Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Annual spatiotemporal migration schedules in three larger insectivorous birds: European nightjar, common swift and common cuckoo

Annual spatiotemporal migration schedules in three larger insectivorous birds: European nightjar,... Background: Knowledge of spatiotemporal migration patterns is important for our understanding of migration ecology and ultimately conservation of migratory species. We studied the annual migration schedules of European nightjar, a large nocturnal insectivore and compared it with two other larger migratory insectivores, common swift and common cuckoo. All species breed in North Europe and winter in sub-Saharan Africa, but estimating their spa- tiotemporal non-breeding distributions from observations is complicated by the occurrence of similar local African species. We used geolocators to track the annual migrations of nightjars and swifts and compared these with satellite tracking of cuckoo migration. Results: Individuals of the three species migrated to wintering grounds centered in Central Africa, except some com- mon swifts that remained in West Africa, crossing or circumventing the Sahara along different routes in spring and fall. Overall, all species showed similar regional and seasonal use of several stopover areas during migration. Among the three species, European nightjars and common cuckoos showed the most similar spatiotemporal migration patterns. The nightjars wintered in SW Central Africa and breeding and wintering made up by far the two longest stationary periods. Swifts were generally more mobile, and some individuals progressively visited areas further east in East Africa during winter and further west in West Africa on spring migration; this species also spent less time on stopovers, but more on wintering areas. Cuckoos were intermediate in their extent of movements. The speed of nightjar spring migration was equal to that of fall migration, in contrast to the two other species where spring return to breeding areas was faster. Conclusions: Ecological requirements are potentially useful for understanding spatiotemporal migration patterns and causes of declines in migratory species. Keywords: Long-distance migration, Migration speed and timing, Large insectivores, European nightjar, Geolocators Background evolution of species-specific migration schedules is sur - Basic knowledge of the ecological requirements of prisingly poor [1]. Evaluating the importance of differ - migratory species throughout their annual cycle and the ent parts of the annual cycle is complicated by the fact that most long-distance migrants are small birds, and thus, their annual migrations to date have been difficult *Correspondence: kthorup@snm.ku.dk to study. However, using satellite telemetry, it is now pos- Deceased Center for Macroecology, Evolution and Climate, Natural History sible to track birds down to the size of cuckoos Cuculus Museum of Denmark, University of Copenhagen, Universitetsparken 15, and small archival light recorders (hereafter referred to 2100 Copenhagen, Denmark Full list of author information is available at the end of the article © The Author(s) 2017. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 3 of 11 as geolocators) enable mapping migration of even small nightjar Caprimulgus europaeus, common cuckoo Cucu- songbirds based on light-level geolocation [2, 3]. lus canorus and common swift Apus apus. Palearctic–African bird migrants presumably track sea- European nightjars and common cuckoos are large sonal resource availability on a continental scale [4, 5], (both have wing lengths around 0.6 m and weigh around but it is less clear what factors shape variation in species- 80 and 110–130  g, respectively) insectivorous species specific and individual migration patterns. Large varia - [20, 21]. The cuckoos feed mainly on caterpillar larvae, tion among species in spatiotemporal schedules is well whereas nightjars catch low flying insects, often caught documented, but poorly understood, with very little the- by hawking from the ground. Swifts are slightly smaller oretical work developed to explain such differences. In (wing 0.45 m and weight 44 g), strictly aerial insectivores general, seasonal species distributions have been mapped foraging exclusively on the wing [22] presumably ena- using observations and ringing [6] supplemented in some bling foraging during migration. In contrast to swifts, cases with estimated distributions based on niche mod- both nightjars and cuckoos forage on much larger insect eling [7]. species [20, 21]. Furthermore, the common cuckoo is an Differences in fundamental niches alone are insuffi - obligate brood parasite [21], so their migration schedules cient to explain, for instance, the striking differences, in are not constrained by parental care. some cases, between the wintering grounds and migra- We used geolocators to map the spatiotemporal migra- tion routes of closely related species, such as found in tion schedule of the European nightjars and compared Ficedula flycatchers [8, 9] and Luscinia nightingales [10, it with that of common cuckoos Cuculus canorus ([23], 11]. Such patterns could be caused by different popula - same data set as included here) and common swifts Apus tion histories (potentially explaining why birds circum- apus ([22], using other data set). Based on the tracking vent the Mediterranean via the eastern or the western data, we describe similarities and differences among the flyway) or competitive exclusion. Detoured routes as seen three species in spatial patterns, length of sedentary stay in red-backed shrikes Lanius collurio breeding in Spain periods, duration of fall and spring migration episodes that migrate to East Africa via the Balkans could be con- as well as speed of migration. We then discuss whether strained by colonization routes [12] but may be optimal these potentially arise due to species-specific characteris - for other reasons such as wind assistance [13, 14]. tics. To compare migration patterns, we focus on migra- Several theoretical studies have dealt with the temporal tion strategies for migrants that spend the winter in the component of migration schedules [15]. Annual sched- same general area. Overall, nightjars, cuckoos and swifts ules need to accommodate key elements of the annual wintered in Central Africa, but some swifts stopped cycle, most importantly breeding, molt and migration migration early and stayed in West Africa for the winter [16]. Optimal scheduling depends on a number of fac- and these were excluded from comparisons. Obviously, tors, with each activity influencing the timing of the the three species studied represent only a small subsam- others [15]. During migration, individuals are, from ple of the Afro-Palearctic migrants, and thus, our results an optimality perspective, expected to minimize time, cannot necessarily be generalized to other migratory energy and/or predation [17]. Optimal migration theory species. commonly assumes that time minimization is important for migrants and this has also been reported empirically Methods [18]. As a consequence, migrants would be expected to Using playback and mist nets, we trapped 10 male Euro- stopover slightly longer and undertake longer migration pean nightjars in June and July 2010 and 14 males in June legs (hence making fewer stops) than expected from an and July 2011 in Northern Jutland, Denmark (57.06°N, energy minimization strategy where the costs of carry- 9.13°E). Birds were ringed with a metal ring and fitted ing extra fuel offset the gain of traveling faster. In general, with geolocators (Mk 10_S; 1.1  g; British Antarctic Sur- smaller species are expected to migrate faster by flap - vey, BAS [24]). The light sensor was placed on a 0.8  cm ping flight than larger species [19] and species with high stalk, in order to raise the sensor above the plumage. We settling costs and low fueling efficiency are expected to attached the loggers as a backpack using a full-body loop undertake longer stopovers. harness (comparable to the wing harness in [25]) made Species-specific biology has rarely been considered from a 2-mm-wide braided nylon string. Females were explicitly when explaining migration schedules, although released without geolocators because in previous years these are likely to be important. Our objective here is to we recaptured a lower proportion of females than males. describe the migration schedules of ecologically simi- Geolocators from seven male nightjars were retrieved by lar species and to discuss whether ecological similarities July 2013 and data were available from six (2010–2011: might lead to similarity in migration patterns. We focus 4; 2011–2012: 2). In three devices, logging had stopped on three species of larger insectivorous birds, European before the bird returned to the breeding site (on 28 Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 4 of 11 March, 18 April and 7 May). An extra effort was devoted that corresponds to a local minimum in latitude variation to catch birds from territories that were provided with during a stopover. We tested angles between 0° and −8° geolocators, presumably resulting in a slightly higher and found no minimum in four birds; in the two birds recapture rate (29%) than the rate of 20% of ringed-only where we found a minimum in latitude variation, the male nightjars (own data). estimated sun elevation angle was within 0.5° of the one A total of 25 common swifts were trapped during obtained using breeding latitude calibration. breeding at two Danish locations. At Bjødstrup, eastern The cuckoo data are from eight published satellite Jutland (56.29°N, 10.53°E), 10 and 7 individuals were fit - tracked birds with completed fall migration in 2010 of four ted with geolocators in 2010 and 2011, respectively, of males and two females, and three males and one female in which 6 individuals were re-trapped (2010–2011: 3; spring 2011 ([23]; Fig.  2; Additional file  2). The common 2011–2012: 3). At Nyborg, Fyn (55.30°N, 10.82°E), 8 indi- swift data from light-level-based geolocators were ana- viduals were fitted with geolocators in 2010 of which 3 lyzed using a threshold of 2, a sun angle of −5° and exclud- returned in 2011 and 1 returned in 2012. The recap - ing three weeks around the equinox. To calibrate the sun ture rate (40%), was below the rate reported in a similar elevation angle we used Hill–Ekström calibration [27] dur- study (75% [22]). This is potentially due to birds switch - ing the longest non-breeding stopover. All resulting sun ing between local nest sites from year to year (as we have elevation angles were between −4.5° and −5.5°, and we observed in previous cases but no quantitative data are used −5° for all individuals (Additional file 2 ). available), and we could not obtain permissions to recap- ture birds located at nearby nest sites located in adjacent Breeding, stopover and wintering periods: location, timing private houses. We used geolocators (Mk 20, 0.6  g from and distances BAS) fitted on the birds using a body loop harness made We used departure in one year and return in the follow- from 1-mm braided nylon string. Birds were trapped ing year to calculate duration of breeding in all three between sunset and full darkness in mist nets close to the species. nesting site or in the nest box. For geolocator data (nightjars and swifts), we defined The Copenhagen Bird Ringing Centre with permission a stopover as when the birds interrupted their migration from the Danish Nature Agency