The Mysteries of Bird Migration – Still Much to Be Learnt

Home , Bird ringing, Northern wheatear, Wheatear

The mysteries of bird migration – still much to be learnt Franz Bairlein ABSTRACT Bird ringing has unveiled many mysteries of avian migrations, notably routes and destinations. However, there is still much to be explored by the use of ringing and other marking techniques. Satellite tracking, geolocation and global positioning systems are new tools, as well as particular chemical and molecular markers which appear to be very useful in the study of bird migration by delineating origin of birds and connectivity between breeding and non-breeding grounds. Understanding of bird migrations also gained much from captive studies about the internal mechanisms in the control of bird migration, but we still lack knowledge about external factors, such as food availability, weather, competitors, parasites or diseases.This paper summarises ongoing studies on Northern Wheatears Oenanthe oenanthe to illustrate the benefit of such an integrated approach. Future migration research must aim much more at comparative research and a more integrated approach at various spatial and temporal scales, and linking various sub-disciplines. It is also important to realise that migration is only one part of the life-cycle of a migrating species.Thus, linking migration and breeding is another future challenge, for both basic science and conservation of migratory birds.

or centuries, the seasonal arrival and The study of bird migration by ringing departure of bird species to and from Our knowledge of bird migration improved Ftheir breeding grounds remained a dramatically with the development of bird mystery. Although these events were described ringing, first practised by the Danish school by early observers such as the wall and floor teacher Hans Christian Cornelius Mortensen in painters of ancient Egypt, Aristotle and the 1899 (Jespersen & Tåning 1950; Bairlein 2001). Emperor Friedrich von Hohenstaufen (who For the first time, birds were ringed in a con- recognised that, in the northern hemisphere, certed effort to unravel the mystery of their birds moved south in autumn and north in movements, although bird ringing had occa- spring), very little information was available sionally been used for that purpose in earlier about particular origins or destinations, times (Bairlein 1999). After Mortensen, system- although there were some intriguing clues. One atic ringing for the study of bird migration was example concerns a female White Stork Ciconia first introduced by Johannes Thienemann in ciconia that was observed and later shot at its 1903, at the newly founded (in 1901) ‘Vogel- nest in Mecklenburg, northeast Germany, on warte Rossitten’ on the Courish Spit, on the east 21st May 1822; embedded in the bird was an shore of the Baltic (Stresemann 1951). 80-cm long spear of the type used by tribes in Although ringing was much criticised by animal central Africa (plate 48; Kinzelbach 2005). This welfare protestors at that time, its development was the first evidence that migrating White for the study of migration continued, and with Storks winter in tropical Africa. great success. Also in 1903, bird ringing began

68 © British Birds 101 • February 2008 • 68–81 The mysteries of bird migration at the Hungarian Centre for Ornithology. In showing some 9,200 recoveries of 151 species. 1909, Hugo Weigold started ringing birds on After about 100 years of bird ringing, various the island of Helgoland, Germany, while H. F. migration atlases have now been published, Witherby and A. Landsborough Thomson first either nationally (Yamashina Institute for introduced ringing in Great Britain that same Ornithology 1996; Fransson & Pettersson 2001; year. It was also in 1909 that the first birds were Wernham et al. 2002; Bakken et al. 2003 & ringed in the USA, but systematic ringing in 2006; Bønløkke et al. 2006) or internationally North America started only in 1920, with the (Zink 1973–1985; McClure 1974; Zink & Bair- collaboration of the US Fish and Wildlife lein 1995); these and the many papers which Service and the Dominion Wildlife Service of have analysed recoveries of single species or Canada. Organised ringing began in Switzer- species groups have unveiled many of the land in 1911, in Sweden in 1912, in The Nether- former mysteries of bird movements (Bairlein lands and France in 1914, and in Finland in 2001). 1916. Subsequently, bird ringing gradually The hundreds of thousands of recoveries of developed as a routine technique used by avian ringed birds are supplemented by abundant scientists worldwide and many countries observational data on the occurrence and founded ‘Ringing Centres’. spatial and temporal distribution patterns of Since migrating birds ignore political migratory species (e.g. Glutz von Blotzheim & boundaries, international collaboration in the Bauer 1966–1998, Moreau 1972, BWP, Keast & study of bird migration was essential. Conse- Morton 1980, Curry-Lindahl 1981, Brown et al. quently, in 1963, the national Ringing Centres 1982–2000, Hagan & Johnston 1992, Poole et al. in Europe founded the ‘European Union for 1992–2002, Rappole et al. 1995, Greenberg & Bird Ringing’ (EURING); they agreed on a Marra 2005, Wisz et al. 2007). Consequently, the common code to computerise ringing and annual movements of many bird species, recovery data, and to gather recovery data in a together with their non-breeding distribution, centralised database. The EURING database at least for birds breeding in the northern hemi- was established and maintained at the Dutch sphere, are fairly well known. Ringing Centre in Heteren until 2005, when it was moved to the BTO in Thetford, Norfolk. Currently, the EURING database contains details of some 2.3 million recoveries (Chris du Feu pers. comm.), an extraordinary resource for the analysis of bird movements (for details see www.euring.org). Such an effort would not have been possible without the many enthusiastic volunteer ringers whose spare-time activities are so important for avian science. These volunteers are trained to extremely high levels, the training being co-ordinated by national ringing centres, and participate in targeted scientific projects. This degree of collaboration between profes- sional and amateur ornithologists is unique among biological sciences worldwide. For many decades, the major interest in and objective of bird ringing was to understand the migration routes and non-breeding distribution of birds. As early as 1910, Thienemann published the first 35 recoveries of ringed White Storks, while, in 1929, von Lucanus compiled several hundred recoveries of 127 species and identified the major migration Courtesy of Prof. Dr R. Kinzelbach, Rostock, Germany routes of European birds. The first ‘Migration 48. White Stork Ciconia ciconia carrying central- Atlas’ was published by Ernst Schüz and Hugo African spear.This individual was shot at its nest in Weigold in 1931, and contained 262 maps Mecklenburg, northeast Germany, in May 1822.

