[MITgcm-support] Thermal wind in ocean cause too large velocity?

[Jody Klymak]

If you start your model with zero velocity and a large horizontal temperature gradient, it will undergo a Rossby adjustment first which can have large wave velocities associated with it, though 4 m/s seems large. The Coriolis force takes at least an inertial period to become important.  

If you want your model to be in geostrophic balance at startup, you need to specify the velocities in the initial conditions as well.  

Cheers,   Jody
 

On May 21, 2012, at  4:38 AM, Chun-Yan Zhou wrote:

> Based on
> The temperature difference in horizontal direction can cause large velocity because of thermal wind.
> And I did a test where the initial temperature field changed form 20-13 degree along the X direction, no other forces are applied.
> Large velocity is expected, but the maximum velocity is 4m/s, and the minimum is -4m/s, most of  the velocity is around 1.5m/s. Is the velocity should be so large?
> Any suggestion is appreciated.
> chunyan
> 
> # ====================
> # | Model parameters |
> # ====================
> #
> # Continuous equation parameters
> &PARM01
> tRef=20*20.,
> sRef=20*35.,
> viscAh=4.E-2,
> viscAz=4.E-2,
> no_slip_sides=.FALSE.,
> no_slip_bottom=.FALSE.,
> diffKhT=4.E-2,
> diffKzT=4.E-2,
> f0=0.E-4,
> beta=0.E-11,
> tAlpha=2.0E-4,
> sBeta =0.,
> tempAdvScheme=77,
> gravity=10.,
> rhoConst=1000.,
> rhoNil=1000.,
> heatCapacity_Cp=4000.,
> #rigidLid=.TRUE.,
> implicitFreeSurface=.TRUE.,
> #exactConserv=.TRUE.,
> eosType='LINEAR',
> nonHydrostatic=.TRUE.,
> saltStepping=.FALSE.,
> staggerTimeStep=.TRUE.,
> &
> 
> # Elliptic solver parameters
> &PARM02
> cg2dMaxIters=1000,
> cg2dTargetResidual=1.E-9,
> cg3dMaxIters=1000,
> cg3dTargetResidual=1.E-9,
> &
> 
> # Time stepping parameters
> &PARM03
> nIter0=0,
> #endTime=43200.,
> nTimeSteps=21600,
> deltaT=10.,
> abEps=0.1,
> pChkptFreq=43200.,
> chkptFreq=7200.,
> dumpFreq=1800.,
> monitorFreq=600.,
> monitorSelect=1,
> monitorFreq=1.,
> &
> 
> # Gridding parameters
> &PARM04
> usingCartesianGrid=.TRUE.,
> dXspacing=100.,
> dYspacing=100.,
> delZ=50*20.,
> &
> 
> # Input datasets
> &PARM05
> hydrogThetaFile='T.bin',
> &
> %%%%%%%%%%%gendata.m
> var=zeros(100,100,50);
> for i=1:100
>    var(i,:,:)=20-0.07*(i-1);
> end
> fid=fopen('T.bin','w','b'); fwrite(fid,var,'real*4'); fclose(fid);
> 
> 
> 
> Dear all,
> 
> 
> 
> 
> The University of Dundee is a registered Scottish Charity, No: SC015096
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support

--
Jody Klymak    
http://web.uvic.ca/~jklymak/