HELP! I need to consider 'edge effects' when creating a tool for focal statistics

[philcass2]

Hello - I am very new to ARCMAP and the learning curve is steep!

I am trying to create a tool in ARCMAP for focal statistics. I am struggling how to consider possible edge effects and make sure the compactness is correctly calculated for pixels near the boundary of the study area. 

Does anyone know of a way to do this?

[DanPatterson]

This applies whether arcmap or Pro

How Focal Statistics works—ArcGIS Pro | Documentation

So if you don't like the way edge effects are handled, then you have two other possible options

• enlarge the area being studies beyond your current area of interest, then clip the focal statistics results after
• keep your area of interest and clip the focal statistics results to an area smaller than that influenced by edges

[philcass2]

Thank you for you're reply!!!!!!!! 🙂

"enlarge the area being studies beyond your current area of interest, then clip the focal statistics results after"

are there any decent steps or tutorials that can help?

[DanPatterson]

Clip (Analysis)—ArcGIS Pro | Documentation  for vector geometry (eg polygons)

Clip Raster (Data Management)—ArcGIS Pro | Documentation for rasters

is all you need.  If you start with vector, clip the source to a larger area, run focal statistics, then clip raster to remove a desired amount from around the edges.

If you have raster only, the clip raster for both.

If you don't have any more area other than what is shown, you omit using a larger input area, and just trim the focal statistics result with clip raster.

In short, the edge effects are generally well handled and well accepted so you may be making unnecessary extra work for yourself