SEARCH

SEARCH BY CITATION

Keywords:

  • annual movements;
  • conservation;
  • forest habitat;
  • life history;
  • technologies

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

1. Tracking return migrations in songbirds has been impossible until recently when miniaturization of light-level loggers enabled observation of the first complete round trip. Although geolocators are extensively used on animals at sea, little is known about how accurate geolocators are for tracking terrestrial or forest-dwelling migrants.

2. To test the accuracy of geolocators for tracking migratory songbirds living in forested habitat, we calibrated geolocators to a source population located in central Europe and collected location estimates based on the source population calibration from stationary geolocators deployed over an 800 km NE to SW gradient in Western Europe. Additionally, we fit non-migratory songbirds (European blackbirds, Turdus merula) with geolocators for 12 months to compare known locations of individuals with locations estimated by geolocators.

3. We found an average error ±95% CI of 201 ± 43 km in latitude for stationary geolocators in forest habitat. Longitude error was considerably lower (12 ± 03 km). The most accurate geolocator was on average 23 km off target, the worst was on average 390 km off.

4. The winter latitude estimate error for geolocators deployed on sedentary birds was on average (±95% CI) 143 ± 62 km when geolocators were calibrated during the breeding season and 132 ± 75 km when they were calibrated during the winter. Longitude error for geolocators deployed on birds was on average (±95% CI) 50 ± 34 km.

5. Although we found error most likely due to seasonal changes in habitat and behaviour, our results indicate that geolocators can be used to reliably track long-distance forest-dwelling migrants. We also found that the low degree of error for longitude estimates attained from geolocators makes this technology suitable for identifying relatively short-distance movements in longitude.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Songbirds are model study systems in many biological disciplines and provide important ecosystem services (Newton 2008). Many songbird species are critically endangered, especially migrants (Wilcove 2008). However, the destinations of migrants among seasons are largely unknown. It is clear that events in one season influence life history, morphological and behavioural traits in subsequent seasons (Marra, Hobson, & Holmes 1998; Webster et al. 2002). The annual survival of migratory songbirds is potentially impacted by events occurring during migration (Sillett & Holmes 2002). It is thus of great scientific but also societal importance to better understand songbird migrations on the level of the individual, particularly to fully understand their annual cycle (Webster & Marra 2005; Wilcove & Wikelski 2008).

Until recently, the tracking of individual migrants has been limited to large animals capable of bearing the load of satellite transmitters and GPS loggers. However, Stutchbury et al. (2009), a landmark study in movement ecology, reported the first complete migratory track for a songbird using light-level loggers (geolocators). Significant findings of this study include an individual purple martin (Progne subis) travelling 577 km day−1.

Geolocators have rapidly become small enough (0·6 g) to fit a significant number of small songbird species. Tracking migrants with light-level loggers is achieved using archived light intensity levels to estimate sunrise and sunset times to calculate day length and the time of midday (Wilson et al. 1992; for complete review of theory see Hill 1994). Sunrise and sunset times are identified using a light intensity threshold determined during a pre-deployment calibration. Error in location estimates during deployment can be due to shading at sunrise and sunset from clouds and landforms, atmospheric aberrations and low variation in day length around the equinoxes. Uncertainties in archived light values are known to cause smaller errors in longitude than latitude, regardless of the season (Ekstrom 2004). Species that live in flat, open habitat with unobstructed views of the horizon are well suited for geolocators (Croxall et al. 2005; Egevang et al. 2010).

An added problem when using geolocators in terrestrial habitats, many of which are at least partially forested, is that foliage causes increased shading. Location estimates could thus be systematically biased if the archived light intensity at sunrise and sunset is different from the threshold identified during calibration. Therefore, geolocators should be calibrated in the typical habitat of the species being tracked. Further, it has long been recognized that the daily activity patterns of birds can change throughout the annual cycle (Daan & Aschoff 1975). Daan and Aschoff identified seasonal changes in the timing of the onset and end of daily activity for several temperate avian species. These differences in behaviour resulted in seasonal differences in subjective day length. Seasonal changes in subjective day length could result in increased error in latitude estimates during the winter if calibrations are performed during the summer. Additionally, seasonal changes in habitat shading either because of leaf loss, or if the bird changes habitat types as it migrates, could result in increased error.

