Cumulative area distribution

[PhilipWarner]

Hi,

I'm carrying out a geomorphological analysis in a mountain area and have calculated some parameters I'd like to analyse. From a DEM I have calculated the drainage areas contributing to each cell of the DEM and also the local (cell) slope. It is well known that there is a relationship between these two variables, and would like to figure out the shape of this relationship for my study area. My final goal is to identify changes in the relationship which would indicate the shift from one type of predominant erosion phenomenon to another (for instance diffuse erosion to channeled erosion). This is a fairly consolidated methodology. However, I'm struggling with creating the graph that would allow the investigation of this relationship in ArcGIS.

In short: I'd like to build a graph showing the cumulative area distribution graph against the drainage area. The graph is computed by arranging all drainage areas say in a basin or catchment in numerical order and computing the probability of exceedence for any given area value.

Any idea on how to do this? Any help is appreciated.
Philip

[curtvprice]

If I'm understanding this right, the Slice tool could be used to find the area value at a given percentile. However, this could be derailed by the fact that the distribution of cell drainage areas is dramatically skewed, to the point where calculating a area distribution curve may be problematic. Most cells in a basin will have very small drainage areas, and just a few (in the drainage) have very large drainage areas.

[PhilipWarner]

Thanks for your answer curtvprice. I have tested the slice tool, and it's certainly a step forward. However, the output of the tool is the number of cells within each interval I think, while I was after the exceedance probability for any given drainage area value. Am I missing something or there is some additional steps I need to add in my workflow?

Thanks also for pointing out the risk of skewed results because of few cells with extremely high value.

Any help is appreciated.
Philip