Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

RRTMGP introduces errors in restarted simulations #3330

Closed
Sbozzolo opened this issue Sep 23, 2024 · 1 comment · Fixed by #3361
Closed

RRTMGP introduces errors in restarted simulations #3330

Sbozzolo opened this issue Sep 23, 2024 · 1 comment · Fixed by #3361
Assignees

Comments

@Sbozzolo
Copy link
Member

Testing with this configuration:

                    test_dict = Dict(
                        "test_dycore_consistency" => true, # We will add NaNs to the cache, just to make sure
                        "check_nan_every" => 3,
                        "log_progress" => false,
                        "moist" => "equil",
                        "precip_model" => "0M",
                        "config" => configuration,
                        "topography" => "NoWarp",
                        "turbconv" => "diagnostic_edmfx",
                        "dt" => "1secs",
                        "bubble" => bubble,
                        "viscous_sponge" => true,
                        "rayleigh_sponge" => true,
                        "insolation" => "timevarying",
                        "rad" => "clearsky",
                        "dt_rad" => "1secs",
                        "surface_setup" => "DefaultMoninObukhov",
                        # "implicit_diffusion" => true,
                        "call_cloud_diagnostics_per_stage" => true,  # Needed to ensure that cloud variables are computed
                        "t_end" => "3secs",
                        "dt_save_state_to_disk" => "1secs",
                        "enable_diagnostics" => false,
                        "output_dir" => joinpath(output_loc, job_id),
                    )

Here are some of the differences (there's more)

Test Failed at /home/sbozzolo/repos/ClimaAtmos.jl/test/restart.jl:42
  Expression: maximum(error) <= 100 * eps(eltype(arr1))
   Evaluated: 0.3669424f0 <= 1.1920929f-5
     Context: name = integrator.u.c.uₕ.components.data.1
              what = reading and simulating
              config = sphere equil 0M NoWarp

Test Failed at /home/sbozzolo/repos/ClimaAtmos.jl/test/restart.jl:42
  Expression: maximum(error) <= 100 * eps(eltype(arr1))
   Evaluated: 0.3669423f0 <= 1.1920929f-5
     Context: name = integrator.u.c.uₕ.components.data.2
              what = reading and simulating
              config = sphere equil 0M NoWarp

Test Failed at /home/sbozzolo/repos/ClimaAtmos.jl/test/restart.jl:42
  Expression: maximum(error) <= 100 * eps(eltype(arr1))
   Evaluated: 1.6163609f-5 <= 1.1920929f-5
     Context: name = integrator.u.f.u₃.components.data.1
              what = reading and simulating
              config = sphere equil 0M NoWarp

Test Failed at /home/sbozzolo/repos/ClimaAtmos.jl/test/restart.jl:42
  Expression: maximum(error) <= 100 * eps(eltype(arr1))
   Evaluated: 0.3669424f0 <= 1.1920929f-5
     Context: name = integrator.p.precomputed.ᶜu.components.data.1
              what = reading and simulating
              config = sphere equil 0M NoWarp

Test Failed at /home/sbozzolo/repos/ClimaAtmos.jl/test/restart.jl:42
  Expression: maximum(error) <= 100 * eps(eltype(arr1))
   Evaluated: 0.3669423f0 <= 1.1920929f-5
     Context: name = integrator.p.precomputed.ᶜu.components.data.2
              what = reading and simulating
              config = sphere equil 0M NoWarp

Test Failed at /home/sbozzolo/repos/ClimaAtmos.jl/test/restart.jl:42
  Expression: maximum(error) <= 100 * eps(eltype(arr1))
   Evaluated: 0.3669424f0 <= 1.1920929f-5
     Context: name = integrator.p.precomputed.ᶜuʲs.1.components.data.1
              what = reading and simulating
              config = sphere equil 0M NoWarp

Test Failed at /home/sbozzolo/repos/ClimaAtmos.jl/test/restart.jl:42
  Expression: maximum(error) <= 100 * eps(eltype(arr1))
   Evaluated: 0.3669423f0 <= 1.1920929f-5
     Context: name = integrator.p.precomputed.ᶜuʲs.1.components.data.2
              what = reading and simulating
              config = sphere equil 0M NoWarp

Test Failed at /home/sbozzolo/repos/ClimaAtmos.jl/test/restart.jl:42
  Expression: maximum(error) <= 100 * eps(eltype(arr1))
   Evaluated: 0.3669424f0 <= 1.1920929f-5
     Context: name = integrator.p.precomputed.ᶜu⁰.components.data.1
              what = reading and simulating
              config = sphere equil 0M NoWarp

Test Failed at /home/sbozzolo/repos/ClimaAtmos.jl/test/restart.jl:42
  Expression: maximum(error) <= 100 * eps(eltype(arr1))
   Evaluated: 0.3669423f0 <= 1.1920929f-5
     Context: name = integrator.p.precomputed.ᶜu⁰.components.data.2
              what = reading and simulating
              config = sphere equil 0M NoWarp
@Sbozzolo Sbozzolo self-assigned this Sep 23, 2024
@akshaysridhar
Copy link
Member

Are these differences also true for scalars? Do we get exact matches across states if the default rad = nothing is used?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants