Possible zero in the denominator in MOM_mixed_layer_restrat #1168
Comments
|
At first glance this is a good candidate for using Adcroft's rule of reciprocals, namely Another approach might be to take the viscous limit on the mixing rate as floor, namely something like |
|
@alperaltuntas, can you please post here your initial fix to this issue so we can compare all the alternatives? |
|
@gustavo-marques and @alperaltuntas, do you have an update on this issue? |
|
@Hallberg-NOAA, we haven't implemented a fix for this yet. IIRC, the solution I temporarily tested was to skip the computations on land cells, although I am not sure which mask2d is an appropriate test here. Again, this issue occurs only when ustar==0.0, so our current temporary fix is to run with small GUST_CONST when we have to( i.e., when any grid cell corners coincide with the Equator). Also, I am happy to test any of the solutions you proposed. if I interpret your first solution proposal correctly, the fix is to (effectively) set |
Added the runtime parameters KV_RESTRAT and RESTRAT_USTAR_MIN, to build on the improvements in github.com/NOAA-GFDL/pull/251, and to provide run-time physical parameters to avoid the potential division by zero in the mixed_layer_restrat code noted at github.com/mom-ocean/issues/1168. Once this PR is merged onto the main branch of MOM6, that issue can be closed. By default, these do not change answers in the MOM6-examples test suite, but the default value for RESTRAT_USTAR_MIN was taken from the hard-coded value in PR that PR. The six copies of the eddy growth rate timescale calculations were consolidated into a new internal function, growth_time, with some other related minor refactoring of the code. Also, mixedlayer_restrat_register_restarts now takes a unit_scale_type arguments like many other analogous routines. All answers are bitwise identical, but there are new runtime parameters or comments that lead to changes in the MOM_parameter_doc files. Also clarified in the comments sent to the MOM_parameter_doc files how VISBECK_L_SCALE works as a dimensional scaling factor when it is given a negative value, and rescaled its units when read as though it were always in m.
Added the runtime parameters KV_RESTRAT and RESTRAT_USTAR_MIN, to build on the improvements in github.com/NOAA-GFDL/pull/251, and to provide run-time physical parameters to avoid the potential division by zero in the mixed_layer_restrat code noted at github.com/mom-ocean/issues/1168. Once this PR is merged onto the main branch of MOM6, that issue can be closed. By default, these do not change answers in the MOM6-examples test suite, but the default value for RESTRAT_USTAR_MIN was taken from the hard-coded value in PR that PR. The six copies of the eddy growth rate timescale calculations were consolidated into a new internal function, growth_time, with some other related minor refactoring of the code. Also, mixedlayer_restrat_register_restarts now takes a unit_scale_type arguments like many other analogous routines. All answers are bitwise identical, but there are new runtime parameters or comments that lead to changes in the MOM_parameter_doc files. Also clarified in the comments sent to the MOM_parameter_doc files how VISBECK_L_SCALE works as a dimensional scaling factor when it is given a negative value, and rescaled its units when read as though it were always in m.
Added the runtime parameters KV_RESTRAT and RESTRAT_USTAR_MIN, to build on the improvements in github.com/NOAA-GFDL/pull/251, and to provide run-time physical parameters to avoid the potential division by zero in the mixed_layer_restrat code noted at github.com/mom-ocean/issues/1168. Once this PR is merged onto the main branch of MOM6, that issue can be closed. By default, these do not change answers in the MOM6-examples test suite, but the default value for RESTRAT_USTAR_MIN was taken from the hard-coded value in PR that PR. The six copies of the eddy growth rate timescale calculations were consolidated into a new internal function, growth_time, with some other related minor refactoring of the code. Also, mixedlayer_restrat_register_restarts now takes a unit_scale_type arguments like many other analogous routines. All answers are bitwise identical, but there are new runtime parameters or comments that lead to changes in the MOM_parameter_doc files. Also clarified in the comments sent to the MOM_parameter_doc files how VISBECK_L_SCALE works as a dimensional scaling factor when it is given a negative value, and rescaled its units when read as though it were always in m.
