[R-sig-Geo] Home range overlaps calculated using BA are sometimes >1

[Charlotte Regan]

Hi there,

I'm using kerneloverlaphr from adehabitatHR to estimate the home range  
overlap between pairs of sheep. I really need an overlap measure that  
scales between 0 and 1 and therefore settled on BA, but I'm finding  
that in a few cases the overlap is greater than 1. It seems to be  
happening when an individual's UD is compared to itself.

My code is as below:

spdf<-SpatialPointsDataFrame(census[,12:13],census[,1:2])
UDs<-kernelUD(spdf[,1],same4all=TRUE,grid=200,extent=4)
overlaps<-kerneloverlaphr(UDs,method="BA",percent=70)

I noticed that there's been a post asking the same question before  
(https://stat.ethz.ch/pipermail/r-sig-geo/2011-February/010875.html),  
but an answer wasn't posted to the thread.

Apologies if it's just some failure in my coding/understanding, but  
thank you very much for any help you can provide!

Thanks,

Charlotte Regan

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.