Closed spencerkclark closed 1 year ago
@lharris4 thanks for quickly providing the suggestion for the fix!
FYI my force push was only to update the commit message to indicate all values of kord_wz
for which this is relevant, instead of referring just to kord_wz = -9
.
FYI my force push was only to update the commit message to indicate all values of kord_wz for which this is relevant, instead of referring just to kord_wz = -9.
Apologies for potentially interfering with the CI with this rapid re-push.
@spencerkclark No problem, thank you for taking care of this.
Description
Following the suggestion from @lharris4 in https://github.com/NOAA-GFDL/GFDL_atmos_cubed_sphere/issues/301#issuecomment-1775426281, this assigns values to
gam(i,km)
consistent with the interior of the column wheniv == -3
incs_profile
. It should be expected to change answers for non-hydrostatic simulations wherefv_core_nml.kord_wz < -7
.Fixes #301
How Has This Been Tested?
This has not been tested for scientific correctness, though I have tested that the change proposed here addresses the non-reproducibility issue identified in #301. I now get bitwise-identical results every time I run the model in this configuration.
Checklist:
Please check all whether they apply or not