[MITgcm-support] Model blowing due to extreme values

[Jody Klymak]

Hi Kunal, 

Its a good idea to plot your data as soon as possible in the run, and maybe to dump a few of the first output steps (i.e. set dumpFreq=180).  Something I do every year or two is supply the model with upside down water column, and it immediately starts convecting, and of course the CFL criteria is violated.  It is also very useful to look at your STDOUT.0000 and see what cfl criteria was violated.

I also note that you have set deltaTfreesurf; I’ve never set that - are you sure 3600 s is appropriate?  Similarly your tracers are set to 3600, and your grids are 7200 m, so you don’t need currents much larger than 1 m/s to violate the CFL criteria.  Finally, are you sure you have your initial velocities in the right units?

Good luck

Cheers,   Jody

> On May 24, 2022, at 1:07 PM, kunal madkaiker <kunal.madkaiker02 at gmail.com> wrote:
> 
> Hi All,
> 
> I have a model domain at 7km resolution in north Indian ocean with a southern open boundary. I get an error when the simulation begins.
> Error: 'MOM_IMPLICIT_R: error when solving 3-Diag problem.'
> Upon further investigation, the error message changes to: 'STOP ABNORMAL END: S/R MON_SOLUTION, stops due to EXTREME Pot.Temp' ( I did check and found extreme +ve and -ve values for temperature and salinity at deeper depths)
> 
> From previous support resolutions, I know it has to do with CFL criteria, and I have tried changing deltaT, viscAh/viscAhgrid, viscAz, initial conditions files and even delZ levels. Somehow I am not able to figure it out.
> 
> I have a second question. In calc_oce_mxlayer.F, if I am not mistaken, model will calc 'hMixCriteria' first and based on its value, will decide to go for method 1 or 2. If so, how can I set the 2nd method for MLD calculation?
> 
> 
> # ====================
> # | Model parameters |
> # ====================
> #
> # Continuous equation parameters
>  &PARM01
> 
> ## viscAhgrid=9.8E-7,
>  viscAh=10.E3,
>  viscAz=1.E-4,
> ## viscAz=1.E-7,
> ## viscA4=0.,
>  diffKhT=1.,
>  diffKzT=1.E-5,
> ## diffK4T=0.,
>  diffKhS=1.,
>  diffKzS=1.E-5,
> ## diffK4S=0.,
> 
> ## useFullLeith=.TRUE.,
> ## This uses the Strain-Tension form for calculating viscous terms
> ## useStrainTensionVisc=.TRUE.,
> 
> ## This is the standard Leith Viscosity coefficient, 
> ## visc=(visc2leith/pi)**3 grad(curl v_h)
> ## viscC2leith=2.0, 
> 
> ## This is an additional Leith Viscosity, prop to grad(div.v_h)
> ## It traps instabilities that have no signal in grad(curl v_h)
> ## viscC2leithD=2.0,
> 
> ## This is the max Viscosity used, be it Smag or Leith, and it scales 
> ## with grid size and timestep
> ## viscAhGridMax=1.E-2,
> ## viscA4GridMax=1.E-9,
> ##
>  no_slip_sides=.FALSE.,
>  no_slip_bottom=.TRUE.,
>  tempAdvScheme=33,
>  saltAdvScheme=33,
>  staggerTimeStep=.TRUE.,
> #
>  tAlpha=2.E-4,
>  sBeta =7.E-4,
>  gravity=9.81,
>  f0=0.0000101456,
>  beta=2.2209E-11,
>  rigidLid=.FALSE.,
>  implicitFreeSurface=.TRUE.,
>  implicitDiffusion=.TRUE.,
>  implicitViscosity=.TRUE.,
>  nonHydrostatic=.FALSE.,
>  bottomDragLinear=0.0002,
>  eosType='MDJWF',
>  useRealFreshWaterFlux=.TRUE.,
>  exactConserv=.TRUE.,
> ## nonlinFreeSurf=4,
> ## select_rStar=2,
>  hFacMin=0.5,
> # hFacMinDz=5.,
>  hFacInf=0.2,
>  hFacSup=2.0,
>  useSingleCpuIO=.TRUE
>  readBinaryPrec=64,
>  writeBinaryPrec=64,
>  &
> 
> # Elliptic solver parameters
>  &PARM02
>  cg2dMaxIters=1000,
>  cg2dTargetResidual=1.E-13,
>  cg3dMaxIters=400,
>  cg3dTargetResidual=1.E-13,
> ## cg2dUseMinResSol=1,
>  &
> 
> # Time stepping parameters
>  &PARM03
>  startTime= 0.,
>  endTime= 86400.,
>  deltaTmom= 180.0,
>  deltaTtracer   = 3600.0,
>  deltaTClock    = 3600.0,
>  deltaTfreesurf = 3600.0,
>  pChkptFreq= 86400.,
>  dumpFreq= 7200.,
> # monitorFreq= 180.0,
>  &
> 
> # Gridding parameters
>  &PARM04
>  usingCurvilinearGrid=.FALSE.,
>  usingCartesianGrid=.TRUE.,
>  usingSphericalPolarGrid=.FALSE.,
>  dxSpacing=7695.06,
>  dySpacing=7045.008,
>  delZ= 5.034 , 10.067 , 10.118 , 10.139 , 10.218 , 10.277 , 10.409 , 10.541 , 10.774 , 11.046,
> 11.473 , 12.010 , 12.802 , 13.838 , 15.307 , 17.259 , 19.951 , 23.514 , 28.291 , 34.483,
> 42.457 , 52.357 , 64.361 , 78.256 , 93.721 , 109.99 , 126.30 , 141.72 , 155.66 , 167.62,
> 177.57 , 185.50 , 191.73 , 196.46 , 200.07 , 202.72 , 204.71 , 206.14 , 207.22 , 207.97,
> 208.55 , 208.93 , 209.25 , 209.44 , 209.61 , 209.70 , 209.80 , 209.84 , 209.9,
> 
>  radius_fromHorizGrid=6370.E3,
>  &
> 
> # Input datasets
>  &PARM05
>  hydrogThetaFile='Twoa_dec_init_UPD_7km.bin',
>  hydrogSaltFile='Swoa_dec_init_UPD_7km.bin',
>  uVelInitFile='Usoda_dec_init_UPD_7km.bin',
>  vVelInitFile='Vsoda_dec_init_UPD_7km.bin',
>  bathyFile='bathy_NIO_7km_UPD.bin',
>  checkIniTemp=.false.,
>  checkIniSalt=.false.,
>  &
> 
> Kindly advise. Thanks
> Kunal
> 
> 
> <data.exf><data.obcs>_______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support