Improve design of nonorographic gravity wave parameterization #3226
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
src/parameterized_tendencies/gravity_wave_drag/non_orographic_gravity_wave.jl
Outdated
Show resolved
Hide resolved
| FT = eltype(ᶜρ) | ||
| ρ_end = Fields.level(ᶜρ, Spaces.nlevels(axes(ᶜρ))) | ||
| ρ_endm1 = Fields.level(ᶜρ, Spaces.nlevels(axes(ᶜρ)) - 1) | ||
| Boundary_value = Fields.Field( |
There was a problem hiding this comment.
Can we give this a more specific name? And use snake_case for the name?
Comment on lines
346
to
378
| u_end = Fields.level(ᶜu, Spaces.nlevels(axes(ᶜu))) | ||
| u_endm1 = Fields.level(ᶜu, Spaces.nlevels(axes(ᶜu)) - 1) | ||
| Boundary_value = Fields.Field( | ||
| FT(2) .* Fields.field_values(u_end) .- Fields.field_values(u_endm1), | ||
| axes(u_end), | ||
| ) | ||
| R1 = Operators.RightBiasedC2F(; top = Operators.SetValue(Boundary_value)) | ||
| R2 = Operators.RightBiasedF2C(;) | ||
| ᶜu_p1 .= R2.(R1.(ᶜu)) | ||
|
|
||
| v_end = Fields.level(ᶜv, Spaces.nlevels(axes(ᶜv))) | ||
| v_endm1 = Fields.level(ᶜv, Spaces.nlevels(axes(ᶜv)) - 1) | ||
| Boundary_value = Fields.Field( | ||
| FT(2) .* Fields.field_values(v_end) .- Fields.field_values(v_endm1), | ||
| axes(v_end), | ||
| ) | ||
| R1 = Operators.RightBiasedC2F(; top = Operators.SetValue(Boundary_value)) | ||
| R2 = Operators.RightBiasedF2C(;) | ||
| ᶜv_p1 .= R2.(R1.(ᶜv)) | ||
|
|
||
| Boundary_value = Fields.level(ᶜbf, Spaces.nlevels(axes(ᶜbf))) | ||
| R1 = Operators.RightBiasedC2F(; top = Operators.SetValue(Boundary_value)) | ||
| R2 = Operators.RightBiasedF2C(;) | ||
| ᶜbf_p1 .= R2.(R1.(ᶜbf)) | ||
|
|
||
| z_end = Fields.level(ᶜz, Spaces.nlevels(axes(ᶜz))) | ||
| z_endm1 = Fields.level(ᶜz, Spaces.nlevels(axes(ᶜz)) - 1) | ||
| Boundary_value = Fields.Field( | ||
| FT(2) .* Fields.field_values(z_end) .- Fields.field_values(z_endm1), | ||
| axes(z_end), | ||
| ) | ||
| R1 = Operators.RightBiasedC2F(; top = Operators.SetValue(Boundary_value)) | ||
| R2 = Operators.RightBiasedF2C(;) |
There was a problem hiding this comment.
It looks like there's a lot of code duplication here, can we somehow reduce it?
Xiaoyang-Xie
force-pushed
the
xxy/gravity_wave
branch
from
c1f5529 to
acd7398
Compare
|
I need to update some other packages too. I can update dependencies. |
Xiaoyang-Xie
force-pushed
the
xxy/gravity_wave
branch
3 times, most recently
from
3161a64 to
453fa3b
Compare
| ), | ||
| Val(nc), | ||
| ) | ||
| Bsum = sum(abs.(B0)) |
There was a problem hiding this comment.
This will allocate, but I think we can write this in a non-allocating way
| Hb = (z_kp1 - z_k) / log(ρ_k / ρ_kp1) # density scale height | ||
| alp2 = FT1(0.25) / (Hb * Hb) | ||
| ω_r = sqrt((bf_kp1 * bf_kp1 * k2) / (k2 + alp2)) # omc: (critical frequency that marks total internal reflection) | ||
| fm = 0.0 |
There was a problem hiding this comment.
Let's use FT to make sure that we have type-stable results for Float32. (here and everywhere in src/)
Xiaoyang-Xie
force-pushed
the
xxy/gravity_wave
branch
2 times, most recently
from
e6d2cb6 to
a0bd8f0
Compare
Xiaoyang-Xie
force-pushed
the
xxy/gravity_wave
branch
from
a0bd8f0 to
3df4b53
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Purpose
Part of #897
Content
1.The function non_orographic_gravity_wave_forcing function has been entirely refactored, now it is more efficient with much less allocations.
2. All three tests for non_orographic_gravity_wave in "/Users/xiexiaoya/ClimaAtmos.jl/test/parameterized_tendencies/gravity_wave/non_orographic_gravity_wave" have been rewriten.
To-do
Change the type of the variable "mask" to be Static Bitvector （This is a newly defined data type), since a vector frequently modified within the column_accumulate function will significantly impact the performance of the program on the GPU.