Take the 2-minute tour ×
Geographic Information Systems Stack Exchange is a question and answer site for cartographers, geographers and GIS professionals. It's 100% free, no registration required.

Working in ArcGIS 10, I have a set of contour lines representing density of commercial square footage, and I have a set of points representing local maxima for each peak of the dataset. For each maximum I would like to find the largest contour line that contains only that maximum and no others.

My strategy is to convert the contours to polygons, spatially join the polygons to the points, choose only the polygons that overlapped 1 point, and for each maximum choose the largest polygon, an operation that will probably happen in MS Access.

I've run into one problem - when I convert the contours to polygons using "feature to polygon", the polygons don't overlap, so my maxima only join to one polygon, breaking the method.

How can I convert polylines to overlapping polygons? If possible, I'd prefer a solution I can fold into an ArcPy script.

share|improve this question
    
what do you mean by a set of contour lines, is this one contour layer or more than one? A single contour layer lines should not cross/overlap. –  artwork21 Aug 10 '11 at 11:24
1  
Artwork, I believe he wants the polygons to be complete discs, not rings (no interior "hole"). So even though the source contour lines didn't cross, the resulting polygons would overlap. –  user3461 Aug 10 '11 at 12:39
    
Yes, that's correct. In many forums it seems as though many people have had the opposite problem of trying to convert overlapping polygons to cocentric, non-overlapping ones. (including this particularly thorough one from our own whuber: link) –  Patrick Aug 10 '11 at 14:06
    
I've accepted whuber's answer, as it gives me exactly what I was looking for, but I think there may be other answers for the general case. Several I had found include using XTools or ET Geowizards to convert the polylines to polygons, or else removing the holes from the non-overlapping polygons. –  Patrick Aug 10 '11 at 15:44
add comment

1 Answer 1

up vote 2 down vote accepted

You started off with a kernel density grid. Negate it (thereby flipping it upside down) and fill the sinks. Comparing the filled grid with the original will identify all areas of fill: their boundaries are the desired contours.

share|improve this answer
    
This is exactly what I needed, and in fact it's much better than what I hoped for. The watershed tool actually gives me exactly the polygons I was looking for in both questions - a polygon that separates the kernel density raster into regions. The method I was attempting here would have only approximated them using the largest contour that only enclosed one maxima, but it would have left gaps. This is perfect. Thank you! –  Patrick Aug 10 '11 at 15:10
    
@Patrick You have anticipated my answer to your previous question! Since the watershed has worked for you, please consider updating that other thread so others can see your solution. –  whuber Aug 10 '11 at 15:12
    
This answer perfectly solves my problem from yesterday as well. If you post it there, I will accept it as the best answer. –  Patrick Aug 10 '11 at 15:14
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.