To test the accuracy of geolocators for tracking terrestrial and partially forest-dwelling migratory birds, we calibrated geolocators to a source population located in central Europe and collected location estimates based on the source population calibration from stationary geolocators deployed over an 800 km NE to SW gradient in Western Europe. Additionally, we analysed location estimates for six non-migratory European blackbirds (Turdus merula merula) we fitted with geolocators to identify error caused by individual and seasonal behavioural differences and seasonal changes in forest shading.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We used MK 10S geolocators (1·2 g) developed by the British Antarctic Survey. All stationary geolocators were deployed between 7 December 2008 and 28 April 2009 from southern Germany (47·88°N, 11·10°E) to south-western France (44·02°N, 2·20°E) at 15 locations (locations were roughly 50 km apart). Two replicates were placed within 5 m of each other in similar forest structure. Geolocators were attached 2–3 m high in trees and shrubs in mixed coniferous/deciduous forests, always with the light sensor facing upward. Forest locations were selected that were suitable roosting habitat for blackbirds (Hill & Cresswell 1997). The transect was selected because it is within the wintering range of migratory blackbirds from southern Germany.

We captured two blackbirds in September 2008 and four in the spring of 2009 (total of six blackbirds) at two locations in southern Germany (47·77°N, 9·00°E; 47·77°N 9·04°E) using mist nets. Prior to capture, Mk 10S geolocators were connected to radiotransmitters (2·6 g; Sparrow Systems, Fisher, IL, USA) with heat-shrink tubing (0·4 g), and leg-loop harnesses were connected to the transmitters through pre-fabricated tubing in the transmitters. A range of harness sizes were built from 1-mm elastic beading cord to fit the naturally occurring body masses of blackbirds in the study population using the method of Naef-Daenzer (2007). All transmitter-geolocator backpacks weighed <5% of the body mass of the individual that it was deployed on. Once a harness was fitted to a bird, it was inspected for appropriateness of fit. All birds were observed for as long as possible immediately after release and throughout deployment whenever possible to ensure normal flight behaviour.

Birds were confirmed to be present throughout deployment by automated receiving units (Sparrow Systems) and manual tracking with a hand-held three element Yagi antenna (AF Antronics, Inc., Urbana, IL, USA) and an AR 8200 MKIII hand-held receiver (AOR U.S.A., Inc., Torrance, CA, USA) from the first date of capture until recapture. In 2009 and 2010, birds were located 2–3 times per week until December and then at least one time per week until recapture. Once a bird was located by manual tracking, latitude and longitude were obtained using a hand-held GPS. All birds were also monitored by automated receiving units throughout deployment. The automated receiving units searched for each frequency every 60 s throughout the testing period. The automated receivers were connected to H antennas (ATS, Isanti, MN, USA), mounted 3–6 m high. The maximum distance at our study site that an automated receiver was capable of receiving a signal from a transmitter was 850 m. The two individuals that were originally captured in 2008 were not tracked manually on a regular basis but were monitored daily with an automated receiving unit. Upon recapture, birds were immediately released after removal of their backpack.

Before deployment, all geolocators were placed outdoors with a similar view of the horizon for 7 days to affirm similar light sensitivity. After deployment, raw data were corrected for clock drift using Bastrak (British Antarctic Survey). To calculate a light threshold and sun elevation angle for sunrise and sunset transitions for stationary geolocators, light data from six stationary geolocators placed in forest habitat at three locations (47·44°N, 8·7°E; 47·77°N, 9·00°E; 47·88°N, 11·10°E) were analysed. These six geolocators were deployed prior to the others at the northeast end of the 800 km transect and were meant to simulate the source population. We visually inspected sunrise and sunset transitions for the six geolocators from 26 November 2008 to 10 December 2008 (pre-deployment calibration) and from 1 April 2009 to 15 April 2009 (post-deployment calibration) using TransEdit2 to identify the average light-level value when the light changed most rapidly, excluding any anomalous transitions. Using typical transition events during the calibration period, a light-level threshold value of 16 was identified. The corresponding average sun elevation angle for the six stationary geolocators during the calibration period (26 November 2008 to 10 December 2008 and 1 April 2009 to 15 April 2009) was −2·84°±0·18 (SE; n = 12).

