[MITgcm-cvs] MITgcm/verification/offline_exf_seaice/input.thermo CVS Commit

[Gael Forget]

Update of /u/gcmpack/MITgcm/verification/offline_exf_seaice/input.thermo
In directory baudelaire:/srv/scratch/gforget/MITgcm/verification/offline_exf_seaice/input.thermo

Modified Files:
	data.seaice 
Log Message:

- switch to HO = 0.2
  To be better consistent with thsice and initial condition.
  With SEAICE_areaLossFormula = 1, HO has a large influence on 
  the ~ equilibrium solution. The three terms of melt/growth
  act separately on ice concentration, which stays below 1.
- switch to SEAICE_areaLossFormula = 2
  Leads to a ~ equilibrium solution (after 3 years run)
  that is ~1 for ice concentration (in cold atm sector)
  similar to the thsice solution.
- switch to SEAICE_growMeltByConv = .TRUE.
  Has little influence with SEAICE_areaLossFormula = 2
- switch to SEAICE_doOpenWaterMelt = .TRUE.
  Has little influence with SEAICE_areaLossFormula = 2
  over most of the domain. But there were spurious 500m ice
  in the extreme cold part of the domain (top right corner,
  where the ocean is below freezing) that are 
  removed when using SEAICE_doOpenWaterMelt.
- these parameter choices do not seem to affect 
  the adjoint accuracy much.