Estimating total population size for Songbirds
This article is part of a larger document. View the larger document here.Abstract
A conviction has developed during the past few years within the avian conservation community that estimates of total population size are needed for many species, especially ones that warrant conservation action. For example, the recently completed monitoring plans for North American shorebirds and landbirds establish estimating population size as a major objective. Obtaining these estimates rigorously, however, is difficult. Rosenberg and Blancher (this volume) describe one approach that generates point estimates based on several assumptions. Here, I describe an alternate approach which incorporates uncertainty about the assumptions and establishes a range for the true population size. It is illustrated by estimating population size for four shrubsteppe species across a large portion of their range using Breeding Bird Survey (BBS) data. The BBS consists of roadside routes randomly selected within one-degree blocks throughout the United States (except Hawaii) and southern Canada. Each route has 50 stations regularly spaced at 0.5-mile intervals, and is surveyed once during the breeding season (mainly in June) starting 30 minutes before dawn. Observers record all birds detected for 3 minutes at each station. The Survey has been widely hailed as one of the best wildlife monitoring programs in the world (e.g. Ralph et al. 1995). Its results are used each year in dozens of publications for both applied and theoretical purposes (Sauer et al. 1999). An estimate of population size, based on BBS data, may be made by writing down an algebraic expression that describes the relationship between mean birds/ BBS route and total population size, and then estimating the terms in this relationship. This approach is followed below.