To correct for behavioural and habitat choices of wild birds, we calibrated geolocators deployed on blackbirds during deployment. To identify the effect of seasonal changes in shading, we calibrated all geolocators during two periods. The breeding calibration period started on the first date that we observed a migratory blackbird from the population return to the breeding site (March 24) and ended on the first date that we observed a migrant blackbird depart during the autumn (October 2). All six individuals in the current study were sedentary throughout the deployment. The winter calibration period started on November 1 and ended on February 15. All six geolocators on blackbirds were deployed throughout the entire winter calibration period. Capture and recapture dates varied, and so, calibration data were not available for the entire breeding calibration period for all individuals. The breeding calibration dates for the six individuals were the following (15 May 2009 to 2 October 2009, 24 March 2010 to 24 April 2010; 12 June 2009 to 2 October 2009, 24 March 2010 to 21 April 2010; 24 June 2009 to 2 October 2009, recaptured before spring; 18 June 2009 to 2 October 2009, 24 March 2010 to 24 April 2010; 10 September 2008 to 2 October 2008, 24 March 2009 to 21 April 2009; 11 September 2008 to 2 October 2008, recaptured before spring). Using a light threshold of 16, the average sun elevation angle for the breeding calibration period was −3·45°±0·19 (SE; n = 10) and −3·57°±0·25 (SE; n = 6) for winter.

Transitions for all geolocators were calculated using TransEdit2, and anomalous transitions were rejected from analysis (mean 40 days ± 26·3 SD, excluding 3 weeks before and after equinox events). Sunsets were retarded by 10 min. Latitude and longitude were calculated using Locator (British Antarctic Survey). Both midnight and noon locations were used. We did not compensate for movement because all geolocators were either stationary or deployed on non-migratory birds throughout deployment.

Before analysis, all error values were converted to distance (km) in latitude and longitude based on the actual location of the individual geolocator to make the results transferable to global applications. To identify location estimate error because of seasonal and individual behavioural differences, we calculated the average latitude and longitude estimate ±95% CI for each month of deployment for geolocators deployed on blackbirds using the breeding calibration angle. Additionally, we calculated the monthly coefficient of variation (CV) of latitude and longitude estimate error for sedentary birds using the average breeding calibration angle. For stationary geolocators, we calculated the monthly coefficient of variation of latitude and longitude estimate error using a sun elevation angle of −2·84. Estimates for the equinoxes included 3 weeks before and after (autumnal equinox = 22 September, vernal equinox = 20 March). Data were only used for one time unit; therefore, days in October, February and April that were included in an equinox event were not used to calculate the error for that month. To compare regional differences in geolocator accuracy, we calculated the average winter location estimate ±95% CI (10 December to 31 January) for stationary geolocators deployed along the migratory trajectory. We also calculated the average winter error ±95% CI of latitude and longitude for stationary geolocators over the same period. For geolocators deployed on sedentary blackbirds, we calculated the average winter error ±95% CI (blackbird winter = 1 November to 15 February) using both the breeding and winter calibrations.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

The average monthly location estimate error (±95% CI) of latitude and longitude for individual geolocators deployed on blackbirds is presented in Table 1. The monthly mean location estimate of each bird (± 95% CI) is presented in Fig. S1. The average latitude winter error (±95% CI) for sedentary blackbirds using the breeding calibration was 143 ± 62 km and 132 ± 75 km using the winter calibration. The average winter error for longitude for sedentary blackbirds (±95% CI), using both the breeding and winter calibration, was 50 ± 34 km. The average winter location error (±95% CI) for stationary geolocators was 201 ± 43 km for latitude and 12 ± 03 km for longitude (Fig. 1). The monthly coefficient of variation of error for latitude and longitude for sedentary birds and stationary geolocators is presented in Fig. 2. One stationary geolocator was removed from the analysis because it was a statistical outlier.

