1

I am using this tool. I am an experienced GIS analyst and have read the documentation, but I can't seem to understand the output of the tool.

In my data, there is a remote area that only has 5 points. I have the search area radius set to be small enough so that only these 5 points should be in the search radius. Population field is set to none. So, I would expect the output values to be the area of the cell divided by the number of points in the search radius. (10,000m / 5 points in this case). Also, all the points in this area should have the same value, and yet they are all slightly different by a few decimal points. The math ain't mathing.

screenshot of my results

What units are the output of this tool in? According to the documentation it's density (points per square kilometer), but that doesn't make match the results I'm actually getting. What's really happening in this tool?

1 Answer 1

2

You are correct that the default output of the Kernel Density tool in ArcGIS Pro is points per unit area, which depends on the linear unit of the projection you are using. For example, if your linear unit is meters the default is points/km2.

Even though all your points may fall within the search radius, you would not necessarily expect all the resulting raster cells to have the same value. See this page that explains how Kernel Density is calculated. The key idea is that:

Conceptually, a smoothly curved surface is fitted over each point. The surface value is highest at the location of the point and diminishes with increasing distance from the point, reaching zero at the Search radius distance from the point.

In other words, the influence of each point on the calculated density values is highest at that location, and decreases as you approach the search radius (where it is 0). Since your points are unevenly spaced, those that are closer together will generate raster cells with higher calculated kernel densities.

If you want a more straightforward count per unit area, I would suggest creating a grid with the Create Fishnet tool and then the Aggregate Points tool to count the number of points in each grid square.

2
  • Thanks. I get how it works now. This does make it really hard to use these values in a real map. The numbers are not really points per unit area, but like weighted points per unit area.
    – yeet_man
    May 17 at 18:27
  • Exactly @yeet_man. The values are weighted based on distance from each input point. The function used to determine that weighting is the "kernel" in kernel density. If you want a more straightforward count per unit area, I would suggest creating a grid with the Fishnet tool and then the Aggregate Points tool to count the number of points in each grid square.
    – lambertj
    May 18 at 14:04

Your Answer

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

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