Tailoring a shapefile
I'm a new ArcGIS 10 user, and I have a question for the community. I would like to take my Ohio county shapefile and create regions for our Alcohol, Drug Addiction and Mental Health Services (ADAMHS) Boards. This process should involve combining 2 or more counties into one unit. I would still like the county lines to be visible, but I want the regional boundaries to have a thicker line. Any help would be greatly appreciated.
Mapping Center Answer:
This is a pretty simple process if you remember that your counties are the starting point and that they have an attribute that identifies which region they each belong to. So, you would use the Dissolve tool to aggregate the counties based on a specified common attribute (your region code or name) and this creates a new layer in your Table of Contents. If you position your new regions layer above the counties layer and symbolize the region polygon boundaries with the thicker line symbol you want to use then that will mask the thinner county lines underneath and give you the effect you are after. (If you still want to see the thinner county lines under the thicker region lines, you could apply a transparency to the regions layer - right click the regions feature class in the Table of Contents, click Properties, and on the Display tab, set the transparency level you want). There is a lot more information on how Dissolve works at the ArcGIS Resource Center on how the Dissolve tool works
Dissolve is a powerful tool for building new datasets from existing ones and works great when all features are contiguous. You may end up with multipart features if you have noncontiguous elements which is fine as long as you remember it is represented in the new attribute table as one record (instead of several for each noncontiguous element).
If you want to clean the resulting data after Dissolve to remove superfluous endpoints you can additionally use the Unsplit Line tool but for your purposes this step probably won't be necessary.
If you would like to post a comment, please login.