Just off a splinter of land sitting midstream in Brazil’s Rio Negro, a crew of scientists in two boats intently scan the sky. Silhouetted against twilit lavender and golden clouds, the crown of a Macacarecuia tree droops with fist-size seedpods. Like other trees nearby, Macacarecuia tolerate the river’s seasonal flooding. The island, known locally as Ilha do Comaru, is submerged as it is every year in March, and only the trees’ crowns poke above the surface.
A handful of Purple Martins dart by, their notched tails and angular wings splitting the heavy air. Then a flock amasses into avian clouds, the birds’ tempestuous movements like particles in Brownian motion. Suddenly, one throng forms a synchronous swirl. With a woosh, the birds drop like a shower of black sleet against the crepuscular horizon, alighting in the canopy just a few feet from the boats. More birds pour down, encrusting the trees, their clamorous calls rising. In a matter of minutes the spectacle has ended, leaving the sky still but for the emergence of stars.
The tiny island—just 12 acres, or slightly larger than Yankee Stadium—attracts an outsize number of the shimmery swallows. Host to a concentration of roughly 250,000 birds between February and April, it’s one of the largest Purple Martin roosts ever discovered. Its significance isn’t just its size, however, but also the pivotal role the roost may play in the bird’s long-distance migration. Comaru could be the staging ground, or launchpad, the scientists suspect, for many of 9.3 million Purple Martins that funnel from South to North America.
Mario Cohn-Haft, the curator of birds at the National Institute of Amazonian Research in Manaus, believes the island could be a key to understanding the species’ steady decline as well. “It is the single biggest window we have into what Purple Martins are doing in South America in the winter,” he says. The North American breeding population is thought to have decreased 25 percent since 1966; in some areas the drop has been even steeper, and in others the birds have disappeared altogether. Little is known about the challenges Purple Martins encounter when they head south each fall. “If we can trace their movements, figure out what they’re eating, and analyze whether they’ve been contaminated by pesticides and other pollutants, we can learn something about how they’re doing down here,” he says.
Cohn-Haft watches the birds’ restless silhouettes, his binoculars slung over his shoulder. He is among a dozen American and Brazilian scientists who have gathered here to observe and perform the most wide-ranging study yet of Comaru. By taking a close look at the birds packed onto this small isle, they hope to glean insights that can help secure the future of the entire species.
he arrival of Purple Martins each spring is eagerly anticipated across North America. The birds’ breeding range extends from Canada to Mexico and consists primarily of the United States east of the Rockies. Apart from small populations on the West Coast, in the southern Rockies, and in Southwestern deserts, they nest exclusively in structures—from hollowed-out gourds to miniature condo complexes—that humans erect to welcome them. The birds often return to the same yard, and even the same martin house, each year. If they show up a little late, “our phone rings off the hook with folks beside themselves, worried that their babies aren’t going to come back,” says Joe Siegrist, president of the Purple Martin Conservation Association, a nonprofit advocacy and research organization.
This close relationship didn’t always exist. Natural cavities, such as tree hollows, once offered ample nesting sites for Purple Martin colonies, but the loss of those—coupled with competition from aggressive, non-native species like the European Starling—has made the birds “100 percent reliant on humans providing housing for them to reproduce,” Siegrist says. As a result, Purple Martins have also become valued research subjects; thousands are banded every year. And more recently the birds’ site fidelity has made them ideal for carrying tracking devices, many of which must be recovered to retrieve information.
Although Purple Martins have been studied extensively in North America, scientific knowledge tapers off when they fly south. Just where they go, what routes they take, and what critical habitat lies along the way remain largely a mystery. Purple Martins are aerial insectivores—they hunt insects on the wing—and that places them among the fastest-declining avian groups. (Insect populations have also dropped, due to causes such as pesticide use.) But the birds’ risk is compounded by their marathon journeys.
Purple Martins had long been seen in the Amazon, but conventional wisdom held that they were only passing through. Some of the first evidence to the contrary came in 2007, when Bridget Stutchbury, a biologist at York University in Toronto, equipped the first songbirds—20 Purple Martins and 14 Wood Thrushes—with geolocators. By measuring light levels, the devices could determine a bird’s location to within about 200 miles. Stutchbury recovered two geolocators from martins the following spring and discovered that one bird had spent the winter in the Amazon while the other wintered much farther south.
Seven years later Stutchbury’s postdoc Kevin Fraser, now an ornithologist at the University of Manitoba, equipped 105 Purple Martins with more advanced trackers: GPS devices that could place a bird to within about 30 feet. The 14 loggers he retrieved revealed that all but one had overwintered in the Amazon and five had spent a lot of time near Manaus crammed into an area no larger than a suburban home lot. Siegrist, who collaborates closely with Fraser, says this almost certainly proved the birds had gathered in a roost—and if the scientists could locate it, they might begin to understand what role wintering grounds play in Purple Martins’ survival.
