[MITgcm-support] T & rho!

[Van Thinh Nguyen]

Hi Martin & others,

Thanks Martin for your explain.
It's true, below 18.5 then N2=0 due to rho=const. This density profile 
looks like 2 layers, I would like to study solitary waves generating by 
tidal flows over a depression bump (tidal current u=1.5*sin(omega*t), 
omega=M2). The model has OBCs at west & east, here are some specific 
options what I used to setup to the model:

-Domain: delX=10000*20., delY=200., delZ=120*1. (m) (2D)
  Gausian bump at centre (after X axis) and h(x,z)=-80.-40./(cosh(x/L));

-OBCs at west & East, using obcs_cal.F to define the tidal current and
  OBCs for U,V W and Eta; using obcs_sponge.F to reduce reflecting waves
  from OBCs

-using OBCs, KPP packages

-applying viscC2Leith=4. and viscC2LeithD=4. due to viscAz=1.E-5 (or
  1.E-6)

I could reduce a huge noise, however a bit of noises and reflecting waves
from OBCs are still remained.

If someone may have any experience about such model, please help me an 
advice.

Thanks a lot

Vth

------------------------------------------------------

On Wed, 6 Sep 2006, Martin Losch wrote:

> Hi VanThinh,
>
> if hydrogThetaFile is not set, then tRef is used to initialize the 
> temperature.
> If sBeta=0 (no salinity), you then treat pot temperature as if it were 
> density.
> You can do as you described, but your density profile leads to N2 = 0 below 
> 18.5 meters, doesn't it? I suggest Alistair's approach, see
> http://forge.csail.mit.edu/pipermail/mitgcm-support/2004-June/002269.html
>
> Martin
>
> On Sep 5, 2006, at 9:23 PM, Van Thinh Nguyen wrote:
>
>> Hi Martin,
>> 
>> I looked in your discuss on T & rho in MITgcm-support archives (June 2004), 
>> but I am really not sure what I am using in my model:
>> 
>> I have a vertical density profile as:
>> 
>> rho(z)=0.001*(18.5-2.75*(1+tanh((z+10)/2)))
>> 
>> From exp1 using linear EOS, we have:
>>
>>    d(rho)/dz = - rho_0*tAlpha*d(T)/dz
>> 
>> Integrating this relation, I get T(z)= - rho(z)/(rho_0*tAlpha)
>> (here rho_0=1000kg/m3, tAlpha=2.E-4 and rho(z) given as above)
>> 
>> This T(z) can be used as tRef in data?
>> 
>> Thanks so much for any help,
>> 
>> Van Thinh
>> ------------------------------------------------------
>> 
>> _______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-support
>
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support