approved capturing and for more than 5 days and calculated an estimated stopo- tagging of nightjars and swifts. ver position by averaging the longitudes and latitudes during this period. We estimated stopover and move- Location estimation ment periods based on changes in latitude and longitude Data for nightjars were downloaded and analyzed using except during equinox, when only longitude was con- the BASTrak software suite [26]. We defined sunrises and sidered. Dictated by the uncertainty of light-level data sunsets using the threshold method with a sun angle esti- [27], we required consecutive stopovers to be separated mated for each individual. There was an unusually high by more than four degrees either latitude or longitude; level of noise in the light measurements on many days, otherwise, the two locations were considered part of the possibly caused by shading from either the vegetation same stopover and the location was acquired as the aver- that the birds hide in during day or from the feathers on age throughout both stops. Departure and arrival dates the back of the birds. To account for this noise, we only were taken as the last and first, respectively, days that used data from days with little noise (measured as stand- were within two degrees longitude of the average location ard deviation in the light measurements between sunrise estimate (Additional file 3). and sunset). To decide the acceptable level of noise, we In the nightjar, some of the fall and spring stopovers looked for the level under which the estimated latitudes occurred during equinox, making latitude estimation were stable during breeding (Additional file  1: Fig. S1). impossible. To estimate these stopovers for illustration Because we filtered out the days with most noise, we (Fig.  1), we calculated theoretically realistic latitudi- only needed to discard all latitudes within two weeks nal spans based on travel speeds. These were calculated on each side of the equinoxes. The remaining data were using the preceding or following stopover (if the preced- calibrated to the breeding latitude during their breed- ing was not available), travel speeds of 200–500  km/day ing site attendance and the derived sun angles (−2.2° to [3, 23] and the longitude estimated from the light data −5.0°) were used to estimate positions (Additional file  2). during the stopover. We calculated the overall migrat- We compared this to using Hill–Ekström calibration ing distance as the minimum distance between breeding [27] during the longest non-breeding stopover, but pre- site, stopover sites and winter area. For this calculation, sumably because of the amount of noise in the data this we also included estimated stopovers during equinox was only possible for two birds. Hill–Ekström calibra- which provided data on individual movement, despite tion requires that there is a specific sun elevation angle being disregarded because of their positional uncertainty. Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 5 of 11 Fig. 1 Main stopovers used by six European nightjars during migration. Colored circles are light-level-based geolocation stopover positions esti- mated using breeding site calibration. Positions are means of high-quality latitude and longitude estimates (see “Methods” section) during the full length of the stopovers. For stopovers latitudes during Equinox, intervals assuming realistic migration speeds (dotted lines correspond to speeds of 200–500 km/day; squares correspond to 350 km/day) rather than positions are indicated Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 6 of 11 Migration speed was calculated as migration segment in stopover use has also been reported in other long- distance divided by duration of the migration segment. distance migrants, for example red-backed shrikes [10] All durations of segments include days stopping over. although their timing appears more flexible. Some might only be a few days (shorter than what we The three species showed overall similarities in the consider a stopover), making some migration speeds routes and timing taken: all undertook stopovers in the close to travel speeds. Sahel after leaving Europe, subsequently spending the For satellite tracking data (cuckoos), we defined stopo - major part of the non-breeding season, i.e., wintering, in vers, dates and distances overall similarly. However, the Central Africa and at stopover sites in West Africa before higher accuracy of the satellite tracking data theoretically returning to the breeding sites. We therefore compared the allows for separation of small-scale changes between timing of departure and arrival in these four places: breed- sites. However, we grouped cuckoo stopovers in spe- ing, Sahel (8–20°N), wintering (south of 5°N and east of cific geographical regions at a spatial scale similar to that 10°E) (for swifts this could include trips to eastern Africa) which we obtained for the other two species. Individual and West Africa (west of 10°E and 0–20°N) in spring. We stopover locations were estimated as averages within also compared durations of stopovers as well as migration these regions as described in [22]. Because of non-daily speed. Roaming behavior during the non-breeding sea- transmissions, departure and arrival dates are rarely son was quantified as the number of stationary periods as identifiable and we used last and first, respectively, days well as the distance traveled among stationary sites dur- within the region/site in question (Additional file 3). ing winter. Because we aim to describe differences in spa - Geolocators in most cases allow for accurate determi- tiotemporal migration patterns in species with otherwise nation of stopover timing to within 1 day. Because of the similar constraints, we focused our analyses of the win- 10  h on 48  h off duty cycle of the satellite transmitters, tering period to birds wintering in the same general area, estimation of departures from stopovers was less accu- thus using only the seven swifts traveling to Central Africa rate for satellite data and transmissions were occasion- (three remained in West Africa all winter). Clearly, the ally missed, resulting in even less accurate estimates. We short-stopping swifts saved time and potentially energy by based stopover duration only on known locations result- using a shorter-distance migration strategy that potentially ing in potential underestimation of stopover duration influences their ability to fatten up before travel and speed (and overestimation of travel time). Missing transmis- of migration depending on local seasonal conditions. sions only occurred relatively frequently in Europe dur- We compare the three species fitting separate ANO - ing fall, potentially leading to underestimates of stopover VAs for each of the parameters above as a function of duration of cuckoos here. However, we found that cuck- species. Because of the large numbers of comparisons, oos staged for longer periods than the other species in Bonferroni-corrected significance levels are also indi - this region anyway. Therefore, we conclude that our com - cated. Overall, results should be interpreted with cau- parisons of stopover duration are likely conservative. tion due to generally low sample sizes. Because of varying sample sizes, tests were not directly comparable and in Comparison of European nightjar with common swift general, we only present comparisons from significant and common cuckoo tests. All models were fitted using general linear model We compare the migration routes and the timing of the (GLM) in R 3.1.0 [28]. nightjars to those of common cuckoos and common swifts. The study of migration in all three species is chal - Results lenging given their relatively small size, long-distance Spatial patterns migration and difficult study conditions in the winter - The six tracked nightjars performed a clockwise loop ing area (ranging from being nocturnal to highly aerial migration from breeding grounds in Denmark to win- often in highly remote areas). To make the data compa- ter grounds in southwest Central Africa traveling a rable among species, we included only adult birds (but mean total distance of 16,636  ±  1514  km (mean  ±  SD; were not able to separate sexes in analyses). Thus, our range 13,993–18,200  km) with no significant difference sample sizes are necessarily small. The different species between fall and spring distances (Matched pair test, were mostly tracked in the same year but some night- p  =  0.34, n  =  6) (Fig.  1; Table  1). The nightjars stopped jars and swifts were from a year later than the cuckoos. over two or three times in fall and three times in spring As birds’ schedules could vary as a response to the vari- (Fig.  1). The main wintering area was in Democratic able conditions among years, comparing schedules from Republic of Congo, Angola and northern Namibia/ the same year would have been preferable. At least at the Botswana. regional scale that we are considering, we do not expect Among the 10 tracked common swifts, most individu- any major differences among years. Similar consistency als (n  =  8) also migrated to the Sahel before continuing Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 7 of 11 Table 1 Mean, standard deviation (SD) and  number of  individuals [n] in  each analysis of  timing, duration, distance and speed for different legs of the migration for each of the three different species Parameter Nightjar (SD) [n] Cuckoo (SD) [n] Swift (SD) [n] P model P Ni_Cu P Ni_Sw P Cu_Sw Timing (day of year one) Dep. breeding 250 (7) [6] 191 (13) [8] 222 (8) [9] 0.000 0.000 0.000 0.000 Arr. Sahel 280 (5) [6] 263 (15) [6] 244 (11) [8] 0.000 0.018 0.000 0.006 Dep. Sahel 294 (3) [6] 301 (12) [6] 280 (15) [8] 0.014 0.371 0.040 0.005 Arr. C. Africa 301 (3) [6] 307 (14) [6] 288 (16) [7] 0.044 0.451 0.080 0.017 Dep. C. Africa 449 (15) [6] 450 (21) [5] 471 (11) [7] 0.036 0.897 0.021 0.036 Arr. W. Africa 455 (16) [6] 458 (18) [5] 481 (11) [7] 0.015 0.753 0.008 0.021 Dep. W. Africa 475 (10) [4] 487 (12) [4] 495 (4) [10] 0.001 0.035 0.000 0.083 Arr. breeding 493 (12) [3] 505 (12) [3] 506 (6) [10] 0.104 0.120 0.038 0.812 Duration (days) Breeding Sahel 30 (10) [6] 57 (15) [3] 27 (10) [7] 0.004 0.004 0.649 0.002 Sahel–C. Africa 7 (2) [6] 6 (4) [5] 8 (5) [6] 0.875 0.755 0.837 0.612 Fall 52 (9) [6] 113 (31) [5] 69 (21) [8] 0.001 0.000 0.161 0.002 Fall stops 31 (8) [6] 88 (30) [5] 43 (20) [8] 0.001 0.000 0.323 0.001 C. Africa–W. Africa 6 (2) [6] 8 (4) [5] 9 (6) [7] 0.463 0.549 0.222 0.573 W. Africa breeding 23 (6) [3] 18 (2) [3] 11 (5) [10] 0.005 0.270 0.003 0.038 Spring 55 (13) [3] 65 (7) [3] 34 (9) [7] 0.001 0.206 0.008 0.001 Spring stops 40 (11) [3] 48 (7) [3] 20 (8) [6] 0.002 0.255 0.009 0.001 Breeding 127 (15) [3] 49 (3) [3] 82 (11) [9] 0.000 0.000 0.000 0.001 C. Africa 148 (15) [6] 139 (23) [5] 183 (13) [7] 0.001 0.422 0.002 0.000 Migration 100 (16) [3] 185 (25) [3] 105 (19) [5] 0.001 0.001 0.758 0.001 Non-migration 265 (16) [3] 180 (25) [3] 260 (19) [5] 0.001 0.001 0.758 0.001 Distance (km) Breeding Sahel 4656 (925) [6] 5048 (89) [6] 6693 (675) [5] 0.000 0.323 0.000 0.001 Sahel–C. Africa 3560 (479) [6] 1822 (503) [6] 2310 (516) [5] 0.000 0.000 0.001 0.128 Fall 8215 (760) [6] 6870 (528) [6] 7954 (1379) [10] 0.085 0.041 0.639 0.063 C. Africa–W. Africa 1658 (906) [5] 1986 (572) [5] 2912 (963) [7] 0.056 0.553 0.025 0.084 W. Africa breeding 5555 (815) [5] 5691 (146) [3] 5906 (291) [10] 0.413 0.705 0.203 0.507 Spring 7180 (1015) [6] 7898 (376) [3] 7944 (1729) [10] 0.572 0.485 0.313 0.961 Speed (km/day) Breeding Sahel 178 (92) [6] 92 (24) [3] 316 (76) [4] 0.010 0.153 0.021 0.004 Sahel–C. Africa 577 (258) [6] 371 (120) [5] 553 (370) [4] 0.406 0.212 0.885 0.316 Fall 162 (30) [6] 65 (17) [5] 161 (103) [9] 0.065 0.044 0.981 0.032 C. Africa–W. Africa 281 (177) [5] 363 (328) [3] 369 (163) [7] 0.750 0.596 0.477 0.964 W. Africa breeding 236 (63) [3] 313 (29) [3] 666 (305) [10] 0.033 0.720 0.024 0.055 Spring 131 (52) [3] 122 (17) [3] 331 (131) [10] 0.012 0.924 0.017 0.014 P model is the p value of a general linear model with species as predictor, and the other three p values are testing pairwise differences between species (Ni nightjar, Cu cuckoo, Sw swift). p values <0.05 are in italics. p values <0.0016 (applying a Bonferroni correction) are in italic underline to Central Africa, but the swifts migrated via the Iberian in the Central African wintering range. The eight tracked Peninsula, followed the northwest African coast and then common cuckoos (Fig.  2) showed a clockwise loop migra- turned east toward the central Sahel (Fig.  2). Three swifts tion to the eastern Sahel, Central Africa, West Africa and spent the entire winter in West Africa, whereas the remain- back to Europe through Tunisia. In spring, all three spe- ing seven wintered in Central Africa. While in Central cies travelled to West Africa before going back to Europe, Africa, the birds used on average four different locations but swifts travelled further west than the other two, and (±0.9 SD) and four of the birds even visited East and South- cuckoos further west than nightjars (mean longitude swift: east Africa. In contrast, the cuckoos and nightjars only −9.3° ± 1.7°, n = 10; cuckoo: −2.0° ± 5.3°, n = 5; nightjar: used two different locations (± 0.5 and ±1, respectively) 7.6° ± 4.8°, n = 5; ANOVA: F  = 8.6, p = 0.002). 2,17 Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 8 of 11 Fig. 2 Comparison of migration routes between European nightjars, common cuckoos and common swifts. Circles indicate stopover areas used for at least 5 days. Species are shown in different colors (green = cuckoo, brown = nightjar, blue = swift). Nightjars and swifts were tracked with light- level-based geolocators with considerable uncertainty, especially for latitudes; cuckoos were tracked with high accuracy satellite transmitters [22]. Several of the nightjar locations included here (for illustration purposes) are extrapolated from known longitude and latitude based on an estimated flying speed of 350 km/day (see Fig. 1 for full latitude span). a Complete annual migration. Lines connect stopover locations and do not necessarily reflect route taken (for swift in fall, the easternmost point of the Sahara crossing is included for illustration of the Zugknick here) with darker shading indicating fall and lighter spring migration, b locations month by month. Background layer is vegetation greenness in each 2° latitude × 2° longi- tude cells averaged for each period 2000–2010 derived from on NOAA-16 Normalized Difference Vegetation Index (NDVI), AVHRR NDVI3 g.v0 [29] Overall, the spatial migration pattern of cuckoos was similar time to nightjars, but swifts stayed for substan- very similar to that of the nightjars, both showing clock- tially longer (183 ±  13  days, n = 7). Furthermore, swifts wise loop migration, differing from that of swifts (Fig. 2). traveled further (4180 ± 2059 km, n = 7) within the win- ter region than nightjars and cuckoos (1280 ±  1210  km, Sedentary stays n  =  6; 1086  ±  667  km, n  =  5, respectively). The night - The timing of nightjars (Additional file  1: Tables S1, jars spent on average 127 ± 15 days on the breeding site S2) was roughly similar to the timing of both swifts (using departure and arrivals from different years, n  = 3). and cuckoos, although most similar to that of cuck- Time spent on the breeding site differed among species oos (Fig.  3; Table  1). Nightjar southbound migration (p  <  0.001; Table  1), with swifts (82  ±  11  days, n  =  9) from the breeding grounds started in September (mean spending less time than nightjars; cuckoos (49 ±  3  days, date 6 September, range 30 August–17 September) with n = 3) spent least time of all species. arrival to the wintering area in October/November Departure from the breeding grounds and arrival to (mean date 28 October, range 25 October–3 November). the Sahel differed among the three species (p  <  0.001). They stayed in Central Africa on average 148  ±  15  days Cuckoos were the first to depart breeding grounds (11 (range 135–174, n  =  6; Fig.  1) and northbound migra- Jul ± 13 days, n = 8) followed by swifts (10 Aug ± 8 days, tion started in March/April (mean 25 March, range n  =  9) then nightjars (6 Sep  ±  7  days, n  =  6). Swifts 13 March–17 April). The three species differed in the arrived to the Sahel first (1 Sep  ±  11  days, n  =  8) fol- number of days they spent in Central Africa (p  <  0.001; lowed by cuckoos (20 Sep  ±  15  days, n  =  6) and night- Table  1), where cuckoos (139  ±  23  days, n  =  5) spent jars (7 Oct ± 5 days, n = 6). Departure from West Africa Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 9 of 11 126–214, n = 6) and in spring 131 ± 52 km/day (84–186, n  =  3). Fall migration speed was lowest in cuckoos but did not differ significantly between the three species. Spring migration speed did, however (p = 0.012; Table 1), with swifts (331  ±  131  km/day, n  =  10) migrating faster than both cuckoos (122  ±  17  km/day, n  =  3) and nightjars. Discussion Overall, the three species showed similar migration pat- terns on a regional and monthly scale, in particular com- pared to migration patterns reported for other species [30]. Although the tracked nightjars bred in a relatively restricted area in Denmark (within 65  km), they dis- persed over a much larger area in Africa and spent the winter outside the former known winter areas in Africa Fig. 3 Comparison of timing of departures and arrivals from different which do not include the central parts of the continent geographical regions shared between the three species. Error bars [20, 31] although three European nightjars from England indicate standard deviations. For swifts, timing analyses are based and several from Sweden spent the winter in the same only on individuals reaching Central Africa. Significance tests for each area as the birds from Denmark [32, 33]. Our tracks of state are given in Table 1 swifts were similar to those already published from other North European populations [22]. The similarity in migration patterns suggests that for differed (p  =  0.001) with nightjars (19 Apr  ±  10  days, insectivorous birds of such body mass, the beneficial n = 4) departing earlier than cuckoos (30 Apr ± 12 days, exploitation of the continent-wide seasonal changes in food n = 4) and swifts (10 May ± 4 days, n = 10). supply leaves limited room for spatiotemporal flexibility. We found the migration pattern of European nightjars to Duration and speed of fall and spring migration be more similar to that of common cuckoos than that of The nightjars spent 31  ±  8  days (range 21–44  days, common swifts both with regard to route and with regard n = 6) in total on stopover sites in fall and 40 ± 11 (range to timing though common cuckoos departed breeding 30–51 days, n = 3) in spring. Number of stopover days in grounds earlier. Potentially, this could be related to similar- fall differed between the three species (p   =  0.001; Fig.  3; ity in ecological niche, such as insect prey size. The main Table  1), where nightjars spent substantially less time difference between cuckoos and nightjars was that nightjars on fall stopovers than did cuckoos (88  ±  30  days, range spent more time on the breeding grounds, because cuckoos 53–115 days, n = 5) but a similar amount of time to swifts departed breeding areas earlier possibly because they pro- (43  ±  20  days, range 10–67  days, n  =  8). Spring stopo- vide no parental care. As a result, there were differences in ver duration also differed between species (p   =  0.002) timing before, after and at the breeding site. with nightjars being comparable to cuckoos (48 ± 7 days, The swifts differ from the other two species in showing range 41–55 days, n = 3), and both nightjars and cuckoos large spatial variation among individuals, with some indi- spent substantially more time than swifts (20  ±  8  days, viduals wintering in West Africa, while others wintered in range 7–28  days, n  =  6). The pattern was the same for Central Africa. Seven of ten common swifts covered large total duration of fall and spring migration. Nightjars spent distances spending their “sedentary” wintering period a similar (ANOVA: F  = 0.19, p = 0.67) amount of time roaming over large areas using more sites than the other 1,7 on spring (55 ± 13 days, range 45–70 days, n = 3) and fall species which could result from the fact that swifts forage migration (52 ± 9 days, range 41–65 days, n = 6), whereas on large numbers of small insect prey, often at high alti- in both cuckoos (spring: 65  ±  7  days, range 61–73  days, tudes and quickly traveling large distances in response to n = 3; fall: 113 ± 31 days, range 67–137 days, n = 5) and food availability. These swifts differed temporally from the swifts (spring: 34 ± 9 days, range 19–47 days, n = 7; fall: other species by spending more time in Central Africa, 69  ±  21  days, range 32–94  days, n  =  8) fall migration is with a shorter stopover in West Africa, which resulted in a lasting significantly longer than spring migration (cuckoo: faster overall spring migration in these individuals (spring F  = 6.6, p = 0.04; swift: F  = 16.8, p = 0.001). migration from West Africa was equally fast among the 1,6 1,13 The mean total migration speed of nightjars (includ - swifts that wintered in there). However, fall migration ing stopover sites) in fall was 162  ±  30  km/day (range speeds were similar in the three species. Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 10 of 11 In addition to resource needs, many other factors could Conclusions be causing variation in annual migration schedules. For The spatial migration pattern of the European nightjar example, because avian annual cycles are evolved to was more similar to that of the common cuckoo than that minimize the overlap of energetically demanding events, of the common swift. Our study illustrates the potential birds must fit molt and breeding into the annual sched - for investigating determinants of migratory schedules ule at other times than during migration [34, 35]. As the using a comparative approach. resource needs also dictate timing of molt, we assume Additional files that general migration patterns are similarly affected by periods of resource needs and expenditure. Additional file 1. Supplementary information on European nightjars. The fact that nightjars both forage and travel at night Data on individual birds ( Table S1), stopover duration ( Table S2) and theoretically restricts their migration speeds compared definition of high-quality positions (Fig. S1). to the other species that can forage by day and travel at Additional file 2. Location estimates for individual European nightjars, night (as well as by day). However, we did not find night - common swifts and common cuckoos included in the study. jar migration speeds to be generally slower or faster. In Additional file 3. Estimated stopovers and timing of European nightjars, most avian species studied so far, spring migration occurs common swifts and common cuckoos. faster than fall migration, possibly because of constraints in the arrival time at the breeding area [36]. Norevik et al. Authors’ contributions [33] reported faster fall than spring migration in Euro- LBJ, NOJ, LH, MD, ADF, APT, KT carried out field work on nightjars and swifts. pean nightjars. We found a similar, but non-significant LBJ, NOJ, MW, APT, KT analyzed the data. LBJ, NOJ, MW, APT, KT drafted the manuscript. LBJ, MW, MD, ADF, APT and KT contributed to the final manu- difference, which contrasted the pattern in swifts and script. LBJ, MW, MD, ADF, APT, KT read and approved the final manuscript. cuckoos, where spring migration is faster than fall migra- tion [22, 23]. Compared to the other two species, the Author details Center for Macroecology, Evolution and Climate, Natural History Museum nightjar departed the breeding site substantially later yet of Denmark, University of Copenhagen, Universitetsparken 15, 2100 Copen- arrived in Central Africa at the same time. Nightjars were hagen, Denmark. BirdLife Denmark, Vesterbrogade 140, 1620 Copenhagen the earliest to arrive on the breeding grounds. V, Denmark. Department of Bioscience, Aarhus University, Kalø, Grenåvej 14, 8410 Rønde, Denmark. The migration of the European nightjars followed dif - ferent routes in fall and spring, with a more westerly Acknowledgements spring migration trajectory. Taking a westerly route in We thank Roine Strandberg, Raymond H.G. Klaassen and Thomas Alerstam for permission to use the Swedish cuckoo data. spring may be advantageous because of stable tailwind patterns in West Sahara [37]. Additionally, they can ben- Competing interests efit from foraging in areas where food becomes avail - The authors declare that they have no competing interests. able, following the fall seasonal patterns of rain along Availability of data and material these routes toward the winter areas [14]. Both swifts The data set(s) supporting the conclusions of this article is(are) included and cuckoos also traveled to West Africa before head- within the article (and its additional file(s)). ing north, suggesting that ecological conditions favor this Ethics approval detour. The three species differed substantially, however, This study was carried out in strict accordance with guidelines to the use in the location of their stopover longitude. of wild birds in research of the Ornithological Council [44]. Animal work on European nightjars and common swifts was approved by the Copenhagen Widespread declines have been reported recently in Bird Ringing Centre (J.Nr. SN 302-009) under permission from the Danish For- sub-Saharan migrant species but the causes are to a large est and Nature Agency. degree unknown [38–41]. Migrant species are likely Funding vulnerable to habitat loss because they depend on the KT thanks the Danish Council for Independent Research for support to condition of networks of sites that may be separated by the MATCH Project (1323-00048B). KT, MW, APT thank the Danish National thousands of kilometers [42]. As migratory animals show Research Foundation for support to the Center for Macroecology, Evolution and Climate (DNRF96). declines around the world [43] there is a growing need to understand potential reliance of migrants on staging Received: 2 September 2016 Accepted: 5 January 2017 regions or areas and for conservation initiatives to pro- tect such networks [42]. An essential starting point is to understand the spatiotemporal migration patterns within and across migratory species. The comparative long References duration of stay on the breeding and wintering grounds 1. Bowlin MS, Bisson IA, Shamoun-Baranes J, Reichard JD, Sapir N, Marra PP, obviously direct conservation research toward these two Kunz TH, Wilcove DS, Hedenström A, Guglielmo CG, Åkesson S, Ramenof- sky M, Wikelski M. Grand challenges in migration biology. Integr Comp periods and areas, but short stops at critical staging areas Biol. 2010;50:261–79. doi:10.1093/icb/icq013. may also be just as crucial. Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 11 of 11 2. Stutchbury BJM, Tarof SA, Done T, Gow E, Kramer PM, Tautin J, Fox JW, Afa- 23. Willemoes M, Strandberg R, Klaassen RHG, Tøttrup AP, Vardanis Y, Howey nasyev V. Tracking longdistance songbird migration by using geolocators. PW, Thorup K, Wikelski M, Alerstam T. Narrow-front loop migration Science. 2009;323:896. in a population of the common cuckoo Cuculus canorus, as revealed 3. Tøttrup AP, Klaassen RHG, Strandberg R, Thorup K, Kristensen MW, by satellite telemetry. PLoS ONE. 2014;9:e83515. doi:10.1371/journal. Jørgensen PS, Fox J, Afanasyev V, Rahbek C, Alerstam T. The annual pone.0083515. cycle of a trans-equatorial Eurasian–African passerine migrant: different 24. Afanasyev V. A miniature daylight level and activity data recorder spatio-temporal strategies for autumn and spring migration. Proc R Soc B. for tracking animals over long periods. Mem Natl Inst Polar Res. 2012;279:1008–16. doi:10.1098/rspb.2011.1323. 2004;58:227–33. 4. Moreau RE. Palaearctic-African bird migration systems. London: Academic 25. Thaxter CB, Ross-Smith VH, Clark JA, Clark NA, Conway GJ, Marsh M, Leat Press; 1972. EHK, Burton NHK. A trial of three harness attachment methods and their 5. Thorup K, Tøttrup AP, Willemoes M, Klaassen RHG, Strandberg R, Lomas suitability for long-term use on Lesser Blackbacked Gulls and Great Skuas. Vega M, Dasari HP, Araújo MB, Wikelski M, Rahbek C. Resource tracking Ring Migr. 2015;29:65–76. doi:10.1080/03078698.2014.995546. within and across continents in long-distance bird migrants. Sci Adv. 26. Fox J. Geolocator manual v8. British Antarctic Survey. www.antarctica. 2017;3:e1601360. ac.uk/engineering/geo_downloads/Geolocator_manual_v8.pdf; 2010. 6. Fry CH, Keith S, Urban E. The birds of Africa, vol. 1–8. Princeton: Princeton 27. Lisovski S, Hewson CM, Klaassen RHG, Korner-Nievergelt F, Kristensen Press; 1982–2004. MW, Hahn S. Geolocation by light: accuracy and precision affected by 7. Wisz MS, Walther BA, Rahbek C. Using potential distributions to explore environmental factors. Methods Ecol Evol. 2012;3:603–12. determinants of Western Palaearctic migratory songbird species richness 28. R Development Core Team. R: a language and environment for statistical in sub-Saharan Africa. J Biogeogr. 2007;34:828–41. computing. Vienna: R Foundation for Statistical Computing; 2011. 8. Dowsett RJ. The separate African winter quarters of Pied Flycatcher 29. Pinzon JE, Compton JT. A non-stationary 1981–2012 AVHRR NDVI 3g time Ficedula hypoleuca and Collared Flycatcher F. albicollis. Bull ABC. series. Remote Sens. 2014;6:6929–60. 2010;17:79–81. 30. McKinnon EA, Fraser KC, Stutchbury BJM. New discoveries in landbird 9. Ouwehand J, Ahola MP, Ausems ANMA, Bridge ES, Burgess M, Hahn S, migration using geolocators, and a flight plan for the future. Auk. Hewson CM, Klaassen RHG, Laaksonen T, Lampe HM, Velmala W, Both C. 2013;130:211–22. Light-level geolocators reveal migratory connectivity in European popu- 31. Cleere N, Nurney D. A guide to the nightjars and related nightbirds. lations of pied flycatchers Ficedula hypoleuca. J Avian Biol. 2016;47:69–83. Robertsbridge: Pica Press; 1998. 10. Tøttrup AP, Klaassen RHG, Kristensen MW, Strandberg R, Vardanis Y, 32. Cresswell B, Edwards D. Geolocators reveal wintering areas of European Lindström Å, Rahbek C, Alerstam T, Thorup K. Drought in Africa caused Nightjar (Caprimulgus europaeus). Bird Study. 2013;60:77–86. delayed arrival of European songbirds. Science. 2012;338:1307. 33. Norevik G, Åkesson S, Hedenström A. Migration strategies and annual 11. Hahn S, Amrhein V, Zehtindjiev P, Liechti F. Strong migratory connectiv- space-use in an Afro-Palaearctic aerial insectivore—the European night- ity and seasonally shifting isotopic niches in geographically sepa- jar Caprimulgus europaeus. J Avian Biol. 2017. doi:10.1111/jav.01071. rated populations of a long-distance migrating songbird. Oecologia. 34. King JR. Seasonal allocation of time and energy resources in birds. Publ 2013;173:1217–25. Nuttall Ornithol Club. 1974;15:4–85. 12. Irwin DE, Irwin JH. Siberian migratory birds: the role of seasonal migration 35. Wingfield JC, Farner DS. Control of seasonal reproduction in temperate- in speciation. In: Marra PP, Greenberg R, editors. Birds of two worlds: the zone birds. Progr Reprod Biol. 1980;5:62–101. ecology and evolution of bird migration. Baltimore, MD: Johns Hopkins 36. Nilsson C, Klaassen RHG, Alerstam T. Difference in speed and duration of University Press; 2003. p. 27–40. bird migration between spring and autumn. Am Nat. 2013;181:837–45. 13. Alerstam T. Detours in bird migration. J Theor Biol. 2001;209:319–31. 37. Kranstauber B, Weinzierl R, Wikelski M, Safi K. Global aerial flyways allow doi:10.1006/jtbi.2001.2266. efficient travelling. Ecol Lett. 2015;18:1338–45. 14. Tøttrup AP, Pedersen L, Onrubia A, Thorup K. Migration of red-backed 38. Sanderson FJ, Donald PF, Pain DJ, Burfield IJ, van Bommel FPJ. Long- shrikes from the Iberian Peninsula: optimal or sub-optimal detour? J term population declines in Afro-Palearctic migrant birds. Biol Conserv. Avian Biol. 2017;48. doi:10.1111/jav.01352. 2006;131:93–105. 15. McNamara JM, Houston AI. Optimal annual routines: behaviour in the 39. Wilcove DS, Wikelski M. Going, going, gone: is animal migration disap- context of physiology and ecology. Philos Trans R Soc B. 2008;363:301–19. pearing? PLoS Biol. 2008;6:1361–4. 16. Barta Z, McNamara JM, Houston AI, Weber TP, Hedenström A, 40. Vickery JA, Ewing SR, Smith KW, Pain DJ, Bairlein F, Skorpilov J, Gregory Feró O. Optimal moult strategies in migratory birds. Proc R Soc B. RD. The decline of Afro-Palaearctic migrants and an assessment of 583 2008;363:211–29. potential causes. Ibis. 2014;156:1–22. doi:10.1111/ibi.12118. 17. Alerstam T, Lindström Å. Optimal bird migration: the relative importance 41. Faaborg J, Holmes RT, Anders AD, Bildstein KL, Dugger KM, Gauthreaux of time, energy and safety. In: Gwinner E, editor. Bird migration: the physi- SA, Heglund P, Hobson KA, Jahn AE, Johnson DH, Latta SC, Levey DJ, ology and ecophysiology. Berlin: Springer; 1990. p. 331–51. Marra PP, Merkord CL, Nol E, Rothstein SE, Sherry TW, Sillett S, Thompson 18. Dänhardt J, Lindström Å. Optimal departure decisions of songbirds from FR, Warnock N. Conserving migratory land birds in the New World: do we an experimental stopover site and the significance of weather. Anim know enough? Ecol Appl. 2007;20:398–418. Behav. 2001;62:235–43. doi:10.1006/anbe.2001.1749. 42. Runge CA, Martin TG, Possingham HP, Willis SG, Fuller RA. Conserving 19. Hedenström A, Alerstam T. How fast can birds migrate? J Avian Biol. mobile species. Front Ecol Environ. 2014;12:395–402. doi:10.1890/130237. 1998;29:424–32. 43. Wilcove DS. No way home. Washington: Island Press; 2008. 20. Holyoak DT. Nightjars and their Allies. Oxford: Oxford University Press; 44. Fair J, Paul E, Jones J, editors. Guidelines to the use of wild birds in 2001. research. Washington: Ornithological Council; 2010. 21. Erritzøe J, Mann CF, Brammer F, Fuller RA. Cuckoos of the world. London: Christopher Helm; 2012. 22. Åkesson S, Klaassen R, Holmgren J, Fox JW, Hedenström A. Migration routes and strategies in a highly aerial migrant, the Common Swift Apus apus, revealed by light-level geolocators. PLoS ONE. 2012;7:e41195. doi:10.1371/journal.pone.0041195. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Animal Biotelemetry Springer Journals

Annual spatiotemporal migration schedules in three larger insectivorous birds: European nightjar, common swift and common cuckoo

Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/annual-spatiotemporal-migration-schedules-in-three-larger-hcsu0NAN7g

References (41)

Publisher
Springer Journals
Copyright
Copyright © 2017 by The Author(s)
Subject
Life Sciences; Animal Systematics/Taxonomy/ Biogeography; Conservation Biology/Ecology; Terrestial Ecology; Bioinformatics; Freshwater & Marine Ecology
eISSN
2050-3385
DOI
10.1186/s40317-017-0119-x
Publisher site
See Article on Publisher Site

Abstract

Background: Knowledge of spatiotemporal migration patterns is important for our understanding of migration ecology and ultimately conservation of migratory species. We studied the annual migration schedules of European nightjar, a large nocturnal insectivore and compared it with two other larger migratory insectivores, common swift and common cuckoo. All species breed in North Europe and winter in sub-Saharan Africa, but estimating their spa- tiotemporal non-breeding distributions from observations is complicated by the occurrence of similar local African species. We used geolocators to track the annual migrations of nightjars and swifts and compared these with satellite tracking of cuckoo migration. Results: Individuals of the three species migrated to wintering grounds centered in Central Africa, except some com- mon swifts that remained in West Africa, crossing or circumventing the Sahara along different routes in spring and fall. Overall, all species showed similar regional and seasonal use of several stopover areas during migration. Among the three species, European nightjars and common cuckoos showed the most similar spatiotemporal migration patterns. The nightjars wintered in SW Central Africa and breeding and wintering made up by far the two longest stationary periods. Swifts were generally more mobile, and some individuals progressively visited areas further east in East Africa during winter and further west in West Africa on spring migration; this species also spent less time on stopovers, but more on wintering areas. Cuckoos were intermediate in their extent of movements. The speed of nightjar spring migration was equal to that of fall migration, in contrast to the two other species where spring return to breeding areas was faster. Conclusions: Ecological requirements are potentially useful for understanding spatiotemporal migration patterns and causes of declines in migratory species. Keywords: Long-distance migration, Migration speed and timing, Large insectivores, European nightjar, Geolocators Background evolution of species-specific migration schedules is sur - Basic knowledge of the ecological requirements of prisingly poor [1]. Evaluating the importance of differ - migratory species throughout their annual cycle and the ent parts of the annual cycle is complicated by the fact that most long-distance migrants are small birds, and thus, their annual migrations to date have been difficult *Correspondence: kthorup@snm.ku.dk to study. However, using satellite telemetry, it is now pos- Deceased Center for Macroecology, Evolution and Climate, Natural History sible to track birds down to the size of cuckoos Cuculus Museum of Denmark, University of Copenhagen, Universitetsparken 15, and small archival light recorders (hereafter referred to 2100 Copenhagen, Denmark Full list of author information is available at the end of the article © The Author(s) 2017. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 3 of 11 as geolocators) enable mapping migration of even small nightjar Caprimulgus europaeus, common cuckoo Cucu- songbirds based on light-level geolocation [2, 3]. lus canorus and common swift Apus apus. Palearctic–African bird migrants presumably track sea- European nightjars and common cuckoos are large sonal resource availability on a continental scale [4, 5], (both have wing lengths around 0.6 m and weigh around but it is less clear what factors shape variation in species- 80 and 110–130  g, respectively) insectivorous species specific and individual migration patterns. Large varia - [20, 21]. The cuckoos feed mainly on caterpillar larvae, tion among species in spatiotemporal schedules is well whereas nightjars catch low flying insects, often caught documented, but poorly understood, with very little the- by hawking from the ground. Swifts are slightly smaller oretical work developed to explain such differences. In (wing 0.45 m and weight 44 g), strictly aerial insectivores general, seasonal species distributions have been mapped foraging exclusively on the wing [22] presumably ena- using observations and ringing [6] supplemented in some bling foraging during migration. In contrast to swifts, cases with estimated distributions based on niche mod- both nightjars and cuckoos forage on much larger insect eling [7]. species [20, 21]. Furthermore, the common cuckoo is an Differences in fundamental niches alone are insuffi - obligate brood parasite [21], so their migration schedules cient to explain, for instance, the striking differences, in are not constrained by parental care. some cases, between the wintering grounds and migra- We used geolocators to map the spatiotemporal migra- tion routes of closely related species, such as found in tion schedule of the European nightjars and compared Ficedula flycatchers [8, 9] and Luscinia nightingales [10, it with that of common cuckoos Cuculus canorus ([23], 11]. Such patterns could be caused by different popula - same data set as included here) and common swifts Apus tion histories (potentially explaining why birds circum- apus ([22], using other data set). Based on the tracking vent the Mediterranean via the eastern or the western data, we describe similarities and differences among the flyway) or competitive exclusion. Detoured routes as seen three species in spatial patterns, length of sedentary stay in red-backed shrikes Lanius collurio breeding in Spain periods, duration of fall and spring migration episodes that migrate to East Africa via the Balkans could be con- as well as speed of migration. We then discuss whether strained by colonization routes [12] but may be optimal these potentially arise due to species-specific characteris - for other reasons such as wind assistance [13, 14]. tics. To compare migration patterns, we focus on migra- Several theoretical studies have dealt with the temporal tion strategies for migrants that spend the winter in the component of migration schedules [15]. Annual sched- same general area. Overall, nightjars, cuckoos and swifts ules need to accommodate key elements of the annual wintered in Central Africa, but some swifts stopped cycle, most importantly breeding, molt and migration migration early and stayed in West Africa for the winter [16]. Optimal scheduling depends on a number of fac- and these were excluded from comparisons. Obviously, tors, with each activity influencing the timing of the the three species studied represent only a small subsam- others [15]. During migration, individuals are, from ple of the Afro-Palearctic migrants, and thus, our results an optimality perspective, expected to minimize time, cannot necessarily be generalized to other migratory energy and/or predation [17]. Optimal migration theory species. commonly assumes that time minimization is important for migrants and this has also been reported empirically Methods [18]. As a consequence, migrants would be expected to Using playback and mist nets, we trapped 10 male Euro- stopover slightly longer and undertake longer migration pean nightjars in June and July 2010 and 14 males in June legs (hence making fewer stops) than expected from an and July 2011 in Northern Jutland, Denmark (57.06°N, energy minimization strategy where the costs of carry- 9.13°E). Birds were ringed with a metal ring and fitted ing extra fuel offset the gain of traveling faster. In general, with geolocators (Mk 10_S; 1.1  g; British Antarctic Sur- smaller species are expected to migrate faster by flap - vey, BAS [24]). The light sensor was placed on a 0.8  cm ping flight than larger species [19] and species with high stalk, in order to raise the sensor above the plumage. We settling costs and low fueling efficiency are expected to attached the loggers as a backpack using a full-body loop undertake longer stopovers. harness (comparable to the wing harness in [25]) made Species-specific biology has rarely been considered from a 2-mm-wide braided nylon string. Females were explicitly when explaining migration schedules, although released without geolocators because in previous years these are likely to be important. Our objective here is to we recaptured a lower proportion of females than males. describe the migration schedules of ecologically simi- Geolocators from seven male nightjars were retrieved by lar species and to discuss whether ecological similarities July 2013 and data were available from six (2010–2011: might lead to similarity in migration patterns. We focus 4; 2011–2012: 2). In three devices, logging had stopped on three species of larger insectivorous birds, European before the bird returned to the breeding site (on 28 Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 4 of 11 March, 18 April and 7 May). An extra effort was devoted that corresponds to a local minimum in latitude variation to catch birds from territories that were provided with during a stopover. We tested angles between 0° and −8° geolocators, presumably resulting in a slightly higher and found no minimum in four birds; in the two birds recapture rate (29%) than the rate of 20% of ringed-only where we found a minimum in latitude variation, the male nightjars (own data). estimated sun elevation angle was within 0.5° of the one A total of 25 common swifts were trapped during obtained using breeding latitude calibration. breeding at two Danish locations. At Bjødstrup, eastern The cuckoo data are from eight published satellite Jutland (56.29°N, 10.53°E), 10 and 7 individuals were fit - tracked birds with completed fall migration in 2010 of four ted with geolocators in 2010 and 2011, respectively, of males and two females, and three males and one female in which 6 individuals were re-trapped (2010–2011: 3; spring 2011 ([23]; Fig.  2; Additional file  2). The common 2011–2012: 3). At Nyborg, Fyn (55.30°N, 10.82°E), 8 indi- swift data from light-level-based geolocators were ana- viduals were fitted with geolocators in 2010 of which 3 lyzed using a threshold of 2, a sun angle of −5° and exclud- returned in 2011 and 1 returned in 2012. The recap - ing three weeks around the equinox. To calibrate the sun ture rate (40%), was below the rate reported in a similar elevation angle we used Hill–Ekström calibration [27] dur- study (75% [22]). This is potentially due to birds switch - ing the longest non-breeding stopover. All resulting sun ing between local nest sites from year to year (as we have elevation angles were between −4.5° and −5.5°, and we observed in previous cases but no quantitative data are used −5° for all individuals (Additional file 2 ). available), and we could not obtain permissions to recap- ture birds located at nearby nest sites located in adjacent Breeding, stopover and wintering periods: location, timing private houses. We used geolocators (Mk 20, 0.6  g from and distances BAS) fitted on the birds using a body loop harness made We used departure in one year and return in the follow- from 1-mm braided nylon string. Birds were trapped ing year to calculate duration of breeding in all three between sunset and full darkness in mist nets close to the species. nesting site or in the nest box. For geolocator data (nightjars and swifts), we