Traffic Effects on Bird Counts on North American Breeding Bird Survey Routes

Abstract

ABSTRACT. The North American Breeding Bird Survey (BBS) is an annual roadside survey used to estimate population change in >420 species of birds that breed in North America. Roadside sampling has been criticized, in part because traffic noise can interfere with bird counts. Since 1997, data have been collected on the numbers of vehicles that pass during counts at each stop. We assessed the effect of traffic by modeling total vehicles as a covariate of counts in hierarchical Poisson regression models used to estimate population change. We selected species for analysis that represent birds detected at low and high abundance and birds with songs of low and high frequencies. Increases in vehicle counts were associated with decreases in bird counts in most of the species examined. The size and direction of these effects remained relatively constant between two alternative models that we analyzed. Although this analysis indicated only a small effect of incorporating traffic effects when modeling roadside counts of birds, we suggest that continued evaluation of changes in traffic at BBS stops should be a component of future BBS analyses.