Method for Spatial Analysis of Raptor Predation

764
1
05-24-2017 11:06 AM
AdamEichenwald
New Contributor

Hi everyone, thank you very much in advance for your help!

I currently have presence data on two birds, a falcon species and the grouse that it prefers to hunt. These data are in two point shapefiles, one for each species. The grouse shapefile represents all of the individuals that I've seen during observational transects, while the falcon shapefile represents all falcon nests in my study region.

I have a hypothesis that the grouse species are located in areas where they are near the fewest possible number of falcon nests (there is almost nowhere on the study site completely without falcon nests, but there are locations where nests are more concentrated). I'd like to examine this with some sort of statistical GIS analysis, but I'm not sure what steps I should take. I've been looking into kernel analysis and the near distance table, but I'm having trouble putting things together into a cohesive method.

I'm hoping to get any kind of advice on this. Anything helps, from references you think I should consult all the way to methods you think will work. Thank you!

P.S. My grouse data is biased toward 3 roadways that run for miles through my study site. If possible, I'd like to also entertain method that may take this bias into account -- but only if possible!

0 Kudos
1 Reply
DanPatterson_Retired
MVP Emeritus

Image if you could please. 

So you basically have two point patterns and an assumption that one pattern will have a preferential location at some distance from the other.  There may also be some sampling bias in one of the patterns.

Do you have any other support data, like preferred cover type  or anything on the surface that may provide a suitable 'attractant' which might account for the observed pattern, since you are going to have to balance that with the 'deterent' factor of the other pattern.  As for sampling... that is a big one, so I presume that you can rule out a preference for road networks as an attractant for one pattern.

Anything else would be useful.... and try to forget that one pattern is predator and the other is prey, this might cloud other avenues of explanation.

0 Kudos