No replies to this topic

[Gretchen Peterson]

For a recent wetlands analysis study, a need arose to develop a new GIS method, and I thought I'd share a bit about what I came up with here. Here is a ink to a more detailed report.

The client wanted to determine where, on the landscape, a line representing a certain elevation increase from the river centerline is. Basically, this could be described as a variable-width buffer that depends on elevation. The method that worked the best--though it does have some flaws, discussed in the report linked to above--was to do a Euclidean allocation of the stream centerline elevations and subtract those from the actual elevations at each point. Then it is a simple raster calculation to determine which cells are higher than the elevation threshold and which are lower.

The goal was to get at an approximation of water flow potential that would, down the line, inform on potential wetland loss. This study was input into some other work they did, so it is not to be interpreted in a vacuum. However, it is the first instance of this kind of analysis that I'm aware of. There is a different method that can be used (described here) in cases where you can assume a constant river elevation. However, my study was done on a large portion of the Poudre River, which has quite a bit of elevation change as it descends from the foothills into the plains.