[Mitgcm-support] Re: More results from cubed-sphere run

[mitgcm-support at dev.mitgcm.org]

Cirt,

Convtave is a convective index: fraction of time spent convecting.

It's a 3D field so a bit difficult to plot so I plot this 2D function
instead:

>> t=rdmds('Ttave',Inf);
>> [nx ny nz]=size(t);
>> c=rdmds('Convtave',Inf);
>> G=loadgrid;
>> hml_mn=reshape( reshape(c,[nx*ny nz])*G.drf' ,[nx ny]);
>> hml_mx=reshape( reshape(mask(c),[nx*ny nz])*G.drf' ,[nx ny]);
>> pcol( ( sq(mask(t(:,:,1))).*hml_mn )' );colorbar('h');

It's not the same as mixed layer depth but good enough. It integrates
the frequency of convection over depth; if the model is always convecting
over a certain distance then hml_mn will equal that depth. hml_mx is
an indication of deepest mixing activity but again is not exactly the same as
maximum mixed layer depth.

A. (air not er)

Curtis hIEsey wrote:
> Alister, did you say that you
> have some Matlab code for
> mix layer depth?