Circadian systems are involved in annual migrations in two ways.
First, the circadian clock is used to
measure the gradual changes in daylength (photoperiod), thus giving the animal sufficient time to
prepare for the annual migratory event.
Such preparations include molting (change of feathers or hair) and pre-migratory fattening in birds (sometimes to the extent that
all other internal organs shrink). Both
circadian and circannual (PDF) timers may be involved.
There are also behavioral changes. For instance, migratory animals aggregate into large groups (flocks of birds or herds of wildebeast, for instance). Birds undergo practice flights - you've all seen birds flying in circles above your head and practicing their flying formations.
Most migratory birds are normally diurnal (day-active), but tend to do their migratory flights over nights. Changes in hormone levels associated with seasonal preparation for migration induce a splitting in the circadian output into two components: one guiding daytime behavior (foraging, practice flights) and the other controlling
nocturnal migratory restlesness which is behavioral preparation for night-time migratory flights.
Photoperiod is not the only environmental cue governing timing of seasonal events. Thermoperiod (relative durations of daily warmth and nightly chills), average temperature, food availabilty and rainfall are secondary (or proximal) cues that fine-tune the timing of seasonal events in animal behavior and physiology.
There is a whole range of responses to photoperiod and other cues. In nature, finches in Australia respond only to rainfall, yet are capable of responding to photoperiod in the laboratory. Tropical/equatorial birds, naturally exposed to minimal yearly changes in daylength are
quite capable of (PDF) measuring tiny changes in photoperiod in the lab. On the other end of the spectrum are birds like the famous
swallows of San Juan Capistrano, in which photoperiodism is clearly the most dominant seasonal clue.
Photoperiodic response is genetically determined. Variation in the relative importance of photoperiod versus proximal cues exists not just between species but also within species (this has mostly been studied in mammals, see the work of
Paul Heideman, for instance).
The second way in which the circadian clock is involved in migration is in orientation and navigation. Animals use a
number of clues in the environment to orient to, including landmarks, orientation of the magnetic field of the Earth, sense of smell ("olfactory maps", specifically in salmon who remember the olfactory stamps of their native streams and rivers), position of the stars and position of the Sun. Position of the Sun changes over the course of the day, thus internal clock is used to correct for the Sun's movement across the sky (I will write in detail about the mechanism in the future). Likewise, stars move across the sky during the night and the clock controls for such movement. The intensity of the magnetic field is higher during the night, too.
Thus, clocks help animals decide both when to go and where to go. Both the timing and the
direction of migration are finely tuned by evolution. Migration is a very energetically expensive, as well as a dangerous endeavor. Thus, it is to be expected that natural selection has resulted in quite optimal solutions for both timing and direction of migration in each species.
Now, this is all starting to fall apart due to global warming. Proximal cues, like temperature and food-availability, are beginning to conflict with photoperiodic information. Species in which photoperiod is dominant continue to migrate at the same time and in the same direction. Other species are shifting their timing to later in fall and earlier in spring. As there is
genetic variation within species there is now evidence for change in relative proportions of phenotypes as some strategies are more adaptive than others, namely migrating later, migrating closer, or not migrating at all may be more adaptive than enduring a long dangerous migratory flight.
Some of those changes in bird behavior have already been
reported.
For instance, various species of European
warblers mainly migrate to Africa for the winter. It has been known for a while now that there is a
genetic basis for intra-species variation in migratory direction. There exists a small subgroup that migrates to the South of England instead of Africa. As of very recently, this subpopulation has been doing great and increasing in relative proportion within the species, threatening to completely abolish the trip to Africa from the species' behavioral repertoire. The genetic basis for the trip to Africa may dissappear, and if the global warming is successfully countered and reversed, this species will be unable to migrate to Africa again, leaving the "England-bound" genotype to freeze and starve in the future. And nobody is asking how will the absence of warblers affect the ecosystems in Africa!
Non-migratory species are also affected. For instance,
sparrows in Scotland are not preparing for winter adequately. The falls are mellower and warmer, so they do not prepare for winter in time. At the same time, human activity is limiting their food supply, further diminishing their ability to prepare for cold Scottish winters (and yes, they are still cold, despite global warming, it's just that they come later, and stopping of Gulf Stream will make them VERY much colder).
A
genetic response to global warming in photoperiodic responses in Pitcher-plant mosquitoes and
Mexican Jays as well as
in many other species have already been documented. Effectively,
whole ecosystems are moving North, but some species move faster than the others, thus breaking up old ecosystems and building new ones. A plant predominantly responsive to temperature may move North, but its pollinator-insect may be strongly photoperiodic and lags behind. The plant lacks its pollinator in the North, the insect lacks its plant food in the South. Such times of tumultuous changes often lead to extinctions of species that cannot quickly adapt to such changes.
Many researchers around the world are watching evolution in action right now. Ecosystems break down and new ones get assembled. Migratory patterns change and new predator-prey and pollinator-flower relationships will emerge. Some species will go extinct and others will change so much their entries in ornithology (and entomology, mammalogy, botany, etc) books will have to be substantially re-written.
On one hand, watching evolution in fast action is fascinating (not to mention that it provides plently of new ammunition to counter creaitonists' claims against what they like to call "macroevolution"). On the other hand, watching species go extinct due to unwillingness of some humans to accept responsibility for global warming and to implement strategies to counter it, is more than frustrating. It makes one ashamed to be a Homo sapiens.