British Birds 101 • February 2008 • 68–81 69 The mysteries of bird migration

New techniques In addition to electronic wizardry, recently In recent years, new techniques have supple- established chemical and molecular markers mented bird ringing and even widened its scope may be used to establish the origin of migrants for establishing migratory routes (Bairlein 2003). and to delineate bird migration routes (Webster One of the most widely used of the new method- et al. 2002). The earth’s surface varies in its ologies for tracking the routes of individual chemical composition. Through diet, birds migrants is that of satellite telemetry, and carry a signature of that chemical composition numerous studies have been conducted. This in their tissues. Stable isotopes are found to enables a much more detailed spatial and tem- function as natural markers and provide new poral resolution of avian migrations and helps to insight into the location histories of highly identify migratory routes, stopover sites and win- mobile animals by delineating the origin of tering grounds, especially of birds for which com- birds feeding in areas where diets differ in paratively few recoveries are available or could be isotope composition (e.g. Hobson & Wassenaar obtained, such as larger or rare species. The tech- 1997, Alisauskas et al. 1998, Bensch et al. 1999, nique has so far been applied only to compara- Hobson 1999, Chamberlain et al. 2000, Ruben- tively large species (e.g. storks (Ciconiidae), stein et al. 2002, Hobson 2003, Lott et al. 2003, cranes (Gruidae), geese (Anatidae), raptors), Bearhop et al. 2005, Yohannes et al. 2007). Simi- owing to the weight of the transmitters, but larly, trace-element composition of plumage miniaturisation of transmitters and improved can be used to identify the origins of migrating receiver sensitivity is likely to enable application birds (e.g. Parrish et al. 1983, Szép et al. 2003). to smaller species. Geolocation (GLS) and Global Positioning System (GPS) are two other new tools Innate migratory behaviour to track migrating birds on a worldwide scale In recent decades much has also been revealed (von Hünerbein et al. 2000; Weimerskirch & about the endogenous control of avian migra- Wilson 2000; Gauthier-Clerc & Le Maho 2001; tions (for reviews see Alerstam 1990, Gwinner Wilson 2001). Geolocation is based on real-time 1990, Berthold 1996, 2001, Bairlein 2002, Bair- measurement of ambient light intensity to deter- lein et al. 2002, Berthold et al. 2003). For their mine geographic co-ordinates, while GPS receives first outbound migration, young migrants data from satellites for calculating a bird’s posi- appear to be equipped with an innate knowl- tion. Initially, these techniques required an edge about timing, distance, direction and ener- archival tag on the bird to collect the data and getic demands. They are capable of finding their subsequent recapture of the bird and recovery of way by using external means of orientation – the logger. However, recent developments to link the sun, the stars, or the earth’s magnetic field GLS and GPS to satellite transmitters allow the (e.g. Wiltschko & Wiltschko 2002, 2003). stored data to be downloaded without recapture. The vast quantities of data gathered during a Biotelemetry and bio-logging have become chal- century of modern bird migration research lenging new tools in the study of bird movements might suggest that there is not much left to be and bird behaviour (for reviews see Cooke et al. explained. However, there is still much to be 2004, Ropert-Coudert & Wilson 2005). explored with respect to migration routes and

food physiology habitat weather innate realised migration migration template other species time conspecifics competition season predation age parasites

Fig. 1. A complex set of factors is involved in shaping an innate migration template into realised migration. The order of the factors does not imply a hierarchy of relevance.

70 British Birds 101 • February 2008 • 68–81 The mysteries of bird migration the distribution of migrants, migration systems, habitats are scarce or distributed patchily, or winter ecology of migrants, the integration of because ecological barriers like oceans or deserts migration in the annual cycle of a migratory hinder resting and/or refuelling. Thus, prior to species, and life-history aspects of migration embarking on a flight across an ecological barrier, (Bairlein 2003; Piersma et al. 2005). The birds have to prepare for a long-distance flight by remainder of this paper will focus on a case intense fuelling. For successful fuelling, birds rely study with Northern Wheatears Oenanthe on an appropriate supply of food (in terms of oenanthe, which illustrates that bird migration quality as well as quantity; Bairlein 2002). During research is still a dynamic subject. a migratory stopover, a bird does not necessarily find the kind of habitat which fits all its require- The factors affecting migration – a case study ments. Consequently, after landing, a bird has to with Northern Wheatears establish whether conditions at a site are suffi- In order to understand migration and to reveal cient for refuelling, and it must evaluate the pros different migration strategies, we need to learn and cons of staying there or continuing in the more about the external factors which affect hope of finding better habitat elsewhere. If a bird migration and which mould an innate template does not find adequate conditions for refuelling into actual migration patterns (fig. 1). and/or surviving at a given site, it should leave Recent theories predict that, in order to opti- quickly and this decision is particularly impor- mise their migration, birds should minimise tant for birds facing an ecological barrier. Suc- either the time spent on migration or their total cessful migration also involves predator energy expenditure, and that predation risk is a avoidance and the ability to interpret weather further criterion to be considered (Alerstam & conditions at take-off, but studies examining Lindström 1990; Alerstam & Hedenström these complex inter-relationships are scarce. 1998). While migrating, birds spend about 90% In order to investigate the effects of immi- of the entire migration period at stopovers in nent long-distance flights on stopover behav- order to store or to replenish fuel for the next iour and departure decisions, researchers at the flight (Hedenström & Alerstam 1997); the flight Institute of Avian Research are studying the itself is of only minor importance in terms of Northern Wheatear. This nocturnal, long-dis- time. Consequently, understanding stopovers tance migrant has a nearly circumpolar distri- and how birds adjust stopover decisions with bution and a fascinating migration system (fig. respect to their migration strategy is crucial to 2). On migration it occurs in a variety of an understanding of how migrating birds organise their journey. For migratory birds, the timing of their arrival on the breeding grounds (in relation to their competitors) is an important factor affecting breeding success (e.g. Currie et al. 2000, Smith & Moore 2005). If birds are under time pressure, the rate of fuel deposition and the bird’s departure fuel load are the two major factors affecting departure decisions (Alerstam & Lindström 1990). Birds maximising the speed of migration (i.e. minimising the time spent on migration) should leave a stopover site quickly if food is not easily available – in theory at the point when fuel deposition rate is too low for the bird to reach the expected average speed of migration for the whole journey (Alerstam & Lindström 1990). However, models of optimal migration are based on the assumption that suitable stopover sites are available all along the migration route and that a bird may make a stopover whenever it wants to. Fig. 2. Schematic representation of possible In reality, many species are restricted in their migration routes of Northern Wheatears Oenanthe choice of stopover sites, either because suitable oenanthe. Distribution map from Cramp (1988).

British Birds 101 • February 2008 • 68–81 71 The mysteries of bird migration

Greenland and Iceland (Dierschke & Delingat 2003). Whereas Scandi- navian birds face sea cross- ings of only 50–150 km when heading northeast towards Schleswig-Holstein or Denmark, or a maximum of 500 km when flying to southern Norway, much longer flights are necessary for Greenland/Icelandic birds to reach stopover sites in Scotland (c. 1,000 km) or their breeding areas (up to 2,500 km). Because such Franz Bairlein Franz long flights require suffi- 49. Spring trap with male Northern Wheatear Oenanthe oenanthe. cient preparations, one hypothesis is that leucorhoa adjust their stopover behaviour more carefully for intense fuel deposition than do oenanthe before their (relatively) short-distance flights. Furthermore, leucorhoa would be expected to be more selec- tive in terms of weather conditions at departure because strong headwinds and orientation errors would have a much greater impact on long-distance flights than short-distance ones. As a bird of open landscapes, Northern Wheatear is a convenient study species that can be easily trapped using baited spring traps (plate 49) and many individuals can be identified to subspecies in the hand (Svensson 1992).

