Which tool to look at correlation between landcover and forest fragmentation?

[SophiaKirn]

For my research, I want to see if forest fragmentation (done in Guidos Toolbox) is correlated with specific land cover types. I also want to see if the pattern of forested pixel gain/loss (also done in Guidos) in an area is clustered at all and around what specific land cover types (i.e. urban or agricultural). I am looking at two tools in Arc Pro right now, the Spatial Autocorrelation tool and the Multidimensional Raster Correlation tool to do this, but I am not sure which tool would be best and am unfamiliar with both. Which tool would be best for this, and would I need any other inputs (I have a forest fragmentation raster, NLCD raster, forest pixel gain/loss raster) in order to run the tools? Thank you! 🙂

[MervynLotter]

 Look into using Forest-based and Boosted Classification and Regression (Spatial Statistics)—ArcGIS Pro | Documentati.... It may work for you. It is a rather amazing tool and can be used for prediction only, suggesting how important some variables are in explaining the phenomena you are assessing. Variables can be continuous or categorical.  It can even consider the distance form certain features as input variables (such as your specific land cover types) although you may need to extract these as individual layers. 

Also look into creating scatterplots to simply assess correlations using scatterplots. These work on rasters too. 

[SophiaKirn]

Thank you so much, I will look into that!!