Summary
Variation in the level of competition for mates and territories is likely to influence the behaviour of competitors. The start of the dawn chorus in singbirds is influenced by a variety of internal factors (e.g. circadian rhythms) and external factors (e.g. light levels, social cues). Here we investigate whether the start time of the dawn chorus is influenced by the singing behaviour of conspecific competitors. Using an Acoustic Location System, we recorded the dawn chorus in neighbourhoods of 5–10 black-capped chickadees, Poecile atricapillus. We used playback to simulate an unfamiliar male performing a dawn song bout within an existing male’s territory. Playback began 15 min before the earliest song sung by any male on the preceding day. Focal males began singing a mean ± SE of 4.3 ± 1.6 min earlier on the day of playback (time relative to sunrise), significantly earlier than on the previous day. We also found a significant communication network level response where neighbouring males began singing 2.3 ± 0.8 min earlier in response to playback. Dawn song bouts of males that received playback were longer, but ended at a similar point relative to sunrise. As this effect of a simulated conspecific on chorus start time is on the scale of only a few minutes, other factors probably play a significant role in shaping the timing of dawn chorus onset. Our results show that animals adjust the timing of their sexual communication in response to increased levels of competition.
Methodology
Study Area, Population and Recording Methods
We studied a banded population of black-capped chickadees at Queen’s University Biology Station (44°34′N, 76°19′W), near Kingston, Ontario, Canada from January to July, 2005–2007. Adult birds were captured in winter using treadle-traps baited with sunflower seeds. Birds were banded with a unique combination of three coloured bands and a numbered aluminium Canadian Wildlife Service band (N = 149 birds in 2005, N = 236 in 2006, N = 61 in 2007).
We recorded the entire dawn chorus of black-capped chickadees using a 16-microphone Acoustic Location System capable of recording all males singing within the neighbourhood surrounding playback and providing position information on each singing bird. For a detailed description of the Acoustic Location System set-up, see Mennill et al. (2006) and Fitzsimmons et al. (2008a). From 30 April to 14 May, 2005–2007, we recorded 15 chickadee neighbourhoods of 5–10 territories. A neighbourhood consisted of a cluster of breeding territories with males that were familiar with each other from their previous winter flocks and familiar with males from nearby flocks. The recording area covered by the Acoustic Location System was approximately 160 000 m2. The recording period spanned the female fertile period, up to and including clutch completion of most pairs. In each neighbourhood, we recorded for 2–3 days. The 2005 recordings were used as baseline recordings, and the playback experiment was conducted in 2006–2007.
Playback Procedure
In 2006 and 2007, we used playback to simulate a singing male black-capped chickadee attempting to insert himself into an existing territory during the dawn chorus. We intended for this simulated insertion to be perceived as a strange floater male singing a dawn chorus much earlier than the focal male within the focal male’s established territory boundaries. In our long-term study of this population, we have detected floater males in our population each year, particularly when birds have lost their breeding partner to predation. Therefore, we consider this simulation to be a plausible one. In each neighbourhood, we recorded an unmanipulated dawn chorus on day 1 (also referred to as ‘preplayback’). We then conducted two independent playback trials on days 2 and 3 at opposite ends of the recorded neighbourhood. On day 2, we presented playback to one male on one edge of the recorded neighbourhood, and on day 3, we presented playback to a different male on the opposite edge of the recorded neighbourhood. These two males were not themselves neighbours, did not share neighbours, and could not hear the playback at the opposite end of the recording area on the day they did not receive playback (N = 20 playbacks). Given the size of the neighbourhoods that we recorded with the Acoustic Location System, playbacks performed at opposite sides of the Acoustic Location System would have been heard by different groups of territorial males. The two playbacks (day 2 and 3) in each neighbourhood were considered independent, and presentation of two playbacks per neighbourhood increased the number of playback trials that could be performed each year. When possible, we selected males with an established nest location such that they were likely to sing in the same spot on the playback day as on the control day. Males were thus chosen as playback subjects if they held a territory at the edge of the array, and, when possible, if their female mate had started lining the nest cavity in anticipation of egg laying (Smith 1991). The playback speaker (Sony SRS-77 G) was placed within the subject’s territory at a position near to where we had observed him singing at dawn on day 1.