ABSTRACT
Estimating proportion of sites occupied, or proportion of area occupied
(PAO) is a common problem in environmental studies. Typically, field
surveys do not ensure that occupancy of a site is made with perfect
detection. Maximum likelihood estimation of site occupancy rates when
detection probabilities are less than one have been developed by MacKenzie et al. (2002) under the assumption of a simple random sample. Their procedures are generalized to cover complex survey designs, in particular, survey designs with stratification, cluster sampling, and unequal probability sampling. Three studies are used to motivate the problem and illustrate the estimation problem. One study is of fish species
present/absence in northeastern lakes based on an unequal probability survey design conducted by EMAP in 1992-1996. The other two are from two studies conducted by the US Geological Survey’s Amphibian Monitoring and Research Initiative on the occurrence of specific amphibian species in ponds in Olympic National Park and southeastern Oregon. In both studies a two-stage cluster sample was used where the first stage primary sampling units are 5th field hydrologic units and the second stage are individual ponds located within each selected hydrologic unit. The presentation will describe the survey designs, present PAO estimation for complex survey designs, and illustrate the estimation with data from the studies.

Refreshments will be served at 2:45 p.m. in Room 008 of the Statistics
Building