H. Schmaljohann Once colour-ringed, they are easy to observe at 50. Colour-ringed Northern Wheatear Oenanthe stopover sites owing to their habitat choice and oenanthe perched on a digital balance placed visibility. Moreover, they can be attracted to in the field. remote-controlled baited balances placed in lowland habitats including meadows, arable their habitats (plate 50) so that data on refuel- land, beaches and other habitats with sparse ling can be gathered without retrapping the vegetation (Cramp 1988; Glutz von Blotzheim birds. Helgoland is a small island of some 150 & Bauer 1988). In the northern breeding range, ha in the southeastern North Sea (54°11’N two subspecies are distinguished, both of which 07°55’E), 53 km off the mainland coasts of overwinter in Africa. Nominate oenanthe breeds Schleswig-Holstein and Lower Saxony and 43 in Great Britain and in an area ranging from km from the nearest Wadden Sea island of continental Europe via Siberia as far east as Wangerooge (also part of Germany), which Alaska (Cramp 1988). ‘Greenland Wheatear’ offers various stopover habitats for Northern O. o. leucorhoa breeds in Iceland, Greenland and Wheatears. Apart from the village and some eastern Canada and is one of the few passerine small bushy areas, most of the island is open migrants regularly covering distances of more habitat and Northern Wheatears generally than 1,000 km over sea. occur in two main habitat types: sandy beaches During both autumn and spring migration, with beds of rotting brown algae, with kelp flies the two subspecies occur together at stopover (Coelopidae) and their larvae as the only (but sites in northern and western Europe, including abundant) food supply; and grassland habitats Helgoland, where oenanthe of Scandinavian with interspersed open patches and boulders, origin mingle with leucorhoa breeding in with various ground-dwelling arthropods for

72 British Birds 101 • February 2008 • 68–81 The mysteries of bird migration food (Delingat & Dierschke 2000). These cir- density (birds/ha) of wheatears was generally cumstances and the general habits of Northern much higher in beach habitats. Wheatears make the entire set-up unique for Individually colour-ringed birds revealed examining the stopover behaviour and deci- different patterns in utilisation of the two habi- sions of a migrant species. tats. Grassland birds were transient and explorative and most individuals departed on Phenology the day of arrival; they were characterised by On Helgoland, Northern Wheatears occur from high mobility on the island, having significantly late March to early June, peak spring migration larger dispersal distances than beach birds. In being in early May, and from late July to early contrast, a higher proportion of beach birds November, peak autumn migration being stayed for at least one night and remained between late August and mid September (Dier- rather stationary, often exhibiting territorial schke & Delingat 2003; Dierschke et al. 2005). behaviour, which was rare in grassland birds. Compared with nominate oenanthe, leucorhoa Territories on beaches contained patches of migrate earlier in spring and later in autumn. In wrack (up to 250 m2) and these were defended spring, the median dates of trapped birds are against conspecifics and other passerines. On 2nd May for leucorhoa and 7th May for beaches, 34% of the wheatears showed aggres- oenanthe; in autumn, the comparative dates are sive encounters, but only 3% did so in grass- 11th September and 31st August respectively. In land. Wheatears do not seem to settle in spring, males migrate earlier than females in grassland; instead they switch to the more prof- both subspecies, although significantly so only itable beaches if they remain on the island. A in leucorhoa, for which the median date is 27th greater proportion of Northern Wheatears April for males and 4th May for females. stayed in the beach habitats and stopovers were During the early part of autumn migration, longer on beaches than in grassland (Dierschke including the first peak in late August/early Sep- 2003). These data clearly reveal the role of tember, oenanthe occurs almost exclusively, habitat quality in stopover decisions. while from mid September onwards leucorhoa is more prominent and outnumbers oenanthe in Length of stopover and body condition the latter part of the autumn. Food availability at stopover sites and the resulting gain in body mass per day (fuel depos- Habitat use ition rate) are expected to play a major role in As described above, two different habitats are optimal behaviour decisions. Time-minimising available to migrant Northern Wheatears on migrants aim to feed as quickly as they can and Helgoland: beach and grassland. In terms of move on rapidly, and would thus be expected to their suitability for stopover wheatears, they show a positive correlation between fuel depos- differ in various respects. In spring, the two ition rate and departure fuel loads; while habitats were used equally but in autumn an energy-minimising migrants will remain at a increasing proportion of wheatears in beach stopover site for as long as it takes to reach their habitats was attributed to declining food sup- optimum fuel load for the next leg of the plies in grassland habitats compared with wrack journey, regardless of fuel deposition rate (Aler- beds on beaches with their abundant kelp flies stam & Lindström 1990). (Delingat & Dierschke 2000). The proportion of In spring, the two subspecies of Northern wheatears found in grassland was significantly Wheatear on Helgoland differ in the respective correlated with the number of invertebrates col- proportion of birds staying on the island, and lected in net sweeps; and the latter measure their length of stopover. In oenanthe, 9% of declined during autumn migration. Conversely, males and 14% of females did not depart on the the proportion of wheatears observed in beach day of ringing, while in leucorhoa 40% of males habitats increased with estimated abundance of and 30% of females stayed on the island for at kelp fly larvae. In spring, pecking rates tended least one day. However, a greater proportion of to be higher in grassland (6.4 pecks/two-minute early migrating oenanthe stayed compared with period in grassland compared with 4.4 on the late migrating oenanthe (Dierschke & Delingat beach), but were significantly higher on the 2001). For birds which stayed on the island, the beach in autumn (9.8 on beaches versus 6.5 in length of stopover did not differ significantly grassland; Delingat & Dierschke 2000). The between the subspecies, although many

