Abstract

In field surveys, environmental variables influence whether an observer is able to re-cord an animal present at a given site. Despite the growing interest in the study of birds in urban ecosystems, the effect of environmental variables on the probability of detection has been scarcely explored. To provide scientific evidence that contributes to an adequate assessment of birds in urban environments, we evaluated the effect of noise, tree cover, and time of the day on the probability of bird detection in the city of Santiago, Chile. At 35 sites, we sampled four points that varied in noise levels. At each point, we recorded birds using distance sampling and measured noise levels as well as habitat attributes. We used the N-mixture model to model detection probability for six bird species: Zenaida auriculata, Elaenia albiceps, Troglodytes aedon, Turdus falcklandii, Zonotrichia capensis, and Passer domesticus. Noise exhibited a negative effect on the probability of detection of E. albiceps and P. domesticus. Tree cover had a negative effect on the probability of detection of T. falcklandii and Z. auriculata. The probability of detection of E. albiceps, P. domesticus and Z. auriculata was higher early in the morning and decreased considerably toward noon. For all six species the probability of detection was very low at distances more than 30 meters from the observer, indicating that corrections for imperfect detection may be necessary when surveys involve greater distances. Our results showed that it is necessary to consider the species-specific effect of environmental variables, which can affect the detection of birds in urban ecosystems.