search

MPAS-Dev / MPAS

Repository for private MPAS development prior to the MPAS v6.0 release.
Other
4 stars 0 forks source link

merge develop, update surface restoring #1427

Closed mark-petersen closed 6 years ago

mark-petersen commented 6 years ago

Merging develop into ocean/develop. This brings many associated changes in the surface restoring. With these changes, we should be able to have ACME G cases with surface salinity restoring on, that pass bit-for-bit restart tests, and bit-for-bit decomposition tests.

mark-petersen commented 6 years ago

@vanroekel Current code compiles and passes nightly test suite. But the surface restoring is all disabled because the subroutine calls changed. Here are my notes to myself:

shared/mpas_ocn_tracer_ecosys.F:1692:!mrp 171002 delete for framework merge: config_forcing_restart_file)
shared/mpas_ocn_tracer_ecosys.F:1871:!mrp 171002 This interface has changed. Need to fix this later.
shared/mpas_ocn_tracer_ecosys.F:2041:      !mrp 171002 the restart file was removed in framework:
shared/mpas_ocn_tracer_short_wave_absorption_variable.F:279:!mrp 171002 delete for framework merge: config_forcing_restart_file)
shared/mpas_ocn_tracer_short_wave_absorption_variable.F:314:!mrp 171002 This interface has changed.  Need to put this back later.
shared/mpas_ocn_tracer_short_wave_absorption_variable.F:505:          !mrp 171002 the restart file was removed in framework:
shared/mpas_ocn_tracer_surface_restoring.F:267:!mrp 171002 delete for framework merge: config_forcing_restart_file)
shared/mpas_ocn_tracer_surface_restoring.F:285:!mrp 171002 This interface has changed.  Figure it out later.
shared/mpas_ocn_tracer_surface_restoring.F:417:      !mrp 171002 the restart file was removed in framework:
vanroekel commented 6 years ago

@mark-petersen the code here now passes my stand-alone BFB decomposition tests with salinity restoring on. Should be ready for ACME testing now.

mark-petersen commented 6 years ago

Rebased with newly merged develop. Compiled with intel/debug, passed nightly regression suite.

vanroekel commented 6 years ago

@mark-petersen this branch is ready to go from my perspective, but I would like to wait for @maltrud to have a chance to sign off.

vanroekel commented 6 years ago

@maltrud I assume this is in the salinity restoring? If so, could you note which part of the code? I'll compare again to the current head of ocean/develop.

maltrud commented 6 years ago

@vanroekel yes it is the salinity restoring. the bits where deltaS is calculated and then used to calculate the mean.

mark-petersen commented 6 years ago

Just rebased on MPAS-Dev/ocean/develop, which includes the recent FCT merge #1393. Tested head of this PR, passes nightly test suite with intel/debug and gnu. Nightly test suite is bfb with current MPAS-Dev/ocean/develop.

vanroekel commented 6 years ago

@mark-petersen I have fixed an issue in the salinity restoring based on discussion with @maltrud

mark-petersen commented 6 years ago

Passed all tests in MPAS and ACME. See https://github.com/ACME-Climate/ACME/pull/1823.