Unless you have to characterize the transition zone (ie low relief reservoir, low perm) under your conditions I'd probably just go with log derived saturations cross plot against log perm and characterize the irreducible water saturations

If transition zone is important, again, you can perhaps go to your log derived saturations and log derived perm - taken in conjunction with the height above FWL you may be able to reverse out a generic relationship

There is no robust generic correlation that I am aware of.