Update maint1.2: 3D varying GM #487
Conversation
In the GM bolus calculation, there is a specified diffusivity and phase speed (see Ferrari et al. 2010). Currently MPAS-O assumes these values are fixed in time and space. This PR allows separate flags to enable 2D varying (+time) phase speed (based on the first baroclinic mode phase speed) and a separate flag for 3D (+time) varying diffusivity.
Removes the vertical averaging of the GM kappa bolus diffusivity. Testing shows this functionality is not beneficial to simulations
…te_maint1.2 Adds options for 3d varying GM bolus and 2d varying phase speed MPAS-Dev#288 In the GM bolus calculation, there is a specified diffusivity and phase speed (see Ferrari et al. 2010). Currently MPAS-O assumes these values are fixed in time and space. This PR allows separate flags to enable 2D varying (+time) phase speed (based on the first baroclinic mode phase speed) and a separate flag for 3D (+time) varying diffusivity. * pr_test2:dd Removes 2D GM code and debug statements Adds 2d variable GM and fixes phase speed bug Adds options for 3d varying GM bolus and 2d varying phase speed
…_update_maint1.2 Fix threading issue in MPAS-O GM routine MPAS-Dev#376 This PR fixes a threading issue that was introduced in E3SM PR #3057 that brought in the initial implementation of 3D varying GM bolus. After that commit, some E3SM threading tests were observed to fail with different results. Tested with: SMS_P12x2.T62_oQU240.CMPASO-NYF.compy_intel - BFB results 10 times SMS_P12x2.ne4_oQU240.A_WCYCL1850.compy_intel.allactive-mach_mods - BFB results 10 times This last test is one that was failing on sandiatoss3.
…bolus velocity * temperature to eddy stats
…ate_maint1.2 Add GM bolus eddy stats MPAS-Dev#339 Added monthly mean zonal and meridional bolus velocities as well as bolus velocity times temperature to eddy stats for Labrador Sea bias/ GM testing
|
I used cherry-pick on these commits, and there were no conflicts. Compiles with gnu and intel. Are there any other PRs needed from |
| @@ -375,6 +375,22 @@ | |||
| description="If true, the standard GM for the tracer advection and mixing is turned on. This includes the redi portion which is captured in hmix." | |||
| possible_values=".true. or .false." | |||
| /> | |||
| <nml_option name="config_gm_lat_variable_c2" type="logical" default_value=".false." units="NA" | |||
There was a problem hiding this comment.
@jonbob note there are 6 new GM flags added to maint-1.2 with this update.
There was a problem hiding this comment.
@mark-petersen - I'm only seeing 4? Aren't the others just new arrays?
There was a problem hiding this comment.
Yes, just 4. My mistake.
|
@mark-petersen I think we need #392 to prevent build errors when using python 3. |
…nt1.2 Fix python 3 in BGC and CVMix makedep.py MPAS-Dev#392 This is needed for E3SM builds of maint-1.0 with python 3.
|
@darincomeau thanks, I just added that. @xylar can you think of anything else we might need for maint-1.2? |
|
@mark-petersen, yes, I believe we need: #447 This disables computing the land-ice mask in forward mode, which I think is necessary with the new initial condition we're about to set up. |
|
Actually, given the complexity of COMPASS changes in #447, you might want to just cherry-pick xylar@da1f7c9. |
With this merge, landIceMask is no longer computed based on landIceFraction. Instead, it is always read in from an input file. This allows landIceMask for global_ocean test cases to be computed from geometric_features while landIceFraction is computed as before during initialization.
…atures' into ocean/gm_update_maint1.2 Only used the single commit needed to remove landIceMask computation from diagnostics: Read in landIceMask from input
|
@mark-petersen , @darincomeau - what about changing ocean output to single precision in this? It would help with run storage... |
Done, thanks. I'll merge this in and set up an E3SM PR. |
@jonbob good idea, I was thinking the same thing. I can add that on the E3SM PR in a minute. |
|
@mark-petersen - I was just looking at PR's from the last year or so. Do any of these make sense to include: |
|
@xylar or @matthewhoffman, it looks like we should include |
…ocean/gm_update_maint1.2 bug fixes for the nonlocal source term in KPP MPAS-Dev#305 This PR fixes several bugs in the calculation of the nonlocal source term in KPP. for temperature, we need to subtract off the contribution from thickness fluxes that don't change buoyancy. for salinity, we need to add virtual salinity fluxes that do change buoyancy. these fixes have been tested in E3SM G-cases and behave as expected.
|
@matthewhoffman will have a better sense than me. It seems like including this would do no harm. It only matters if we set |
Yes, this is the namelist option used in ISMF compsets. |
@jonbob thanks for the suggestion. I added the KPP nonlocal fix, which is that 3056 (MPAS PR #305) |
|
@mark-petersen - just let me know when you think you're done and we can start the corresponding E3SM PR |
This update adds the following PRs from ocean/develop: