To cross the Sahara every autumn and spring, a little songbird snaps out of its nocturnal travel habits and flies for 40-60 hours, a study has found.
Scientists caught the pied flycatcher in the act using tiny light-logging tags, which recorded sustained periods of sunlight during its seasonal trek.
The findings appear in Biology Letters.
Previous evidence painted an unclear picture of whether small birds like this use daytime rests or non-stop travel to negotiate the desert.
"It was a bit of a controversy, as to what was going on," said Janne Ouwehand, a PhD student at the Unviersity of Groningen in the Netherlands.
Many birds, including small songbirds like the flycatcher, breed in Europe during summer and spend their winters in central Africa. So twice a year, they must cross the vast and inhospitable Sahara Desert.
Back in the 1970s, British ornithologist Reginald Moreau suggested that a non-stop flight of 40-60 hours was the obvious solution.
But more recent observations have suggested that breaks are involved. Groups of birds can be seen resting in the heat of the day, for example, and radar measurements have spied many more small birds traversing the region's skies at night than during the day.
The radar data, however, cannot track individuals or identify different species. That sort of detailed observation is made much more feasible, even for small birds, by new logging and tracking technology.
Ms Ouwehand and her colleagues attached tiny loggers to 80 pied flycatchers at a breeding ground in the Netherlands, late in the summer of 2013.
These birds are smaller than a house sparrow and weigh about 12g - less than three one-teaspoon sugar cubes. In the summer they nest right across Europe and this particular population spends its winters in the Ivory Coast and Guinea.
The 0.5g loggers are like "a very tiny backpack" for the birds, Ms Ouwehand said, and make no measurable impact on the little animals' performance. They record the light level and the temperature every 5-10 minutes, for months on end.
A lot of information can be gleaned from those light cycles.
"You can make a nice daylight curve, and you can get a rough indication of their position," Ms Ouwehand told the BBC News website.
"You have the day length, which gives you an indication of the latitude, and the midpoint of the day and the night gives you an indication of the longitude."
The following summer, she and her colleagues retrieved 27 of the gadgets and started to look at the data.
"It's always extremely exciting when they return - how many will return, which birds do we see back. And you have to wait for a whole year; you don't get any data in the meanwhile so you just hope everything will be fine."
Of the 27 returned loggers, 15 contained data on both the spring and autumn migration periods. But as it turned out, the pied flycatchers' distinctive behaviour made the light curves rather messy and it was difficult to glean coordinates from them.
"You get a lot of spikiness in your data: very bright periods and very dark periods. That's because they're in the shade, in woody habitats and things like that," Ms Ouwehand explained.
On two occasions however, coinciding with the migrations, something odd happened in the data. For up to two whole days, that up-and-down, light-and-shade spikiness was completely absent.
"We discovered this very clear pattern twice a year and we thought, what are the birds doing here?"
At the same time, the temperature readings were cooler than during the days on either side.
This can only mean one thing, Ms Ouwehand says: a long-haul, high-altitude flight that traverses the Sahara in one hit. These tiny birds, which usually do their migrating under cover of darkness, are crossing the desert in broad daylight.
Long way round
"We see not only that they can make these flights, but also they use a different strategy in autumn than in spring."
The pattern of light and temperature changes is very different in the spring, she explained, suggesting an alternative route for the birds' return flight to Europe - possibly going a longer way around to spend more of the journey over the ocean.
This is not among the longest of bird migrations, which can span whole oceans in a single flight. But it is perhaps the clearest indication yet of the different strategies small birds use to cross the world's biggest desert.
As technology continues to improve, Ms Ouwehand said, and more researchers enlist individual small birds to gather their own backpacks full of data, we will learn more and more about those specific strategies.
"We are just at the moment when we are really able to show these patterns," Ms Ouwehand said.
"I find that pretty exciting and I'm very curious about what we will see in other, comparable species. Which are the individuals that do stop? Is it age differences, ecological differences, species differences?
"That's really now what we can start to understand."
Follow Jonathan on Twitter