My observation, noted in one of your previous threads, is that the first image (e5m9dr) is already coded by elevation of the surface, with red being lowest and blue being highest. I gave specific coordinate examples of depressions (which could be small pores) in the previous discussion.
What you need to calculate is relative elevation changes: elevation compared to what is nearby. That is what a gradient plot is.
As I noted before, though, there are a minimum of two directions that the gradient needs to be measured in for each point, but you only have one plot location per point. You must decide how you want to handle having two simultaneous direction deltas in one location. I already gave a suggestion about one representation method.