British Birds 101 • February 2008 • 68–81 73 The mysteries of bird migration

25 leucorhoa (26.8 ± 5.7 g, n=138) than in oenanthe (24.4 ± 2.1 g, 20 n=210). When foraging in beach habitats, wheatears gained mass 15 by 1.7 g per day in spring (fig. 3) and 1.8 g per day in autumn, 10 which is close to the maximum rate of mass increase in a 5 passerine of that size (Lindström body mass change (g) 1991). Since leucorhoa stay 0 2 4 6 8 10 12 14 16 longer, the difference in body -5 length of stopover (days) mass between the two subspecies at departure is much larger than Fig. 3. Body mass gain of Northern Wheatears Oenanthe oenanthe on on the day of ringing. Helgoland during spring migration (from Dierschke & Delingat 2001). oenanthe stayed for only one day, while most of Fuel deposition rate and departure fuel load the long-stayers were leucorhoa. Changes in body mass during stopovers are At first capture, most oenanthe showed low generally difficult to measure using capture and to moderate fat scores (scores 1–4 on a scale of recapture efforts so, to examine the relationship 0–9; Kaiser 1993), whereas 15% of leucorhoa between food availability, fuel deposition rate were very fat and scored 5–7, although the and departure fuel load, supplementary food average fat scores did not differ significantly (bowls with mealworms) was provided. These between subspecies (Dierschke & Delingat bowls were attached to digital scales and the 2001). Moreover, the breast muscle score (Bair- body mass of colour-ringed birds visiting these lein 1994) did not differ between subspecies. feeders could be read from a distance to the However, standardised body mass (adjusted for nearest 0.1 g, using binoculars or a telescope. size differences, since leucorhoa is on average During the experiment, the scales were larger than oenanthe) was significantly higher in observed daily throughout most daylight hours. During their stay, wheatears used this unlimited food supply 1.2 O. o. leucorhoa males O. o. leucorhoa females for fuel deposition. The relation- 1.0 O. o. oenanthe both sexes ship between fuel deposition rate and departure fuel load in spring 0.8 differed between the two subspecies, and tended to differ also 0.6 between males and females in leucorhoa (Delingat et al. 2006; 0.4 fig. 4). For male leucorhoa, the departure fuel load relationship between fuel deposi- 0.2 tion rate and departure fuel load is strongly positive (and almost 0.0 -0.06 -0.02 0.02 0.06 0.10 0.14 0.18 0.22 statistically significant), so this fuel deposition rate group can be considered time- minimisers (Alerstam & Lind- Fig. 4. Relationship between departure fuel load (in relation to lean body mass) and total daily fuel deposition rate in Northern Wheatears ström 1990). Female leucorhoa Oenanthe oenanthe on Helgoland in spring (after Dierschke et al. 2005, tended to show a weaker rela- Delingat et al. 2006). Departure fuel load and fuel deposition rate are tionship between fuel deposition both relative measures – see p. 75. Migrants which are energy-minimisers rate and departure fuel load, would be expected to show a flat line in this graph because, irrespective reflecting a compromise between of feeding rate, they keep feeding until they have reached their optimal weight for the next stage of migration. In contrast, time-minimisers time- and energy-minimising; would be expected to show a positive relationship between these two while oenanthe leave the island at variables as they feed as fast as they can before moving on. Hence, as a particular departure fuel load, the results here show, oenanthe are energy-minimisers, male leucorhoa irrespective of fuel deposition are time-minimisers, while female leucorhoa show a compromise between the two strategies. rate, an energy-minimising

74 British Birds 101 • February 2008 • 68–81 The mysteries of bird migration strategy. But female leucorhoa leave the island compared with females, males were more often with a higher fuel load than oenanthe, which dominant at the feeding stations or held territo- reflects the difference in onward migration dis- ries, refuelling patterns could not be explained tance. by dominance. Subordinate or non-territorial The two subspecies are able to accumulate birds did not refuel at a lower rate or depart fuel at the same rate (fuel deposition rate; with lower fuel loads than dominant or territo- Delingat et al. 2006) but show significant differ- rial birds. In non-territorial birds, the restricted ences in their departure fuel loads. Most leu- access to feeding stations was compensated by corhoa left the island with a much higher fuel larger doses of food taken per visit, leading to load compared with their arrival, but male the same energy intake as that of dominant and oenanthe (there are few data for females) territorial birds. Therefore, competition during increased their stores only slightly (Dierschke et stopover could be eliminated as the reason for al. 2005; fig. 5). The rate of refuelling for the differential timing of migration of males and whole stay was similar in both male (0.133 females (of course, this result may be species- g/day) and female (0.135 g/day) leucorhoa,but specific). lower in male oenanthe (0.083 g/day). (Note that fuel deposition rate as given here is a rela- Time allocation tive measure of body mass gain, calculated by In order to identify factors which influence time dividing body mass gain by lean body mass. budgets, and thus possibly limit the refuelling of Lean body mass is the estimated body mass passerine migrants, the time allocation of without visible fat – see Delingat et al. 2006.) In Northern Wheatears on Helgoland was exam- leucorhoa with a stopover length of more than ined (Dierschke et al. 2003). Full-day observa- two days, departure fuel load ranged from 0.497 tions revealed that Northern Wheatears on to 1.102 in males (mean 0.856) and from 0.554 stopover spent 51–67% of the daylight period to 0.828 in females (mean 0.695) (note that foraging. Large parts of the day were also spent departure fuel load is also a relative measure: a resting or being vigilant, whereas flying, fuel load of 0.5 means that the bird’s weight is preening and aggressive behaviour were of 50% above lean body mass). Departure fuel minor importance. The density of wheatears load was not correlated with wing length (body did not influence the time devoted to foraging size) nor with a dominance index which reflects and aggressive behaviour, and the time spent the wins and losses in intraspecific aggressive resting/being vigilant was not correlated with interactions observed at the mealworm bowls (Dierschke et 0.8 al. 2005). Departure fuel load in female leucorhoa was lower than 0.7 for males, but nonetheless sufficient to enable them to bypass 0.6 stopover sites en route. Thus, time selection seems to be more 0.5 pronounced in males and may be the reason why males migrate 0.4 earlier. However, females are not fuel load able to reach Greenland without 0.3 additional refuelling elsewhere, which supports the idea that 0.2 female leucorhoa adopt a compromise strategy of time- and 0.1 energy-minimising. Intraspecific 0 aggressive interactions between leucorhoa males leucorhoa females oenanthe males colour-ringed birds were predominantly won by the initiator, Fig. 5. Arrival fuel load (white) and departure fuel load (grey) of by males and by larger birds; fuel subspecies and sex classes of Northern Wheatear Oenanthe oenanthe load and subspecies did not on Helgoland in spring, for birds staying at least two days (after Dierschke et al. 2005). Departure fuel load is a relative measure affect the outcome. Although, – see text, above left.

British Birds 101 • February 2008 • 68–81 75 The mysteries of bird migration predation risk (measured by an index of fly- lasts 1–7 minutes, but can last up to 33 minutes over raptors). Several observations showed that after a raptor flight. Predation risk was assessed refuelling on beach habitats, which presented by recording all raptors posing a threat to the most favourable feeding conditions and passerines. The daily threat from raptors fluctu- allowed high rates of body mass gain, was meta- ated between 0 and 4.7 raptor flights per hour, bolically limited. The total time devoted to for- with a maximum of 53 flights per day. Eurasian aging was independent of day length, and Sparrowhawk Accipiter nisus was the most supplementary food (mealworms) was com- abundant species, accounting for 75% of all pletely ignored, indicating that reduced for- flights. aging effort would not improve net energy gain. The results showed that predation risk influ- In the poorer grassland habitat, in contrast, enced the stopover decision. Birds experiencing there was a marked response to supplementary more danger showed lower rates of refuelling food. Although this suggests that refuelling is (fig. 6), indicating that danger indirectly limited by the amount of food available and the affected the stopover pattern via the effect on costs of obtaining it, foraging times were the fuel deposition rate – time-minimisers avoid same as on the beach. In grassland, the behav- stopover sites with low fuel deposition rate iour pattern of birds refuelling was probably when better refuelling conditions may be distorted by a high proportion of transient and expected elsewhere (Schmaljohann & Dierschke explorative individuals. 2005). Lighter birds were more likely to be pre- dated than heavy birds (Dierschke 2003), indi- Predation risk and stopover cating that the role of fuel load with respect to Experiments testing predictions of optimal predation (in theory, heavy birds should be less migration theory have so far concentrated on able to escape from predators as they are less time and energy as the elements that birds strive manoeuvrable) is less important than exposure to minimise during migration. Taking advan- to predators (in theory, birds in poor condition tage of the great variation in numbers of are more vulnerable to predators because they migrating raptors over Helgoland, a field are forced to spend more time foraging and less experiment looking at predation risk as a pos- time being vigilant). sible factor in stopover decisions of migrating In addition to this field experiment, a labo- Northern Wheatears was carried out (Schmaljo- ratory experiment was carried out (Dierschke & hann & Dierschke 2005). Wheatears show time- Walter in prep.). Northern Wheatears were consuming antipredator behaviour: they either trapped under licence and, in the laboratory, stop feeding when detecting birds of prey and were exposed to different degrees of predation try to hide behind or under stones, or they danger using a Sparrowhawk model, and either ‘freeze’ by staying motionless. Freezing usually unlimited or restricted (5 g mealworms) food supply, during which time their 0.12 diurnal and nocturnal activity was recorded. Whether wheatears 0.10 were active or inactive at night, resembling ‘departing’ or 0.08 ‘staying’, was dependent on fuel 0.06 deposition rate, fuel stores, and predation risk. The proportion of 0.04 ‘departing’ birds was significantly

