Refinements on Creating Vertical Buffer Based on Ridgelines

[MarthaSelig1]

I need some help with a workflow, please, as my raster analysis experience is limited. I read and tested Bill Hubers' write-up on this (http://forums.esri.com/Thread.asp?c=93&f=995&t=155595&mc=3#msgid455712) and it worked very well for the single ridgeline I tried it on, but not for the larger ridgeline file. I need to make some additional refinements and I'm not entirely sure how to proceed.

In my case, the file contains only selected ridgelines. I need to define 50' and 85' vertical buffers around these, but I am only to consider the area around each ridge down to the point where elevation begins to rise again.

Once I ran the Euclidean Allocation and did my subtraction, I realized there were lots of areas where the elevation was higher than my ridge points, yielding negative values that met my criteria of <= 50. I don't want to exclude all values < 0 , however, because I don't know that the ridgeline file I've been given actually tracks to the highest points along each ridge. Similarly, just because a point is lower than the ridge doesn't mean it isn't on the upslope of another ridge.

I thought that perhaps a basin analysis would assist me in identifying these low point barriers but even if I could create one (I got the flow direction fine, but the basin tool has failed twice with a very helpful 999999 error), I'm not entirely sure what to do with it. My idea is that I want the allocation to be NoData beyond these low points.

If someone could please help me with a procedure, I believe I can make my way through it.

Thank you for your assistance. I love new stuff.
Martha Selig

[MarthaSelig1]

OK, with some further research, here's the workflow I've devised that I think will do the job:

1. Invert the DEM by multiplying by -1
2. Follow the steps in the help system's hydrologic analysis sample applications topic to produce a watershed analysis.
3. Convert the watershed to polygons and select those containing my ridgelines.
4. Use this new layer as a mask and continue with the original set of instructions to derive my buffers.

Next up, understanding pour points and how they affect the watershed.