[MITgcm-support] Floats Advection

[Sergio Jaramillo]

Chris,

Thanks for the advice. I already did what you suggested, but what I 
think is my problem is that in flt_functions.F  when you use " if ( 
xlo.le.rG. and . xhi.gt.rG)" to find the float, xlo < xhi is always 
assumed, but  in my polar grid xlo (=xc(i))  sometimes is greater than 
xhi (xc(i+1)), so the subroutine never actually find the floats. I also 
have problems with the delX(iG) and delY(iG) that are used there, 
because they are not related to  dxf or dyF (which change in space 
according to DXF.bin, DYF.bin). Am I too  lost?

Thanks,

Sergio Jaramillo




Chris Hill wrote:

>Sergio,
>
> The coords in XC and YC should be valid even in curvilinear
>corrdinates. Have you got some examples of XC and YC values you get
>together with the local_i,local_j numbers you see.
>
>Thanks,
>
>Chris Hill 
>
>  
>
>>-----Original Message-----
>>From: mitgcm-support-bounces at mitgcm.org 
>>[mailto:mitgcm-support-bounces at mitgcm.org] On Behalf Of 
>>Sergio Jaramillo
>>Sent: Thursday, July 31, 2003 7:35 PM
>>To: mitgcm-support at mitgcm.org
>>Subject: [MITgcm-support] Floats Advection
>>
>>
>>Hi,
>>
>>I am setting MITgcm to model the flow in a rotating table. 
>>First I tried 
>>an example with  cartesian coordinates and it worked very well, I put 
>>some floats and they are advected in the way I expected. But now I am 
>>running the model with curvilinear coordinates, and even 
>>though I have 
>>good results regarding the flow, my floats don't move  at all 
>>this time. 
>>The positions of the floats that I used are in XC, YC 
>>coordinates and I 
>>don't know if this is correct. I checked in flt_functions.F and the 
>>global2local_i, and global2local_j that come out are totally 
>>unrealistic.
>>
>>I hope that some one can help me with this problem.
>>
>>
>>Thanks,
>>
>>
>>Sergio Jaramillo
>>  Oceanography, Dept. of Earth and Ocean Sciences, Univ. of British 
>>Columbia,
>>  6270 University Blvd., Vancouver, B.C. CANADA V6T 1Z4
>>  phone: (604) 822 - 3911 email:     sjaramillo at eos.ubc.ca
>>                                     sju at eos.ubc.ca
>>
>>
>>_______________________________________________
>>MITgcm-support mailing list
>>MITgcm-support at mitgcm.org 
>>http://dev.mitgcm.org/mailman/listinfo/mitgcm-> support
>>
>>    
>>