Covariate Adjustments to Effective Area in Variable-Area Wildlife Surveys

Publisher Website
Google Scholar
Abstract
The problem is that of adjusting estimates of effective area surveyed in line transect or variable circular plot wildlife surveys for covariates that are known to influence detectability. We present a model where this can be accomplished within a general linear models framework (Nelder and Wedderburn, 1982, Journal of the Royal Statistical Society, Series A 135, 370-384). The model treats effective area as a scale parameter for detection areas, and uses the logarithm of effective area as a link to the covariates. Maximum likelihood reults re presented along with least squares estimates that provide good starting values for iterative searches. We apply the methodology to a situation arising in an avian survey of Saipan Island.

This publication has 2 references indexed in Scilit: