Timing of departures in nocturnally migrating songbirds
In a new paper entitled "Ecological factors influence timing of departures in nocturnally migrating songbirds at Falsterbo, Sweden" in Animal Behaviour, we (myself, Thomas Alerstam, Susanne Åkesson and Rachel Muheim) investigated the departure timing of four species of nocturnally migrating songbirds using the automated radiotelemetry system in Falsterbo. We found that the variation in departure time in relation to sunset was large, indicating that departure timing is affected by several factors and not only a behaviour triggered by the setting sun. The birds departed sooner after sunset during spring than autumn, and different species departed at different times in relation to sunset. In addition, birds departed earlier when nights were shorter, suggesting that night duration is an important factor that may drive much of the observed timing differences between seasons and species. Lean birds delayed their departures compared to fat individuals. When birds experienced favourable wind conditions (tailwind or weak winds) at sunset, they departed earlier. Thus, it appears that the decision to take off for a long-distance flight is adapted both with regard to body condition and wind conditions. Timing of departure was not correlated with sun elevation, which would have been expected if availability of specific orientation cues (sun, skylight polarization pattern, stars) acted as triggers for departures.
To the paper!
Time of departures of nocturnal passerine migrants (four species as indicated by different colour codes given in Fig. 1) at Falsterbo, Sweden, as recorded by automated radio telemetry. Circular diagrams a–h show departure timing for each species and season during the 24 h of day/night with 0000 CET at the top and 1200 CET at the bottom of each circle. Each symbol outside the circle refers to the departure time of one individual bird (black = records included in analyses of timing in relation to sunset; grey = records excluded from those analyses, cf. text). Mean vectors (arrows) with length proportional to the mean vector length (cf. Batschelet, 1981) show the mean timing with 95% angular confidence intervals indicated by the dotted lines (cf. Table A1). Shaded sectors refer to the night (between mean times of sunset and sunrise) with the range of variation in sunset and sunrise times indicated by hatched sectors. Histograms i and j show the distribution of departure times in relation to sunset during autumn and spring, respectively. Mean departure time (±SE) in relation to sunset and season is plotted relative to average night duration for each species (k). Mean sun elevation at departure (±SE, degrees below horizon) is plotted relative to night duration (l), grey areas indicate nautical twilight (light grey; ends at sun elevation –12°) and astronomical twilight (dark grey; ends at sun elevation –18°). Data in i–l are restricted to the nocturnal and near-nocturnal period 1700–0500 CET.