Finding the Lowest Elevation Inside of a Polygon

[bydesign99]

I have a need to find the lowest elevation inside of a small rectangular polygon (6'x7'), then assign the polygon to that elevation. This is fairly easily done manually with feature lines, however I need to repeat the operation about 3,000 times to create a design grid (therefore, not obliged to using all of the design budget to do so). Does anyone know of a method to automate this? I employed the GIS department to run the analysis, which they can do, however they can't rotate the DEM grid output to match my project because it doesn't run parallel to north/south, thus skewing the results.

Ideas, commands, lisp routines, scripts, anything?

[Neilw_05]

I do think this would be better handled by GIS. The GIS guys should be able to densify the grid so you get more sample points inside the polygons.

Do they have spatial analyst?

[bydesign99]

The GIS analyst managed to get it done, but since the grid doesn't align there is a slight loss in accuracy. It's probably not enough to be concerned about, but thought it would be a good tool to have as we are using this more frequently now for some specific design tasks.

You confirmed my suspicion, thanks for the reply.

[Neilw_05]

Won't you have the same grid alignment problem in C3D? How would you propose to address that?

[tcorey]

A few months ago, I wrote a program that analyzes a surface and finds the high and low points. You might be able to modify the code to do what you're attempting. If you want a copy to play with, just send me an email: tcorey at shasta dot com.

Are your surfaces TIN surfaces or Grid surfaces? I might get time next week to make an attempt at this for you, assuming Christmas week is as quiet as usual.

Tim

[bydesign99]

We are using TIN surfaces. When I exported a DEM to the GIS analyst it was a grid with 1ft spacing.

We actually got ahold of a LISP routine that can set a feature line to the lowest elevation vertice. By using that we set the feature line over the grid, captured elevations from the surface, then ran the LISP. Voila. The grids are small enough where this works wonderfully.

Thanks for all of the input!