An Arctic tern’s year, recorded in light
The Arctic tern makes the longest known yearly round trip of any animal, from the north to the edge of Antarctica and back. A light logger on one bird from a Swedish island shows what that trip is like from the inside: in one year it saw about 40% less night than a person in London.
Landöskärgården is a scatter of small islands near Kristianstad, in the far south of Sweden. Between 20 and 45 pairs of Arctic terns nest there each summer. An adult weighs about 108 grams.
From 2008 a team of Swedish biologists caught some of them on the nest and fitted each with a light logger weighing 1.5 to 2.0 grams, less than 2% of the bird. The logger records how bright it is, day and night, until the bird returns and is caught again, sometimes years later.
This tern carried logger K526, fitted in June 2014. It measured the light every minute and kept the brightest reading of every five.
The band at the bottom of the screen is that record. Each thin column is one day: noon at the top and bottom edges, midnight in the middle (Greenwich time). Gold where the logger saw daylight, dark blue where it saw night. At the colony in June, night lasted about five hours.
The globe shows the Earth as it was at midnight where the bird was.
By 12 July it had left the Baltic. It flew out across the North Sea to the open Atlantic and on down past Africa.
Once out in the Atlantic, the terns of the study sped up: from there to the equator they covered about 500 km a day, the fastest of their year.
It crossed the equator around 10 August. In the tropics day and night are about the same length all year, and the dark band in the middle of the record widens to almost half of each column.
These are the longest nights of the bird’s year, about eleven hours.
A light logger finds latitude from the length of the day. Around the September equinox the day is about twelve hours long everywhere on Earth, so from 2 September to 8 October the record cannot say how far north or south the bird was.
The time of noon still gives the longitude. For these weeks the globe shows only the meridian the bird was on.
When the record can place it again, the tern is far east of Africa, at about 35° south in the Indian Ocean, nearer to Australia than to Sweden.
Here the terns of the study flew more slowly, about 175 km a day, stopping to feed.
Around 6 November its estimated track reached the belt of Antarctic pack ice. It was now about 14,600 km from its nest.
Here spring is turning into summer, and the nights are getting shorter every day. On the ice the logger can no longer fix the latitude well: the study’s authors place the birds at 60° south or farther, so the line on the globe, drawn dashed from here, runs too far north.
On 20 November the logger recorded no darkness at all. Until 11 February it never did: 84 days in a row in which the light never fell to the logger’s level for night.
Around midnight the light dimmed, the way it does in twilight, then rose again. The band at the bottom is gold from edge to edge.
The terns spend about four months in the belt of pack ice, moving slowly, about 60 km a day, while they moult and feed.
This one worked its way west along the ice across 115 degrees of longitude. Its most southerly estimate, on 16 February, puts it at 68° south or farther, on the eastern side of the Weddell Sea.
In early March it turned north, and the spring equinox hid it for another month. On 26 April it was back in the Baltic.
That took under eight weeks. In a straight line, its last estimated position on the ice is about 13,600 km from the Baltic; its real route was longer.
By June it was back at the colony, a year after the logger was fitted. In that year it lived through two summers, one in the north and one at the edge of Antarctica.
Its logger counted 2,256 hours of night. A year in London has 3,820 hours in which the Sun is more than six degrees below the horizon, which is about where the logger reads dark.
Night here means the Sun more than six degrees below the horizon, counted in ten-minute blocks over the same twelve months.
Four loggers recorded the year cleanly (one stopped a week early). All four tell the same story: 2,256 to 2,368 hours of night in a year, and 42 to 86 days with no night at all.
The records of three other birds show weeks with no night where the Sun clearly set, which points to a failing logger. They are left out of these numbers.
The terns of this study flew about 50,000 km a year, between 45,000 and 60,000, more than once around the Earth. These are their routes, eight birds over a year each.
Every one of them went from the northern summer to the southern one, and back in time for the next.
The route and the light record are the published data of Alerstam and colleagues (2019), released with no restrictions (CC0) on Dryad. Positions are estimated from the light, twice a day, and are uncertain by about 150 to 200 km; here they are averaged over five days. A ten-minute block counts as night when every reading in it was at or below the logger’s dark level (2 lux for K526; 3 on the 0–64 scale of the older loggers). The study calibrated this level at the colony against a Sun 6.3 degrees below the horizon for K526, so places are compared against a Sun six degrees down. Three loggers recorded no night on days when the Sun at their estimated position went more than twelve degrees below the horizon, or for weeks outside the Antarctic summer with no position at all; those three are left out of the comparisons. From November to March the study estimated latitudes with a higher light threshold, and the authors write that they are biased north to an unknown degree; they only establish that the birds were at about 60° south or farther. The Earth’s daylight side is NASA’s Blue Marble; the night side is NASA’s Black Marble (2016). The Sun’s position is computed with the low-precision formulas of the Astronomical Almanac.