The fluke broke the surface off the Abrolhos Bank on a summer morning in 2003, the photographer reaching for a camera almost by reflex. Twenty-two years later, on a cold September day in 2025, a different observer on a whale-watching vessel in Hervey Bay framed almost the same patch of black and white skin on the underside of a humpback's tail. The two photographs, separated by more than 15,000 kilometres of open ocean and the better part of a human generation, sat in different libraries on different continents for two decades. They only met in May 2026, inside an algorithm and a database run from a small office in Lismore, New South Wales.
That match is the centrepiece of a paper published on 19 May 2026 in the journal Royal Society Open Science [1]. Its authors, led by Dr Cristina Castro of the Pacific Whale Foundation and co-led by Stephanie Stack, a PhD candidate at Griffith University, report the first confirmed evidence of bidirectional exchange between the humpback breeding populations of eastern Australia and Brazil. Two animals, identified only by the unique patterns on their tail flukes, had moved between breeding grounds separated by more than 14,000 kilometres of ocean. One of them holds the new record for the longest distance ever documented between sightings of the same individual humpback whale.
The two travellers
The first whale, designated in the paper as Whale A, was photographed in Hervey Bay in 2007 and again in 2013. In 2019, the same individual was identified off São Paulo, Brazil, a minimum straight-line distance of 14,200 kilometres, roughly the distance from Sydney to London. The second animal, Whale B, is the record-holder. It was first photographed in 2003 at the Abrolhos Bank, off the coast of Bahia, in a group of nine adults. In September 2025, twenty-two years later, the same whale was identified again, this time alone, in Hervey Bay. The minimum great-circle distance between the two sightings is 15,100 kilometres [1][2][3].
That second figure has occasionally been reported as 13,000 kilometres in earlier coverage; the peer-reviewed paper and the Griffith University press release settle on 15,100 [1][2]. A small correction note appended to the Phys.org version of the story initially muddied the year of the resighting, but the 22-year arithmetic only fits 2025, and that is the date the authors, the university, and independent reporters all now agree on [1][2][3][4].
A fluke-print in a global database
The matching itself was done by Happywhale, the citizen-science platform co-founded in 2015 by Ted Cheeseman, a doctoral student at Southern Cross University in Lismore, and Ken Southerland. Happywhale works the way a fingerprint database does: every humpback carries a unique pattern of pigment and scarring on the underside of its tail, recognisable for life [3]. Anyone with a camera and a clear view of a whale's fluke can upload the photograph. An automated image-recognition algorithm sorts the submissions against the rest of the catalogue, and a human reviewer, working in front of a screen, makes the final call on every possible match [1][2][5].
The dataset behind the new paper is enormous: 19,283 high-quality fluke photographs collected between 1984 and 2025 from eastern Australia and Latin America, representing almost 20,000 individually identified whales. Within that, the platform found just two animals that had been seen on both sides of the South Atlantic, a frequency of roughly 0.01 per cent of catalogued individuals [1].
Cheeseman, who has spent years on small expedition ships in the Southern Ocean and is now finishing his PhD, told NOAA Fisheries in 2024 that the project had not originally been built with climate science in mind. "We didn't anticipate that climate change would lower the capacity of the ocean for humpbacks," he said, after a separate Happywhale-based study documented a 20 per cent crash in the North Pacific population [5]. The platform now holds more than 60,000 individually identified humpback whales, the largest photographic dataset assembled for any cetacean species, and its scope has quietly expanded from a conservation side project to a central piece of infrastructure for southern-hemisphere whale research [6].
The Southern Ocean Exchange
The mechanism behind the crossing is the heart of the paper. Castro and Stack lean on a hypothesis that has been quietly gaining ground for more than a decade, the Southern Ocean Exchange. Under this model, humpbacks from different ocean basins, eastern Australia, New Caledonia, the west coast of South America, the sub-Antarctic islands, all converge on the same broad feeding grounds in Antarctic waters during the southern summer. Most of them migrate back to their natal breeding ground the following autumn. But occasionally an animal follows a different crowd home, and over time a few of these "vagrant" individuals may persist in their new population, or return to it again and again [1][7].
"Occasional individuals moving between distant breeding grounds can help maintain genetic diversity across populations and may even carry new song styles from one region to another," Stack said [1][3][4]. The genetic case is the easier one to argue. Humpback populations were reduced to a small fraction of their pre-whaling size during the industrial era, and a pulse of new arrivals from another ocean is, in the language of population genetics, a free infusion of variation. The second-order consequence, the one Stack hints at, is stranger and more interesting.
Songs that travel with the whales
Humpback song is one of the great puzzles of animal behaviour. Males produce long, structured, evolving songs that change across a season and spread, sometimes within months, from one population to the next. The 2013 paper by Ellen Garland and colleagues in PLOS ONE established the foundational evidence: song recorded on the Southern Ocean feeding grounds in Antarctic Area V could be matched to the eastern Australian and New Caledonian breeding populations, with shared themes from a given year appearing in both [7]. The mechanism, the authors argued, was cultural transmission on the feeding grounds. Whales from different breeding populations hear each other sing. They take the new song home. A hit single becomes a regional cover version in a single season.
If a small number of individuals are quietly switching breeding populations, the implications for humpback culture, not just humpback genetics, are real. A whale that learned the eastern Australian song off the east coast of Australia, spent a summer feeding in the same Antarctic waters as a Brazilian whale, and then migrated back to Brazil the following autumn, would carry a piece of Australian song with it. Over decades, even a handful of such exchanges could be enough to keep the cultural river flowing in both directions.
A 2024 paper in the same journal, Royal Society Open Science, had already documented the longest previously recorded great-circle distance between sightings of an adult humpback on two breeding grounds, an animal moving between the eastern South Pacific and the southwest Indian Ocean [10]. The new Castro and Stack paper pushes the record further still, and does so with photographs taken 22 years apart.
A warming ocean, a shifting menu
There is a third reason the paper matters, and it is the one that is hardest to fit into a tidy narrative. As the climate warms, the Southern Ocean is changing in ways that are pushing humpback feeding grounds around. Antarctic sea ice is retreating. Antarctic krill, the small, shrimp-like crustaceans at the base of the southern food web, are shifting southward in response, and their overall biomass is declining [8]. A single adult humpback needs roughly three tonnes of krill a day during the feeding season [9]. When krill moves, whales move to follow it, and when whales from different breeding populations are crowded into the same shifting patches of feeding habitat, the odds of an exchange go up.
The signals from the North Pacific are an early warning of what climate-driven prey collapse can do. There, the Hawaii-born humpback population grew from about 16,875 animals in 2002 to a peak of 33,488 in 2012, a recovery that briefly approached what researchers call the ocean's carrying capacity. Between 2012 and 2021, a prolonged marine heatwave, sometimes called "The Blob", reduced prey availability, and the population fell by about 7,000 animals, a drop of roughly 20 per cent [5]. The Castro and Stack paper does not predict an Australian or Brazilian equivalent, but it makes plain that the same physics, shrinking sea ice, moving krill, animals in search of a meal, is now operating in the southern hemisphere as well.
A week with the catalogue
What stays with a reader who spends a week with the catalogue is something physical. The 2003 photograph of Whale B at the Abrolhos Bank sits in a library somewhere. A different photograph, taken twenty-two years later in a different ocean, captured almost the same animal. The photographers who took those frames were not chasing a record. They were just paying attention.
The exchange is rare, perhaps 0.01 per cent of identified individuals, but it is real, and it suggests that the boundaries between humpback populations are softer than the catalogues once implied. In a warming ocean, that porosity is not a footnote. It may be the thing that lets the song continue.