fuel deposition rate higher among those wheatears 0.02 exposed to the Sparrowhawk model than those in the other 0.00 groups. Birds with unlimited 0 0.5 1 1.5 2 2.5 predation risk (raptor flights per hour) food were recorded ‘departing’ to a lesser degree than those with Fig. 6. Correlation between total daily fuel deposition rate and the total restricted food availability; while rate of raptor flights for Northern Wheatears Oenanthe oenanthe staying on Helgoland more than three days beyond the day of arrival (from ‘departing’ birds had higher Schmaljohann & Dierschke 2005, modified). Fuel deposition rate is a evening fuel loads and lower fuel relative measure – see p. 75. deposition rates than ‘staying’

76 British Birds 101 • February 2008 • 68–81 The mysteries of bird migration birds. Moreover, diurnal activity was significantly higher in birds with low food supply than in those with unlimited food. These data suggest that migrating Northern Wheatears try to leave a stopover site with inappropriate fuelling conditions, even during the daytime, despite the fact that they are normally nocturnal migrants.

Weather and departure As well as being influenced by stopover-site characteris- tics, the decision to embark on migratory flight is affected by weather: strong Markus Varesvuo headwinds or drift will 51. Male Northern Wheatear Oenanthe oenanthe of the nominate subspecies, which migrate somewhat later in spring and earlier in autumn compared with increase fuel consumption, the Greenland/Icelandic subspecies leucorhoa; Helsinki, Finland,April 2004. while overcast conditions may compromise orientation. During spring affected by those factors; this may be attributed migration, wind conditions did not seem to to the differences in onward migration flight. play a major role in the departure decisions of In summary, almost all oenanthe departed Northern Wheatears on Helgoland (Dierschke quickly, irrespective of refuelling and weather & Delingat 2001; Dierschke 2006). However, conditions, whereas many (but not all) when a comparison was made between birds leucorhoa seemed to prepare for a long-distance staying or departing, cloud cover was signifi- flight and carefully adjusted departure to cantly greater for the former group (for both weather conditions. subspecies) and the majority of stays coincided with a nearly completely overcast sky. Visibility Departure direction seems to be an important factor in the decision A release experiment was conducted to study to depart, which is in line with previous results the departure direction of Northern Wheatears that visual cues are important in the orientation from Helgoland (Dierschke & Delingat 2003). of migrating birds in general (Åkesson & Wheatears trapped in spring during the day Bäckman 1999). This was further illustrated were caged until the evening (food and water when combining tailwind conditions and cloud were provided in captivity). One hour before cover. The percentage of departing leucorhoa sunset, the cages were exposed to the natural was considerably lower with a completely over- sky. When the sky was completely dark, the cast sky. When both weather variables were birds were equipped with a 0.16-g activated unfavourable, only a few leucorhoa left the green lightstick, taped to the two outermost tail island. By contrast, most oenanthe departed feathers. The birds were then released and their irrespective of weather conditions. Only when departure direction observed and measured to both weather variables were favourable were no the nearest 5° with a compass. Cloud cover, differences between subspecies observed in the wind force and wind direction at release were percentage of departing birds. recorded. These results suggest that factors which are Both subspecies showed the same propor- probably important in the decision to depart or tion of birds departing immediately after release stay differed between subspecies. In leucorhoa, but differed significantly in their departure few birds departed with bad or deteriorating direction. Most of the Scandinavian oenanthe weather conditions (wind and cloud), whereas departed northwards, while Greenland/Ice- departures of oenanthe seemed to be little landic leucorhoa headed predominantly north-

British Birds 101 • February 2008 • 68–81 77 The mysteries of bird migration

N European patterns of Northern Wheatear migration In order to look at spring migration patterns of oenanthe leucorhoa Northern Wheatears on a wider scale, departure fuel loads at eight stopover sites across Europe were recorded, and related to flight distance and optimality models (Delingat et al. W E 2006). Mean fuel loads of wheatears at various stopover sites in western Europe were generally rather low, with variability being highest in the North Sea (Wilhelmshaven, northern Germany, and Fair Isle, as well as Helgoland; fig. 8). Results showed that leucorhoa carried higher S fuel loads than oenanthe; differences were moderate when migrating over land but more Fig. 7. Initial departure directions of Northern Wheatears Oenanthe oenanthe in a release pronounced when approaching the sea crossing experiment on Helgoland in spring (after at Helgoland and on Fair Isle. Individual leu- Dierschke & Delingat 2003). corhoa on Helgoland were recorded with fuel loads of more than 90% of lean body mass west (fig. 7). Departure was affected by the without supplementary feeding. Flight range weather – many more birds departed with a estimates for oenanthe showed that most birds clear sky and departure was then faster than on trapped during migration were probably in evenings with cloud cover. sufficient condition for a ‘night-long’ flight. These data reveal that the southern North Mean fuel loads (see definition on p. 75) were Sea is at the northeastern edge of the flyway of 0.05, which would be sufficient for a 7-hour leucorhoa and one of the last areas where they flight. Fuel loads below 0.11 were shown by switch their northerly migration, which starts in 75% of all oenanthe and, consequently, they western Africa, towards the northwest to reach could fly less than c. 600 km in 15 hours. Only their breeding grounds. A few may even con- the upper 5% showed fuel loads of more than tinue to southern Norway before they switch 0.23, which would provide sufficient energy to direction, but (as ringing recoveries reveal) fly about 1,200 km in 28 hours. In other words, most leucorhoa change their spring migration most birds on migration over the European direction at lower latitudes (Zink 1973; continent deposit sufficient fuel to fly at least a Wernham et al. 2002). few hours each night. Very few birds, especially in southern Europe, deposited 100 sufficient fuel stores to enable them to migrate more than two 90 successive nights without refuel- 80 ling during the day; 95% would 70 have to refuel after one night of 60 migration, before dusk the fol- 50 lowing day at the latest. Flight

