Creating random steps for home ranges NOT steps

[vladimirovav]

Hi Johannes,

I've spoken to you previously. Your package is amazingly useful for my project, but I am planning to use it for home range shifts as opposed to steps (like normally you'd use it for). My issue is that unlike with steps, home ranges don't have nice and neat polygons that you can extract. Home ranges of caribou (my study animal) are very patchy and not uniform (created kernel Brownian bridge using adehabitatHR package). When you have something that patchy it is difficult to figure out where to put the random kernels/home ranges to create available areas that I can compare with the selected ones. I can not use MCP (minimum convex polygon - that would have nice one polygon per year/animal) because they were shown to not be a good representative of a habitat selection. In addition, caribou never leave completely their home range, which means that the area between the two years always overlaps, creating a problem when deciding how to create random home ranges. What would you advise me to do?

Thanks,

Varvara

p.s. I have attached a photo of the polygons that I have so you can see roughly what I'm talking about.