Extracting overlap area between home ranges

[Jenny Hansen]

Hi Chris,

   I was wondering if there is a way to extract the overlapping area between two akde home ranges? I can make spatial polys of individual home ranges, but I am hoping to get spatial polys of overlap areas of 50% and 95% home ranges between individuals. I have used sf & sp to overlap adehabitat home ranges before, but I was curious if there was a native method within ctmm.

Cheers,

Jenny

[Christen Fleming]

Hi Jenny,

There is a more rigorous overlap measure implemented in the package, with confidence intervals and some bias correction.

• https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.13027
• https://ctmm-initiative.github.io/ctmm/reference/overlap.html
  

Best,

Chris

[Jenny Hansen]

Hi Chris,

   Sorry for the late reply. I should have been more specific. I definitely use the overlap function to get the overlap estimates with confidence intervals; that is one variable I use in my analyses. What I am trying to do is get the spatial area as a polygon to use in a resource selection function analysis. So, for example, I want to run five RSFs based on mother-daughter dyads from my dataset. One in each of their core ranges, their 95% ranges, and one in the overlapping area between the two individuals. Poking around the ctmm package functions, it looks like I need to export the UDs as SpatialPolys and derive the overlapping and non-overlapping polys using sf or sp. This is not a problem. I was just curious if there was a function within ctmm that was like st_overlap(), over(), or gIntersection(). Thanks for your response. :)

Cheers,

Jenny

[Christen Fleming]

Hi Jenny,

Yeah, I don't have any built in functions for that. Also, even though it's very common to do RSFs inside home-range contours, and even though I've been on papers where my co-authors have done that, I don't recommend it. Assuming you don't have any spatially constraining terms in your RSFs and the polygon is what is constraining the estimated distribution, then the results are generally very sensitive to the choice of polygon, which isn't being fit within the RSF. I think it's a better idea to throw in spatially constraining terms like distance to a stream that the individual likes or x, y, x^2+y^2 for a smooth circular constraint. But just note that these kinds of terms can be colinear if they are describing the same effect.

Best,

Chris