[MITgcm-support] instability at Eta

[Martin Losch]

Hi Kaveh,

you already got a comment about balancing the flow.

The extra column/row that you get for u and v is a feature of the netcdf output. The mnc package writes U(1:Nx+1,1:Ny) and V(1:Nx,1:Ny+1) to the netcdf file. The Nx+1st column and Ny+1st row are the same as the 1st column and row, respectively because of the default double periodicidy of the model, but with OBCS these values are never used (actually overwritten by obcs-values before they are used), so you can just disregard them. Sometimes it useful to have them, e.g. in more complicated domains like the cubed sphere grids.

M.

> On 07 Mar 2016, at 14:51, kaveh Purkiani <kavehpurkiani at googlemail.com> wrote:
> 
> Hello everybody,
> 
> I have a configuration to simulate the circulation in tropical Pacific Ocean. It is a simple box model with 20 grid cells in horizontal and 200 in vertical.
> Daily averaged temp, salt, u and v velocities are downloaded from HYCOM. As to OBCS, all the open boundary files are written as Nx/y*Nz*time format 
> (here is 20*200*232). 
> 
> Salinity, temperature, u and v outputs of the model at boundaries are read correctly by the model. The CFL number (eq.3.28) is extremely below the
> criteria.
> As simulation forwards, Eta decreases drastically until I get Nan values and so negative values for temperature. How to solve this? 
> 
> Besides that I am having another problem. Although the temperature and salinity forcing seem to behave as they are, the forcing of the u and v
> velocities are not. In the NetCDF files I get, T and S of dimension 20 * 20, u of dimension 21 x 20 and v of dimension 20 x 21 which 
> 
> make sense. Therefore, I have two columns of boundary values for U values at western boundary, similar happens at eastern boundary.
> The penultimate column is fine but the next one (last column) is duplication of western boundary. Similar for V values at Northern and Southern boundaries happen. 
> Does this cause an error in reading boundaries? If yes, how to solve this. I already tried with different OLx and OLy but did not help.
> Does anyone have any ideas about what could be my mistake?
> 
> 
> below are my data, data.exf and data.obcs files:
> 
> # ====================
> # | Model parameters |
> # ====================
> #
> # Continuous equation parameters
>  &PARM01
> # tRef= 200*8,
> # sRef= 200*34,
>  implicitDiffusion=.TRUE.,
>  implicitViscosity=.TRUE.,
>  viscAr=1.E0,
>  viscAh=1.E0,
> #- put small value (<< stab.limit ~ 3.e10) only to test biharmonic-viscosity
>  viscA4=1.E8,
>  no_slip_sides=.FALSE.,
>  no_slip_bottom=.FALSE.,
>  diffKhT=1.E3,
>  diffKrT=1.E-5,
>  diffKhS=1.E3,
>  diffKrS=1.E-5,
>  saltAdvScheme=4,
>  eosType='LINEAR',
>  tAlpha=2.E-4,
>  sBeta =0.E-4,
>  gravity=9.81,
>  f0=1.e-4,
>  beta=0.E-11,
>  nonHydrostatic=.FALSE.,
>  rigidLid=.FALSE.,
>  implicitFreeSurface=.TRUE.,
>  exactConserv=.TRUE.,
>  hFacMin=0.2,
>  readBinaryPrec=64,
>  nonlinFreeSurf = 4,
> # hFacInf=0.0002,
> # hFacSup=2.5,
>  &
> 
> # Elliptic solver parameters
>  &PARM02
>  cg2dMaxIters=1000,
>  cg2dTargetResidual=1.E-13,
>  &
> 
> # Time stepping parameters
>  &PARM03
>  nIter0=0,
>  nTimeSteps=288000,
>  baseTime=0.,
>  deltaT=10.0,
>  abEps=0.1,
>  momDissip_In_AB=.FALSE.,
>  pChkptFreq=0.0,
>  chkptFreq=0.0,
>  dumpFreq=900.0,
>  monitorSelect=2,
>  monitorFreq=1.,
> # for time dependent open boundary conditions, activate the following 3 lines:
> # periodicExternalForcing=.FALSE.,
> # externForcingPeriod=86400.,
> # externForcingCycle =20044800.,
>  &
> 
> # Gridding parameters
>  &PARM04
>  usingCartesianGrid=.FALSE.,
>  usingSphericalPolarGrid=.TRUE.
>  ygOrigin=24.5,
>  xgOrigin=117.525,
>  delX=20*0.01,
>  delY=20*0.01,
>  delR= 200*20,
>  &
> 
> # Input datasets
>  &PARM05
>  bathyFile = 'topognew-nobump.bump',
>  hydrogThetaFile='initial_temp.bin',
>  hydrogSaltFile= 'initial_salt.bin',
> &
> 
> data.exf:
> 
>                                                                                                         
> 
> # External Forcing Data
> # *********************
>  &EXF_NML_01
>  exf_iprec         = 64,
>  useExfCheckRange=.TRUE.,
>  &
> # *******************
>  &EXF_NML_02
>  &
> # ***************
>  &EXF_NML_03
>  &
> #
>  &EXF_NML_04
>  &
>  &EXF_NML_OBCS
>  obcsNstartdate1=20070101,
>  obcsNstartdate2=000000,
>  obcsNperiod=86400,
> #
>  obcsEstartdate1=20070101,
>  obcsEstartdate2=000000,
>  obcsEperiod=86400,
> #
>  obcsSstartdate1=20070101,
>  obcsSstartdate2=000000,
>  obcsSperiod=86400,
> #
>  obcsWstartdate1=20070101,
>  obcsWstartdate2=000000,
>  obcsWperiod=86400,
>  &
> 
> 
> data.obcs:
> 
> 
> # Open-boundaries
>  &OBCS_PARM01
> # This flag turns off checking and fixing problematic topography across
> # open boundaries.
>  OBCSfixTopo=.TRUE.,
>  OB_Jnorth= 20*-1,
>  OB_Jsouth = 20*1,
>  OB_Ieast = 20*-1,
>  OB_Iwest = 20*1,
> useOBCSprescribe = .TRUE.,
>  OBNuFile  = 'OB_NorthU.bin',
>  OBNvFile  = 'OB_NorthV.bin',
>  OBNsFile  = 'OB_NorthS.bin',
>  OBNtFile  = 'OB_NorthT.bin',
>  OBSuFile  = 'OB_SouthU.bin',
>  OBSvFile  = 'OB_SouthV.bin',
>  OBSsFile  = 'OB_SouthS.bin',
>  OBStFile  = 'OB_SouthT.bin',
>  OBEuFile  = 'OB_EastU.bin',
>  OBEvFile  = 'OB_EastV.bin',
>  OBEsFile  = 'OB_EastS.bin',
>  OBEtFile  = 'OB_EastT.bin',
>  OBWuFile = 'OB_WestU.bin',
>  OBWvFile = 'OB_WestV.bin',
>  OBWsFile = 'OB_WestS.bin',
>  OBWtFile = 'OB_WestT.bin',
> &
> 
> 
> 
> Thank you very much for your attention in advance.
> 
> regards,
> 
> Kaveh.
> 
> 
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support