Table 1.   Average monthly error (km) ±95% CI of latitude (upper row) and longitude estimates (lower row) for six geolocators on sedentary blackbirds in southern Germany. Equinox calculations include 3 weeks before and 3 weeks after (22 September 2008 & 2009 and 20 March 2009 & 2010)
JuneJulyAugustAutumnal equinoxOctoberNovemberDecemberJanuaryFebruaryVernal equinoxApril
  1. n, north; s, south; e, east; w, west.

   s 582 ± 916 e 73 ± 16s 375 ± 193 e 70 ± 36s 258 ± 43 e 23 ± 16s 159 ± 32 e 49 ± 14s 154 ± 36 e 61 ± 15s 318 ± 99 e 59 ± 16s 1429 ± 1039 e 46 ± 13n 184 ± 65 e 31 ± 12
   s 723 ± 837 e 75 ± 15s 455 ± 144 e 46 ± 15s 202 ± 47 e 62 ± 18s 139 ± 45 e 97 ± 21s 107 ± 38 e 95 ± 16s 287 ± 100 e 98 ± 26  
s 76 ± 92 e 162 ± 47s 108 ± 49 e 148 ± 33s 9 ± 79 e 40 ± 27n 23 ± 1511 e 16 ± 44n 756 ± 483 w 41 ± 103n 231 ± 92 w 16 ± 29n 167 ± 56 w 32 ± 36n 96 ± 61 e 34 ± 31s 22 ± 150 e 34 ± 30s 80 ± 777 w 13 ± 16n 186 ± 64 w 12 ± 18
s 19 ± 49 e 173 ± 38s 95 ± 45 e 170 ± 23s 127 ± 94 e 108 ± 28s 1343 ± 894 e 31 ± 23n 15 ± 113 e 51 ± 24s 76 ± 69 e 50 ± 20s 39 ± 33 e 37 ± 16s 92 ± 39 e 15 ± 19s 94 ± 149 e 72 ± 31  
n 55 ± 51 e 145 ± 30n 113 ± 34 e 41 ± 24n 40 ± 57 e 80 ± 20s 455 ± 755 e 50 ± 17s 216 ± 162 e 49 ± 26s 165 ± 56 e 18 ± 17s 96 ± 37 e 29 ± 25s 122 ± 37 e 37 ± 17s 506 ± 120 e 20 ± 15s 1015 ± 819 w 0 ± 11n 311 ± 118 e 23 ± 33
s 71 ± 33 e 191 ± 42s 16 ± 30 e 45 ± 40s 511 ± 115 e 4 ± 35s 1781 ± 847 e 30 ± 22n 343 ± 145 e 67 ± 28n 61 ± 63 e 69 ± 19n 115 ± 40 e 74 ± 23n 137 ± 36 e 107 ± 24n 134 ± 153 e 107 ± 29s 144 ± 845 e 35 ± 19n 219 ± 53 e 5 ± 13
image

Figure 1.  Error of winter locations for stationary geolocators in forest habitat. Shown are average location estimates of 30 geolocators distributed in Western Europe from 10 December 2008 to 31 January 2009 (winter). One outlier had an average winter location error of 565 km (*) Locations (black dots) along the transect are the actual locations of two replicates (a). Winter location estimates with 95% CI are connected to actual locations (dotted lines). Panel b shows the average winter error at 15 locations along the transect for latitude and longitude. Negative values indicate error south and west. The outlier indicated in panel a was removed when calculating the average error of latitude of that location in panel b.

