Experimental traffic noise attracts birds during the breeding season

Abstract

Understanding how anthropogenic disturbance affects animal behavior is challenging because observational studies often involve co-occurring disturbances (e.g., noise, lighting, and roadways), and laboratory experiments often lack ecological validity. During the 2016 and 2017 avian breeding seasons, we investigated the effects of anthropogenic noise and light on the singing and spatial behavior of wild birds by independently manipulating the presence of each type of disturbance at 89 sites in an otherwise undisturbed boreal forest in Labrador, Canada. Each treatment was surrounded by an eight-channel microphone array that recorded and localized avian vocalizations throughout the manipulation. We analyzed the effects of noise and light on the timing of the first vocalizations of each species at each array during the dawn chorus, and on the proximity of the vocalizing birds to the disturbance when those songs were produced. We analyzed all species combined, and then conducted separate analyses for the six most common species: boreal chickadee, dark-eyed junco, ruby-crowned kinglet, Swainson’s thrush, white-throated sparrow, and yellow-rumped warbler. When all species were analyzed together, we found that traffic noise attracted vocalizing birds. There was some evidence that light repelled birds, but this evidence was inconsistent. In our species-specific analyses, yellow-rumped warbler sang earlier in response to noise; Swainson’s thrush was attracted to noise and the combination of noise and light but repelled by light alone. Our study provides some of the first experimental evidence of the independent and combined effects of traffic noise and light on the vocal and spatial behavior of wild birds and suggests that breeding birds may be attracted to noisy roads where they could be exposed to additional forms of disturbance.

Acoustic localization of terrestrial wildlife: Current practices and future opportunities

“…Six studies calculated animal abundance, either by direct calculation of number of individuals (Frommolt & Tauchert, 2014; Hedley, Huang, & Yao, 2017; Spillmann et al, Willems, van Noordwijk, Setia, & van Schaik, 2017; Wahlberg et al, 2003; Wilson & Bayne, 2018) or indirectly by calibration of acoustic indices, as described by Stevenson et al (2015) (Thompson et al, 2009). Five studies used localization to infer territory boundaries or habitat use, including assessing animals' relationships with anthropogenic or natural habitat features (Ethier & Wilson, 2019; Hennigar, Ethier, & Wilson, 2019; Kershenbaum et al, 2019; Spillmann et al, 2017; Wilson & Bayne, 2018). Three studies separated animal sounds from background noise to improve species classification (Kojima, Sugiyama, Hoshiba, Suzuki, & Nakadai, 2017; Kojima, Sugiyama, Suzuki, Nakadai, & Taylor, 2016; Suzuki, Matsubayashi, Nakadai, & Okuno, 2016).…”

“…In addition to deploying the microphone array, practitioners must measure the local temperature in order to accurately estimate speed of sound. During longer deployments, temperature and other climactic values such as humidity and wind speed may be obtained from a weather logger (e.g., Hennigar et al, 2019; Wahlberg et al, 2003) or nearby public weather stations. Many other preparations are required for setting up ARUs in general, such as selecting a power source and choosing a bit depth for recordings, but we do not focus on these here.…”

“…Ethier and Wilson (2019) tested a similar array of four ARUs positioned at the points of a 40 m × 40 m square, with each ARU containing two microphones separated vertically by 2–3 m, but found that this vertical separation was insufficient for accurate vertical position estimation. Only four studies achieved three‐dimensional positioning of audible sound using hyperbolic localization (Harlow et al, 2013; Hennigar et al, 2019; Spiesberger, 1999; Stepanian et al, 2016).…”

Singing behaviour of ruby‐crowned kinglets Regulus calendula in relation to time‐of‐day, time‐of‐year and social context

“…Each array was set up in the morning or early afternoon (8 am to 2 pm), and then left to record in the absence of human observers until at least 5 pm the following day; therefore, recordings sampled the evening chorus on the day of set‐up plus the dawn chorus the following day. Arrays were left recording for an additional day as part of another experiment (Hennigar et al ). If there was heavy precipitation (> 5 mm) during the first or second day, the array was left recording for an additional day.…”

“…This second step was repeated 100 times, with each iteration using a finer‐resolution lattice and a smaller study area (equal to the spatial resolution of the previous iteration) centred on the estimated origin of the song from the previous iteration. (Wilson et al , Ethier and Wilson , Hennigar et al ).…”