fuel load 40 range estimates suggest that 30 wheatears in general refuel every 20 day after nocturnal flights and 10 do not prepare for longer, non- stop flights as long as no signifi- 0 GIB FEU VEN WHV HEL FI cant barrier has to be crossed. Fig. 8. Mean and upper range (value) of departure fuel loads of The data on leucorhoa suggest Northern Wheatears Oenanthe oenanthe trapped on spring migration that this subspecies in general at different sites in southern and western Europe. GIB: Gibraltar; also migrates over continental FUE: Fuentes de Nava, western Spain; VEN: Ventotene, Italy; WHV: Europe using short flights. Only Wilhelmshaven, northern Germany; HEL: Helgoland; FI: Fair Isle (from an imminent barrier crossing Delingat et al. 2006, modified). Not all ringing sites captured both subspecies. Departure fuel load is a relative measure – see p. 75. forces them to deposit large fuel

78 British Birds 101 • February 2008 • 68–81 The mysteries of bird migration loads, exceeding by far those that were observed This study confirmed model predictions for for oenanthe. Regarding the importance of time- and energy-minimised migration, but it arrival date and condition at the breeding also showed that the effect of weather on grounds for migrating passerines, it seems that stopover and departure decisions should not be selection acts on migratory behaviour to favour underestimated and might lead to shorter or a ‘numerous-stops-and-flights strategy’ on longer stopovers under favourable or migration over continental Europe (Delingat et unfavourable weather conditions, respectively, al. 2006). than predicted by the current optimality models (e.g. Alerstam & Lindström 1990, Hedenström Conclusion and perspectives & Alerstam 1997, Alerstam & Hedenström Since their introduction to bird migration 1998, Weber et al. 1999). Moreover, flight route research by Alerstam & Lindström (1990), and length of impending flight have to be taken models of optimal bird migration have been into consideration when modelling optimal tested by very few field studies (Carpenter et al. bird migration. 1983; Lindström & Alerstam 1992; Fransson This study is the first that links manifold 1998a,b). The ongoing study on Northern quantitative field and laboratory studies under Wheatears summarised here is the first of suffi- controlled conditions, so providing a unique cient complexity to deal concurrently with opportunity to investigate the interplay various factors that might be involved in the between internal (genetic) and external factors organisation of migration. It is also the first in the control of avian migration. This will that relies largely on colour-ringed birds and also help to explain more about flexibility in their individually assigned behaviour rather migratory behaviour and the adaptability of than looking in general at non-marked individ- this behaviour to a changing environment – uals. for example, habitat changes on a local, Some of the key findings of this study can be regional and global scale, and climate change. summarised briefly: Climate change is affecting bird migration • Compared with Scandinavian-bound (e.g. Walther et al. 2002, Hüppop & Hüppop oenanthe, Greenland-/Iceland-bound 2003, Bairlein & Hüppop 2004, Thorup et al. leucorhoa migrate earlier in spring and later 2007), although in a very complex manner, in autumn. and as well as affecting the process of migra- • Habitat quality of stopover sites is impor- tion itself, this will have consequences for sub- tant; wheatears did not settle in grassland, sequent breeding and demography (e.g. Both but used the more profitable beaches, where et al. 2006a,b). Migration is an integral part of stopovers were longer. the annual life-cycle and the life-history of a • A greater proportion of leucorhoa than migrant species, and future research should oenanthe stayed on the island for at least one emphasise the relationship between migratory day. performance and reproductive performance, • Male leucorhoa showed a strong positive and vice versa (Bairlein 2003; Drent et al. relationship between fuel deposition rate 2006). These studies would greatly benefit and departure fuel load and can be consid- from improved miniaturisation and avail- ered time-minimisers; female leucorhoa ability of remote data loggers and receiver tended to show a compromise between time- platforms so that many birds could be tagged and energy-minimising; while oenanthe are and tracked, and the connectivity between energy-minimisers. migration and breeding performances evalu- Competition during stopover was not • ated through data based upon individual responsible for the differential timing of animals. migration of males and females. • Predation risk influenced stopover decisions Acknowledgments and birds experiencing more danger showed I wish to thank the entire ‘wheatear group’ at the Institute lower rates of refuelling. of Avian Research very much, namely Julia Delingat,Volker • Few leucorhoa departed in bad or deteri- Dierschke, Ivan Maggini, Bettina Mendel, Heiko orating weather conditions (wind and Schmaljohann and Annegret Walter. Volker Dierschke made valuable comments on an earlier draft of the cloud), whereas departures of oenanthe manuscript. The project is supported by the Deutsche seemed to be little affected by those factors. Forschungsgemeinschaft.

