Skip to content
NatureHQ

Behaviour and cognitionbehaviour

Migration

A migrating animal is one that has switched off the responses that would normally make it stop. That, and not distance, is what makes a journey a migration.

Migration is not defined by distance. It is a change of behaviour: an animal stops responding to the things that would normally stop it — food, company, a good place to be — and keeps going in one direction until a different state takes over.

The definition matters more here than in most subjects, because the popular one is about distance and the useful one is not. An aphid flying two hundred metres with its feeding responses switched off is migrating. A wandering albatross covering a thousand kilometres in search of squid is not: it is foraging, and it will stop the moment it finds some. What distinguishes them is the suppression — during migration the ordinary business of noticing that here is quite nice is temporarily turned off, which is why migrating animals cross places they would otherwise settle in. Once that is clear, the variety stops looking chaotic. Some populations migrate every year and neighbouring ones do not; in many species some individuals leave and their siblings stay, which is partial migration and is commoner than the tidy version implies. Some movements are irruptions, triggered by a failed seed crop rather than a season. Some are vertical rather than horizontal — up a mountain in summer, down in winter, or the daily migration of ocean plankton that is the largest animal movement on Earth and covers a few hundred metres. And migration is not fixed: it evolves and disappears within decades, which is why a field guide written a generation ago can be wrong about where a species spends its winter.

Developed coverage · 79% complete · reviewed 2026-09-03

What this page covers

Migration in the behavioural sense occurs across birds, mammals, fish, insects, crustaceans and reptiles. The evidence is heavily weighted towards birds, because birds are the group tracking technology reached first.

Often confused with: Dispersal, which is a one-way movement away from a birthplace and usually happens once; Nomadism, where animals move to wherever conditions are good with no fixed destination; Any long journey — distance is not what makes a movement a migration

Quick facts

What defines it
Directed movement with the responses that would arrest it suppressed
Not one pattern
Seasonal, partial, irruptive, altitudinal, developmental — often within one species
How the route is known
A direction can be inherited; position generally has to be learned
How we know any of it
Rings, radar and tags — each with its own blind spot

Where this appears

Every organism below has been linked to this page because the evidence links them. Each one carries its own evidence, and its own limits.

What makes a journey a migration

The definition is behavioural, and it excludes several things people expect it to include.

The short answer

What actually counts as migration?

Persistent movement in one direction with the animal’s ordinary stop-here responses switched off. Distance is irrelevant: an aphid can migrate two hundred metres and an albatross can cover a thousand kilometres without migrating at all.

The behavioural definition exists because distance-based ones split cases that belong together and join cases that do not. A foraging albatross and a migrating godwit are both crossing an ocean; only one of them will stop when it finds food, and that difference is what predicts everything else about the two journeys — how they are timed, how they are fuelled, and what happens if you move the animal sideways. Dispersal is different again: a young spider ballooning away from where it hatched is going somewhere, once, and is not coming back. Nomadism is different from both, because there is no destination at all: the animal is following conditions. The categories grade into each other at the edges, and the grading is real rather than a failure of definition.

Check it for yourself

Migration is not defined by how far an animal goes. It is defined by a change in behaviour: movement that keeps going in one direction while the responses that would normally stop it are switched off.

Established

Specialists would state this without hedging. Multiple independent lines of evidence agree.

Migration is characterised behaviourally as persistent, straightened-out movement with temporary suppression of station-keeping responses to resources and conspecifics, distinguishing it from dispersal, ranging and nomadism irrespective of the distance covered.

Who this applies to
A definition applied across animal groups, from aphids moving hundreds of metres to whales crossing ocean basins.
Studied in
Animalia
Why we rate it this way, and what the caveats are
EstablishedHigh confidence

A definitional rather than an empirical claim, and the behavioural definition is the one the primary literature now works with. Alternative definitions exist and are narrower rather than contradictory.

How far it can be extended

The behavioural criterion was developed to unify insect, bird, fish and mammal cases that distance-based definitions split.

Caveats

  • The suppression of station-keeping responses is easy to state and hard to observe directly in a large animal, so in practice migration is often inferred from the shape of a track.
  • Other definitions remain in use, particularly in fisheries and in insect work, and they draw the boundary in different places.

