Clausen et al. 2003
Butler et al. 2003
Fox et al. 2003
Javed et al. 2000
All figures are mean values except where otherwise indicated. Details are only from birds that were followed over one or more complete journey (i.e. part-journeys excluded), and were calculated in the same way for all studies. In some cases, therefore, the figures may differ from those given in the original publications. All the species shown migrate by self-powered flapping flight. N: number of birds tracked. aMainly over water.
bAutumn movement: moult migration, so not comparable to spring migration. Wintering area includes moulting area.
Brent Geese Branta bernicla migrating mainly over land between the Wadden Sea and the Taimyr Peninsula covered the mean flight distance of 5004 km in an average of 42 days (range 34-52 days), with a mean speed of 118 km per day (range 97-148) (Green et al. 2002). The slowest birds thus took half as long again as the fastest. Seven males on their longest apparent non-stop flight covered an average of 1056 km (range 768-1331 km). Much greater individual variation was recorded among Barnacle Geese Branta leucopsis migrating largely over water from Svalbard via Bear Island to Scotland in autumn over a total distance of 2500-3000 km. At one extreme, some birds completed this journey in as little as 2-3 days - more than 1000 km per day - during which their stops were of such short duration that they gave little chance to replenish reserves (Butler et al. 1998). At the other extreme, four birds held up en route by unfavourable winds took 9-36 days over the journey, giving mean migration speeds of 306-76 km per day. Overall journey time thus varied at least 12-fold between individuals. Possibly most birds left Svalbard with enough fuel to complete the journey to Scotland if they could do it within 2-3 days. But if they were held up by weather, their reserves would have become depleted, requiring additional feeding.
Other relevant data have been obtained from Houbara Bustards Chlamydotis undulata, which also migrate using flapping flight (Table 8.2). Of eight individuals trapped in Kazakhstan, two flew directly to their wintering areas with no stopover longer than one day (Combreau et al. 1999). They covered distances of 1600 and 1970 km in 20 and 13 days respectively, giving mean speeds of 24 and 151 km per day (6.3-fold). Six other birds made a longer stop, lasting more than a week, and one bird stopped twice for 51 and 16 days. With such wide variations, the duration of autumn migration varied from 13 to 73 days, depending largely on the time spent on stopovers, mean travel speeds ranging between 24 and 151 km per day. These birds also showed a roughly two-month spread in departure dates from breeding areas (15 July-18 September) and of arrival dates in wintering areas (14 September-14 November).
Although many seabirds travel by soaring flight, few estimates have been made of their migration speeds. With an average flight cost estimated at only 3 X BMR, and a rate of energy accumulation of 2 X BMR, and assuming that they were on the wing for 50% of the time, albatrosses are expected on theoretical grounds to migrate at average speeds in the range 440-880 km per day (Hedenstrom 1993). This exceeds by a broad margin the average migration speeds of birds using flapping flight, discussed above (Alerstam 2003). Tracking studies have confirmed that albatrosses achieve overall average speeds between 220 and 950 km per day on trips over distances of 3000-25 000 km (Jouventin & Weimerskirch 1990, Prince et al. 1992, Weimerskirch et al. 1993, Croxall et al. 2005). Some Greyheaded Albatrosses Thalassarche chrystostoma that flew around the earth included some extraordinary examples of flight performance (Croxall et al. 2005). Typical journeys from South Georgia to the southwest Indian Ocean took 6.2 days at 950 km per day; the second leg to the southwest Pacific lasted 13.2 days at 950 km per day, and the last leg back to South Georgia 10.3 days at 750 km per day. Without stopping, a complete circumnavigation of the Southern Ocean could, in theory, be completed in 30 days; which provides a context for the exceptional performance of one bird that made this journey in just 46 days. Based on average rather than occasional speeds, albatrosses that travel by gliding flight are evidently among the fastest of all long-distance animal travellers.
Estimates of migration speeds were also obtained from a smaller species, the Sooty Shearwater Puffinus griseus, by use of geolocation tracking tags (Shaffer et al. 2006). On their figure-eight migrations around the Pacific, 19 tagged birds travelled an average of 64 037 km in 198 days, giving a mean speed of 323 km per day. On parts of the journey, speeds rose to an average of 910 km per day, not very different from the highest recorded from albatrosses.
Turning to species that migrate by soaring-gliding migration over land, many species of raptors have been fitted with radio-tags and monitored by satellite on their journeys (Table 8.3, Box 8.3, Figures 8.6 and 8.7). Most of these birds made long journeys, some from one continent to another. Their migration was limited to the period each day during which thermal soaring was possible - at most about 9 hours (Spaar & Bruderer 1996). Their need for thermals also led many such birds to take long indirect routes to avoid long sea-crossings. Their journeys were thereby lengthened by up to 50% over the shortest (great circle) routes.
The shortest migration recorded was undertaken by the Bald Eagle Haliaetus leucocephalus mentioned earlier, which spent less than one day over its 435 km journey (Grubb et al. 1994). The longest migration recorded was undertaken by Swainson's Hawks Buteo swainsoni which travelled an average of 13 500 km (maximum 15 000 km) between breeding areas in western North America and wintering areas in southern South America (Box 8.3; Fuller et al. 1998). Overall, excluding the Bald Eagle mentioned above, mean migration speeds of raptors varied between 12 and 294 km per day in autumn, and between 23 and 196 km per day in spring (Table 8.3). Not unexpectedly, migration was more rapid over unavoidable seas and deserts, but otherwise daily rates were much affected by weather, being greatly reduced by strong winds, dense cloud and rain.
As expected, the time spent on migration varied with the length of the journey, a trend apparent both within and between species, but with considerable individual variation (Figure 8.8). Within species, this was most apparent among Peregrines Falco peregrinus and Ospreys Pandion haliaetus, large numbers of which have been tracked from different parts of the breeding range. Like some passerines that travel by flapping flight, soaring species tended to migrate faster on longer journeys, but nevertheless no significant correlation emerged between migration speed and distance travelled (Figure 8.8). This was presumably because soaring species required much less food to fuel the flights, and did not need to stop for long feeding periods during the journey (although some individuals clearly did). Similarly, among 24 juvenile Steller's Sea Eagles Haliaeetus pelagicus tagged at nests in various parts of the breeding range, mean speed per day increased with length of journey (as measured by degrees of latitude travelled) (McGrady et al. 2003). Theory predicts that, in soaring birds, migration speed should increase with body mass (see above), and although this relationship was found for the crosscountry speeds of raptors studied in Israel (Figure 7.4), no such relationship was apparent for the whole migrations of species in Table 8.3. This may be because, while larger species can glide more rapidly than small ones, small ones can utilise weaker thermals, and so can migrate over a longer part of each day (Chapter 3).
Was this article helpful?