Opened 5 years ago
Closed 5 years ago
#2479 closed Bug (fixed)
Inconsistent time scale in aEVP rheology
Reported by: | jchanut | Owned by: | systeam |
---|---|---|---|
Priority: | high | Milestone: | |
Component: | SI3 | Version: | v4.0.* |
Severity: | major | Keywords: | |
Cc: |
Description
The adaptive rheology solver (ln_aEVP=T) is based on the modified EVP solver of Bouillon et al (2013) in which time stepping of rheology is replaced by an iterative process. In that context the time scale involved in the equations is the "slow" ice time step, "rdt_ice", in contrary to the standard EVP method which considers the substep "rdt_ice/nn_nevp".
NEMO makes no disctinction so that the time stepping, varying alpha/beta coefficients are all wrong. Replacing "zdtevp" by "rdt_ice" at the beginning of the rheology subroutine ("ice_dyn_rhg_evp") solves the problem.
There are potentially other, but minor, inconsistencies related to aEVP such that the missing spatial interpolation of alpha/beta at velocity points.
Commit History (1)
Changeset | Author | Time | ChangeLog |
---|---|---|---|
13064 | clem | 2020-06-08T15:43:28+02:00 | fix ticket #2479. This was a major bug preventing aEVP to work as it should. + Add output of beta |
Change History (2)
comment:1 Changed 5 years ago by clem
comment:2 Changed 5 years ago by clem
- Resolution set to fixed
- Status changed from new to closed
In 13064: