Abrupt switch to migratory night flight in a wild migratory songbird

Abstract

Every year, billions of wild diurnal songbirds migrate at night. To do so, they shift their daily rhythm from diurnality to nocturnality. In captivity this is observed as a gradual transition of daytime activity developing into nocturnal activity, but how wild birds prepare their daily rhythms for migration remains largely unknown. Using an automated radio-telemetry system, we compared activity patterns of free-living migrant and resident European blackbirds (Turdus merula) in a partially migratory population during the pre-migratory season. We found that activity patterns between migrant and resident birds did not differ during day and night. Migrants did not change their daily rhythm in a progressive manner as has been observed in captivity, but instead abruptly became active during the night of departure. The rapid shift in rhythmicity might be more common across migratory songbird species, but may not have been observed before in wild animals due to a lack of technology.

Figure 1. Double-plotted actograms (48 hours) showing activity of a ‘‘resident’’ captive, “migrant” captive, free-living resident and free-living migrant European blackbirds (Turdus merula) during autumn.

Captive “resident” (a) and captive “migrant” (b) were exposed to natural photoperiod during the autumn (Sep. 1–Oct. 31). These two actograms were plotted using raw data from Partecke & Gwinner (2007). In the case of the captive “migrant” (b), around mid-September a morning component of activity moves gradually into the night-time, developing nocturnal activity (Zugunruhe). Zugunruhe peaks the night of October 2. Activity of a free-living resident bird (c) was recorded using the automated telemetry system (ARU) continuously from the autumn until the consecutive spring. (d) Activity of a free-living migrant bird was recorded also using the ARU, during the autumn until its departure. The departure time of the free-living migrant is indicated by *. After departure, the ARU showed some false positive inactivity due to noise in the recording. Day and night time activity coloured red and blue, respectively.

Figure 2. Activity pattern of migrant and resident birds seven days before the departure of migrants.

Mean activity % and standard error of migrant (blue) and resident (orange) European blackbirds (Turdus merula) in 30 minute intervals four days prior to departure. White and grey backgrounds represent mean day and night-time respectively. Dashed vertical lines represent the variation in dawn and dusk onset during the time of the study. Sample size: 21 migrant individuals (11 females, 8 males and 2 unknown sex birds; 2 individuals with multiple departure events) and 23 resident individuals (3 females and 20 males).

Figure 3. Results of the Generalized Additive Mixed Model (GAMM) on daily activity seven days before the departure of migrants.

Predicted daily activity pattern by GAMM of migrant (blue) and resident (orange) individuals. Solid lines represent the mean fitted values of the model (smooth term) with 95% confidence interval (shaded area). Dots correspond to the raw observations. White and grey rectangles represent mean day and night time respectively. Dashed vertical lines represent the variation of dawn and dusk onset during the time of the study.