Opened 4 years ago
Closed 4 years ago
#2553 closed Bug (fixed)
restartability broken when LEV melt pond activated (ln_pnd_LEV)
Reported by: | mathiot | Owned by: | systeam |
---|---|---|---|
Priority: | low | Milestone: | |
Component: | SI3 | Version: | trunk |
Severity: | minor | Keywords: | restartability, |
Cc: |
Description
Context
Despite restartability test OK with ORCA2_ICE_PISCES in SETTE, it appears ln_pnd_LEV break restartability because the melt pond variables are not read from the ice restart in icedyn_adv_pra.F90.
Analysis
Ckeck of melt pond variable presence in icedyn_adv_pra is using the wrong file id, so SI3 assume the restart does not contains the melt pond variables.
It affects NEMO4 and trunk.
Fix
Replace numror by numrir when checking presence of sxap and sxvl.
-
icedyn_adv_pra.F90
1005 1005 END DO 1006 1006 ! 1007 1007 IF( ln_pnd_LEV ) THEN ! melt pond fraction 1008 IF( iom_varid( numr or, 'sxap', ldstop = .FALSE. ) > 0 ) THEN1008 IF( iom_varid( numrir, 'sxap', ldstop = .FALSE. ) > 0 ) THEN 1009 1009 CALL iom_get( numrir, jpdom_auto, 'sxap' , sxap , psgn = -1._wp ) 1010 1010 CALL iom_get( numrir, jpdom_auto, 'syap' , syap , psgn = -1._wp ) 1011 1011 CALL iom_get( numrir, jpdom_auto, 'sxxap', sxxap ) … … 1023 1023 ENDIF 1024 1024 ! 1025 1025 IF ( ln_pnd_lids ) THEN ! melt pond lid volume 1026 IF( iom_varid( numr or, 'sxvl', ldstop = .FALSE. ) > 0 ) THEN1026 IF( iom_varid( numrir, 'sxvl', ldstop = .FALSE. ) > 0 ) THEN 1027 1027 CALL iom_get( numrir, jpdom_auto, 'sxvl' , sxvl , psgn = -1._wp ) 1028 1028 CALL iom_get( numrir, jpdom_auto, 'syvl' , syvl , psgn = -1._wp ) 1029 1029 CALL iom_get( numrir, jpdom_auto, 'sxxvl', sxxvl )
Commit History (2)
Changeset | Author | Time | ChangeLog |
---|---|---|---|
13637 | mathiot | 2020-10-19T16:16:19+02:00 | fix ticket #2553 in trunk |
13634 | mathiot | 2020-10-19T16:12:54+02:00 | fix ticket #2553 in NEMO4 |
Change History (3)
comment:1 Changed 4 years ago by mathiot
comment:2 Changed 4 years ago by mathiot
In 13637:
comment:3 Changed 4 years ago by mathiot
- Resolution set to fixed
- Status changed from new to closed
Clem agreed it is a bug but we really did not understand why SETTE didn't catch this bug. I didn't go further in trying to understand why sette missed this bug (no melt pond at time of restart ? something else ?)
In 13634: