Revisit flux climatology #493

navidcy · 2025-04-30T14:45:24Z

This PR revisits #340 and enhances it to also compute statistics of the components of the heat fluxes.

glwagner · 2025-04-30T14:49:57Z

I think we should add this as an example. It should not implement any custom types --- those should all go into the source code and become tested. What do you think?

navidcy · 2025-04-30T14:51:43Z

Yes. I totally agree!

navidcy · 2025-05-03T06:15:26Z

using Oceananigans#main I get:

julia> v = FieldTimeSeries(v_meta, arch; time_indices_in_memory=13)
ERROR: ArgumentError: Cannot specify north boundary condition ValueBoundaryCondition: Oceananigans.BoundaryConditions.PolarValue{OffsetArrays.OffsetArray{Float32, 3, Array{Float32, 3}}, Symbol} on a field at Face()!
Stacktrace:
  [1] validate_boundary_condition_location(bc::BoundaryCondition{…}, loc::Face, side::Symbol)
    @ Oceananigans.Fields ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:57
  [2] validate_boundary_conditions(loc::Tuple{…}, grid::LatitudeLongitudeGrid{…}, bcs::FieldBoundaryConditions{…})
    @ Oceananigans.Fields ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:76
  [3] #apply_regionally!#58
    @ ~/.julia/packages/Oceananigans/WEWMN/src/Utils/multi_region_transformation.jl:121 [inlined]
  [4] apply_regionally!
    @ ~/.julia/packages/Oceananigans/WEWMN/src/Utils/multi_region_transformation.jl:118 [inlined]
  [5] macro expansion
    @ ~/.julia/packages/Oceananigans/WEWMN/src/Utils/multi_region_transformation.jl:206 [inlined]
  [6] Field(loc::Tuple{…}, grid::LatitudeLongitudeGrid{…}, data::OffsetArrays.OffsetArray{…}, bcs::FieldBoundaryConditions{…}, indices::Tuple{…}, op::Nothing, status::Nothing)
    @ Oceananigans.Fields ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:101
  [7] Field(loc::Tuple{…}, grid::LatitudeLongitudeGrid{…}, T::DataType; indices::Tuple{…}, data::OffsetArrays.OffsetArray{…}, boundary_conditions::FieldBoundaryConditions{…}, operand::Nothing, status::Nothing)
    @ Oceananigans.Fields ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:194
  [8] Field(loc::Tuple{…}, grid::LatitudeLongitudeGrid{…}, T::DataType)
    @ Oceananigans.Fields ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:185
  [9] #_#12
    @ ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:182 [inlined]
 [10] Field
    @ ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:178 [inlined]
 [11] (Field{…})(grid::LatitudeLongitudeGrid{…})
    @ Oceananigans.Fields ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:178
 [12] empty_field(metadata::Metadata{ECCO4Monthly, DateTime}; architecture::CPU, horizontal_halo::Tuple{Int64, Int64})
    @ ClimaOcean.DataWrangling ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/DataWrangling/metadata.jl:257
 [13] empty_field
    @ ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/DataWrangling/metadata.jl:230 [inlined]
 [14] FieldTimeSeries(metadata::Metadata{ECCO4Monthly, StepRange{DateTime, Month}}, architecture::CPU; kw::@Kwargs{time_indices_in_memory::Int64})
    @ ClimaOcean.DataWrangling ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/DataWrangling/restoring.jl:115
 [15] top-level scope
    @ REPL[11]:1
Some type information was truncated. Use `show(err)` to see complete types.

