Adding temperature restore capability for SPEAR. #36
Conversation
Added parameter SPEAR_ECDA_SST_RESTORE_TFREEZE to allow activation of sea surface salinity based modification of restoring of temperature. The formula used is different from the Millero (default in SPEAR runs) scheme.
Codecov Report
@@ Coverage Diff @@
## dev/gfdl #36 +/- ##
=========================================
Coverage 28.96% 28.96%
=========================================
Files 240 240
Lines 71212 71212
=========================================
Hits 20623 20623
Misses 50589 50589 Continue to review full report at Codecov.
|
| if ( CS%trestore_SPEAR_ECDA ) then | ||
| do j=js,je ; do i=is,ie | ||
| if (abs(data_restore(i,j)+1.8)<0.0001) then | ||
| data_restore(i,j) = -0.0539*sfc_state%SSS(i,j) |
There was a problem hiding this comment.
This particular expression has a hard-coded constant that may be specific to the particular choice of the expression (here linear) for the freezing point. Would it be better to replace this with a call to the MOM6 code that actually calculates the freezing point based on the salinity?
Also, elsewhere in this routine, I think that it would be a good to make it clear in the comments that this option is only changing the target temperatures when they appear to be very close to the freezing point.
There was a problem hiding this comment.
I see how this could be useful, but before accepting these changes, please consider making the following changes:
- Expand the comments to make it clear that these changes only apply to temperatures near the nominal freezing point.
- Add run-time parameters to specify the dimensional constants used by this new scheme. These parameters should only be logged if this new option used.
We are trying to actively discourage the introduction of any new hard-coded dimensional constants, and we will be moving to replace all existing significant dimensional constants (i.e., not absurdly huge or tiny fill values) in the MOM6 code with run-time variables. The hard-coded dimensional values are fine as the defaults for the run-time variables, because then they will have an explicit context and units that should be obvious to anyone reading the code.
The freezing temperature came from SIS2 code. Changing the default value here to be consistent with that. (-0.054 vs -0.0539) The salinity restoring code used the -0.0539 value also so answers may change if using that code (RESTORE_SALINITY=T)
The freezing temperature came from SIS2 code. Changing the default value here to be consistent with that. (-0.054 vs -0.0539) The salinity restoring code used the -0.0539 value also so answers may change if using that code (RESTORE_SALINITY=T)
There was a problem hiding this comment.
With the more recent updates, I think that this commit is looking right, and I agree that it should be merged in after the pipeline and TC testing have passed on the updated version.
It should be noted that this PR will change the MOM_parameter_doc.all files for some experiments, so there will be a manual update required for MOM6-examples as soon as this PR is accepted.
|
Gaea regression: https://gitlab.gfdl.noaa.gov/ogrp/MOM6/-/pipelines ✔️ 🟡 New parameter:
Squash-merging this one, since subsequent commit logs look incremental. |
Added parameter SPEAR_ECDA_SST_RESTORE_TFREEZE to allow activation of
sea surface salinity based modification of restoring of temperature.
The formula used is different from the Millero (default in SPEAR runs)
scheme.