[MITgcm-support] cal, exf, obcs: problem at the tile boundaries

[Nico Wienders]

On Jun 4, 2009, at 2:57 AM, Martin Losch wrote:

>
> going back to your initial message, if think that this is the clue:
>> EXF WARNING: vstress out of range for bi,bj,i,j,it=       1       
>> 1     90    267      0   0.260417995106866E+31
> exf_check_range often complains about too high heat fluxes (and then  
> they are too high by maybe a factor of 2 or so), but I have never  
> seen it complain about u or vstress, and 1e30 points to a single/ 
> double or little/big endian issue. I think it is the former, because  
> I think that you are using the default value of  exf_iprec = 32  
> (exf_readparms.F), but you said that your input fields are real*8  
> (which would correspond to exf_iprec=64, which should show up in  
> your data.exf file).


Martin,

Shame on me, i had never seen this little guy. This was the trick.
You are my savior. Thank you so much.  I owe you a drink (I mean many).