#285 closed Bug (fixed)
Heat transfer coefficient in CLIO formulation
Reported by: | vancop | Owned by: | nemo |
---|---|---|---|
Priority: | low | Milestone: | |
Component: | OCE | Version: | v3.0 |
Severity: | Keywords: | CLIO, ORCA2, turbulent heat fluxes | |
Cc: |
Description
Salut, hello, bonjour, privet, etc...
Origin:
In the ORCA2 configuration and using the CLIO forcing, sometimes, for an unknown reason, the latent and sensible heat fluxes blow up in the equator region near the east coast of continents (e.g., east of South America near the Amazone region). I could not figure out why, but I guess the new surface module is more explicit than it used to be and hence, more unstable. The present bugfix is a dirty plaster to prevent that. But I argue that a dirty plaster on a dirty formulation of the forcing is not a problem.
Proposed solution:
The heat transfer coefficient zchcm blows up.
Let's limit its value!
Code processing:
in sbcblk_clio.F90
replace this line :
zchcm = zcmcmn / ( 1. - zchn * zpsih / ( vkarmn * zcmn ) )
by those lines (much more beautiful, no ?) :
! sometimes the ratio zchn * zpsih / ( vkarmn * zcmn ) is too close to 1
! and zchcm becomes very very big
zcmax = 0.1 ! choice for maximum value of the heat transfer coefficient, guided by my intuition
zrmax = 1 - 3.e-4 / zcmax ! maximum value of the ratio
zchcm = zcmcmn / ( 1. - MIN ( zchn * zpsih / ( vkarmn * zcmn ) , zrmax ) )
Effect:
Very small on the results. But you need this to run more than 5 years with CLIO on the NEC.
A bientot,
Martin!
Commit History (1)
Changeset | Author | Time | ChangeLog |
---|---|---|---|
1269 | ctlod | 2009-01-14T17:16:05+01:00 | limit the heat transfer coefficient zchcm value, see ticket: #285 |
Change History (2)
comment:1 Changed 15 years ago by ctlod
- Resolution set to fixed
- Status changed from new to closed
comment:2 Changed 8 years ago by nicolasmartin
- Milestone New Surface Module deleted
Milestone New Surface Module deleted