NCAR / amwg_dev

Repo to store model sandboxes and cases used for CAM development
8 stars 2 forks source link

L32 with CAM6 physics #531

Open adamrher opened 2 weeks ago

adamrher commented 2 weeks ago

To test if precip biases in E.Pac are due to cam_dev physics.


Identical to the recent L32 run (https://github.com/NCAR/amwg_dev/issues/524), but substitute %CAM60 for %CAM_DEV in a long name compset that is otherwise identical to BLT1850. Doing it this way (rather than just setting -phys cam6) will ensure that the gust mods aren't turned on.


Case directory:

cecilehannay commented 2 weeks ago

Hi Adam, The longname in (https://github.com/NCAR/amwg_dev/issues/524) uses CAM%DEV%LT%GHGMAM4 Do you want to replace this by CAM60%LT%GHGMAM4 or just CAM60.

cecilehannay commented 2 weeks ago

Also, there was some namelist mods in (https://github.com/NCAR/amwg_dev/issues/524). Do we want these or not?

! dust tuning
dust_emis_fact = 1.3

! clubb diffusion on
clubb_l_do_expldiff_rtm_thlm = .true.

! tuning
microp_aero_wsub_scale  = 1.0
clubb_C8=4.25

! revert NorESMs mods
cldfrc2m_do_avg_aist_algs = .false.
cldfrc2m_qist_min = 1.e-7

bnd_topo = "/glade/campaign/cgd/amp/pel/topo/files/se/ne30pg3_gmted2010_modis_bedmachine_nc3000_Laplace0100_noleak_20240117.nc"
se_hypervis_subcycle = 2
adamrher commented 2 weeks ago

Yes, I would keep those settings except I would not explicitly set dust_emis_fact or clubb_C8. Let's just run with the default values for those two parameters. Thanks!

cecilehannay commented 2 weeks ago

Sounds good. Also, I have skip all the SourceMods/src.cam from (https://github.com/NCAR/amwg_dev/issues/524). They were related to gustiness, pumas and fast dycore.

Last thing, here is the way I set the longname. I replaced: CAM%DEV%LT%GHGMAM4 -> CAM60

I want to make sure you didn't want: CAM60%LT%GHGMAM4

adamrher commented 2 weeks ago

Correct, just CAM60. None of that other stuff.