[MITgcm-support] KPP diffusivities

[Dimitris Menemenlis]

Hi Stefano, I do not have any experience with KPP using vertical levels 
of 0.5 m.  the smallest I have used is 10 m near the surface and usually 
much coarser near the bottom.  But I am curious to try and understand 
your problem.  I have following questions:

Do you have currents near the bottom of your gulf?  What you may be 
seeing is mixing due to shear instability.  This happens when the local 
Richardson Number is greater than Riinfty (default value is Riinfty=0.7).

You can test above hypothesis by recompiling your code with:
#define EXCLUDE_KPP_SHEAR_MIX
in KPP_OPITONS.h

Or if you do not want to recompile you can either set Riinfty to some 
ridiculously large value or set:

       difm0   = 0.0
       difs0   = 0.0
       dift0   = 0.0

in the data.kpp runtime file.

Or if you do not want to rerun, you can plot vertical shear near the 
bottom of your Gulf, to see if it has a daily cycle that can explain the 
eddy diffusivity that you observe.

Cheers, Dimitris

-- 
Dimitris Menemenlis <menemenlis at jpl.nasa.gov>
Jet Propulsion Lab, California Institute of Technology
MS 300-323, 4800 Oak Grove Dr, Pasadena CA 91109-8099
tel: 818-354-1656;  fax: 818-393-6720