[R] accumulation curve for kernel density using adehabitatHR

[Bernardo Santos]

Dear all,

I am trying to create an accumulation curve for kernel density estimation (KDE) of home range size, with kernel density in the y-axis and number of telemetry fixes in the x-axis. I am using "kernelUD" and "getverticeshr" functions from adehabitatHR package.

However, I have a problem: when re-sampling a few number of points to calculate KDE, the function getverticeshr (and sometimes kernelUD also) returns an error (maybe because the fixes are too close and it is impossible to normalize the surface, but I am not sure), and then it is impossible to build the accumulation curve.

I tried to "throw out" this cases in which the functions return errors and sample once again, but it produces a bias towards high and unrealistic values of home range estimation.

Have anyone worked with that before, or have any cues on how to solve that?

Thank you very much.
Bernardo Niebuhr

	[[alternative HTML version deleted]]