Estimating animal densities and home range in regions with irregular boundaries and holes: A lattice-based alternative to the kernel density estimator

摘要

Density estimates based on point processes are often restrained to regionswith irregular boundaries or holes. We propose a density estimator, thelattice-based density estimator, which produces reasonable density estimatesunder these circumstances. The estimation process starts with overlaying theregion with nodes, linking these together in a lattice and then computing thedensity of random walks of length k on the lattice. We use an approximation tothe unbiased crossvalidation criterion to find the optimal walk length k. Thetechnique is illustrated using walleye (Sander vitreus) radiotelemetryrelocations in Lake Monroe, Indiana. We also use simulation to compare thetechnique to the traditional kernel density estimate in the situation wherethere are no significant boundary effects.