Catching the same Alaskan warbler twice is the price of reading its migration

An archival tag changes what a sample size means. A satellite transmitter sends its data home; a multisensor geolocator writes to its own memory and waits until someone catches the bird again ^[5]. Retrieval means fieldwork: find the male you banded the previous year, raise a mist net in his territory, and play his own song underneath it until he drops down to evict the intruder he thinks he hears ^[4]. Every track therefore costs a full annual cycle plus one successful recapture, which puts the earliest possible readout roughly twelve months after deployment ^[21]. Dozens of tags went out on warblers breeding at Anchorage ^[1][3]. How many came back is not in the account, though that is the number the statistics rest on ^[22].

What the barometer buys is time stamps. A bird climbing into migratory flight produces a steep pressure drop, and the tag records the timing of departure and arrival for every stop along the route ^[11]. Between flights the pressure reading reflects the elevation where the bird is sitting, which narrows the candidate locations ^[12], and the program GeoPressureR matches the recorded series against global weather data to find places whose pressure history fits, with light and wind data able to sharpen the estimate further ^[13]. Set that against the older option for birds this small, which infers position from the timing of sunrise and sunset ^[9]: those estimates can be wrong by hundreds of miles, and near the equinoxes, when day length is the same worldwide, the uncertainty becomes impractically large ^[10]. The blind spot is fixed to two dates in the calendar rather than to the bird's schedule, so its cost depends on when a given population happens to be flying.

The question the tags were sent to answer is old. In 1899 the ornithologist Richard McGregor found myrtle warblers wintering on the coast of California and proposed that they bred in Alaska and British Columbia ^[16]. Myrtle warblers in the northeast are known to winter along the Atlantic and Gulf coasts; the northwestern birds' destination has stayed unclear ^[15]. The tags were deployed to test whether Anchorage breeders take the shortest route to winter in California ^[17].

The available text stops one sentence short of the answer, at "To our surprise" ^[18], while both phys.org and The Conversation carry the account under a headline describing these migrations as sometimes circuitous ^[19][20]. The headline points toward a direction for the result while withholding the substance behind it. Absent are the route itself, the number of tags recovered ^[22], any error figure for the pressure-based positions ^[23], and any discussion of migration corridors or conservation planning ^[24]. A headline adjective describes a claim, not an effect size, and one surprising track from one Alaskan population would not by itself rewrite what banding data implied about songbird routes. What the material does support is narrower and still worth having: a device light enough for birds in a weight class the account puts below a spoonful of sugar ^[8] can now date each leg of a journey, though where those legs went comes out of a model fitting pressure to weather, and the tracks are only as good as that fit.