British Birds 101 • February 2008 • 68–81 79 The mysteries of bird migration

References gain and adjustment of feeding territory size in migrant Åkesson, S., & Bäckman, J. 1999. Orientation in Pied hummingbirds. Proc. Natl.Acad. Sci USA 80: 7259–7263. Flycatchers: the relative importance of magnetic and Chamberlain, C. P., Bensch, S., Feng, X., Åkesson, S., & visual information at dusk. Anim. Behav. 57: 819–828. Andersson,T. 2000. Stable isotopes examined across a Alerstam,T. 1990. Bird Migration. Cambridge UP, migratory divide in Scandinavian Willow Warblers Cambridge. Phylloscopus trochilus trochilus and P.t. acredula reflect — & Hedenström, A. 1998.The development of bird their African winter quarters. Proc. Roy. Soc. Lond. B 267: migration theory. J.Avian Biol. 29: 343–369. 43–48. — & Lindström, Å. 1990. Optimal bird migration: the Cooke, S. J., Hinch, S. G.,Wikelski, M., Swedish Andrews, relative importance of time, energy, and safety. In: R. D., Kuchel, L. J.,Wolcott,T. G., & Butler, P.J. 2004. Gwinner, E. (ed.), Bird Migration: physiology and Biotelemetry: a mechanistic approach to ecology. TREE 19: 334–343. ecophysiology: 331–351. Springer-Verlag, Berlin. Cramp, S. (ed.) 1988. Handbook of the Birds of Europe, the Alisauskas, R.T., Klaas, E. E., Hobson, K. A., & Ankney, C. D. Middle East and North Africa.Vol. 5. OUP,Oxford. 1998. Stable-carbon isotopes support use of Currie, D.,Thompson, D.B.A., & Burke,T. 2000. Patterns of adventitious color to discern winter origins of Lesser territory settlement and consequences for breeding Snow Geese. J. Field Ornithol. 69: 262–268. success in the Northern Wheatear Oenanthe oenanthe. Bairlein, F. 1994. Manual of Field Methods. European-African Ibis 142: 389–398. Songbird Migration Network. Institut für Curry-Lindahl, K. 1981. Bird Migration in Africa. Academic Vogelforschung,Wilhelmshaven. Press, London. — 1999. Hundert Jahre wissenschaftliche Vogelberingung: Delingat, J., & Dierschke,V. 2000. Habitat utilisation by Rückblick – Einblick – Ausblick. Der Falke 46: 260–268. Northern Wheatears Oenanthe oenanthe stopping over — 2001. Results of bird ringing in the study of migration on an offshore island during migration. Vogelwarte 40: routes. Ardea 89 (special issue): 7–19. 271–278. — 2002. How to get fat: nutritional mechanisms of —, —, Schmaljohann, H., Mendel, B., & Bairlein, F. 2006. seasonal fat accumulation in migratory songbirds Daily stopovers as optimal migration strategy in a long- (review). Naturwissenschaften 89: 1–10. distance migrating passerine: the Northern Wheatear — 2003.The study of bird migrations – some future Oenanthe oenanthe.Ardea 94: 593–605. perspectives. Bird Study 50: 243–253. Dierschke,V. 2003. Predation hazard during migratory — & Hüppop, O. 2004. Migratory fuelling and global stopover: are light or heavy birds under risk? J.Avian Biol. climate change. Advances Ecol. Research 35: 33–47. 34: 24–29. —, Elkins, N., & Evans, P.2002.Why and how do birds — 2006. Factors determining stopover decisions in migrate? In:Wernham, C.V.,Toms,M. P., Marchant, J. H., migrating passerines on an offshore island. Acta Zool. Clark, J. A., Siriwardena, G. M., & Baillie, S. R. (eds.), Sinica 52 (Suppl.): 594–598. The Migration Atlas: movements of the birds of Britain — & Delingat, J. 2001. Stopover behaviour and departure and Ireland: 23–43. Poyser, London. decision of Northern Wheatears Oenanthe oenanthe Bakken,V., Runde, O., & Tjørve, E. 2003 & 2006. Norsk facing different onward non-stop flight distances. Behav. Ringmerkingsatlas.Vols. 1 & 2. Stavanger Museum, Ecol. Sociobiol. 50: 535–545. Stavanger. — & — 2003. Stopover of Northern Wheatears Bearhop, S., Fiedler,W., Furness, R.W.,Votier, S. C.,Waldron, Oenanthe oenanthe at Helgoland: where do the S., Newton, J., Bowen, G., Berthold, P., & Farnsworth, K. migratory routes of Scandinavian and Nearctic birds 2005. Assortative mating as a mechanism for a rapid join and split? Ornis Svecica 13: 53–61. evolutionary divide. Science 310: 502–504. —, —, & Schmaljohann, H. 2003.Time allocation in Bensch, S., Andersson,T., & Åkesson, S. 1999. Morphological migrating Northern Wheatears (Oenanthe oenanthe) and molecular variation across a migratory divide in during stopover: is refuelling limited by food availability Willow Warblers Phylloscopus trochilus. Evolution 53: or metabolically? J. Ornithol. 144: 33–44. 1925–1935. —, Schmaljohann, H., & Mendel, B. 2005. Differential timing Berthold, P.1996. The Control of Bird Migration. Chapman & of spring migration in Northern Wheatears: hurried Hall, London. males or weak females? Behav. Ecol. Sociobiol. 57: — 2001. Bird Migration: a general survey. OUP,Oxford. 470–480. —, Gwinner, E., & Sonnenschein, E. 2003. Avian Migration. Drent, R., Fox,T., & Stahl, J. (eds.) 2006.Travelling to breed. Springer-Verlag, Berlin. Ardea 94: 299–713. Bønløkke, J., Madsen, J. J.,Thorup, K., Pedersen, K.T., Bjerrum, Fransson,T. 1998a. Patterns of migratory fuelling in M., & Rahbek, C. 2006. The Danish Bird Migration Atlas. Whitethroats Sylvia communis in relation to departure. Forlaget Rhodos A/S & Zoologisk Museum, University J.Avian Biol. 29: 569–573. of Copenhagen. — 1998b. A feeding experiment on migratory fuelling in Both, C., Sanz, J. J., Artemyev, A.V., Blaauw, B., Cowie, R. J., Whitethroats Sylvia communis.Anim. Behav. 55: 153–162. Dekhuisen, A. J., Enemar,A., Järvinen, A., Nyholm, N. E. I., — & Pettersson, J. 2001. Swedish Bird Ringing Atlas.Vol. 1. Potti, J., Ravussin, P.-A.,Silverin, B., Slater, F. M., Sokolov, Naturhistoriska riksmuseet and Sveriges Onritologiska L.V.,Visser, M. E.,Winkel,W.,Wright, J., & Zang, H. 2006a. Förening, Stockholm. Pied Flycatchers Ficedula hypoleuca travelling from Gauthier-Clerc, M., & Le Maho,Y.2001. Beyond bird Africa to breed in Europe: differential effects of winter marking with rings. Ardea 89 (special issue): 221–230. and migration conditions on breeding date. Ardea 94: Glutz von Blotzheim, U., & Bauer, K. M. 1966–1998. 511–525. Handbuch der Vögel Mitteleuropas. AULA-Verlag, —, Bouwhuis, S., Lessells, C. M., & Visser, M. E. 2006b. Wiebelsheim. Climate change and population declines in a long- Greenberg, R., & Marra, P.P.2005. Birds of Two Worlds. distance migratory bird. Nature 441: 296–298. Smithsonian,Washington D.C. Brown, L. H., Urban, E. K., & Newman, K. 1982–2000. The Gwinner, E. 1990. Bird Migration: physiology and Birds of Africa. Academic Press, London. ecophysiology. Springer-Verlag, Heidelberg. Carpenter, F. L., Paton, D. C., & Hixon, M. A. 1983.Weight Hagan III, J. M., & Johnston, D.W. 1992. Ecology and