I was under the impression that CliMA/Oceananigans.jl#4457 and CliMA/Oceananigans.jl#4463 dealt with this, no?

navidcy · 2025-05-03T06:42:02Z

using Oceananigans#main I get:

julia> v = FieldTimeSeries(v_meta, arch; time_indices_in_memory=13)
ERROR: ArgumentError: Cannot specify north boundary condition ValueBoundaryCondition: Oceananigans.BoundaryConditions.PolarValue{OffsetArrays.OffsetArray{Float32, 3, Array{Float32, 3}}, Symbol} on a field at Face()!
Stacktrace:
  [1] validate_boundary_condition_location(bc::BoundaryCondition{…}, loc::Face, side::Symbol)
    @ Oceananigans.Fields ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:57
  [2] validate_boundary_conditions(loc::Tuple{…}, grid::LatitudeLongitudeGrid{…}, bcs::FieldBoundaryConditions{…})
    @ Oceananigans.Fields ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:76
  [3] #apply_regionally!#58
    @ ~/.julia/packages/Oceananigans/WEWMN/src/Utils/multi_region_transformation.jl:121 [inlined]
  [4] apply_regionally!
    @ ~/.julia/packages/Oceananigans/WEWMN/src/Utils/multi_region_transformation.jl:118 [inlined]
  [5] macro expansion
    @ ~/.julia/packages/Oceananigans/WEWMN/src/Utils/multi_region_transformation.jl:206 [inlined]
  [6] Field(loc::Tuple{…}, grid::LatitudeLongitudeGrid{…}, data::OffsetArrays.OffsetArray{…}, bcs::FieldBoundaryConditions{…}, indices::Tuple{…}, op::Nothing, status::Nothing)
    @ Oceananigans.Fields ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:101
  [7] Field(loc::Tuple{…}, grid::LatitudeLongitudeGrid{…}, T::DataType; indices::Tuple{…}, data::OffsetArrays.OffsetArray{…}, boundary_conditions::FieldBoundaryConditions{…}, operand::Nothing, status::Nothing)
    @ Oceananigans.Fields ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:194
  [8] Field(loc::Tuple{…}, grid::LatitudeLongitudeGrid{…}, T::DataType)
    @ Oceananigans.Fields ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:185
  [9] #_#12
    @ ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:182 [inlined]
 [10] Field
    @ ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:178 [inlined]
 [11] (Field{…})(grid::LatitudeLongitudeGrid{…})
    @ Oceananigans.Fields ~/.julia/packages/Oceananigans/WEWMN/src/Fields/field.jl:178
 [12] empty_field(metadata::Metadata{ECCO4Monthly, DateTime}; architecture::CPU, horizontal_halo::Tuple{Int64, Int64})
    @ ClimaOcean.DataWrangling ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/DataWrangling/metadata.jl:257
 [13] empty_field
    @ ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/DataWrangling/metadata.jl:230 [inlined]
 [14] FieldTimeSeries(metadata::Metadata{ECCO4Monthly, StepRange{DateTime, Month}}, architecture::CPU; kw::@Kwargs{time_indices_in_memory::Int64})
    @ ClimaOcean.DataWrangling ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/DataWrangling/restoring.jl:115
 [15] top-level scope
    @ REPL[11]:1
Some type information was truncated. Use `show(err)` to see complete types.

I was under the impression that CliMA/Oceananigans.jl#4457 and CliMA/Oceananigans.jl#4463 dealt with this, no?

Ignore this; I wasn't on main!

navidcy · 2025-05-03T07:04:13Z

I can run now. I hit this error:

