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.

Sign up for GitHub

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

Jump to bottom

Add a legacy version of diabatic_driver + updates in KPP and tidal mixing #777

Closed
wants to merge 67 commits into from
Closed

Add a legacy version of diabatic_driver + updates in KPP and tidal mixing #777

wants to merge 67 commits into from

Conversation

gustavo-marques
Copy link
Collaborator

A new input parameter has been added (USE_LEGACY_DIABATIC_DRIVER). If true, the model will
use a legacy version of the diabatic driver (module MOM_legacy_diabatic_driver). This is
temporary and is needed to avoid change in answers while MOM_diabatic_driver is been
restructured.

In addition, this PR also adds the following capabilities:

KPP BLD smoothing:

  • The i,j loop to compute (1) KPP Boundary layer depth and (2) diffusivities is divided into two loops such that BLD is computed over the entire grid before diffusivities get computed. This is because BLD of neighboring cells is necessary to apply smoothing. The new loop containing KPP BLD computation is placed in a new subroutine called KPP_compute_BLD.
  • A new logical called smoothBLD now controls whether a Laplacian Filter is to be applied on BLD after BLD is computed by KPL and before diffusivities are computed.
  • A new subroutine, KPP_smooth_BLD, is added that implements a 1-1-4-1-1 Laplacian filter one time on BLD to reduce any horizontal two-grid-point noise.

Schmittner tidal mixing

  • In tidal mixing module, the capability of reading 3d tidal energy dissipation data (in ER03 format) is added as an option.
  • CVMix Schmittner tidal mixing scheme is added as an option and all the necessary CVMix subroutine calls are added.
  • Diagnostics are added for Schmittner tidal mixing.

This PR does not change answers (only Intel was tested) recorded for dev-master-candidate-2018-05-15.

