O1.5.x Make a good multiyear simulation before 7/4

[szy21]

Purpose

We need to show a (good) multiyear AMIP simulation during the VESRI meeting.

People

@szy21 @Sbozzolo @juliasloan25 and others

Target

Resolution: h_elem = 16 (~200km), z_elem = 63
Simulation year: 2010-2014
Atmos: all-sky radiation, diagnostic EDMF, 0-moment microphysics, no aerosols and rrtmg default ozone
Land: bucket

Stretch target

Resolution: h_elem = 30 (~100km), z_elem = 63
Simulation year: 2005-2014
Atmos: all-sky radiation, diagnostic EDMF, 1-moment microphysics, prescribed aerosols and ozone (with radiation)
Land: bucket

Tasks

Simulation

Beta Give feedback

1. https://github.com/CliMA/ClimaModules-Derecho/issues/1
2. 5-year helem 16 diagnostic EDMF + 0M
3. 5-year helem 16 diagnostic EDMF + 0M + better cloud diagnostics
4. 5-year helem 16 diagnostic EDMF + 0M + better cloud diagnostics + aerosols

Implementation

Beta Give feedback

1. Add more aerosol data ClimaArtifacts#23
   +0
   
2. add radiative fluxes to leaderboard #846
   +0
   
3. Add SSPKnoth and make it default ClimaAtmos.jl#3019
   +0
   
4. precipitation is not defined for 1M microphysics #849
   bug +1
   
5. Fix land albedo regridding #764
   🔥 Urgent bug +2
   
6. check fields regridded with TempestRemap #860
   🔥 Urgent 🍃 leaf +2
   
7. Add support for aerosol optics. RRTMGP.jl#510
   Breaking 💥 +1
   
8. Add cloud diagnostics for diagnostic edmf ClimaAtmos.jl#3126
   Microphysics enhancement +2
   

Package update

Beta Give feedback

1. Update RRTMGP ClimaAtmos.jl#3098
   +0
   
2. Update RRTMGP again ClimaAtmos.jl#3114
   +0
3. Use ClimaComms0.6, ClimaAtmos0.26, and update ClimaEarth experiments #864
   +0
4. Fix for ClimaLand 0.12.4 #868
   +0

Issues

Beta Give feedback

1. Unphysical values in TOA downward SW flux #851
   bug +1
   
2. NaN in longwave radiation flux #869
   bug +1
   
3. Many aquaplanet simulations are broken ClimaAtmos.jl#3110
   ASAP +1
   
4. No output in netcdf when cpu run breaks #858
   +0
   
5. Stabilize diagnostic EDMF + 0M AMIP run #865
   +0
   
6. Water is not conserved when precipitation is present ClimaAtmos.jl#3113
   Numerics +1
   
7. make tests GPU-compatible #571
   3 of 3
   GPU +1
   

[szy21]

cc @tapios

[tapios]

cc @tapios

I'd prefer the option 2 configuration, even it's for a shorter simulation (5 years).

[szy21]

I'd prefer the option 2 configuration, even it's for a shorter simulation (5 years).

Sounds good. I'll see if I can get 1-moment microphysics to run stably by the end of this week. And if I can get someone to work on prescribing ozone then it's possible (although for a five-year simulation I don't think there would be much difference in the statistics between prescribed or default ozone profile?).

[tapios]

Ozone variations with latitude would be good to have, but it's not crucial relative to, say, EDMF.

[szy21]

Here are some plots from the 5-year diagnostic edmf and 0-moment microphysics simulation:

I will close this issue soon and move the remaining tasks elsewhere.