[MITgcm-support] Re: viscosity questions

[Dimitris Menemenlis]

Martin, thanks for taking a look at modified growth.F
I am afraid I have no experience with horizontal resolutions that you are aiming 
to work with.  Parameters that Chris and I have used successfully, i.e., no grid 
scale noise, for up to 1/16 deg integrations are:

http://mitgcm.org/cgi-bin/viewcvs.cgi/MITgcm_contrib/high_res_cube/input/data_hr
http://mitgcm.org/cgi-bin/viewcvs.cgi/MITgcm_contrib/quarter_degree_global/input/data_NCEP_16

Especially look at first data file, which is applied to more recent code.
Some differences that I note from your data file are:

1. more conservative shaving:
  hFacMinDr=50.,
  hFacMin=0.3,
  hFacInf=0.1,
  hFacSup=5.,

2. nonlinear free surface
  select_rStar=2,
  nonlinFreeSurf=4,

3. different advection scheme, etc.
  useAreaViscLength=.TRUE.,
  highOrderVorticity  = .TRUE.
  tempAdvScheme=33,
  saltAdvScheme=33,
  multiDimAdvection=.true.,
  vectorInvariantMomentum=.TRUE.,

But not sure which if any parameters matter.  Mostly arrived at above 
configuration via lots of trial and error.

D.