Still unanswered

  • Whether partial migration — where some individuals of a population move and others do not — is better treated as one strategy with variation or as two.

Last reviewed 2026-09-03

The evidence (2 studies)

  • Supports · primary

    What Is Migration?

    Dingle and Drake, 2007 · BioScience

    The paper that sets out the behavioural definition and argues against distance.

  • Supports · supporting

    Long-distance migration: evolution and determinants

    Alerstam et al., 2003 · Oikos

    Treats migration as a strategy under ecological and energetic constraint rather than as a distance category.

Diagram

Four journeys, and which of them are migrations

Schematic. Track shapes are illustrative rather than measured.

Four journeys, and only one is a migrationSchematic. Track shapes are illustrative.MigrationPast resources it would stop for, then back.ForagingStops at the first patch that is good enough.DispersalOne way, once. No return leg at all.NomadismFollows conditions. There is no destination.The test is behavioural: are the stop-here responses switched off?Distance appears nowhere in the definition.
The same explanation in words

Four movement patterns drawn side by side. The first, labelled migration and marked as counting, shows a directed track running straight past patches of resources without stopping, then a return track the other way. The second, labelled foraging or ranging and marked as not counting, shows a track that wanders and settles wherever a resource patch appears. The third, labelled dispersal and marked as not counting, shows a single one-way track leaving a birthplace with no return. The fourth, labelled nomadism and marked as not counting, shows an irregular track moving between scattered patches with no fixed destination and no return. A note explains that the difference is whether the responses that would normally stop the animal are switched off, and that distance does not appear anywhere in the definition.

Words used here
Station keeping
The ordinary behaviour that keeps an animal in a good place — feeding, defending a patch, staying with company. Migration is partly defined by these being switched off.
Dispersal
A one-way movement away from where an animal was born or bred, usually once in a lifetime and with no return.

There is no single migratory pattern

There is no single migratory pattern. Some populations go every year and some only in bad years; in many species some individuals leave and their neighbours stay.

Established

Specialists would state this without hedging. Multiple independent lines of evidence agree.

Migratory strategies include obligate seasonal return migration, partial migration in which a fraction of a population migrates, irruptive movement driven by resource failure, nomadism without a fixed destination, and altitudinal and developmental migrations, with the strategy varying between populations of the same species.

Who this applies to
Documented across birds, mammals, fish and insects.
Studied in
Animalia
Why we rate it this way, and what the caveats are
EstablishedHigh confidence

The variation is directly observed and much of it is measurable within a single population in a single season.

How far it can be extended

Partial migration alone is documented in hundreds of bird species and in fish and ungulates.

Caveats

  • The categories are not sharp: partial migration grades into irruption, and irruption into nomadism.
  • Which strategy a population uses can change within decades, so field guides written a generation ago can be wrong about it.

Still unanswered

  • What determines which individuals in a partially migratory population are the ones that leave.

Last reviewed 2026-09-03

The evidence (3 studies)

  • Supports · primary

    Long-distance migration: evolution and determinants

    Alerstam et al., 2003 · Oikos

    Sets out migration as an outcome of optimisation under constraint, which is why populations of one species reach different answers.

  • Supports · supporting

    What Is Migration?

    Dingle and Drake, 2007 · BioScience

    Separates migration from nomadism and ranging, which the popular usage merges.

  • Supports · supporting

    Genetic Basis of Migratory Behavior in European Warblers

    Berthold and Querner, 1981 · Science

    Shows how quickly the migratory tendency itself can shift under selection, which is why it varies between neighbouring populations.

The categories grade into one another; the boundaries are useful rather than sharp.
PatternWhat happensWhere you meet it
Seasonal returnThe same journey out and back each yearThe version everyone pictures: swallows, wildebeest, humpbacks
Partial migrationSome individuals of a population leave; others stay putVery common in birds, and in fish and ungulates
IrruptionMass movement in some years and none in others, when food failsWaxwings, crossbills, some owls
NomadismContinuous movement following conditions, with no destinationDesert birds, some ungulates — and it is not migration
AltitudinalUp a mountain and back down, over a few hundred metresMany mountain birds and mammals
DevelopmentalA one-way move between life stages rather than seasonsEels and salmon, in opposite directions

Partial migration is worth pausing on, because it breaks the habit of talking about what a species does. In a partially migratory population the migrants and the residents are the same species in the same place, and which individuals go can depend on age, sex, condition or dominance. Whatever a field guide says about where the species winters is therefore true of some of them.

