New Zealand banded dotterels have reached Australia about four days later each year since 2014

The arrival figure and the stay figure check against each other. Eleven annual steps of about four days come to roughly 44 days ^[12]. The estimated stay fell by 44 days, from 173 to 129 ^[13], a drop of 25.4% from the 2014 value ^[14]. If departure held still, as the authors report ^[2], later arrival accounts for the whole loss. The match is approximate, since the source gives the yearly shift only as about four days.

Both dates are inferred. For each Australian area of about 23,000 km2, the team read arrival and departure off the way eBird sightings built up and tailed off ^[4]. The summary of the study does not say how changes in the number or location of birdwatchers over the years were handled, and that is the first thing we would check in the paper itself.

The thing this doesn't tell you is what any one bird did. Bart Kempenaers, who heads the department, said: "We want to know whether individuals change their timing of migration, and whether some birds switch their strategy from being a migrant to becoming a year-round resident." ^[8] The team is now following individual dotterels with satellite tags, light loggers or unique color-band combinations ^[9].

The climate link is a correlation. Each year's arrival date was set against monthly readings of the Southern Annular Mode, and later arrivals lined up with milder weather in New Zealand toward the end of the breeding season ^[5]. That shows the two move together. It does not isolate wind as the cause. The bird series starts in 2014 ^[1], while the mild-weather trend it is tied to runs back to the 1950s ^[15], so the window covers only the recent end of that trend.

The forecast rests on a trade-off between winter conditions and crossing risk. First author Bashar Jarayseh said: "Most banded dotterels that migrate come from harsher inland alpine habitats, escaping the New Zealand winter for milder conditions in Australia." ^[11] He added that "the ocean crossing is dangerous, and if their stay in Australia shortens, the benefits of migration may shrink while that risk remains, so within a few decades migrating individuals may be at a disadvantage." ^[16]

In our view, the shortened stay is the finding to rely on, provided the sighting method holds up against changes in birder effort. The end of migration is a projection. Kempenaers put the open question this way: "It remains to be seen whether there could be a tipping point beyond which the crossing is altogether abandoned." ^[18]