defined The Copenhagen Bird Ringing Centre with permission a stopover as when the birds interrupted their migration from the Danish Nature Agency approved capturing and for more than 5 days and calculated an estimated stopo- tagging of nightjars and swifts. ver position by averaging the longitudes and latitudes during this period. We estimated stopover and move- Location estimation ment periods based on changes in latitude and longitude Data for nightjars were downloaded and analyzed using except during equinox, when only longitude was con- the BASTrak software suite [26]. We defined sunrises and sidered. Dictated by the uncertainty of light-level data sunsets using the threshold method with a sun angle esti- [27], we required consecutive stopovers to be separated mated for each individual. There was an unusually high by more than four degrees either latitude or longitude; level of noise in the light measurements on many days, otherwise, the two locations were considered part of the possibly caused by shading from either the vegetation same stopover and the location was acquired as the aver- that the birds hide in during day or from the feathers on age throughout both stops. Departure and arrival dates the back of the birds. To account for this noise, we only were taken as the last and first, respectively, days that used data from days with little noise (measured as stand- were within two degrees longitude of the average location ard deviation in the light measurements between sunrise estimate (Additional file 3). and sunset). To decide the acceptable level of noise, we In the nightjar, some of the fall and spring stopovers looked for the level under which the estimated latitudes occurred during equinox, making latitude estimation were stable during breeding (Additional file  1: Fig. S1). impossible. To estimate these stopovers for illustration Because we filtered out the days with most noise, we (Fig.  1), we calculated theoretically realistic latitudi- only needed to discard all latitudes within two weeks nal spans based on travel speeds. These were calculated on each side of the equinoxes. The remaining data were using the preceding or following stopover (if the preced- calibrated to the breeding latitude during their breed- ing was not available), travel speeds of 200–500  km/day ing site attendance and the derived sun angles (−2.2° to [3, 23] and the longitude estimated from the light data −5.0°) were used to estimate positions (Additional file  2). during the stopover. We calculated the overall migrat- We compared this to using Hill–Ekström calibration ing distance as the minimum distance between breeding [27] during the longest non-breeding stopover, but pre- site, stopover sites and winter area. For this calculation, sumably because of the amount of noise in the data this we also included estimated stopovers during equinox was only possible for two birds. Hill–Ekström calibra- which provided data on individual movement, despite tion requires that there is a specific sun elevation angle being disregarded because of their positional uncertainty. Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 5 of 11 Fig. 1 Main stopovers used by six European nightjars during migration. Colored circles are light-level-based geolocation stopover positions esti- mated using breeding site calibration. Positions are means of high-quality latitude and longitude estimates (see “Methods” section) during the full length of the stopovers. For stopovers latitudes during Equinox, intervals assuming realistic migration speeds (dotted lines correspond to speeds of 200–500 km/day; squares correspond to 350 km/day) rather than positions are indicated Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 6 of 11 Migration speed was calculated as migration segment in stopover use has also been reported in other long- distance divided by duration of the migration segment. distance migrants, for example red-backed shrikes [10] All durations of segments include days stopping over. although their timing appears more flexible. Some might only be a few days (shorter than what we The three species showed overall similarities in the consider a stopover), making some migration speeds routes and timing taken: all undertook stopovers in the close to travel speeds. Sahel after leaving Europe, subsequently spending the For satellite tracking data (cuckoos), we defined stopo - major part of the non-breeding season, i.e., wintering, in vers, dates and distances overall similarly. However, the Central Africa and at stopover sites in West Africa before higher accuracy of the satellite tracking data theoretically returning to the breeding sites. We therefore compared the allows for separation of small-scale changes between timing of departure and arrival in these four places: breed- sites. However, we grouped cuckoo stopovers in spe- ing, Sahel (8–20°N), wintering (south of 5°N and east of cific geographical regions at a spatial scale similar to that 10°E) (for swifts this could include trips to eastern Africa) which we obtained for the other two species. Individual and West Africa (west of 10°E and 0–20°N) in spring. We stopover locations were estimated as averages within also compared durations of stopovers as well as migration these regions as described in [22]. Because of non-daily speed. Roaming behavior during the non-breeding sea- transmissions, departure and arrival dates are rarely son was quantified as the number of stationary periods as identifiable and we used last and first, respectively, days well as the distance traveled among stationary sites dur- within the region/site in question (Additional file 3). ing winter. Because we aim to describe differences in spa - Geolocators in most cases allow for accurate determi- tiotemporal migration patterns in species with otherwise nation of stopover timing to within 1 day. Because of the similar constraints, we focused our analyses of the win- 10  h on 48  h off duty cycle of the satellite transmitters, tering period to birds wintering in the same general area, estimation of departures from stopovers was less accu- thus using only the seven swifts traveling to Central Africa rate for satellite data and transmissions were occasion- (three remained in West Africa all winter). Clearly, the ally missed, resulting in even less accurate estimates. We short-stopping swifts saved time and potentially energy by based stopover duration only on known locations result- using a shorter-distance migration strategy that potentially ing in potential underestimation of stopover duration influences their ability to fatten up before travel and speed (and overestimation of travel time). Missing transmis- of migration depending on local seasonal conditions. sions only occurred relatively frequently in Europe dur- We compare the three species fitting separate ANO - ing fall, potentially leading to underestimates of stopover VAs for each of the parameters above as a function of duration of cuckoos here. However, we found that cuck- species. Because of the large numbers of comparisons, oos staged for longer periods than the other species in Bonferroni-corrected significance levels are also indi - this region anyway. Therefore, we conclude that our com - cated. Overall, results should be interpreted with cau- parisons of stopover duration are likely conservative. tion due to generally low sample sizes. Because of varying sample sizes, tests were not directly comparable and in Comparison of European nightjar with common swift general, we only present comparisons from significant and common cuckoo tests. All models were fitted using general linear model We compare the migration routes and the timing of the (GLM) in R 3.1.0 [28]. nightjars to those of common cuckoos and common swifts. The study of migration in all three species is chal - Results lenging given their relatively small size, long-distance Spatial patterns migration and difficult study conditions in the winter - The six tracked nightjars performed a clockwise loop ing area (ranging from being nocturnal to highly aerial migration from breeding grounds in Denmark to win- often in highly remote areas). To make the data compa- ter grounds in southwest Central Africa traveling a rable among species, we included only adult birds (but mean total distance of 16,636  ±  1514  km (mean  ±  SD; were not able to separate sexes in analyses). Thus, our range 13,993–18,200  km) with no significant difference sample sizes are necessarily small. The different species between fall and spring distances (Matched pair test, were mostly tracked in the same year but some night- p  =  0.34, n  =  6) (Fig.  1; Table  1). The nightjars stopped jars and swifts were from a year later than the cuckoos. over two or three times in fall and three times in spring As birds’ schedules could vary as a response to the vari- (Fig.  1). The main wintering area was in Democratic able conditions among years, comparing schedules from Republic of Congo, Angola and northern Namibia/ the same year would have been preferable. At least at the Botswana. regional scale that we are considering, we do not expect Among the 10 tracked common swifts, most individu- any major differences among years. Similar consistency als (n  =  8) also migrated to the Sahel before continuing Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 7 of 11 Table 1 Mean, standard deviation (SD) and  number of  individuals [n] in  each analysis of  timing, duration, distance and speed for different legs of the migration for each of the three different species Parameter Nightjar (SD) [n] Cuckoo (SD) [n] Swift (SD) [n] P model P Ni_Cu P Ni_Sw P Cu_Sw Timing (day of year one) Dep. breeding 250 (7) [6] 191 (13) [8] 222 (8) [9] 0.000 0.000 0.000 0.000 Arr. Sahel 280 (5) [6] 263 (15) [6] 244 (11) [8] 0.000 0.018 0.000 0.006 Dep. Sahel 294 (3) [6] 301 (12) [6] 280 (15) [8] 0.014 0.371 0.040 0.005 Arr. C. Africa 301 (3) [6] 307 (14) [6] 288 (16) [7] 0.044 0.451 0.080 0.017 Dep. C. Africa 449 (15) [6] 450 (21) [5] 471 (11) [7] 0.036 0.897 0.021 0.036 Arr. W. Africa 455 (16) [6] 458 (18) [5] 481 (11) [7] 0.015 0.753 0.008 0.021 Dep. W. Africa 475 (10) [4] 487 (12) [4] 495 (4) [10] 0.001 0.035 0.000 0.083 Arr. breeding 493 (12) [3] 505 (12) [3] 506 (6) [10] 0.104 0.120 0.038 0.812 Duration (days) Breeding Sahel 30 (10) [6] 57 (15) [3] 27 (10) [7] 0.004 0.004 0.649 0.002 Sahel–C. Africa 7 (2) [6] 6 (4) [5] 8 (5) [6] 0.875 0.755 0.837 0.612 Fall 52 (9) [6] 113 (31) [5] 69 (21) [8] 0.001 0.000 0.161 0.002 Fall stops 31 (8) [6] 88 (30) [5] 43 (20) [8] 0.001 0.000 0.323 0.001 C. Africa–W. Africa 6 (2) [6] 8 (4) [5] 9 (6) [7] 0.463 0.549 0.222 0.573 W. Africa breeding 23 (6) [3] 18 (2) [3] 11 (5) [10] 0.005 0.270 0.003 0.038 Spring 55 (13) [3] 65 (7) [3] 34 (9) [7] 0.001 0.206 0.008 0.001 Spring stops 40 (11) [3] 48 (7) [3] 20 (8) [6] 0.002 0.255 0.009 0.001 Breeding 127 (15) [3] 49 (3) [3] 82 (11) [9] 0.000 0.000 0.000 0.001 C. Africa 148 (15) [6] 139 (23) [5] 183 (13) [7] 0.001 0.422 0.002 0.000 Migration 100 (16) [3] 185 (25) [3] 105 (19) [5] 0.001 0.001 0.758 0.001 Non-migration 265 (16) [3] 180 (25) [3] 260 (19) [5] 0.001 0.001 0.758 0.001 Distance (km) Breeding Sahel 4656 (925) [6] 5048 (89) [6] 6693 (675) [5] 0.000 0.323 0.000 0.001 Sahel–C. Africa 3560 (479) [6] 1822 (503) [6] 2310 (516) [5] 0.000 0.000 0.001 0.128 Fall 8215 (760) [6] 6870 (528) [6] 7954 (1379) [10] 0.085 0.041 0.639 0.063 C. Africa–W. Africa 1658 (906) [5] 1986 (572) [5] 2912 (963) [7] 0.056 0.553 0.025 0.084 W. Africa breeding 5555 (815) [5] 5691 (146) [3] 5906 (291) [10] 0.413 0.705 0.203 0.507 Spring 7180 (1015) [6] 7898 (376) [3] 7944 (1729) [10] 0.572 0.485 0.313 0.961 Speed (km/day) Breeding Sahel 178 (92) [6] 92 (24) [3] 316 (76) [4] 0.010 0.153 0.021 0.004 Sahel–C. Africa 577 (258) [6] 371 (120) [5] 553 (370) [4] 0.406 0.212 0.885 0.316 Fall 162 (30) [6] 65 (17) [5] 161 (103) [9] 0.065 0.044 0.981 0.032 C. Africa–W. Africa 281 (177) [5] 363 (328) [3] 369 (163) [7] 0.750 0.596 0.477 0.964 W. Africa breeding 236 (63) [3] 313 (29) [3] 666 (305) [10] 0.033 0.720 0.024 0.055 Spring 131 (52) [3] 122 (17) [3] 331 (131) [10] 0.012 0.924 0.017 0.014 P model is the p value of a general linear model with species as predictor, and the other three p values are testing pairwise differences between species (Ni nightjar, Cu cuckoo, Sw swift). p values <0.05 are in italics. p values <0.0016 (applying a Bonferroni correction) are in italic underline to Central Africa, but the swifts migrated via the Iberian in the Central African wintering range. The eight tracked Peninsula, followed the northwest African coast and then common cuckoos (Fig.  2) showed a clockwise loop migra- turned east toward the central Sahel (Fig.  2). Three swifts tion to the eastern Sahel, Central Africa, West Africa and spent the entire winter in West Africa, whereas the remain- back to Europe through Tunisia. In spring, all three spe- ing seven wintered in Central Africa. While in Central cies travelled to West Africa before going back to Europe, Africa, the birds used on average four different locations but swifts travelled further west than the other two, and (±0.9 SD) and four of the birds even visited East and South- cuckoos further west than nightjars (mean longitude swift: east Africa. In contrast, the cuckoos and nightjars only −9.3° ± 1.7°, n = 10; cuckoo: −2.0° ± 5.3°, n = 5; nightjar: used two different locations (± 0.5 and ±1, respectively) 7.6° ± 4.8°, n = 5; ANOVA: F  = 8.6, p = 0.002). 2,17 Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 8 of 11 Fig. 2 Comparison of migration routes between European nightjars, common cuckoos and common swifts. Circles indicate stopover areas used for at least 5 days. Species are shown in different colors (green = cuckoo, brown = nightjar, blue = swift). Nightjars and swifts were tracked with light- level-based geolocators with considerable uncertainty, especially for latitudes; cuckoos were tracked with high accuracy satellite transmitters [22]. Several of the nightjar locations included here (for illustration purposes) are extrapolated from known longitude and latitude based on an estimated flying speed of 350 km/day (see Fig. 1 for full latitude span). a Complete annual migration. Lines connect stopover locations and do not necessarily reflect route taken (for swift in fall, the easternmost point of the Sahara crossing is included for illustration of the Zugknick here) with darker shading indicating fall and lighter spring migration, b locations month by month. Background layer is vegetation greenness in each 2° latitude × 2° longi- tude cells averaged for each period 2000–2010 derived from on NOAA-16 Normalized Difference Vegetation Index (NDVI), AVHRR NDVI3 g.v0 [29] Overall, the spatial migration pattern of cuckoos was similar time to nightjars, but swifts stayed for substan- very similar to that of the nightjars, both showing clock- tially longer (183 ±  13  days, n = 7). Furthermore, swifts wise loop migration, differing from that of swifts (Fig. 2). traveled further (4180 ± 2059 km, n = 7) within the win- ter region than nightjars and cuckoos (1280 ±  1210  km, Sedentary stays n  =  6; 1086  ±  667  km, n  =  5, respectively). The night - The timing of nightjars (Additional file  1: Tables S1, jars spent on average 127 ± 15 days on the breeding site S2) was roughly similar to the timing of both swifts (using departure and arrivals from different years, n  = 3). and cuckoos, although most similar to that of cuck- Time spent on the breeding site differed among species oos (Fig.  3; Table  1). Nightjar southbound migration (p  <  0.001; Table  1), with swifts (82  ±  11  days, n  =  9) from the breeding grounds started in September (mean spending less time than nightjars; cuckoos (49 ±  3  days, date 6 September, range 30 August–17 September) with n = 3) spent least time of all species. arrival to the wintering area in October/November Departure from the breeding grounds and arrival to (mean date 28 October, range 25 October–3 November). the Sahel differed among the three species (p  <  0.001). They stayed in Central Africa on average 148  ±  15  days Cuckoos were the first to depart breeding grounds (11 (range 135–174, n  =  6; Fig.  1) and northbound migra- Jul ± 13 days, n = 8) followed by swifts (10 Aug ± 8 days, tion started in March/April (mean 25 March, range n  =  9) then nightjars (6 Sep  ±  7  days, n  =  6). Swifts 13 March–17 April). The three species differed in the arrived to the Sahel first (1 Sep  ±  11  days, n  =  8) fol- number of days they spent in Central Africa (p  <  0.001; lowed by cuckoos (20 Sep  ±  15  days, n  =  6) and night- Table  1), where cuckoos (139  ±  23  days, n  =  5) spent jars (7 Oct ± 5 days, n = 6). Departure from West Africa Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 9 of 11 126–214, n = 6) and in spring 131 ± 52 km/day (84–186, n  =  3). Fall migration speed was lowest in cuckoos but did not differ significantly between the three species. Spring migration speed did, however (p = 0.012; Table 1), with swifts (331  ±  131  km/day, n  =  10) migrating faster than both cuckoos (122  ±  17  km/day, n  =  3) and nightjars. Discussion Overall, the three species showed similar migration pat- terns on a regional and monthly scale, in particular com- pared to migration patterns reported for other species [30]. Although the tracked nightjars bred in a relatively restricted area in Denmark (within 65  km), they dis- persed over a much larger area in Africa and spent the winter outside the former known winter areas in Africa Fig. 3 Comparison of timing of departures and arrivals from different which do not include the central parts of the continent geographical regions shared between the three species. Error bars [20, 31] although three European nightjars from England indicate standard deviations. For swifts, timing analyses are based and several from Sweden spent the winter in the same only on individuals reaching Central Africa. Significance tests for each area as the birds from Denmark [32, 33]. Our tracks of state are given in Table 1 swifts were similar to those already published from other North European populations [22]. The similarity in migration patterns suggests that for differed (p  =  0.001) with nightjars (19 Apr  ±  10  days, insectivorous birds of such body mass, the beneficial n = 4) departing earlier than cuckoos (30 Apr ± 12 days, exploitation of the continent-wide seasonal changes in food n = 4) and swifts (10 May ± 4 days, n = 10). supply leaves limited room for spatiotemporal flexibility. We found the migration pattern of European nightjars to Duration and speed of fall and spring migration be more similar to that of common cuckoos than that of The nightjars spent 31  ±  8  days (range 21–44  days, common swifts both with regard to route and with regard n = 6) in total on stopover sites in fall and 40 ± 11 (range to timing though common cuckoos departed breeding 30–51 days, n = 3) in spring. Number of stopover days in grounds earlier. Potentially, this could be related to similar- fall differed between the three species (p   =  0.001; Fig.  3; ity in ecological niche, such as insect prey size. The main Table  1), where nightjars spent substantially less time difference between cuckoos and nightjars was that nightjars on fall stopovers than did cuckoos (88  ±  30  days, range spent more time on the breeding grounds, because cuckoos 53–115 days, n = 5) but a similar amount of time to swifts departed breeding areas earlier possibly because they pro- (43  ±  20  days, range 10–67  days, n  =  8). Spring stopo- vide no parental care. As a result, there were differences in ver duration also differed between species (p   =  0.002) timing before, after and at the breeding site. with nightjars being comparable to cuckoos (48 ± 7 days, The swifts differ from the other two species in showing range 41–55 days, n = 3), and both nightjars and cuckoos large spatial variation among individuals, with some indi- spent substantially more time than swifts (20  ±  8  days, viduals wintering in West Africa, while others wintered in range 7–28  days, n  =  6). The pattern was the same for Central Africa. Seven of ten common swifts covered large total duration of fall and spring migration. Nightjars spent distances spending their “sedentary” wintering period a similar (ANOVA: F  = 0.19, p = 0.67) amount of time roaming over large areas using more sites than the other 1,7 on spring (55 ± 13 days, range 45–70 days, n = 3) and fall species which could result from the fact that swifts forage migration (52 ± 9 days, range 41–65 days, n = 6), whereas on large numbers of small insect prey, often at high alti- in both cuckoos (spring: 65  ±  7  days, range 61–73  days, tudes and quickly traveling large distances in response to n = 3; fall: 113 ± 31 days, range 67–137 days, n = 5) and food availability. These swifts differed temporally from the swifts (spring: 34 ± 9 days, range 19–47 days, n = 7; fall: other species by spending more time in Central Africa, 69  ±  21  days, range 32–94  days, n  =  8) fall migration is with a shorter stopover in West Africa, which resulted in a lasting significantly longer than spring migration (cuckoo: faster overall spring migration in these individuals (spring F  = 6.6, p = 0.04; swift: F  = 16.8, p = 0.001). migration from West Africa was equally fast among the 1,6 1,13 The mean total migration speed of nightjars (includ - swifts that wintered in there). However, fall migration ing stopover sites) in fall was 162  ±  30  km/day (range speeds were similar in the three species. Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 10 of 11 In addition to resource needs, many other factors could Conclusions be causing variation in annual migration schedules. For The spatial migration pattern of the European nightjar example, because avian annual cycles are evolved to was more similar to that of the common cuckoo than that minimize the overlap of energetically demanding events, of the common swift. Our study illustrates the potential birds must fit molt and breeding into the annual sched - for investigating determinants of migratory schedules ule at other times than during migration [34, 35]. As the using a comparative approach. resource needs also dictate timing of molt, we assume Additional files that general migration patterns are similarly affected by periods of resource needs and expenditure. Additional file 1. Supplementary information on European nightjars. The fact that nightjars both forage and travel at night Data on individual birds ( Table S1), stopover duration ( Table S2) and theoretically restricts their migration speeds compared definition of high-quality positions (Fig. S1). to the other species that can forage by day and travel at Additional file 2. Location estimates for individual European nightjars, night (as well as by day). However, we did not find night - common swifts and common cuckoos included in the study. jar migration speeds to be generally slower or faster. In Additional file 3. Estimated stopovers and timing of European nightjars, most avian species studied so far, spring migration occurs common swifts and common cuckoos. faster than fall migration, possibly because of constraints in the arrival time at the breeding area [36]. Norevik et al. Authors’ contributions [33] reported faster fall than spring migration in Euro- LBJ, NOJ, LH, MD, ADF, APT, KT carried out field work on nightjars and swifts. pean nightjars. We found a similar, but non-significant LBJ, NOJ, MW, APT, KT analyzed the data. LBJ, NOJ, MW, APT, KT drafted the manuscript. LBJ, MW, MD, ADF, APT and KT contributed to the final manu- difference, which contrasted the pattern in swifts and script. LBJ, MW, MD, ADF, APT, KT read and approved the final manuscript. cuckoos, where spring migration is faster than fall migra- tion [22, 23]. Compared to the other two species, the Author details Center for Macroecology, Evolution and Climate, Natural History Museum nightjar departed the breeding site substantially later yet of Denmark, University of Copenhagen, Universitetsparken 15, 2100 Copen- arrived in Central Africa at the same time. Nightjars were hagen, Denmark. BirdLife Denmark, Vesterbrogade 140, 1620 Copenhagen the earliest to arrive on the breeding grounds. V, Denmark. Department of Bioscience, Aarhus University, Kalø, Grenåvej 14, 8410 Rønde, Denmark. The migration of the European nightjars followed dif - ferent routes in fall and spring, with a more westerly Acknowledgements spring migration trajectory. Taking a westerly route in We thank Roine Strandberg, Raymond H.G. Klaassen and Thomas Alerstam for permission to use the Swedish cuckoo data. spring may be advantageous because of stable tailwind patterns in West Sahara [37]. Additionally, they can ben- Competing interests efit from foraging in areas where food becomes avail - The authors declare that they have no competing interests. able, following the fall seasonal patterns of rain along Availability of data and material these routes toward the winter areas [14]. Both swifts The data set(s) supporting the conclusions of this article is(are) included and cuckoos also traveled to West Africa before head- within the article (and its additional file(s)). ing north, suggesting that ecological conditions favor this Ethics approval detour. The three species differed substantially, however, This study was carried out in strict accordance with guidelines to the use in the location of their stopover longitude. of wild birds in research of the Ornithological Council [44]. Animal work on European nightjars and common swifts was approved by the Copenhagen Widespread declines have been reported recently in Bird Ringing Centre (J.Nr. SN 302-009) under permission from the Danish For- sub-Saharan migrant species but the causes are to a large est and Nature Agency. degree unknown [38–41]. Migrant species are likely Funding vulnerable to habitat loss because they depend on the KT thanks the Danish Council for Independent Research for support to condition of networks of sites that may be separated by the MATCH Project (1323-00048B). KT, MW, APT thank the Danish National thousands of kilometers [42]. As migratory animals show Research Foundation for support to the Center for Macroecology, Evolution and Climate (DNRF96). declines around the world [43] there is a growing need to understand potential reliance of migrants on staging Received: 2 September 2016 Accepted: 5 January 2017 regions or areas and for conservation initiatives to pro- tect such networks [42]. An essential starting point is to understand the spatiotemporal migration patterns within and across migratory species. The comparative long References duration of stay on the breeding and wintering grounds 1. Bowlin MS, Bisson IA, Shamoun-Baranes J, Reichard JD, Sapir N, Marra PP, obviously direct conservation research toward these two Kunz TH, Wilcove DS, Hedenström A, Guglielmo CG, Åkesson S, Ramenof- sky M, Wikelski M. Grand challenges in migration biology. Integr Comp periods and areas, but short stops at critical staging areas Biol. 2010;50:261–79. doi:10.1093/icb/icq013. may also be just as crucial. Jacobsen et al. Anim Biotelemetry (2017) 5:4 Page 11 of 11 2. Stutchbury BJM, Tarof SA, Done T, Gow E, Kramer PM, Tautin J, Fox JW, Afa- 23. Willemoes M, Strandberg R, Klaassen RHG, Tøttrup AP, Vardanis Y, Howey nasyev V. Tracking longdistance songbird migration by using geolocators. PW, Thorup K, Wikelski M, Alerstam T. Narrow-front loop migration Science. 2009;323:896. in a population of the common cuckoo Cuculus canorus, as revealed 3. Tøttrup AP, Klaassen RHG, Strandberg R, Thorup K, Kristensen MW, by satellite telemetry. PLoS ONE. 2014;9:e83515. doi:10.1371/journal. Jørgensen PS, Fox J, Afanasyev V, Rahbek C, Alerstam T. The annual pone.0083515. cycle of a trans-equatorial Eurasian–African passerine migrant: different 24. Afanasyev V. A miniature daylight level and activity data recorder spatio-temporal strategies for autumn and spring migration. Proc R Soc B. for tracking animals over long periods. Mem Natl Inst Polar Res. 2012;279:1008–16. doi:10.1098/rspb.2011.1323. 2004;58:227–33. 4. Moreau RE. Palaearctic-African bird migration systems. London: Academic 25. Thaxter CB, Ross-Smith VH, Clark JA, Clark NA, Conway GJ, Marsh M, Leat Press; 1972. EHK, Burton NHK. A trial of three harness attachment methods and their 5. Thorup K, Tøttrup AP, Willemoes M, Klaassen RHG, Strandberg R, Lomas suitability for long-term use on Lesser Blackbacked Gulls and Great Skuas. Vega M, Dasari HP, Araújo MB, Wikelski M, Rahbek C. Resource tracking Ring Migr. 2015;29:65–76. doi:10.1080/03078698.2014.995546. within and across continents in long-distance bird migrants. Sci Adv. 26. Fox J. Geolocator manual v8. British Antarctic Survey. www.antarctica. 2017;3:e1601360. ac.uk/engineering/geo_downloads/Geolocator_manual_v8.pdf; 2010. 6. Fry CH, Keith S, Urban E. The birds of Africa, vol. 1–8. Princeton: Princeton 27. Lisovski S, Hewson CM, Klaassen RHG, Korner-Nievergelt F, Kristensen Press; 1982–2004. MW, Hahn S. Geolocation by light: accuracy and precision affected by 7. Wisz MS, Walther BA, Rahbek C. Using potential distributions to explore environmental factors. Methods Ecol Evol. 2012;3:603–12. determinants of Western Palaearctic migratory songbird species richness 28. R Development Core Team. R: a language and environment for statistical in sub-Saharan Africa. J Biogeogr. 2007;34:828–41. computing. Vienna: R Foundation for Statistical Computing; 2011. 8. Dowsett RJ. The separate African winter quarters of Pied Flycatcher 29. Pinzon JE, Compton JT. A non-stationary 1981–2012 AVHRR NDVI 3g time Ficedula hypoleuca and Collared Flycatcher F. albicollis. Bull ABC. series. Remote Sens. 2014;6:6929–60. 2010;17:79–81. 30. McKinnon EA, Fraser KC, Stutchbury BJM. New discoveries in landbird 9. Ouwehand J, Ahola MP, Ausems ANMA, Bridge ES, Burgess M, Hahn S, migration using geolocators, and a flight plan for the future. Auk. Hewson CM, Klaassen RHG, Laaksonen T, Lampe HM, Velmala W, Both C. 2013;130:211–22. Light-level geolocators reveal migratory connectivity in European popu- 31. Cleere N, Nurney D. A guide to the nightjars and related nightbirds. lations of pied flycatchers Ficedula hypoleuca. J Avian Biol. 2016;47:69–83. Robertsbridge: Pica Press; 1998. 10. Tøttrup AP, Klaassen RHG, Kristensen MW, Strandberg R, Vardanis Y, 32. Cresswell B, Edwards D. Geolocators reveal wintering areas of European Lindström Å, Rahbek C, Alerstam T, Thorup K. Drought in Africa caused Nightjar (Caprimulgus europaeus). Bird Study. 2013;60:77–86. delayed arrival of European songbirds. Science. 2012;338:1307. 33. Norevik G, Åkesson S, Hedenström A. Migration strategies and annual 11. Hahn S, Amrhein V, Zehtindjiev P, Liechti F. Strong migratory connectiv- space-use in an Afro-Palaearctic aerial insectivore—the European night- ity and seasonally shifting isotopic niches in geographically sepa- jar Caprimulgus europaeus. J Avian Biol. 2017. doi:10.1111/jav.01071. rated populations of a long-distance migrating songbird. Oecologia. 34. King JR. Seasonal allocation of time and energy resources in birds. Publ 2013;173:1217–25. Nuttall Ornithol Club. 1974;15:4–85. 12. Irwin DE, Irwin JH. Siberian migratory birds: the role of seasonal migration 35. Wingfield JC, Farner DS. Control of seasonal reproduction in temperate- in speciation. In: Marra PP, Greenberg R, editors. Birds of two worlds: the zone birds. Progr Reprod Biol. 1980;5:62–101. ecology and evolution of bird migration. Baltimore, MD: Johns Hopkins 36. Nilsson C, Klaassen RHG, Alerstam T. Difference in speed and duration of University Press; 2003. p. 27–40. bird migration between spring and autumn. Am Nat. 2013;181:837–45. 13. Alerstam T. Detours in bird migration. J Theor Biol. 2001;209:319–31. 37. Kranstauber B, Weinzierl R, Wikelski M, Safi K. Global aerial flyways allow doi:10.1006/jtbi.2001.2266. efficient travelling. Ecol Lett. 2015;18:1338–45. 14. Tøttrup AP, Pedersen L, Onrubia A, Thorup K. Migration of red-backed 38. Sanderson FJ, Donald PF, Pain DJ, Burfield IJ, van Bommel FPJ. Long- shrikes from the Iberian Peninsula: optimal or sub-optimal detour? J term population declines in Afro-Palearctic migrant birds. Biol Conserv. Avian Biol. 2017;48. doi:10.1111/jav.01352. 2006;131:93–105. 15. McNamara JM, Houston AI. Optimal annual routines: behaviour in the 39. Wilcove DS, Wikelski M. Going, going, gone: is animal migration disap- context of physiology and ecology. Philos Trans R Soc B. 2008;363:301–19. pearing? PLoS Biol. 2008;6:1361–4. 16. Barta Z, McNamara JM, Houston AI, Weber TP, Hedenström A, 40. Vickery JA, Ewing SR, Smith KW, Pain DJ, Bairlein F, Skorpilov J, Gregory Feró O. Optimal moult strategies in migratory birds. Proc R Soc B. RD. The decline of Afro-Palaearctic migrants and an assessment of 583 2008;363:211–29. potential causes. Ibis. 2014;156:1–22. doi:10.1111/ibi.12118. 17. Alerstam T, Lindström Å. Optimal bird migration: the relative importance 41. Faaborg J, Holmes RT, Anders AD, Bildstein KL, Dugger KM, Gauthreaux of time, energy and safety. In: Gwinner E, editor. Bird migration: the physi- SA, Heglund P, Hobson KA, Jahn AE, Johnson DH, Latta SC, Levey DJ, ology and ecophysiology. Berlin: Springer; 1990. p. 331–51. Marra PP, Merkord CL, Nol E, Rothstein SE, Sherry TW, Sillett S, Thompson 18. Dänhardt J, Lindström Å. Optimal departure decisions of songbirds from FR, Warnock N. Conserving migratory land birds in the New World: do we an experimental stopover site and the significance of weather. Anim know enough? Ecol Appl. 2007;20:398–418. Behav. 2001;62:235–43. doi:10.1006/anbe.2001.1749. 42. Runge CA, Martin TG, Possingham HP, Willis SG, Fuller RA. Conserving 19. Hedenström A, Alerstam T. How fast can birds migrate? J Avian Biol. mobile species. Front Ecol Environ. 2014;12:395–402. doi:10.1890/130237. 1998;29:424–32. 43. Wilcove DS. No way home. Washington: Island Press; 2008. 20. Holyoak DT. Nightjars and their Allies. Oxford: Oxford University Press; 44. Fair J, Paul E, Jones J, editors. Guidelines to the use of wild birds in 2001. research. Washington: Ornithological Council; 2010. 21. Erritzøe J, Mann CF, Brammer F, Fuller RA. Cuckoos of the world. London: Christopher Helm; 2012. 22. Åkesson S, Klaassen R, Holmgren J, Fox JW, Hedenström A. Migration routes and strategies in a highly aerial migrant, the Common Swift Apus apus, revealed by light-level geolocators. PLoS ONE. 2012;7:e41195. doi:10.1371/journal.pone.0041195.

Journal

Animal Biotelemetry – Springer Journals

Published: Feb 8, 2017

There are no references for this article.