gustavo-marques and others added 30 commits April 10, 2018 12:54
@gustavo-marques
Adding option to smooth Ri using a 1-2-1 filter
99ca9a7
@gustavo-marques
Fill Ri_grad in vanished layers with adjacent value, just when Ri smo…
b14213a 
…oth is enabled
@gustavo-marques
Update CVMix
c4f1f55
@gustavo-marques
Always allocate CS%OBLdepth since other modules may need to know OBLd…
0c363ae 
…epth
@alperaltuntas
distinguish profiles and cvmix schemes
8a4a5ed
@gustavo-marques
Initialize visc%Kv_slow and update halos
60f7d66
@gustavo-marques
Add "slow" vertical viscosity in vertvisc_coef
d774e5f
@gustavo-marques
Set visc%Kv_slow to zero in diabatic
fd23f91
@gustavo-marques
Add option to save Kv_slow and Kv
1488b78
@gustavo-marques
Remove trailing space
69c2c96
@gustavo-marques
Attempt to add diag. for total vertical visc.
d5be1f8 
* I believe there is something wrong (halo updates?) with the way
this is being done. It needs to be fixed!
@alperaltuntas
add capability to read ER03 energy file
20cc62b
@gustavo-marques
Bug fix in MOM_bkgnd_mixing
dcfb722 
The background vertical diff was being added to itself, which is
wrong and was leading to a increase in kd_bkgnd over time. This
commit fixes this problem.
@gustavo-marques
Add option to diagnose Kv at u and v points
d09e809 
Total vertical viscosity can now be diagnosed at u and v points.
@alperaltuntas
add call to compute_Schmittner_invariant
f132940
@gustavo-marques
Add option to post diags for bkgnd_mixing
901c301
@gustavo-marques
Merge branch 'dev/ncar' into adding_cvmix_part2
f6a21e7
@gustavo-marques
Fix bug in MOM_cvmix_conv
a0358fb
@gustavo-marques
Rename variable in register_diag_field
9d6cb46 
Variable names are now consistent with what is used in
other modules.
@gustavo-marques
Add missing halo update for visc%Kv_slow
78656bb
@gustavo-marques
Initialize Kd, Kd_int and Kv_slow using interior values specified by …
3385857 
…user
@gustavo-marques
Add a factor of 2 when adding Kv_slow into Kv_add
191b5be
@alperaltuntas
add cvmix_compute_SchmittnerCoeff
0e4dce5
@alperaltuntas
resolve conflict
b280bfd
@alperaltuntas
Merge remote-tracking branch 'altuntas/initialize_tidal_ctrl_vars' in…
9590f3b 
…to introduce_cvmix_tidal
@alperaltuntas
remap tidal energy from input coord to model coord
9703e7c
@alperaltuntas
call cvmix_coeffs_tidal_schmittner
93f242f
@alperaltuntas
add diagnostics for schmittner
c04f455
@alperaltuntas
debug 3d tidal energy remapping
05a2e59
@alperaltuntas
merge with dev/ncar
bab2e43
alperaltuntas and others added 21 commits May 15, 2018 14:09
@alperaltuntas
move KPP_compute_BLD to streamline merging
bfa2613
@alperaltuntas
1/2 - merge with candidata may15
e0a5538
@alperaltuntas
2/2 - merge with candidate may15
91e459a
@gustavo-marques
Merge pull request #57 from alperaltuntas/merge_with_candidate_may15
439041a 
Merge with candidate 15 May 2018
@alperaltuntas
Update halo OBLdepth before smoothing
90e8f93
@alperaltuntas
Merge remote-tracking branch 'origin/dev/ncar' into HEAD
adc133e
@gustavo-marques
Merge pull request #58 from alperaltuntas/update_halo_OBLdepth
5a58c34 
Update halo obldepth
@gustavo-marques
Doxygenize subroutine differential_diffuse_T_S
02acd12
@gustavo-marques
First version of Double-diffusion via CVMix
507d34e 
* Delete the old double-diffusion code
@gustavo-marques
Doxygenize MOM_diabatic_aux
81265e6
@gustavo-marques
Fix indentation
05daede
@gustavo-marques
Avoid NaNs when computing stratification parameter
d5ce7a4
@gustavo-marques
Move description to the end of the module
abd620c
@gustavo-marques
Change cvmix to CVMix
152a707
@gustavo-marques
Clean up spaces and comments
cc27362
@gustavo-marques
Merge branch 'dev/ncar' into double_diffusion_cvmix
148a944
@gustavo-marques
Add a flag to control if visc%Kv_slow is used
70d88e4 
This commit adds a flag ADD_KV_SLOW (default is FALSE) that controls if
the background vertical viscosity in the interior (i.e., tidal + background
+ shear + convenction) is addded when computing the coupling coefficient.
The purpose of this flag is to be able to recover previous answers and it will
likely be removed in the future since this option should always be true.
@gustavo-marques
Rename MOM_cvmix_ddiff.F90 -> MOM_CVMix_ddiff.F90
8877c17
@gustavo-marques
Clean the ddiff code and improve comments
bf6c003
@gustavo-marques
Add a legacy version of diabatic_driver
387f4e6 
A new input parameter has been added (USE_LEGACY_DIABATIC_DRIVER). If true, the model will
use a legacy version of the diabatic driver (module MOM_legacy_diabatic_driver). This is
temporary and is needed to avoid change in answers while MOM_diabatic_driver is been
restructured.
@gustavo-marques
Merge pull request #59 from gustavo-marques/add_legacy_diabatic_driver
4f5dee8 
Add a legacy version of diabatic_driver
@adcroft
Copy link
Collaborator

adcroft commented May 22, 2018

@awallcraft, @kshedstrom, This is another PR from NCAR following the discussion during Monday's developer call.

I'm evaluating this at GFDL in https://gitlab.gfdl.noaa.gov/ogrp/MOM6/pipelines/4286

@adcroft
Copy link
Collaborator

adcroft commented May 23, 2018

@gustavo-marques Can you make a branch for the PR so we don't have a moving target? The GFDL tests all passed but you just went and changed the PR so we're back to scratch.

image

@gustavo-marques
Copy link
Collaborator Author

Yes, I will do that. Sorry, forgot that sending PR's using dev/ncar is susceptible to this moving target issue.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@gustavo-marques @adcroft @alperaltuntas