Download figure to PowerPoint

image

Figure 2.  Monthly coefficient of variation for stationary geolocators and geolocators deployed on sedentary blackbirds in latitude (a) and longitude (b). The monthly error was pooled for all geolocators in each category (geolocators on blackbirds = solid lines, stationary geolocators = dotted lines).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Tracking individual migratory songbirds across seasons will provide researchers with the opportunity to answer important questions about the basic biology of a multitude of species, which have already been studied exclusively on either their breeding or wintering grounds. Unfortunately, there is still a large gap in our understanding of the movement ecology of migrant songbirds and the effect of varying movement strategies on future life-history stages. Most importantly, little is known about the connectivity of breeding and wintering populations (Webster et al. 2002). The introduction of geolocators as a method for tracking songbirds throughout their annual cycle has understandably received considerable attention (Robinson et al. 2009; Bowlin et al. 2010a,b). However, as we proceed with any new method, we must also learn its limitations.

By analysing location estimates for stationary geolocators calibrated to a source population and deployed along an 800 km terrestrial transect through Western Europe, we have identified location estimate error potentially because of landscape and habitat differences. Variation in topography and vegetation density across south-western Europe could result in different intensities of shading at sunrise and sunset. These differences can occur because of slight changes in microhabitat structure and large-scale changes in the landscape. If the light-level threshold used for trajectory calculations is reached at a different sun elevation angle during deployment than identified during calibration owing to differences in shading, error is expected. Alternatively, the variation in accuracy of geolocators along the transect could be due to regional differences in shading by clouds. This is unlikely given that all stationary geolocators were analysed over the same time frame and that the distance between individual geolocators was relatively short (roughly 50 km).

For many migrant species, changes in habitat use throughout the year are expected (Rivera et al. 1998; Pagen, Frank, & Burhans 2000). This is particularly true for long-distance migrants. Estimates of accuracy on breeding grounds might not reflect accuracy at wintering grounds if animals change habitat seasonally, such that sun elevations calculated from breeding habitat bias locations estimated in wintering habitat. To better interpret winter location estimates for migratory wood thrush, a species which has high site fidelity to their winter territory, Stutchbury et al. (2011) compared the wintering locations of individual wood thrushes over multiple years.

The average winter location error for stationary geolocators that we report for latitude (201 ± 43 km) is similar to the error reported previously for geolocators in forest habitat during the spring and summer; the average winter location error for longitude (12 ± 3 km) reported is considerably lower than previous estimates (Stutchbury et al. 2009, 2011). However, we should be cautious when interpreting results from geolocators not attached to live birds, as presented as a part of this study, because of potential influences of behaviour on location estimates. The most realistic measure of error is recorded from animals at known locations, as Stutchbury et al. (2009, 2011) did for the summer months only, when their individuals were still on the breeding grounds. In this study, we found lower average error for latitude estimates derived from geolocators deployed on live birds than stationary geolocators. Latitude estimates are sensitive to differences in shading between deployment and calibration. Non-migratory blackbirds roost in similar habitat throughout their annual cycle (A. Fudickar, pers. obs.) and therefore experience relatively similar shading throughout the year. However, migratory birds could experience a multitude of different levels of shading from stopover to wintering grounds. The results from our 800 km transect might reflect the potential variation in shading experienced by a bird over a relatively short distance.

We found slightly lower average winter error for latitude estimates derived from winter calibrations than estimates derived from summer calibrations for sedentary blackbirds. This difference in error could be due to seasonal changes in forest shading. Additionally, the decrease in error could be partially due to behavioural differences between the breeding season and the winter, which are compensated for when calibrations are performed during the winter. It is important to note that for long-distance migrants, potentially experiencing a much broader range of habitats and climates, larger seasonal differences in sun elevations should be expected. Error reduction should be possible for long-distance migrants by calibrating geolocators in typical wintering habitat for the focal species. Stutchbury and co-authors are now performing winter calibrations on wood thrush in tropical forests in Costa Rica and Belize to better estimate the winter territory locations of birds that breed in temperate deciduous forests in eastern Canada and the United States (B. Stutchbury pers. comm.).

