I have a large forest habitat raster dataset (several watersheds in state of Oregon) that I need to convert into a shapefile. The raster has 70 different categories of habitat types.

Steps: I've converted the raster into a a polygon using the "Raster to Polygon" tool and placed it in a GDB. The resulted feature class ended up with over 6 million records. Next, I tried to dissolve the feature class based on the Habitat type, which should result with 70 records total (out of the 6m).

Problem: The dissolve tool seems to work, but its not including all the habitat types. It looks like it stops mid point and I end up with 30 habitat type vs. the 70 I should get.

Is this a simple issue of lack of memory? I am working with a powerful machine (64GB RAM).

During the conversion from raster to feature class, is there a way to avoid getting 6 million records and reducing that number somehow?

enter image description here


1 Answer 1


I found a solution/work-around:

I've simply split my polygon (that contains 6+ million records) into smaller parts (sub-watersheds in my case) and dissolved the smaller feature classes (which varied between 300k records to 1.2 million records) Next, I took those sub-watersheds feature classes and merged them all together. Lastly, I dissolved the merged file one more time to get my final results.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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