Point process estimation using distance measurements
by
Gavin Melville
NSW Agriculture
Coauthors: P. Hall(Australian National University), A. Welsh (University of Southampton)

Many of the objects studied in ecology may be represented as a point process. Examples include trees, shrubs, weeds, fungi, and nests belonging to animals and birds. Measurements between objects of the point process are often used to estimate the intensity of the process. Although this works quite well for a Poisson process in an infinite region, in practice most ecological populations are aggregated and they are always measured in a finite region. This results in a bias in the intensity estimates, due both to edge effects and also to the non-random nature of the process.

This talk focuses an a little-used sampling procedure that is designed to quickly and easily collect distance measurements in the field. Methods of obtaining unbiased estimates are presented, using a nonparametric bootstrap to simultaneously remove both of the above sources of bias. Examples are given from simulated and actual plant data to illustrate the properties of the estimates and variance estimates.

Date received: September 4, 2002

Copyright © 2002 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cajn-26.