Where birds went in winter was genuinely unknown until surprisingly recently, and the methods developed to answer it now produce data at a scale that has transformed the field.
Ringing
The oldest systematic method, attaching a uniquely numbered ring to a bird's leg.
Which produces information only when the bird is recovered, either recaptured or found dead.
Recovery rates are very low for small birds, which means the method requires enormous numbers of birds ringed to produce a modest number of recoveries.
It has nonetheless produced the foundational knowledge of migration routes, and long-running schemes provide survival and population data no other method matches.
Geolocators
Small devices recording light levels against time.
Day length and the timing of sunrise and sunset allow approximate position to be calculated.
Which is imprecise, particularly for latitude near the equinoxes, and it is small and cheap enough to fit on small birds.
The device must be recovered to retrieve the data, which means the bird must be caught again.
Their use revealed migration routes for small species that had been entirely unknown.
Satellite transmitters
Transmit position directly, which removes the recovery requirement.
Which produces detailed near-real-time tracks and requires devices heavy enough that they are limited to larger birds.
Solar charging has extended their operating life considerably.
Miniaturisation continues, and the size threshold has fallen steadily.
Radio tracking networks
Automated receiving stations detecting small tags as birds pass.
Which allows very small tags, since they only need to reach a nearby receiver rather than a satellite.
Continental-scale networks have been established, and the coverage determines what can be detected.
Weather radar
Radar detects birds as well as precipitation.
Which means the existing weather radar network has been recording migration for decades, and analysis of archived data has revealed long-term trends.
It measures aggregate movement rather than individual birds, and the scale is enormous — entire migration events across a continent, quantified.
Studies using this have documented substantial declines in migratory biomass over recent decades.
Stable isotopes
Chemical signatures in feathers reflect the location where they grew.
Which allows origin to be inferred from a feather sample without tracking the bird at all.
The resolution is coarse, indicating broad regions rather than points, and it works for any bird that can be caught once.
Citizen observation
Structured recording by volunteers, aggregated at scale.
Which has become a major data source, with online platforms collecting hundreds of millions of observations.
The data is uneven — observation effort concentrates near people and in pleasant weather — and statistical methods account for much of this.
The volume compensates for the unevenness in ways that professional surveys cannot match.
What has been learned
Journeys far longer and faster than expected, including non-stop flights of thousands of kilometres.
Extreme physiological adaptations, including organs shrinking before departure and regrowing on arrival.
Precise navigation using multiple cues including magnetic field, stars and landmarks.
And that the stopover sites where birds refuel are as critical as the breeding and wintering areas, which has changed conservation priorities substantially.
What tracking has changed in conservation
Knowing where birds actually go allows protection to be targeted.
Which revealed that some populations concentrate at a small number of stopover sites, making those sites disproportionately important.
It also identified mortality hotspots — illegal hunting areas, hazardous infrastructure, degraded wintering grounds — that were previously unknown.
Flyway-scale conservation, coordinating across all the countries a population uses, follows directly from this evidence.
Tag effects
Attaching a device affects the bird, and the extent is a genuine research question.
Guidelines specify maximum weight as a proportion of body mass, based on studies of effects on survival and breeding.
Which is why device miniaturisation matters ethically as well as practically, and why studies increasingly report comparisons between tagged and untagged birds.
Light pollution
Migrating birds are disoriented by artificial light, and collisions with lit buildings kill large numbers.
Which is addressable — programmes reducing building lighting during migration periods have documented substantial reductions in collisions.
Long-term datasets
Ringing schemes running for many decades provide the only evidence for population trends over that period.
Which means their continuity is scientifically critical and their funding is generally precarious.
The value of a monitoring dataset increases with length, and interruptions cannot be recovered.
Data sharing
Tracking data aggregated across studies allows analyses no single project could support.
Which has produced shared repositories and standards, and it required overcoming reluctance to share data before publication.
Sensitive location data for threatened species is generally restricted within these systems.
Hunting along flyways
Legal and illegal hunting during migration remains a substantial mortality source in several regions.
Which tracking has quantified, by identifying where tagged birds stop transmitting.
International agreements address this and enforcement varies considerably by country.
Which makes flyway agreements as important as domestic protection, since a population protected in one country can be lost in another.