Opened 7 years ago
Closed 6 years ago
#2042 closed Task (fixed)
ENHANCE-05_Gurvan-Vertical_Advection
Reported by: | gm | Owned by: | acc |
---|---|---|---|
Priority: | normal | Milestone: | 2018 WP |
Component: | OCE | Version: | trunk |
Severity: | minor | Keywords: | |
Cc: |
Description (last modified by acc)
Context
A 2018 development to add an adaptive-implicit vertical advection based on Shchepetkin 2015 (as used in CROCO).
Proposal
- Option to be selected via a logical set in the namzdf name list (ln_zad_Aimp)
- New routine to compute Courant number criteria and partition wn into wn + wi where wi is to be treated implicitly (hopefully at infrequent and sparse locations). New routine added to sshwzv.F90 and called from step
- wi treated implicitly by adding appropriate terms to the tridiagonal solvers in tra_zdf and dyn_zdf
- include diagnostic output to track activity of the scheme
Development branch:
http://forge.ipsl.jussieu.fr/nemo/browser/NEMO/branches/2018/dev_r9956_ENHANCE05_ZAD_AIMP
Commit History (4)
Changeset | Author | Time | ChangeLog |
---|---|---|---|
10364 | acc | 2018-11-30T18:42:51+01:00 | Introduce Adaptive-Implicit vertical advection option to the trunk. This is code merged from branches/2018/dev_r9956_ENHANCE05_ZAD_AIMP (see ticket #2042). The structure for the option is complete but is currently only successful with the flux-limited advection scheme (ln_traadv_mus). The use of this scheme with flux corrected advection schemes is not recommended until improvements to the nonoscillatory algorithm have been completed (work in progress elsewhere). The scheme is activated via a new namelist switch (ln_zad_Aimp) and is off by default. |
10019 | acc | 2018-07-31T14:27:50+02:00 | Branch: dev_r9956_ENHANCE05_ZAD_AIMP. Minor corrections to sshwzv.F90, dynzdf.F90 and trazdf.F90 to provide a draft working set. Early indications are that the scheme is now working for the simple test cases tried so far. See ticket #2042 |
10008 | acc | 2018-07-26T17:58:13+02:00 | Branch: dev_r9956_ENHANCE05_ZAD_AIMP. Fix bug in trazdf.F90 to get closer to maintaining constant salinity in a simple GYRE test with uniform salinity. Also tidy dynzdf.F90 so results are the same with wi=0 and ln_zad_Aimp=T/F. See ticket #2042 |
9976 | acc | 2018-07-19T17:58:12+02:00 | Branch: dev_r9956_ENHANCE05_ZAD_AIMP. First set of changes to implement an adaptive-implicit vertical advection option (see ticket #2042). This code compiles and runs but has issues when the new option is activated. |
Change History (10)
comment:1 Changed 7 years ago by gm
- Type changed from Enhancement to Task
comment:2 Changed 7 years ago by nicolasmartin
- Owner set to gm
- Status changed from new to assigned
comment:3 Changed 6 years ago by acc
- Description modified (diff)
- Owner changed from gm to acc
comment:4 Changed 6 years ago by acc
- Description modified (diff)
comment:5 Changed 6 years ago by acc
comment:6 Changed 6 years ago by acc
Branch: dev_r9956_ENHANCE05_ZAD_AIMP. Fixes to dynzdf.F90 so that it gives the same results with or without ln_zad_Aimp set true if vertical velocity partitioning is disabled
Changeset: 9979
comment:7 Changed 6 years ago by acc
In 10008:
comment:8 Changed 6 years ago by acc
In 10019:
comment:9 Changed 6 years ago by acc
In 10364:
comment:10 Changed 6 years ago by acc
- branch_review set to passed
- mp_status set to yes
- Resolution set to fixed
- Status changed from assigned to closed
- wp_comment set to Code successfully merged into trunk. Works successfully with flux limited scheme (ln_traadv_mus) but awaits developments elsewhere to be fully operational. Documentation delayed to discourage activation before final tweaks and testing in 2019.
In 9976: