That sounded like an interesting algorithm development problem to me, so I did a little research and found some suitable-looking algorithms. I even partially implemented one. But I never did find the time to work out all the different pieces, so I never got around to writing about it here.
I finally decided recently that I just need to dive in and try it. And I'm going to do it in front of you readers. I've picked a paper that looks promising (by David Tarboton of the Utah Water Research Laboratory, Utah State University), and for the next few weeks I'm going to try to turn the methods described in this paper into a working MATLAB implementation.
We're going to get into flow direction computations, how to deal with plateaus, and maybe (if I'm guessing right) even some sparse linear algebra.
Next up: the D∞ flow direction method.
To leave a comment, please click here to sign in to your MathWorks Account or create a new one.