There was a sharp increase in the CV of latitude estimate error for geolocators deployed on blackbirds in August and February (Fig. 2a). In August, blackbirds in southern Germany transition from breeding to moulting (Parteke, Van’t Hof, & Gwinner 2005). During this transition, blackbirds often change their daily foraging behaviour. During breeding, blackbirds leave the roost in early morning to sing from the tops of trees and fly to the forest edge to forage. During moult, blackbirds are inconspicuous and forage in dense vegetation most likely to avoid predators during a period of high vulnerability. During February, resident blackbirds in southern Germany begin to prepare for the oncoming breeding season (Parteke, Van’t Hof, & Gwinner 2005). Males start to establish territories, and females can begin nest building. These seasonal changes in behaviour could result in variations in location estimate error if individuals are transitioning between behaviours at different times and if these changes in behaviour are accompanied by changes in habitat use. Interestingly, we also report a high CV for longitude estimates in August for blackbirds. Error in longitude estimates results from the archived midday diverging from the actual midday.

As long as satellite transmitters and GPS loggers are limited to large animals, the potential for the use of geolocators to track long-distance songbird migrants is astounding given how little we actually know about these seasonal movements. However, we think that this technology could also be used for identifying seasonal movements of medium- and short-distance migrants (400–1000 km). Limiting location estimates to data collected during months with lowest error (December and January) could provide the resolution to distinguish differences in overwintering sites within a relatively short range. Further, researchers interested in studying the movements of individuals with seasonal movements directed at least partially east or west could benefit from the low error inherent in longitudinal estimates. Migratory barriers that hinder direct northerly or southerly movements, such as the Himalayas, can create such migratory pathways.