Siegrist and Fraser decided it was time to go have a look around for themselves. They visited Manaus with two other colleagues in November 2016 and again two years later. They checked the sites that Fraser’s birds had visited but found nothing. As it turned out, their timing was off; they were three months early. They also eventually discovered that the birds relocated their roost from Ilha do Papagaio, where Fraser had pinpointed them. “They switched right when we thought we knew where they were,” says Siegrist.
Finally, early in 2019, Cohn-Haft heard from a student about a huge flock of birds that a local had showed her, only 20 miles from Manaus. The guide, named José Francisco dos Santos de Moraes, runs a honky-tonk theme park featuring trained freshwater dolphins. With an eye for new attractions, he had started taking visitors to see the flock as it arrived at dusk. The guide offered to take Cohn-Haft, too. Cohn-Haft knew right away that it was a roost of Purple Martins: “It really was just one of these absolutely goosebump-raising, heartwarming, uplifting, inspiring moments.” And he sensed immediately that the roost could provide a treasure trove of scientific data.
t’s evening at Flutuante Restaurante do Paulão (Big Paul’s Floating Restaurant), anchored a short distance from Comaru. The scientists have rented the whole place as a cafeteria, dormitory, and lab. They sit around a table where they had earlier lunched on feijoada, Brazil’s national meal of pork and beans, recalling the ease with which they have just collected Purple Martins. “You can paddle a boat right up and pick them off the trees like apples,” says Siegrist.
Now the captives dangle in cotton sacks from a taut line like a load of socks hanging out to dry. The scientists remove the feathered subjects one at a time and pass them from person to person in a slow ornithological bucket brigade.
Ramiro Dário Melinski, a Brazilian research assistant at Cohn-Haft’s institute, wipes a bird’s mouth and cloaca with small cotton-tipped swabs. He seals the samples in pinkie-size test tubes and drops them into a thermos of liquid nitrogen. Genetic studies of material collected from the birds will help Erika Hingst-Zaher, a researcher at Brazil’s Instituto Butantan, and her collaborator, C. Loren Buck, a biologist at Northern Arizona University, determine what pathogens the birds carry. Disease, perhaps aggravated by a toxic exposure, could explain the birds’ declining numbers.
Melinski passes the bird to Clarissa Santos, then a research assistant and now a grad student studying ecotoxicology at the University of São Paulo. She clamps a band on its leg and hands it to Cohn-Haft.
Cradling the animal in his palm, Cohn-Haft gently blows its glossy breast feathers aside. He scrutinizes the heft of the pectoral muscles. He fans out a wing like a poker hand. The bird has recently molted and its flight feathers are fresh. “They’re all brand spanking new,” he says. Flight feathers become scruffy with use. That’s good enough for short excursions, but making an intercontinental trip with worn feathers would be like driving cross-country on bald tires. He declares the bird fit to depart for North America.
This is the third year that Cohn-Haft and Siegrist have studied the martins at Comaru. They’ve captured about 100 birds each time, and most of those seem about ready to head north—not just because their feathers are fresh but also because their muscles are bulked up for travel and they have ample fat stores. Radio tags attached to birds in past seasons have shown they stay at most two weeks. This gives Cohn-Haft confidence that many more individuals pass through than the count reflects at any given time—potentially closer to 1.5 million.
The fact that such a large fraction of the world’s Purple Martins relies on a single site raises conservation concerns. Though there are currently no plans to dam the Rio Negro, a future hydroelectric dam could, for instance, wash away the roost. “That tight concentration makes them vulnerable,” says Siegrist.
The research at Comaru could point to more global implications as well. Mercury from natural and human sources, such as artisanal mining operations, transforms in hydroelectric reservoirs and farm ponds into a highly toxic form that may then travel up the food chain to Purple Martins. Hingst-Zaher and Buck, who both joined the team at Comaru, are investigating whether mercury contamination affects the birds’ endocrine systems, reducing their fat stores and making them less fit to migrate. Buck has already found mercury in the feathers of Purple Martins in North America; samples from their wintering ground will help determine the severity of the problem there. A martin contaminated in Brazil might not reach North America for study, he says: “We may never even sample it because it flies and dies.”
The scientists hope any discoveries they make will help uncover what’s behind the decline of other songbirds, especially other aerial insectivores. Siegrist likens Purple Martins to the proverbial canaries in the coal mine. An early run of bad luck made the Purple Martin dependent on the kindness of strangers for survival. But the bird’s adaptability and now the discovery of Comaru could give it a lead role in saving its feathered brethren.
It’s around midnight before the researchers have processed the last bird. Two graduate students lift them all off the line, still trapped inside their cloth bags, and gingerly step into an aluminum boat. Cohn-Haft urges them to hurry up. The birds must get back to their roost so they can have a good night’s rest. After all, they have a long journey ahead.
This story was supported by the Pulitzer Center. It originally ran in the Fall 2022 issue as “Home Away From Home.” To receive our print magazine, become a member by making a donation today.