The groups in detail

Knowing when, which is not the same as knowing where

The cue that starts a migration and the cue that steers it are different systems, and the difference matters as the climate moves.

Departure is timed mostly by day length, which is the one environmental signal that does not drift: the sun does the same thing every year regardless of temperature. That reliability is exactly what makes it a problem now. Day length still says the same thing in April that it said a century ago, while the insects a returning bird eats are emerging earlier — so an unchanged departure date can arrive after the food has peaked. Within the window day length opens, weather does the fine timing: birds wait for a following wind, and depart in pulses when one arrives.

Day length decides roughly when to go, because it is the one signal that never drifts. Weather decides the night: migrants wait, then leave in pulses when the wind turns favourable.

Established

Specialists would state this without hedging. Multiple independent lines of evidence agree.

Migratory departure is primarily under photoperiodic control acting through the endogenous annual cycle, with weather — particularly wind direction and the passage of synoptic systems — determining departure within that window. Across twenty-three years of United States weather-radar data, atmospheric conditions accounted for the great majority of variation in nightly migration intensity.

Who this applies to
Best established in nocturnally migrating birds of the northern temperate zone, which is where both the captive work and the radar coverage are.
Studied in
Aves
Why we rate it this way, and what the caveats are
EstablishedHigh confidence

Two independent lines: captive birds held on manipulated day lengths shift their migratory restlessness, and continental radar shows departure intensity tracking weather with forecastable skill.

How far it can be extended

Photoperiodic control of annual cycles is general across temperate vertebrates; the radar evidence for weather-driven timing is continental but from one network.

Caveats

  • Radar measures departure intensity, not whether any individual left on its optimal night.
  • Photoperiod is the dominant cue in temperate migrants; tropical and irruptive migrants are timed differently and are much less studied.
  • Day length does not drift and springs do — so an unchanged photoperiodic response can now deliver a bird after the food peak it evolved to meet.

Still unanswered

  • How much individual flexibility birds have to adjust departure within the photoperiodic window, which determines how fast populations can track a shifting spring.

Last reviewed 2026-09-03

The evidence (3 studies)

Which is measurable at continental scale, because weather radar sees migration whether or not anyone is watching the sky:

Weather radar sees migration. It cannot tell you which birds, or where they are going — but it counts how many are in the air tonight across a continent, and that turns out to be predictable from the weather.

Established

Specialists would state this without hedging. Multiple independent lines of evidence agree.

Weather-surveillance radar quantifies nightly migration intensity as biomass aloft without species identification; across twenty-three years of United States radar data, atmospheric conditions accounted for the great majority of variation in migration intensity, supporting skilful forecasts several days ahead.

Who this applies to
Nocturnal migration over the continental United States, where the radar network is dense.
Studied in
Aves
Why we rate it this way, and what the caveats are
EstablishedHigh confidence

Two decades of data and a forecasting model with tested skill, which is a stronger form of validation than a single-season correlation.

How far it can be extended

European radar networks have produced comparable measurements; the method needs dense coverage.

Caveats

  • Radar cannot identify species, and insects must be filtered out of the signal.
  • It measures how much is aloft, not where any of it is going or whether it arrives.
  • Coverage is a function of the radar network, so the picture is continental where the network is and absent where it is not.

Still unanswered

  • How to attribute radar-measured biomass to species groups, which would connect these measurements to population trends.

Last reviewed 2026-09-03

The evidence (1 study)

The distinction between timing and direction is not academic. An animal can have a perfectly good compass and a perfectly good map and still fail, if the calendar it starts on has come loose from the season it evolved to meet. That mismatch — the timing question rather than the navigation question — is where most of the conservation concern about migration currently sits.

What the journey costs, and where it is paid

A long migration is an energetics problem before it is a navigation problem. Birds preparing for an ocean crossing roughly double their mass in fat, and some shrink their digestive organs on the way to save weight — an animal that will not eat for a week has no use for a gut. That fuel has to be gathered somewhere, which is why stopover sites matter out of all proportion to the time spent on them: a stretch of mudflat used for three weeks a year can be the reason a flyway exists at all.