To optimize future studies utilizing geolocator technology, we recommend tests to address methods for reducing error, such as those currently being performed by Stutchbury and colleagues mentioned earlier. Ideally, future tests would also be conducted on forest-dwelling migratory species during migration. One approach would be to combine satellite telemetry or GPS loggers with geolocators to compare estimated locations. Given the rapid decrease in the mass of gps loggers, this might be possible on the largest songbirds very soon. Although the monetary expense would be high, additional verification of this method would be valuable for the future of movement ecology. We are optimistic that with increased effort to verify and optimize this powerful tracking technology, long-standing questions about songbird migration, and thus our understanding of the phenomenon of migration in general, can be addressed.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank Catarina Miranda, Davide Dominoni and Carrie Fudickar for help capturing birds and deploying geolocators and Wolfgang Fiedler for help building the maps. Jeff Kelly, Carlos David Santos and Sjouke Kingma provided helpful comments on previous versions of the manuscript. We especially thank Bridget Stutchbury and Heiko Schmaljohann for providing very thoughtful and thorough reviews of the manuscript. Funding was provided by the Max Planck Institute for Ornithology, Department of Migration and Immuno-Ecology.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Bowlin, M.S., Henningsson, P., Muijres, F.T., Vleugels, R.H.E., Liechti, F. & Hedenström, A. (2010a) The effects of geolocator drag and weight on the flight ranges of small migrants. Methods in Ecology and Evolution, 1, 398402.
  • Bowlin, M.S., Bisson, I., Shamoun-Baranes, J., Reichard, J.D., Sapir, N., Marra, P.P., Kunz, T.H., Wilcove, D.S., Hedenstrom, A., Guglielmo, C.G., Akesson, S., Ramenofsky, M. & Wikelski, M. (2010b) Grand challenges in migration biology. Integrative and Comparative Biology, 50, 261279.
  • Croxall, J.P., Silk, J.R.D., Phillips, R.A., Afanasyev, V. & Briggs, D.R. (2005) Global circumnavigations: tracking year-round ranges of nonbreeding albatrosses. Science, 307, 249250.
  • Daan, S. & Aschoff, J. (1975) Circadian-rhythms of locomotor activity in captive birds and mammals: their variations with season and latitude. Oecologia, 18, 269316.
  • Egevang, C., Stenhouse, I.J., Phillips, R.A., Petersen, A., Fox, J.W. & Silk, J.R.D. (2010) Tracking of Arctic terns Sterna paradisaea reveals longest animal migration. Proceedings of the National Academy of Sciences, USA, 107, 20782081.
  • Ekstrom, P.A. (2004) An advance in geolocation by light. Memoirs of the National Institute of Polar Research, 58, 210226.
  • Hill, R.D. (1994) Theory of geolocation by light levels. Elephant Seals: Population Ecology, Behavior, and Physiology (eds J. Burney, B.J. Boeuf & R.M. Laws), pp. 227236. University of California Press, Berkley.
  • Hill, I.F. & Cresswell, B. (1997) The use of a communal summer roost by radio-tagged blackbirds Turdus merula. Bird Study, 44, 114116.
  • Marra, P.P., Hobson, K.A. & Holmes, R.T. (1998) Linking winter and summer events in a migratory bird by using stable-carbon isotopes. Science, 282, 18841886.
  • Naef-Daenzer, B. (2007) An allometric function to fit leg-loop harnesses to terrestrial birds. Journal of Avian Biology, 38, 404407.
  • Newton, I. (2008) The Migration Ecology of Birds. Academic Press, London.
  • Pagen, R.W., Frank, R.T. & Burhans, D.E. (2000) Breeding and post-breeding habitat use by forest migrant songbirds in the Missouri Ozarks. Condor, 102, 738747.
  • Parteke, J., Van’t Hof, T.J. & Gwinner, E. (2005) Underlying physiological control of reproduction in urban and forest-dwelling European blackbirds Turdus merula. Journal of Avian Biology, 36, 295305.
  • Rivera, J.H.V., McShea, W.J., Rappole, J.H. & Haas, C.A. (1998) Pattern and chronology of prebasic molt for the wood thrush and its relation to reproduction and migration departure. Wilson Bulletin, 110, 384392.
  • Robinson, W.D., Bowlin, M.S., Bisson, I., Shamoun-Baranes, J., Thorup, K., Diehl, R.H., Kunz, T.H., Mabey, S. & Winkler, D.W. (2009) Integrating concepts and technologies to advance the study of bird migration. Frontiers in Ecology and the Environment, 8, 354361.
  • Sillett, T.S. & Holmes, R.T. (2002) Variation in survivorship of a migratory songbird throughout its annual cycle. Journal of Animal Ecology, 71, 296308.
  • Stutchbury, B.J.M., Tarof, S.A., Done, T., Gow, E., Kramer, P.M., Tautin, J., Fox, J.W. & Afanasyev, V. (2009) Tracking long-distance songbird migration by using geolocators. Science, 323, 896896.
  • Stutchbury, B.J.M., Gow, E.A., Done, T., MacPherson, M., Fox, J.W. & Afanasyev, V. (2011) Effects of post-breeding moult and energetic condition on timing of songbird migration into the tropics. Proceedings of the Royal Society B, 278, 131137.
  • Webster, M.S. & Marra, P.P. (2005) The importance of understanding migratory connectivity and seasonal interactions. Birds of Two Worlds: Ecology and Evolution of Migration (eds R.S. Greenberg & P.P. Marra), pp. 199209. Johns Hopkins University Press, Baltimore.
  • Webster, M.S., Marra, P.P., Haig, S.M., Bensch, S. & Holmes, R.T. (2002) Links between worlds: unraveling migratory connectivity. Trends in Ecology and Evolution, 17, 7683.
  • Wilcove, D.S. (2008) No Way Home: The Decline of the World’s Great Animal Migration. Shearwater Books, Washington, DC.
  • Wilcove, D.S. & Wikelski, M. (2008) Going, going, gone: is animal migration disappearing. PLoS Biology, 6, e188.
  • Wilson, R.P., Ducamp, J.J., Rees, G., Culik, B.M. & Niekamp, K. (1992) Estimation of location: global coverage using light intensity. Wildlife Telemetry: Remote Monitoring and Tracking of Animals (eds I.G. Priede & S.M. Swift), pp. 131134. Ellis Howard Ltd, Chichester.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Fig. S1. Monthly mean of location estimates with 95% CI for six (Panel a–f) sedentary European blackbirds. All birds were observed at the breeding site (square) in Southwest Germany throughout the entire deployment.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
MEE3_136_sm_FigS1.doc110KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.