I have been working with GIS for many years but I am fairly new to the spatial analyst side of things. I am working with many satellite tracks for many different individual animals and would like to determine a migratory corridor for these animals. We have decided to use line density for the analysis of these data. Our question is: how to determine the search radius, which I have learned is very important?

I found out that one way to do this would be to use the tools on movebank.org like ArcMET and Home Range Tools. However to use these tools, a time/date stamp is required for every point on the satellite track and we do not have this information for all of the points. This is also the reason why we decided to use line density instead of kernel density. So we decided to use line density in ArcGIS.

I found that one way that many people determine a search radius for point density or kernel density is by using the Incremental Spatial Autocorrelation. However to perform this type of spatial statistics, one needs count data/incidence data. Since these are satellite tracks, the data inherently is only one animal per track (line).

So based on these issues, does anyone have any ideas on how to determine the search radius? Any other spatial statistics that we could use?

closed as too broad by PolyGeo Apr 14 '16 at 11:05

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 2
    What uses will you be making of the resulting maps? That should be an important determinant of the search radius, which will strongly influence the appearance of its width. – whuber Dec 15 '14 at 17:20
  • 2
    Thanks for reading and responding! The resulting maps will be used to determine the migration corridor for species which would be used for conservation purposes - determining what percent of the corridor is located in protected areas or managed areas, determining possible new protected areas, what percent of corridor overlaps with threats, etc. We will also overlap this corridor with other species corridors to see which species use the same pathways during migration. – Amanda H Dec 15 '14 at 17:32
  • 1
    Those results may be so sensitive to your choice of search radius that perhaps you should consider alternative methods of modeling the migration corridor altogether, lest your maps ultimately be seen as arbitrary. – whuber Dec 15 '14 at 17:37
  • 1
    The search radius should be defined on a quantitative factor from your study. Example, finding risk areas for tornado paths commonly use a 25 mile search radius. That is because issued watches are based upon the likelihood that a tornado will occur within 25 miles of a given point. Are there any calculated corridor widths or average migration sizes to help figure the search radius value? Perhaps average widths of existing corridors can be used? You may also want to break up your data based on the topography. Open areas vs. mountains most likely influences the corridor as well. – evv_gis Dec 15 '14 at 18:22
  • Thanks whuber! Yes, I am afraid that the results would be sensitive and we definitely would not want our maps to been seen as arbitrary but unfortunately we haven't found a different method to model the corridor. Thanks evv - I will look into this as an idea but unfortunately I think this varies depending on the species, number of tracks and variation of track locations in the study, and topography. Anyone else who has other suggestions for various spatial statistics would be really helpful! Thanks! – Amanda H Dec 15 '14 at 20:28