[MITgcm-support] Stratification creating current without forcing

[Nadine Steiger]

Dear all,

I am using an idealized setup of a v-shaped channel in MITgcm with 
dimensions of 250 km x 140 km x 0.8 km.
So far, I had no stratification in the water and added now a linear 
stratification in salinity (still constant temperature).

To test the setup, I started without any forcing and heat/momentum 
source, to test if the water stays in rest. Surprisingly, it does not 
and I get a circulation that increases in strength with time. In the 
model runs without stratification everything worked fine and the ocean 
stayed in rest.

I checked the first time steps and it looks like the salinity reduces at 
the bottom along the topography, where I also get non-zero velocities. 
It seems like the reduction in salinity happens exactly in the cells 
where HFacC is between 0 and 1.

Does anyone know what is happening there?

Thanks a lot,
Nadine


This is my data-file:

# Continuous equation parameters
  &PARM01
  Tref = 160*-1.,
  Sref = 160*34.5,
  viscAr=1.E-4,
  viscAh=25.,
  viscA4=0.E-10,
  no_slip_sides=.FALSE.,
  no_slip_bottom=.FALSE.,
  diffKrT=1.E-5,
  diffKhT=5.,
  diffKrS=1.E-5,
  diffKhS=5.,
  bottomDragQuadratic=0.,
  eosType='JMD95Z',
  saltAdvScheme=33,
  tempAdvScheme=33,
  HeatCapacity_cp = 3974.0,
  rhoConst=1027.6,
  rhoNil=1027.6,
  gravity=9.81,
  f0=-1.398E-04,
  beta=0.E-11,
  tAlpha=3.67E-5,
  sBeta=7.89E-4,
# tAlpha=2.E-4,
# sBeta =0.E-4,
  convertFW2Salt = 33.4,
  rigidLid=.FALSE.,
  implicitFreeSurface=.TRUE.,
  nonlinFreeSurf = 0.,
  exactConserv=.TRUE.,
  staggerTimeStep=.TRUE.,
  useRealFreshWaterFlux = .TRUE.,
  select_rStar = 0.,
  useSingleCpuIO=.TRUE.,
  hFacMin = 0.5,
  hFacMindr = 0.0,
  hFacInf = 0.2,
  hFacSup = 2.0,
  nonHydrostatic=.FALSE.,
  readBinaryPrec=64,
  useCDScheme = .FALSE.,
  &

# Elliptic solver parameters
  &PARM02
  cg2dMaxIters=1000,
  cg2dTargetResidual=1.E-13,
  &

# Time stepping parameters
  &PARM03
  nIter0=0,
  nTimeSteps=60000.,
  deltaT=100.,
  abEps=0.1,
  cAdjFreq = -1.,
#tauCD = 400000.,
  pChkptFreq=10000.0,
  chkptFreq=1000.0,
# dumpFreq=50.,
  taveFreq=0.0,
  monitorFreq=1.,
  monitorSelect=2,
  &

# Gridding parameters
  &PARM04
  usingCartesianGrid=.TRUE.,
  delX=500*500.,
  delY=240*500.,
  delR=160*5.,
  &

# Input datasets
  &PARM05
  bathyFile='topog.channel',
  hydrogThetaFile='T0.bin',
  hydrogSaltFile='S0.bin',
  &

-- 
Nadine Steiger
PhD Candidate
University of Bergen | Geophysical Institute
Bjerknes Centre for Climate Research