Opened 13 years ago
Closed 12 years ago
#800 closed Bug (fixed)
bug in tra_zdf_imp.F90
| Reported by: | poddo | Owned by: | cetlod |
|---|---|---|---|
| Priority: | low | Milestone: | |
| Component: | OCE | Version: | v3.3 |
| Severity: | Keywords: | ||
| Cc: |
Description
In tra_zdf_imp.F90
The following statment
zwt(:,:,:)=avt(:,:,:)
wich is active in case of no rotation or in cae key_ldfslp not defined, should be modified as follows
DO ki=2,jpkm1
DO jj=2, jpjm1
DO ji=fs_2, fs_jpim1
zwt(ji,jj,jk)=avt(ji,jj,jk)
END DO
END DO
END DO
I think this should be done also for in the salinity section.
Paolo
Commit History (1)
| Changeset | Author | Time | ChangeLog |
|---|---|---|---|
| 2602 | cetlod | 2011-02-21T16:23:38+01:00 | Improvment of surface boundary condition on trazdf_imp.F90 : minor bug correction+style, see ticket #800 |
Change History (8)
comment:1 Changed 13 years ago by poddo
- Version changed from v3.2 to v3.3
comment:2 Changed 13 years ago by poddo
comment:3 Changed 13 years ago by cetlod
- Owner changed from NEMO team to cetlod
comment:4 follow-up: ↓ 5 Changed 13 years ago by cetlod
Thank you Paolo for your remark.
Andrew C. has also found that bug a few days ago and it is because the surface boundary condition on temperature in trazdf_imp.F90 was accidentally removed in version v3.3 (not for salinity or passive tracers)
We've also remarked that this surface boundary condition was unecessary when using rotated lateral mixing (key_ldfslp activated), since the vertical coefficient array zwt was set to zero at surface.
We've re-written the routine ( style + correction of that minor bug ) : see changeset:2602
comment:5 in reply to: ↑ 4 ; follow-up: ↓ 7 Changed 13 years ago by poddo
I think that also:
zwt(:,:,jpk) = 0._wp
should be included just after the
zwt(:,:,1) = 0._wp
statement.
This is because later in the routine you will use zwt(ji,jk,jk+1)
Paolo
Replying to cetlod:
Thank you Paolo for your remark.
Andrew C. has also found that bug a few days ago and it is because the surface boundary condition on temperature in trazdf_imp.F90 was accidentally removed in version v3.3 (not for salinity or passive tracers)
We've also remarked that this surface boundary condition was unecessary when using rotated lateral mixing (key_ldfslp activated), since the vertical coefficient array zwt was set to zero at surface.
We've re-written the routine ( style + correction of that minor bug ) : see changeset:2602
comment:6 Changed 13 years ago by cetlod
Not necessary, because the bottom boundary condition is applied through the masked field avt
comment:7 in reply to: ↑ 5 Changed 13 years ago by cetlod
Replying to poddo:
Not necessary, because the bottom boundary condition is applied through the masked field avt
I think that also:
zwt(:,:,jpk) = 0._wp
should be included just after the
zwt(:,:,1) = 0._wp
statement.
This is because later in the routine you will use zwt(ji,jk,jk+1)
Paolo
Replying to cetlod:
Thank you Paolo for your remark.
Andrew C. has also found that bug a few days ago and it is because the surface boundary condition on temperature in trazdf_imp.F90 was accidentally removed in version v3.3 (not for salinity or passive tracers)
We've also remarked that this surface boundary condition was unecessary when using rotated lateral mixing (key_ldfslp activated), since the vertical coefficient array zwt was set to zero at surface.
We've re-written the routine ( style + correction of that minor bug ) : see changeset:2602
comment:8 Changed 12 years ago by rblod
- Resolution set to fixed
- Status changed from new to closed
The vertical loop should be in "jk" and not in "ki" as written in my previous message