An Observer-Specific, Full-Season, Strip-Map Method for Censusing Songbird Communities

Abstract

Abstract In this paper I describe and demonstrate a strip-map census method designed to provide a complete record of the composition and density of a temperate-zone breeding bird community through a period covering the peaks of breeding activity of all resident species and individuals. A route transecting the area of interest is laid out, flanked on each side by an unmarked strip of a width equal to the distance at which birds can be detected (the detection-threshold distance). All birds detected during uniform listening periods on frequent, closely scheduled, early morning traverses of the route are recorded with appropriate symbols at the moment of detection on simple line maps of the route. Because species differ in the distance at which their calls and songs can be heard and observers differ in their hearing acuity, each observer has an observer-specific strip width for each species. These strip widths are determined directly from the distribution of all detection points on the census maps at the end of the season. Multiplied by the length of the strip, they give the area of the strip for each species. The accumulations of detection-point clusters on summary maps for each species are interpreted as the territories of individual birds. Densities within the study area or segments of it are determined by dividing the number of equivalent territories (Σ wholes plus fractions) within the species-detection strip by the area of the strip. Song frequencies are determined for each individual territory holder by multiplying the proportion of listening periods during which the bird was heard by the proportion of that bird's territory that fell within the species' detectability strip. The method provides a continuous record, divisible into segments of any desired length, of the density, song activity, and distribution of individuals of each species through its breeding season. The detection-threshold distance and song-frequency values obtained can be used as species-specific conversion factors for translating simple detection counts to bird densities.