julia> include("flux_climatology.jl")
[ Info: Initializing simulation...
[ Info: Initializing simulation...
[ Info:     ... simulation initialization complete (28.375 μs)
[ Info: Iter: 0, time: 0 seconds, Δt: 3 hours, max|u|: (0.00e+00, 0.00e+00) m s⁻¹, extrema(T): (0.00, 0.00) ᵒC, wall time: 3.121 seconds
[ Info:     ... simulation initialization complete (592.593 ms)
[ Info: Executing initial time step...
[ Info:     ... initial time step complete (1.441 seconds).
[ Info: Iter: 10, time: 1.250 days, Δt: 3 hours, max|u|: (9.62e-01, 5.25e-01) m s⁻¹, extrema(T): (31.52, -2.00) ᵒC, wall time: 5.634 seconds
[ Info: Iter: 20, time: 2.500 days, Δt: 3 hours, max|u|: (9.43e-01, 5.26e-01) m s⁻¹, extrema(T): (31.51, -2.00) ᵒC, wall time: 4.154 seconds
[ Info: Iter: 30, time: 3.750 days, Δt: 3 hours, max|u|: (9.23e-01, 5.26e-01) m s⁻¹, extrema(T): (31.50, -2.00) ᵒC, wall time: 4.537 seconds
[ Info: Iter: 40, time: 5 days, Δt: 3 hours, max|u|: (9.03e-01, 5.29e-01) m s⁻¹, extrema(T): (31.49, -2.00) ᵒC, wall time: 5.144 seconds
[ Info: Iter: 50, time: 6.250 days, Δt: 3 hours, max|u|: (8.84e-01, 5.33e-01) m s⁻¹, extrema(T): (31.48, -1.99) ᵒC, wall time: 5.742 seconds
[ Info: Iter: 60, time: 7.500 days, Δt: 3 hours, max|u|: (8.64e-01, 5.37e-01) m s⁻¹, extrema(T): (31.48, -1.99) ᵒC, wall time: 5.001 seconds
[ Info: Iter: 70, time: 8.750 days, Δt: 3 hours, max|u|: (8.44e-01, 5.41e-01) m s⁻¹, extrema(T): (31.47, -1.99) ᵒC, wall time: 4.638 seconds
[ Info: Iter: 80, time: 10 days, Δt: 3 hours, max|u|: (8.24e-01, 5.46e-01) m s⁻¹, extrema(T): (31.48, -1.99) ᵒC, wall time: 4.456 seconds
[ Info: Iter: 90, time: 11.250 days, Δt: 3 hours, max|u|: (8.05e-01, 5.50e-01) m s⁻¹, extrema(T): (31.49, -1.99) ᵒC, wall time: 5.032 seconds
[ Info: Iter: 100, time: 12.500 days, Δt: 3 hours, max|u|: (7.85e-01, 5.54e-01) m s⁻¹, extrema(T): (31.49, -1.99) ᵒC, wall time: 6.087 seconds
[ Info: Iter: 110, time: 13.750 days, Δt: 3 hours, max|u|: (7.65e-01, 5.58e-01) m s⁻¹, extrema(T): (31.50, -1.99) ᵒC, wall time: 6.876 seconds
[ Info: Iter: 120, time: 15 days, Δt: 3 hours, max|u|: (7.45e-01, 5.63e-01) m s⁻¹, extrema(T): (31.51, -1.99) ᵒC, wall time: 5.981 seconds
[ Info: Iter: 130, time: 16.250 days, Δt: 3 hours, max|u|: (7.26e-01, 5.67e-01) m s⁻¹, extrema(T): (31.52, -1.99) ᵒC, wall time: 4.903 seconds
[ Info: Iter: 140, time: 17.500 days, Δt: 3 hours, max|u|: (7.24e-01, 5.71e-01) m s⁻¹, extrema(T): (31.52, -1.99) ᵒC, wall time: 5.490 seconds
[ Info: Iter: 150, time: 18.750 days, Δt: 3 hours, max|u|: (7.26e-01, 5.75e-01) m s⁻¹, extrema(T): (31.53, -2.00) ᵒC, wall time: 6.037 seconds
[ Info: Iter: 160, time: 20 days, Δt: 3 hours, max|u|: (7.38e-01, 5.80e-01) m s⁻¹, extrema(T): (31.54, -2.00) ᵒC, wall time: 5.320 seconds
[ Info: Iter: 170, time: 21.250 days, Δt: 3 hours, max|u|: (7.54e-01, 5.84e-01) m s⁻¹, extrema(T): (31.55, -2.01) ᵒC, wall time: 4.888 seconds
[ Info: Iter: 180, time: 22.500 days, Δt: 3 hours, max|u|: (7.70e-01, 5.88e-01) m s⁻¹, extrema(T): (31.55, -2.01) ᵒC, wall time: 5.062 seconds
[ Info: Iter: 190, time: 23.750 days, Δt: 3 hours, max|u|: (7.86e-01, 5.92e-01) m s⁻¹, extrema(T): (31.56, -2.02) ᵒC, wall time: 5.086 seconds
[ Info: Iter: 200, time: 25 days, Δt: 3 hours, max|u|: (8.02e-01, 5.97e-01) m s⁻¹, extrema(T): (31.57, -2.02) ᵒC, wall time: 6.011 seconds
ERROR: LoadError: TaskFailedException

    nested task error: MethodError: no method matching 
    
    Closest candidates are:
      ClimaOcean.OceanSeaIceModels.InterfaceComputations.InterfaceState(::Any, ::Any, ::Any, ::Any, ::Any, ::Any, ::Any, ::Any)
       @ ClimaOcean ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/OceanSeaIceModels/InterfaceComputations/interface_states.jl:341
      ClimaOcean.OceanSeaIceModels.InterfaceComputations.InterfaceState(::FT, ::FT, ::FT, ::FT, ::FT, ::FT, ::FT, ::FT, ::Bool) where FT
       @ ClimaOcean ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/OceanSeaIceModels/InterfaceComputations/interface_states.jl:330
    
    Stacktrace:
     [1] 
       @ ClimaOcean.OceanSeaIceModels.InterfaceComputations ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/OceanSeaIceModels/InterfaceComputations/interface_states.jl:341
     [2] iterate_interface_state
       @ ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/OceanSeaIceModels/InterfaceComputations/compute_interface_state.jl:117 [inlined]
     [3] compute_interface_state
       @ ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/OceanSeaIceModels/InterfaceComputations/compute_interface_state.jl:48 [inlined]
     [4] macro expansion
       @ ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/OceanSeaIceModels/InterfaceComputations/atmosphere_ocean_fluxes.jl:128 [inlined]
     [5] cpu__compute_atmosphere_ocean_interface_state!
       @ ~/.julia/packages/KernelAbstractions/sWSE0/src/macros.jl:304 [inlined]
     [6] cpu__compute_atmosphere_ocean_interface_state!(__ctx__::KernelAbstractions.CompilerMetadata{…}, interface_fluxes::@NamedTuple{…}, interface_temperature::Field{…}, grid::ImmersedBoundaryGrid{…}, clock::Clock{…}, turbulent_flux_formulation::SimilarityTheoryFluxes{…}, interior_state::@NamedTuple{…}, atmosphere_state::@NamedTuple{…}, interface_properties::ClimaOcean.OceanSeaIceModels.InterfaceComputations.InterfaceProperties{…}, atmosphere_properties::@NamedTuple{…}, ocean_properties::@NamedTuple{…})
       @ ClimaOcean.OceanSeaIceModels.InterfaceComputations ./none:0
     [7] __thread_run(tid::Int64, len::Int64, rem::Int64, obj::KernelAbstractions.Kernel{…}, ndrange::Nothing, iterspace::KernelAbstractions.NDIteration.NDRange{…}, args::Tuple{…}, dynamic::KernelAbstractions.NDIteration.DynamicCheck)
       @ KernelAbstractions ~/.julia/packages/KernelAbstractions/sWSE0/src/cpu.jl:144
     [8] (::KernelAbstractions.var"#20#23"{…})()
       @ KernelAbstractions ~/.julia/packages/KernelAbstractions/sWSE0/src/cpu.jl:119
Stacktrace:
  [1] sync_end(c::Channel{Any})
    @ Base ./task.jl:455
  [2] macro expansion
    @ ./task.jl:487 [inlined]
  [3] __run(obj::KernelAbstractions.Kernel{…}, ndrange::Nothing, iterspace::KernelAbstractions.NDIteration.NDRange{…}, args::Tuple{…}, dynamic::KernelAbstractions.NDIteration.DynamicCheck, static_threads::Bool)
    @ KernelAbstractions ~/.julia/packages/KernelAbstractions/sWSE0/src/cpu.jl:118
  [4] (::KernelAbstractions.Kernel{…})(::@NamedTuple{…}, ::Vararg{…}; ndrange::Nothing, workgroupsize::Nothing)
    @ KernelAbstractions ~/.julia/packages/KernelAbstractions/sWSE0/src/cpu.jl:45
  [5] (::KernelAbstractions.Kernel{…})(::@NamedTuple{…}, ::Vararg{…})
    @ KernelAbstractions ~/.julia/packages/KernelAbstractions/sWSE0/src/cpu.jl:38
  [6] _launch!(::CPU, ::ImmersedBoundaryGrid{…}, ::KernelParameters{…}, ::Function, ::@NamedTuple{…}, ::Field{…}, ::Vararg{…}; exclude_periphery::Bool, reduced_dimensions::Tuple{}, active_cells_map::Nothing)
