So, this is for the benefit of a numerical model?
in addressing piping you clearly need to assess the critical gradient, the head drop in the last cell or two and make the comparison against some safety factor. If the soil is truly anisotropic, you can use a numerical model or you can transform the flow net.
If this is all too complicated, you can also simplify the problem statement, see the scale of the concern and still install a toe drain to allow the exit gradient to occur under confinement.
I will remain unclear on the end game. Is this a dam, retaining wall, shoring system, box culvert? Are there boring logs to make you think of an anisotropic setting? Are you striving to use anisotropy to simplify a layered system of soils, etc.
I enjoy these sort of problems!
f-d
ípapß gordo ainÆt no madre flaca!