I'm trying to create a map that has a all layers within a subject area having 0% transparency (full opaque), while features outside that area (defined by a polygon shapefile) having a reduced transparency of say 50%. The ideal result is a map that has a focus on a specific area, but also shows the surrounding area. I'm working with ArcMap 10.

Similar to this image:

enter image description here

I tried using the clip feature in the Data Frame properties but its not what I'm looking for.


  • 3
    Those kinds of images are often drawn by overlaying a mask shapefile on top of the others, drawing it in gray with some alpha transparency. (The mask is the set-theoretic complement of the area of interest and is readily created using geoprocessing operations such as a union.) This requires no change in any of the layers that make up the map itself.
    – whuber
    Feb 28 '12 at 21:16

Similar to whuber suggestion (using geoprocessing tool) you can create this look by:

  1. Create temp polygon shp, and digitize a large feature over all features that you want to mask out.
  2. Use the Erase tool to erase the temp polygon from your AOI layer.
  3. The resulting layer, from the erase operation, you can load into ArcMap and give it a transparency.

This method will not mask labels though. To mask labels you can setup a separate label class (for each feature that has labels) to define those features within and outside your mask.


Here is what I did:

  1. Created big_aoi featureclass
  2. Created small_aoi featureclass (the real area of interest in the middle)
  3. Unioned the two (but Erase will work better as artwork21 points out), got big_aoi_Union featureclass
  4. Deleted the inner polygon from the union to get the donut
  5. Set big_aoi_Union to dark grey with 50% transparency

And I got:

enter image description here

This could probably all be scripted with arcpy I bet.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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