Added the runtime parameters KV_RESTRAT and RESTRAT_USTAR_MIN, to build on the improvements in github.com/NOAA-GFDL/pull/251, and to provide run-time physical parameters to avoid the potential division by zero in the mixed_layer_restrat code noted at github.com/mom-ocean/issues/1168. Once this PR is merged onto the main branch of MOM6, that issue can be closed. By default, these do not change answers in the MOM6-examples test suite, but the default value for RESTRAT_USTAR_MIN was taken from the hard-coded value in PR that PR. The six copies of the eddy growth rate timescale calculations were consolidated into a new internal function, growth_time, with some other related minor refactoring of the code. Also, mixedlayer_restrat_register_restarts now takes a unit_scale_type arguments like many other analogous routines. All answers are bitwise identical, but there are new runtime parameters or comments that lead to changes in the MOM_parameter_doc files. Also clarified in the comments sent to the MOM_parameter_doc files how VISBECK_L_SCALE works as a dimensional scaling factor when it is given a negative value, and rescaled its units when read as though it were always in m.
Added the runtime parameters KV_RESTRAT and RESTRAT_USTAR_MIN, to build on the improvements in github.com/NOAA-GFDL/pull/251, and to provide run-time physical parameters to avoid the potential division by zero in the mixed_layer_restrat code noted at github.com/mom-ocean/issues/1168. Once this PR is merged onto the main branch of MOM6, that issue can be closed. By default, these do not change answers in the MOM6-examples test suite, but the default value for RESTRAT_USTAR_MIN was taken from the hard-coded value in PR that PR. The six copies of the eddy growth rate timescale calculations were consolidated into a new internal function, growth_time, with some other related minor refactoring of the code. Also, mixedlayer_restrat_register_restarts now takes a unit_scale_type arguments like many other analogous routines. All answers are bitwise identical, but there are new runtime parameters or comments that lead to changes in the MOM_parameter_doc files. Also clarified in the comments sent to the MOM_parameter_doc files how VISBECK_L_SCALE works as a dimensional scaling factor when it is given a negative value, and rescaled its units when read as though it were always in m.
Added the runtime parameters KV_RESTRAT and RESTRAT_USTAR_MIN, to build on the improvements in github.com//pull/251, and to provide run-time physical parameters to avoid the potential division by zero in the mixed_layer_restrat code noted at github.com/mom-ocean/issues/1168. Once this PR is merged onto the main branch of MOM6, that issue can be closed. By default, these do not change answers in the MOM6-examples test suite, but the default value for RESTRAT_USTAR_MIN was taken from the hard-coded value in PR that PR. The six copies of the eddy growth rate timescale calculations were consolidated into a new internal function, growth_time, with some other related minor refactoring of the code. Also, mixedlayer_restrat_register_restarts now takes a unit_scale_type arguments like many other analogous routines. All answers are bitwise identical, but there are new runtime parameters or comments that lead to changes in the MOM_parameter_doc files. Also clarified in the comments sent to the MOM_parameter_doc files how VISBECK_L_SCALE works as a dimensional scaling factor when it is given a negative value, and rescaled its units when read as though it were always in m.
|
This issue has been addressed by NOAA-GFDL#266, which has now been merged onto the main branch of MOM6, and hence this issue can be closed. |
We have a global test case where the corner point is on the Equator. Once we set GUST_CONST=0.0, following the discussion about default parameter changes in one of the dev/calls, this test started failing. The reason being a zero in the denominator of the following calculation:
mom_mixrate = (0.41*9.8696)*u_star**2 / & (absf*h_vel**2 + 4.0*(h_vel+dz_neglect)*u_star)See relevant lines of code here
There is no issue when GUST_CONST > 0 because that enforces u_star to be > 0. However, since the corner point is on the Equator
absfis problematic and perhaps must be bounded:absf = 0.5*(abs(G%CoriolisBu(I-1,J)) + abs(G%CoriolisBu(I,J)))@alperaltuntas has come up with a possible way of avoiding this issue, but it would be good to get @Hallberg-NOAA's option before we send a PR.
The text was updated successfully, but these errors were encountered: