It is widely known that, at particular localities, some bird species appear only in the breeding season, and others in the non-breeding season or at times of passage. Watching birds on migration has become a favourite pastime for thousands of bird watchers, and in many countries the concentration points (such as coastal promontories, offshore islets and mountain passes) are now well known. Hawk Mountain in Pennsylvania, which is famous as a viewing site for raptor migration, attracts about 20 000 raptors each autumn, but more than 100 000 human observers. Large numbers of people also visit Cape May in New Jersey each year, the Santa Ana Wildlife Refuge in southwest Texas, and Falsterbo in Sweden. At some sites, diurnal migration has been observed systematically over many years, giving information on the numbers, directions and passage periods of different species, on prevailing weather effects on migration, and on long-term changes in numbers and migration timing (Chapter 7).
For most bird species, however, counts of birds seen on the ground or flying over represent only a small and variable proportion of those passing overhead. This is because most migrating birds fly much too high to be seen with the naked eye or even with binoculars, and in any case many species migrate mainly at night. It is chiefly when they encounter headwinds that birds fly low enough to be easily seen. Migrants come to ground mainly to rest or refuel, or after they have drifted off course because of side winds, or been forced down by headwinds, mist or rain. Hence, visual counts of migrants cannot usually reflect the true volume of migration, or the weather conditions that most favour it (for a critique of observational methods of study, see Kerlinger 1989). On the other hand, any birds seen can usually be identified to species by their appearance or calls. Large-scale studies of visible bird migration, with observers posted at different localities, formed some of the earliest cooperative projects involving bird-watchers.
Watching seabirds on migration is most profitable when onshore winds cause birds to fly closer to land, and often thousands per hour can be seen streaming past headlands. Such counts depend on wind conditions bringing migrating birds within view. In other conditions, migration would occur too far out to be visible to a land-based observer. Also, as seabird species forage at long distances from their nests, it is impossible at certain times of year to distinguish migration from foraging flights.
At night, ground-based observers are much more limited, but on clear nights low-flying birds can be seen as they cross the lit surface of the moon (moon-watching). The drawbacks of the method are that it can only be used near full moon in clear weather and the observation cone has a relatively small angle (on average 0.52°), covering only a tiny portion of the night sky. By adventurous calculations involving the moon's bearing and elevation, counts of birds crossing the face of the moon can be transformed into estimates of the numbers passing over, their direction of movement and even their height and speed (Nisbet 1959a). Using a telescope with 40 X magnification, it was estimated that about 50% of the birds flying at 1.5 km distance from the observer were detected, reducing to zero at 3.5 km, based on comparison with radar and infrared observations (Liechti et al. 1995). The moon-watching method is also hard on the eyes, and ideally needs several observers taking turns. The most impressive large-scale count programme ever undertaken on the basis of moon-watching was in central Asia where, at the time it was done, no other study methods were available there (Box 2.1).
Other observers have used a strong spotlight directed skywards to count the birds passing through the beam. The best device for this purpose is a ceilometer, which is normally used at airports for measuring cloud height. In warm weather, the lower part of the beam tends to be full of insects, but birds seen flying through the upper part can be recorded in the same way as for moon-watching, but with limitations on distance as the beam typically extends only to a few hundred metres. Hebrard (1971) used a horizontally directed portable ceilometer placed on a tower to illuminate birds at they took off from the tree canopy at night.
Other evidence of nocturnal migration can be obtained by listening for the calls of birds as they pass invisibly overhead. The unaided human ear cannot pick up the normal flight calls of birds beyond about 400 m, but use of a parabolic reflector and amplifier can extend the range to 3000 m or more. Birds call more during mist and poor visibility than in clear skies, and some species seem not to call at all, so the numbers of calls heard are only broadly related to the number of birds passing (Farnsworth et al. 2004). Nevertheless, the opportunity that listening affords for identifying species makes it a useful accessory to other methods.
Early indication of the numbers and species of birds migrating at night was provided by 'kills' of low-flying birds attracted to lighthouses and illuminated communication masts (Chapter 28; Gatke 1895, Clarke 1912). Spectacular slaughter has sometimes been recorded at particular sites, such as the 50 000 birds of 53 species killed at one site in Georgia in one night (Johnston & Haines 1957). Some species, such as Common Snipe Gallingo gallingo, Water Rail Rallus aquaticus and Common Grasshopper Warbler Locustella naevia in Europe, seem notoriously prone to such accidents. Mortality occurs mainly on overcast or foggy nights, and the resulting corpses have provided information on the migration seasons, body weights and condition of different species.
Indications that landbirds cross the sea have been obtained from coastal observations of birds flying out to sea, or from ships or oilrigs of birds passing over or stopping by, and from radar observations. The recent series of oil and gas
Box 2.1 The use of 'moon-watching' to assess migratory bird numbers in central Asia.
The most extensive series of counts ever made by moon-watching was obtained by a team of Russian observers stationed at various points across a 2200-km zone stretching from the Caspian Sea to the eastern part of the Tien Shan Mountains in central Asia (Dolnik & Bolshakov 1985). At each point, birds were counted in spring as they crossed the lit surface of the moon, using 30x telescopes. The totals were converted to numbers of birds crossing 1 km of a latitude line, and then extrapolated to the whole 2200 km. Although skies were generally clear throughout the migration season, the observers could not count around times of a new moon, when too little of the moon surface was lit. For those darker nights they assumed that the same numbers of birds passed as on nights around full moon, when counts could be made. All the main types of birds expected were seen, but passerines formed 73-89% of the totals at different localities. Overall, an estimated 731 million birds crossed the 42°N parallel on spring migration. From the directions taken, 85% of these birds came from winter quarters in southern Asia (heading mainly north-northwest), and the rest (110 million) probably mainly from Africa (heading mainly northeast). The volume of migration was not uniform across the whole 2200-km stretch, perhaps because some birds were diverted by topographical features, including the Tien Shan Mountains.
In autumn, the number of migrants would be expected to be at least twice the number counted in spring, owing to reproduction. This was equivalent to three times the autumn estimate made for North Africa which spans approximately double the length of latitude as the Asian survey. The implication was that about six times more migrants left this part of Asia in autumn than entered Africa. The difference may be due partly to the different methods of assessment, each with their own errors; to greater bird densities in central Asia where habitats have been less disturbed than in Europe; and to the fact that many of the Asian birds may have wintered south of 42°N, but at latitudes north of the northern latitude of Africa. The 220 million birds that could have been heading for Africa across this 2200-km front would form nearly 4.5% of the 5000 million Eurasian migrants estimated by Moreau (1972) to enter Africa each autumn.
drilling platforms in the North Sea has provided additional information on the movements of birds between Britain and continental Europe (Bourne et al. 1979, Anderson 1990).
Was this article helpful?