Bar-tailed godwits fly from Alaska to New Zealand in one go — more than ten thousand kilometres over open ocean, without landing, eating or drinking, for something like eight or nine days.

Established

Specialists would state this without hedging. Multiple independent lines of evidence agree.

Satellite tracking of bar-tailed godwits of the baueri subspecies documented continuous southbound transoceanic flights from Alaska to New Zealand and eastern Australia exceeding 10,000 km, undertaken without stopover and timed to depart with favourable synoptic wind conditions.

Who this applies to
One subspecies on the southbound leg of one flyway. The northbound return is not nonstop.
Studied in
Limosa lapponica baueri
Why we rate it this way, and what the caveats are
EstablishedHigh confidence

Direct satellite tracking of individual birds, with continuous positions across the ocean — the flight is observed rather than inferred.

How far it can be extended

Other bar-tailed godwit populations migrate differently, and the record applies to this route.

Caveats

  • Small samples: transmitter mass restricts which individuals can be tagged.
  • The southbound leg is nonstop; the northbound journey is staged through the Yellow Sea, so "never stops" is wrong as a description of the annual cycle.
  • Later tracking has recorded longer individual flights; the record figure is a moving one and NatureHQ states the study it comes from.

Still unanswered

  • How the birds sleep, if they do, across eight days of continuous flapping flight.

Last reviewed 2026-09-03

The evidence (1 study)

How we know

Putting a transmitter on a godwit and waiting nine days

Do bar-tailed godwits cross the Pacific in one flight, or island-hop where nobody is watching?

Bar-tailed godwits in Alaska were fitted with satellite transmitters and tracked as they departed. Because a satellite tag reports position from anywhere, the birds could be followed over open ocean where no observer could see them, and departure timing was compared against weather charts.

What happened

Birds flew from Alaska to New Zealand and eastern Australia without landing — flights of more than ten thousand kilometres lasting on the order of eight or nine days — and departures coincided with favourable wind systems.

What it shows

That the flight is genuinely nonstop, which before satellite tracking was an inference from nobody having seen the birds anywhere in between. It also shows the birds are choosing their weather rather than simply setting off on a date.

What it does not show

Small samples, because a transmitter has to be light enough to carry. It applies to one subspecies on the southbound leg — the northbound return is staged through the Yellow Sea, so "never lands" is wrong about the annual cycle. And a tag is a burden: what it costs the bird carrying it is not fully known.

The controls — what makes this evidence rather than a story
  • Continuous positions over water, which makes an unobserved stopover detectable rather than assumed away.
  • Multiple individuals across seasons.
  • Independent meteorological data, so departure timing can be tested against wind rather than described after the fact.

From Extreme endurance flights by landbirds crossing the Pacific Ocean: ecological corridor rather than barrier?

Diagram

Why a stopover is not a rest

Schematic. Mass changes are indicative of the pattern, not of a species.

A stopover is not a restSchematic. Mass changes show the pattern, not a species.arrive, thinfeeding: mass climbswait for a following windflight: no food, mass fallsarrive, thin againMost of the time at a stopover is spent eating, not resting.Which is why losing one site can break a route used for a few weeks a year.
The same explanation in words

A cycle in four stages drawn around a loop. First, fuelling at a stopover: body mass climbs steeply as fat is laid down, and the bird does little but eat. Second, departure, waiting on a following wind. Third, the flight itself: mass falls steadily as fat and then some muscle and organ tissue are consumed, with no feeding. Fourth, arrival at the next stopover with mass at its lowest, where the cycle begins again. A note beside the loop explains that the time spent at a stopover is mostly spent feeding rather than resting, and that losing one site can break a route even though the animals were only there for a few weeks.

Related

What is still unsettled

  • What decides which individuals in a partially migratory population are the ones that leave?

    Why it matters: It is the difference between migration being a species trait and being an individual decision made each year, and the answer bears directly on how quickly populations can adjust to a changing climate.

    What would settle it: Long-term tracking of marked individuals across several years, which is now becoming possible as tags get small enough.

  • How fast can a migration route change?

    Why it matters: Blackcap wintering behaviour shifted within a few decades. Whether that speed is typical or exceptional determines how much of the current concern about mistimed migrations is warranted.

