybe we can use the initial conditions from the 2000pSASU case that @slevis-lmwg ran recently? Is this consistent with how we typically set up F2000 cases @olyson and @ekluzek?

[wwieder]

Let me know if there's a different place to have this discussion (e.g. AMGW-dev?). I'm relatively new to this cross working group coordination.

[ekluzek]

@wwieder this is really a subtask under #2403. We do have preliminary finidat files for 2000 from previous simulations (but they importantly do NOT have the arctic veg changes in place). So mainly the task here is to update them with the latest model versions (the table in #2403 isn't complete and doesn't show most of the 2000 finidat files).

Since, running with CAM is significantly different than running standalone -- we'll want spunup initial conditions based on a transient run either a B fully coupled case or a F compset with CAM.

So @slevis-lmwg CN-Matrix simulations will be good for I compsets. But, for CAM we'll want something running with CAM for these (at least longer term). But, for the immediate need maybe we should use it? Or is there a fully coupled or CAM F compset simulation that could be used for this purpose? We would want it be with the latest ctsm5.2.0 tag to make sure it has everything needed from CTSM. I'm guessing that might be something that we'll need to run to get something good to use here. We don't want the older finidat files, because we need to have the arctic veg changes in place for these.

Also @cacraigucar (and @adamrher) does CAM want both 2000 and 2010 compsets and finidat files for both? And for that matter it seems like we should maybe update to have 2020 compsets and initial conditions? And maybe sunset the others? Has anyone thought about that?

[cacraigucar]

Also @cacraigucar (and @adamrher) does CAM want both 2000 and 2010 compsets and finidat files for both? And for that matter it seems like we should maybe update to have 2020 compsets and initial conditions? And maybe sunset the others? Has anyone thought about that?

@PeterHjortLauritzen will be a better person than me to answer the question of which variations we will ultimately want for CAM. He indicated yesterday that it wasn't decided yet for 2010 and I didn't ask him about 2020 at all.

[slevis-lmwg]

If you decide to use the file that I generated for I2000 (started with 1850 spin-up with cnmatrix and continued with Hist without cnmatrix), then that finidat is under /inputdata here:
lnd/clm2/initdata_esmf/ctsm5.2/clmi.I2000Clm52BgcCrop.2000-01-01.1.9x2.5_gx1v7_gl4_simyr2000_c240515.nc

[PeterHjortLauritzen]

We are only supporting an F2000climo COMPSET for CESM3 (out-of-the-box).

[ekluzek]

So @PeterHjortLauritzen just to confirm that means you need 2000 finidat files -- but not 2010 then? Are the 2010 compsets/tests going to be removed in CAM?

This would be easier for us as we'd just support 2000 and remove 2010 compsets we had for CAM support.

[PeterHjortLauritzen]

There are some FC2010 compset tests (full chemistry) so removal would need approval from the chemistry working group @tilmes

[slevis-lmwg]

Reminder: I still need to
./rimport -file lnd/clm2/initdata_esmf/ctsm5.2/clmi.I2000Clm52BgcCrop.2000-01-01.1.9x2.5_gx1v7_gl4_simyr2000_c240515.nc
I just tried it on casper (derecho is down) and got this error:
ERROR: /glade/p/cesmdata/cseg/cseg/inputdata/lnd/clm2/initdata_esmf/ctsm5.2/clmi.I2000Clm52BgcCrop.2000-01-01.1.9x2.5_gx1v7_gl4_simyr2000_c240515.nc does not exist
I tried it on izumi and got this error:
svn: E000028: Failed writing updated file: No space left on device

[slevis-lmwg]

@ekluzek @wwieder
I'm clarifying here based on info from today's meetings. Please comment/correct if I misunderstood.

First iteration of generating IC files occurs once the following work gets merged in ctsm master:

• Crop Calendar work
• Excess Ice
• Dust emissions
• Explicit A/C (just needs to be turned on by default)

Second iteration, after the following work gets merged in ctsm master:

• CTSM5.3 datasets
• MEGAN 2.1
• Fang's Fire PRs
• PPE branch

Third iteration, after this also gets in:

• Soil NOx work

[ekluzek]

Yes, we'll need to do a few different iterations of IC files.

And as such we should reduce how many we create until the final step when tuning has happened.

I think we might just need:

1850 spinup for f09 and ne30pg3?

So maybe you can skip doing 2000 IC files for now? @wwieder does that sound right?

[wwieder]

Maybe we make an F2000 ic file after step 2 that Sam mentioned above, when the 5.3 datasets come in? It really depends on how much other WGs want this, though so it's hard to say for sure.

[wwieder]

This seems like it may be a duplicate of #2403, at least regarding 5.3 datasets and initial conditions