    @ Oceananigans.Utils ~/.julia/packages/Oceananigans/pNFg0/src/Utils/kernel_launching.jl:329
  [7] _launch!
    @ ~/.julia/packages/Oceananigans/pNFg0/src/Utils/kernel_launching.jl:306 [inlined]
  [8] launch!
    @ ~/.julia/packages/Oceananigans/pNFg0/src/Utils/kernel_launching.jl:289 [inlined]
  [9] compute_atmosphere_ocean_fluxes!(coupled_model::OceanSeaIceModel{…})
    @ ClimaOcean.OceanSeaIceModels.InterfaceComputations ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/OceanSeaIceModels/InterfaceComputations/atmosphere_ocean_fluxes.jl:44
 [10] #update_state!#4
    @ ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/OceanSeaIceModels/time_step_ocean_sea_ice_model.jl:57 [inlined]
 [11] update_state!
    @ ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/OceanSeaIceModels/time_step_ocean_sea_ice_model.jl:51 [inlined]
 [12] time_step!(coupled_model::OceanSeaIceModel{…}, Δt::Float64; callbacks::Tuple{}, compute_tendencies::Bool)
    @ ClimaOcean.OceanSeaIceModels ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/src/OceanSeaIceModels/time_step_ocean_sea_ice_model.jl:46
 [13] time_step!(sim::Simulation{…})
    @ Oceananigans.Simulations ~/.julia/packages/Oceananigans/pNFg0/src/Simulations/run.jl:149
 [14] run!(sim::Simulation{…}; pickup::Bool)
    @ Oceananigans.Simulations ~/.julia/packages/Oceananigans/pNFg0/src/Simulations/run.jl:105
 [15] run!
    @ ~/.julia/packages/Oceananigans/pNFg0/src/Simulations/run.jl:92 [inlined]
 [16] compute_flux_climatology(earth::Simulation{…})
    @ Main ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/experiments/flux_climatology/flux_climatology.jl:108
 [17] top-level scope
    @ ~/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/experiments/flux_climatology/flux_climatology.jl:239
 [18] include(fname::String)
    @ Base.MainInclude ./client.jl:494
 [19] top-level scope
    @ REPL[9]:1
in expression starting at /Users/navid/Library/CloudStorage/OneDrive-TheUniversityofMelbourne/Documents/Research/ClimaOcean.jl-v4/experiments/flux_climatology/flux_climatology.jl:239
Some type information was truncated. Use `show(err)` to see complete types.

navidcy · 2025-05-04T06:22:34Z

@simone-silvestri, why this choice of albedo/emmisivity?

ClimaOcean.jl/experiments/flux_climatology/flux_climatology.jl

Line 209 in 20db923

    
           earth_model = OceanSeaIceModel(ocean, nothing; atmosphere, radiation = Radiation(ocean_emissivity=0, ocean_albedo=1))

simone-silvestri · 2025-05-05T05:58:16Z

this is the eliminate radiation from the fluxes and account only for the turbulent ones. Let's put some reasonable radiation though (the default one) and see the results

glwagner · 2025-05-05T12:50:21Z

For a climatology the important part is diagnosing all the fluxes separately anyways. But you do sum at the end to see how it adds up.

glwagner · 2025-05-05T12:51:03Z

A key question in my mind is how the fluxes differ between using a monthly vs daily ocean data product, eg capturing variability. One could also average the atmospheric state monthly to see how that changes the results.

glwagner · 2025-05-05T12:51:27Z

Don't we need a separate PR to implement some of the types that were present in the original climatology? Or are you combining the two efforts

navidcy · 2025-05-05T13:20:10Z

I'll do a separate PR in a bit. First I'm testing things within the script and modifying these types etc. When I'm settled with what works I'll implement in a new PR.

