Trying to interpret Kernel Density output

[GinoOddone]

Hi everyone, 

I have done a Kernel density analysis but the results are somewhat confusing. Are these reasonable or am I missing something? 

Any assistance or advice is appreciated. See attached for further information.

 

 

[DanPatterson]

Kernel Density (Spatial Analyst)—ArcGIS Pro | Documentation

magnitude per unit area. 

Your area is quite large (aka the denomitor in the division)

are the roadkills quite small per unit area?

[GinoOddone]

At the chance of sounding ignorant I have no idea for sure. I have 249 incidents in a road network. Here is the map. Also would these challenges be resolvable in ArcPro Gis documentation given that I am using ArcMap? 

[DanPatterson]

The tools are the same/similar in their operation between arcmap and Pro.

Zoom into one of your red areas and see how many points fall within a raster cell... probably not many, hence the low density (area is in meters squared I presume)

[GinoOddone]

The area is in square kms. 

[DanPatterson]

So your output cell size is 1490 sq km? (see your first post)

I would check the inputs and the raster properties.  also 10^6 m^2 per km^2

[GinoOddone]

Is there a guide for how to identify an appropriate output cell size? When doing the optimised hot spot analysis I used a distance band of 70333,51m as one of the parameters. For Raster analysis environments I also notice it does not make any significant difference to the figures.