ESTIMATING POPULATION-CHANGE FROM COUNT DATA - APPLICATION TO THE NORTH-AMERICAN BREEDING BIRD SURVEY

For birds and many other animal taxa, surveys that collect count data form a primary source of information on population change. Because cou nts are only indices to population size, care must be taken in using t hem in analyses of population change. Temporal or geographic differenc es in the proportion of animals counted can be misinterpreted as diffe rences in population size. Therefore, temporally or geographically var ying factors that influence the proportion of animals counted must be incorporated as covariables in the analysis of population parameters f rom count data. We describe the North American Breeding Bird Survey (B BS) for illustration. The BBS is a major, landscape-level survey of bi rds in North America; it is typical of many count surveys, in that the same sample units (survey routes) are sampled each year, and change i s modeled on these routes over time. We identify covariables related t o observer ability, the omission of which can bias estimation of popul ation change from BBS data. Controlling for observer effects or other potential sources of confounding requires the specification of models relating counts to population size. We begin with a partial model spec ification relating expected counts to population sizes; we describe es timators currently in use in relation to this partial specification. A dditional assumptions lead to a class of overdispersed multinomial mod els, for which we describe estimators of population change and procedu res for parsimonious model selection. We illustrate the use of overdis persed multinomial models by an application to data for Carolina Wren (Thryothorus ludovicianus).