We also realised with @simone-silvestri that DataWrangling method hardcodes Float32 for Datasets but it would be nice if we implement an, e.g., a

const dataset_defaults = Oceananigans.Defaults(FloatType=Float32)

so we can have a handle on the float type. We were getting type-instabilities from the iterator.

glwagner · 2025-05-05T21:21:09Z

don't re-use Oceananigand.Defaults because that is intended to be specific to Oceananigans --- for example it will eventually have more dfeaults like gravitational_acceleration, planetary_rotation_rate, etc.

glwagner · 2025-05-05T21:21:48Z

Do we really need a global default for the datasets? Or should we just make this settable in Metadata constructor? Global defaults are not very nice --- we resorted to them in Oceananigans FloatType because we don't have any better idea and the problem is hard. But they should be avoided at all costs because global defaults damage scriptability

navidcy · 2025-05-06T14:56:01Z

Here's the climatology computed with the script on this PR using the defaults as is on main atm (gustiness_parameter = 1):

Here's again when I changed the default to gustiness_parameter = 6.5:

Very different from #340!

simone-silvestri · 2025-05-06T15:00:20Z

looks like gustiness does not change the heat flux as much as I expected. There must be another bug somewhere...

simone-silvestri · 2025-05-06T15:24:26Z

Btw, salinity flux and stresses look similar to #340 but heat flux is completely different, so it might be a temperature - specific problem (like for example radiation)

glwagner · 2025-05-06T15:30:06Z

The heat fluxes are everywhere negative??????

glwagner · 2025-05-06T15:30:37Z

Can you also plot the difference between the atmosphere and ocean surface temperature? The sensible heat flux must have the same sign.

navidcy · 2025-05-07T05:22:03Z

Can you also plot the difference between the atmosphere and ocean surface temperature? The sensible heat flux must have the same sign.

You mean plot: mean(Ta) - mean(To) or mean(Ta - To)?

navidcy · 2025-05-07T05:34:41Z

The heat fluxes are everywhere negative??????

Yes.... no wonder we are seeing warm SSTs....

navidcy · 2025-05-07T08:55:02Z

OK, revisiting this, me and @simone-silvestri noticed that #340 was using radiation with ocean_emissivity=0, ocean_albedo=1. So when I use these params in Radiation I get:

This is very very similar to #340. Does this points to the Radiation being the culprit? Note that the Radiation formulation hasn't changed from #340 to now.

…' into ss-nc/fix-climatoloy-experiment

…imaOcean.jl into ss-nc/fix-climatoloy-experiment

glwagner · 2025-05-07T13:36:26Z

Why would radiation change the sensible heat flux?

navidcy · 2025-05-09T12:28:41Z

simulation gets very warm still..

https://clima.github.io/ClimaOceanDocumentation/previews/PR493/literated/near_global_ocean_simulation/

glwagner · 2025-05-09T17:33:57Z

Could the problem be that we don't have penetrating radiation?

navidcy · 2025-05-11T06:19:16Z

@simone-silvestri, @taimoorsohail pointed to me that there are also daily heat fluxes from ECCO: https://podaac.jpl.nasa.gov/dataset/ECCO_L4_HEAT_FLUX_05DEG_DAILY_V4R4

(this will remove the uncertainty whether we are incorporating nonlinearities by comparing the mean of the fluxes vs the fluxes of the mean)

taimoorsohail · 2025-05-12T04:25:44Z

It might also be worth plotting the surface integrated heat fluxes (i.e. sum(Q*Az)) over time to compare with the ECCO heat fluxes, if we are concerned it's a global bias rather than a regional issue. Might be easier than eyeballing differences in surface flux plots.

simone-silvestri and others added 2 commits April 30, 2025 16:28

start climatoloy

1667e33

few fixes + sensible and latent heat fluxes

e26b67e

This was referenced Apr 30, 2025

flux investigations taimoorsohail/ocean-ensembles#51

Draft

Ocean is too warm in 1 deg Tripolar simulation taimoorsohail/ocean-ensembles#22

Open

navidcy added the atmosphere-ocean coupling ⛅️ 🌊 When the relationship isn't as smooth as expected label Apr 30, 2025

simone-silvestri and others added 4 commits May 1, 2025 10:11

no space

63bd72a

some fixing

9760ef8

run it on GPU for the moment

c8f27ad

Merge branch 'main' into ss-nc/fix-climatoloy-experiment

20db923

navidcy added 4 commits May 6, 2025 14:16

add dataset_type

5edabc0

use dataset_defaults.FloatType

9ccb0cc

add Qc+Qv + better averaging that do not depend on the end_time

894ab39

fixes

f139804

navidcy added 2 commits May 6, 2025 17:58

update figs

e6428da

compute std in the kernel

0d39d05

navidcy and others added 10 commits May 7, 2025 13:06

Merge branch 'main' into ss-nc/fix-climatoloy-experiment

0159623

add all this

0ec0bd4

Merge remote-tracking branch 'origin/ss/add-radiative-flux-diagnostic…

33a7d7f

…' into ss-nc/fix-climatoloy-experiment

easier flux computation by storing the flux

b5dfad2

Merge branch 'ss-nc/fix-climatoloy-experiment' of github.com:CliMA/Cl…

e1e64ba

…imaOcean.jl into ss-nc/fix-climatoloy-experiment

improve stats

1e048c9

Merge branch 'ss-nc/fix-climatoloy-experiment' of github.com:CliMA/Cl…

2a9922b

…imaOcean.jl into ss-nc/fix-climatoloy-experiment

fix the edson stability function

455a12a

some fixes

60969bc

add some fluxes

ac25d0a

navidcy mentioned this pull request May 7, 2025

Fix mistake in EdsonMomentumStabilityFunctions #503

Merged

Merge branch 'main' into ss-nc/fix-climatoloy-experiment

2823180

simone-silvestri added the build docs Add this label to built the docs in a PR label May 8, 2025

simone-silvestri added 2 commits May 8, 2025 15:51

small change to trigger docs build

d47961f

Update interpolate_atmospheric_state.jl

50861af

Revisit flux climatology #493

Are you sure you want to change the base?

Revisit flux climatology #493

Uh oh!

Conversation

navidcy commented Apr 30, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

glwagner commented Apr 30, 2025

Uh oh!

navidcy commented Apr 30, 2025

Uh oh!

navidcy commented May 3, 2025

Uh oh!

navidcy commented May 3, 2025

Uh oh!

navidcy commented May 3, 2025

Uh oh!

navidcy commented May 4, 2025

Uh oh!

simone-silvestri commented May 5, 2025

Uh oh!

glwagner commented May 5, 2025

Uh oh!

glwagner commented May 5, 2025

Uh oh!

glwagner commented May 5, 2025

Uh oh!

navidcy commented May 5, 2025

Uh oh!

glwagner commented May 5, 2025

Uh oh!

glwagner commented May 5, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

navidcy commented May 6, 2025

Uh oh!

simone-silvestri commented May 6, 2025

Uh oh!

simone-silvestri commented May 6, 2025

Uh oh!

glwagner commented May 6, 2025

Uh oh!

glwagner commented May 6, 2025

Uh oh!

navidcy commented May 7, 2025

Uh oh!

navidcy commented May 7, 2025

Uh oh!

navidcy commented May 7, 2025

Uh oh!

glwagner commented May 7, 2025

Uh oh!

navidcy commented May 9, 2025

Uh oh!

glwagner commented May 9, 2025

Uh oh!

navidcy commented May 11, 2025

Uh oh!

taimoorsohail commented May 12, 2025

Uh oh!

Uh oh!

navidcy commented Apr 30, 2025 •

edited

Loading

glwagner commented May 5, 2025 •

edited

Loading