    What would settle it: Comparable long-term records for more species, which exist for very few.

  • How much migratory movement is happening that nobody has measured?

    Why it matters: Radar keeps finding enormous insect movements overhead that were invisible before. The same is likely true of small animals nobody can tag.

    What would settle it: Wider deployment of radar and acoustic monitoring, and methods for attributing what they detect to species.

Claims about this, checked

Things people have heard, and what the evidence actually supports.

The research behind this page

12 studies, newest first. Each one has a page explaining what it found and what it could not show.

2018Science

A continental system for forecasting bird migration

Nightly migration intensity is predictable from weather, with atmospheric conditions accounting for the great majority of variation, and forecasts skilful up to a week ahead.

2018Nature

Long-distance navigation and magnetoreception in migratory animals

Navigation in migratory animals is multi-sensory and redundant: compasses of several kinds are well established, while the mechanism of the magnetic sense and the existence of a true positional map remain unresolved.

2015Current Biology

Eurasian Reed Warblers Compensate for Virtual Magnetic Displacement

Birds shifted their headings in the direction that would compensate for the displacement the simulated field implied, despite no other cue having changed.

2015Science

Terrestrial animal tracking as an eye on life and planet

Tag mass, battery life and data recovery, rather than biology, have set the boundaries of what was knowable about animal movement, and each reduction in tag size opened a new class of species and question.

2015Science

Aquatic animal telemetry: A panoramic window into the underwater world

Water blocks the signals terrestrial tracking depends on, so aquatic tracking works either by recording and later transmitting, or by listening for tagged animals with fixed receiver arrays — which makes coverage a function of where receivers happen to be.

2013Science

Social Learning of Migratory Performance

Deviation from the direct route fell as birds aged, and a group containing older birds performed better than one of the same size without them; the presence of a single older individual accounted for far more improvement than group size did.

2009Proceedings of the Royal Society B: Biological Sciences

Extreme endurance flights by landbirds crossing the Pacific Ocean: ecological corridor rather than barrier?

Birds flew directly from Alaska to New Zealand and eastern Australia over open ocean, without landing, on flights lasting several days and covering distances above ten thousand kilometres; departures coincided with favourable wind systems.

2009Science

Tracking Long-Distance Songbird Migration by Using Geolocators

Birds travelled far faster on spring migration than existing estimates assumed, with individual daily rates several times higher than had been inferred from ringing recoveries.

2007BioScience

What Is Migration?

Migration is better defined by the behaviour of the individual — undistracted, directed movement, with station-keeping responses temporarily switched off — than by distance travelled, seasonality, or whether the animal returns.

2003Oikos

Long-distance migration: evolution and determinants

Long-distance migration is best understood as a solution to seasonal resource variation whose limits are set by energetics and by the geography of barriers, and which evolves and disappears repeatedly rather than being a fixed property of a lineage.

1981Science

Genetic Basis of Migratory Behavior in European Warblers

The amount and timing of migratory restlessness tracked the population of origin rather than the rearing conditions, and responded rapidly to artificial selection.

1958Ardea

Two types of orientation in migrating starlings and chaffinches, as revealed by displacement experiments

Juveniles continued on the original compass heading from the new starting point and ended up in Spain, well outside the normal wintering range.

This page is a stop on a longer route

A guided journey reads several subjects in a deliberate order, with an argument for why one follows another. You can join in the middle.

Where to go from here

Each of these follows from something on this page — a relationship in the evidence, a claim people ask about, or the next mechanism along.

How complete this page is, and what it is still missing

NatureHQ publishes its own gaps. This page is at 79% completeness against what we would call a finished subject, and was last reviewed on 2026-09-03. It carries 7 claims and answers 6 mapped search questions.

  • 5 high-priority search intent(s) not yet covered
  • no research from the last few years is attached — check for newer work
  • Mammal migration — wildebeest, caribou, bats — is named rather than treated, and the ungulate literature is large.
  • Fish migration beyond salmon is absent, including eels, which migrate in the opposite direction and are stranger.
  • Diel vertical migration in the ocean is mentioned once despite being the largest animal movement on the planet.
  • The conservation dimension — flyway protection, stopover loss — is touched on and not developed.