80 British Birds 101 • February 2008 • 68–81 The mysteries of bird migration

Conservation of Neotropical Migrant Landbirds. Schüz, E., & Weigold, H. 1931. Atlas des Vogelzuges. Smithsonian Inst. Press,Washington. R. Friedländer & Sohn, Berlin. Hedenström, A., & Alerstam,T. 1997. Optimal fuel loads in Smith, R. J., & Moore, F. R. 2005. Arrival timing and seasonal migratory birds: distinguishing between time and energy reproductive performance in a long-distance migratory minimisation. J.Theor. Biol. 189: 227–234. landbird. Behav. Ecol. Sociobiol. 57: 231–239. Hobson, K. A. 1999.Tracing origins and migration of wildlife Stresemann, E. 1951. Die Entwicklung der Ornithologie von using stable isotopes: a review. Oecologia 120: 314–326. Aristoteles bis zur Gegenwart. F.W. Peters, Berlin. — 2003. Making migratory connections with stable Svensson, L. 1992. Identification Guide to European isotopes. In: Berthold, P., Gwinner, E., & Sonnenschein, E. Passerines. 4th edn. Privately published, Stockholm. (eds.), Avian Migration: 379–391. Springer, Berlin. Szép,T., Møller,A. P.,Vallner, J., Kovács, B., & Norman, D. — & Wassenaar, L. I. 1997. Linking breeding and wintering 2003. Use of trace elements in feathers of Sand Martin grounds of neotropical migrant songbirds using stable Riparia riparia for identifying moulting areas. J.Avian Biol. hydrogen isotopic analysis of feathers. Oecologia 109: 34: 307–320. 142–148. Thienemann, J. 1910. Der Zug des weißen Storches Hüppop, O., & Hüppop, K. 2003. North Atlantic Oscillation (Ciconia ciconia). Zool. Jahrb. (Suppl.) 12: 665–686. and timing of spring migration in birds. Proc. Roy. Soc. Thorup, K.,Tøttrup, A. P.,& Rahbek, C. 2007. Patterns of Lond. B 270: 233–240. phenological changes in migratory birds. Oecologia 151: Jespersen, P., & Tåning, Å. 1950. Studies in Bird Migration 697–703. being the Collected Papers of H. Chr. C. Mortensen. von Hünerbein, K., Hamann, H.-J., Rüter, E., & Wiltschko,W. Munksgaard, Copenhagen. 2000. A GPS-based system for recording the flight paths Kaiser,A. 1993. A new multi-category classification of of birds. Naturwissenschaften 87: 278–279. subcutaneous fat deposits of songbirds. J. Field Ornithol. von Lucanus, F. 1929. Die Rätsel des Vogelzuges. H. Beyer & 64: 246–255. Söhne, Langensalza. Keast, A., & Morton, E. S. 1980. Migrant Birds in the Walther, G.-R., Post, E., Convey,A., Menzel, A., Parmesan, C., Neotropics: ecology, behavior, distribution, and conservation. Beebee,T. J. C., Fromentin, J.-M., Hoegh-Guldberg, O., & Smithsonian Inst. Press,Washington. Bairlein, F. 2002. Ecological responses to recent climate Kinzelbach, R. 2005. Der Pfeilstorch. Basilisken Presse, change (review). Nature 416: 389–395. Marburg. Weber,T. P.,Fransson,T., & Houston, A. I. 1999. Should I stay Lindström, Å. 1991. Maximum fat deposition rates in or should I go? Testing optimality models of stopover migrating birds. Ornis Scand. 22: 12–19. decisions in migrating birds. Behav. Ecol. Sociobiol. 46: — & Alerstam,T. 1992. Optimal fat loads in migrating birds: 280–286. a test of the time minimisation hypothesis. Amer. Natur. Webster, M. S., Marra, P.P.,Haig, S. M., Bensch, S., & Holmes, 140: 477–491. R.T. 2002. Links between worlds: unraveling migratory Lott, C. A., Meehan,T. D., & Heath, J. A. 2003. Estimating the connectivity. TREE 17: 76–83. latitudinal origins of migratory birds by using hydrogen Weimerskirch, H., & Wilson, R. P.2000. Oceanic respite for and sulfur stable isotopes in feathers: influence of wandering albatrosses. Nature 406: 955–956. marine prey base. Oecologia 134: 505–510. Wernham, C.V.,Toms,M. P., Marchant, J. H., Clark, J. A., McClure, H. E. 1974. Migration and Survival of the Birds of Siriwardena, G. M., & Baillie, S. R. 2002. The Migration Asia. SEATO, Bangkok. Atlas: movements of the birds of Britain and Ireland. Moreau, R. E. 1972. The Palaearctic–African Bird Migration Poyser, London. Systems. Academic Press, London. Wilson, R. P.2001. Beyond rings on birds for the Parrish, J. R., Rogers, D.T., & Prescott, F. 1983. Identification determination of movements: wither the archival tag? of natal locals of Peregrine Falcons (Falco peregrinus) by Ardea 89: 231–240. trace-element analysis of feathers. Auk 100: 560–567. Wiltschko, R., & Wiltschko,W. 2003. Avian navigation: from Piersma,T., Pérez-Tris, J., Mouritsen, H., Bauchinger, U., & historical to modern concepts. Anim. Behav. 65: Bairlein, F. 2005. Is there a ‘migratory syndrome’ 257–272. common to all migrant birds? Ann. New York Acad. Sci. Wiltschko,W., & Wiltschko, R. 2002. Magnetic compass 1046: 282–294. orientation in birds and its physiological basis. Poole, A. F., Stettenheim, P., & Gill, F. B. 1992–2002. The Birds Naturwissenschaften 89: 445–452. of North America.The American Ornithologists’ Union Wisz, M. S.,Walther, B. A., & Rahbek, C. 2007. Using and the Academy of Natural Sciences of Philadelphia, potential distributions to explore determinants of Washington, D.C. & Philadelphia, P.A. Western Palaearctic migrant songbird species richness Rappole, J. H., Morton, E. S., Lovejoy III,T. E., & Ruos, J. L. in sub-Saharan Africa. J. Biogeography 34: 828–841. 1995. Nearctic Avian Migrants in the Neotropics. Yamashina Institute for Ornithology. 1996. Atlas of Japanese Smithsonian Institution, Front Royal. Migratory Birds (Passerines 1961–1995).Yamashina Ropert-Coudert,Y., & Wilson, R. P.2005.Trends and Institute for Ornithology,Abiko. perspectives in animal-attached remote sensing. Front Yohannes,E., Hobson, K. A., & Pearson, D. J. 2007. Feather Ecol. Environm. 3: 437–444. stable-isotope profiles reveal stopover habitat selection Rubenstein, D. R., Chamberlain, C. P., Holmes, R.T., Ayres, and site fidelity in nine migratory species moving M. P.,Waldbauer, J. R., Graves, G. R., & Tuross, N. C. 2002. through sub-Saharan Africa. J.Avian. Biol. 38: 347–355. Linking breeding and wintering ranges of a migratory Zink, G. 1973, 1975, 1981, 1985. Der Zug europäischer songbird using stable isotopes. Science 295:1062–1065. Singvögel: ein Atlas der Wiederfunde beringter Vögel. Lfg. Schmaljohann, H., & Dierschke,V. 2005. Optimal migration 1–4. AULA-Verlag,Wiesbaden. and predation risk: a field experiment with Northern Zink, G., & Bairlein, F. 1995. Der Zug europäischer Singvögel: Wheatears (Oenanthe oenanthe). J.Anim. Ecol. 74: ein Atlas der Wiederfunde beringter Vögel.Vol. 3. AULA- 131–138. Verlag,Weisbaden.

Franz Bairlein, Institute of Avian Research, ‘Vogelwarte Helgoland’, An der Vogelwarte 21, D-26386 Wilhelmshaven, Germany; e-mail: [email protected]

British Birds 101 • February 2008 • 68–81 81