Birds tend to commence migration in large numbers only when they have a favourable tail wind. In North America the winds north in spring and south in autumn are ideal to assist seasonal migrations. Once started however only very bad weather will stop them. Many birds fly high when migrating because of prevailing winds at higher altitudes and also because the cold at these altitudes helps them disperse all the heat being generated by their flight muscles. Many species of wildfowl fly at 6,000m and some have been observed flying at 8,000m, 86mph in temp = -48 degrees C.
Not all birds from a summer breeding site overwinter at the same area. What happens, come autumn, if a male bird meets a female bird in the breeding grounds who has a different overwintering site? Whose site do they go to now they are a pair? In many species the pair bond breaks up at the end of the breeding season, but some like swans mate for life. In the case of the Bewick's Swan the male decides where to fly to for the winter and the female follows him. However, the female decides when it is time to travel back to the tundra for another year's breeding.
The reverse scenario is when birds with different breeding sites overwinter in the same area; if pairing commences on the overwintering ground, whose breeding ground to they return to. The answer may be different for different species. The only example I know of involved Mallards in the USA and in this case the male followed the female.
Timing of migration is a mix of internal stimulus which results in a feeding binge to put on fat to survive the journey and then the tendency to aggregate into flocks. Once the pre-migration flock is gathered, the feeding continues while the birds wait for suitable weather conditions. Thus while the birds' internal clock probably releases the hormonal triggers at a fairly accurate date each year, the availability of food and the presiding weather conditions decide when the migration starts and hence when we see the first spring migrants arrive and the last autumn ones leave.
A 12-year study of Common Terns at Cape Cod showed that an average 75% of birds, and as much as 83%, returned to the same area to nest in successive years. Eighty percent nested within 25 feet of the original nest site. Another study of Layson Albatrosses showed that in the following year a nest was on average only 13 inches away from the previous nest.
For geographical reasons, i.e. mountains, coasts and rivers, many migrating birds travel certain general flyway or routes. In the USA there are 4 main flyways and these are termed the Atlantic, Mississippi, Central and Pacific flyways. In Europe the mountains do not run so regularly north/south as in the USA, therefore certain passes become funnels through which many migratory birds pass. Also the Mediterranean is a major obstacle to many birds and they veer either left or right to avoid crossing it.
Migratory routes are not fixed eternally and in some species part of the population follows one route and part another. Also, some birds travel south by a different route to that which they use to travel north, e.g. Golden Plover.
Some migrants fly very long distances. Some arctic terns fly 11,000 miles each way. Other birds fly lesser distances. Blackpolls from Hudson Bay overwinter in Venezuela, 5,000 miles each way and Golden Plovers fly 2400 miles each way in the USA.
Speed - birds often fly faster when on a migratory flight then they do during ordinary flight. Thus distances of 200 to 400 miles a day are commonplace among long distance migrants. Some birds, however, migrate more slowly, e.g. Robins coming up the Gulf coast average 13 miles a day.
Most flights occur at between 600 and 5,000 ft above sea level with an average height of 1525 ft a.s.l. However, mountains may mean greater heights are needed and heights over 10,000 ft a.s.l. are not uncommon.
Homing - Little is known about how birds navigate. Experiments show that most migratory birds have a built-in sense of direction and know innately which direction they need to travel. First year Starlings in Europe kept in a covered cage and away from birds which have already migrated once or more, still move to the correct side of the cage when the time comes for them to migrate.
Some birds appear to use landmarks and obviously at a height of several thousand feet they can see a considerable distance.
Young crows born and raised in Alberta and then kept caged until after all the population had flown south and the first snows had fallen flew straight to Oklahoma where the rest of their flock was.
Young birds can also learn a lot from travelling with their parents. Mallards are migratory in Finland, but not in England. Young hatched from eggs taken from English Mallards and put under Finish females had no problems migrating with the rest of the population.
Wind direction - Some birds, particularly those flying across the North Sea, navigate at least partially using wind direction. This makes sense when you realise that in many places the wind blows in the same direction in the same season every year.
The Sun - A number of elegant experiments involving ???, and/or displacing birds to different geographical regions have shown that many birds use the sun, at least during the day, as a cue to direction when migrating or homing.
More clever experiments, some involving rearing birds in cages so that they have never seen the night sky, and/or putting them in a planetarium, have shown that many birds orient themselves by means of the stars, but not the moon or planets and that this ability is innate. Birds are known to stop migrating on cloudy nights.
Birds of prey, swallows and crows migrate by day. Thrushes, warblers, cuckoos and woodpeckers migrate by night. Wildfowl migrate both day and night.Most songbirds migrate at night.
There is believed to be some hormonal stimulus to migrate, resulting, at least in the spring, in the development of the gonads. Other stimuli appear to involve temperature, daylight/darkness ratios and an internal clock.