[Mitgcm-support] Re: GM c48

[mitgcm-support at dev.mitgcm.org]

Patrick Heimbach wrote:
> Surprising to me to need Large_SlopeSqr = 1.e48
> even surprising to me that they can be as large as 1.e26
> Doesn't that hint at another problem somewhere in GM?

The slopes are formally allowed to become infinite. In practice
the largest possible value of the slope is determined by the small-number
epsilon:

   max(S) that could be passed to slope_limit() 
            =  max( delta sigma ) / ( Delta x * epsilon )

since sigma_z is limited to be -epsilon.

The value of 1e48 chosen by J-M is larger than this so it should
be doing nothing.

A.