Richard Robinson

Summer 33

Two people have been counting albatrosses on remote islands in the subantarctic for more than three decades. Their research shows that at least one species is en route to extinction. A few changes to the way we fish could save it.

Written by       Photographed by Richard Robinson

The wind was ruthless, the air grey with salt spray whipped off the ocean. The hillside ahead seemed to be moving; it was covered with long grass, waving like kelp in the gale. Graeme Elliott couldn’t see anything. He knew the birds were there, but like any sensible creature, they were hunkered down below the mounds of tussock.

Elliott didn’t want to startle an albatross sitting on its egg; that struck him as unimaginably rude. He could call out to them, he supposed—but what should he say? It wouldn’t do to shout nonsense. This is a subantarctic island, he reminded himself, it doesn’t matter, but some instinct nudged him towards decorum.

Then he had an idea. “Evening Post!” he yelled, just like the newspaper boys in Wellington used to call. Several white heads popped flamingo-like out of the  grass. Long hooked beaks, faces impassive and expressionless. Elliott waded through the tussocks toward one, then hesitated. The Gibson’s albatross regarded him with one dark eye, bright as a pin. Crouched down, he was eye-to-eye with her.

“Hello,” he said.

The chick is losing the last of its down—it’s just a few weeks away from taking off to spend the next few years of its life at sea.

That was during his second summer counting albatrosses. Maybe his third. It’s hard to remember which, after 33 summers counting albatrosses. Neither Elliott nor biologist Kath Walker, his partner in life and work, can quite remember.

First, Elliott and Walker counted them by accident, and then by design, and then for work, and then for nothing. By the start of 2006, 16 summers in, they were paying to do the counting, paying to sail four days’ south of New Zealand through the heaving seas of the Roaring Forties, paying to spend their summers on an archipelago with weather so miserable that its only settlement collapsed in drunkenness and disaster after eight months.

Graeme Elliott and Kath Walker weigh an Antipodean albatross in 1996, Summer 6 of their multi-decade study.

But return they did, and over the past decade, the albatross census has provided an impetus for fisheries reform. It may yet become the record of an extinction event.

[Chapter Break]

The first albatross count was an accident: Elliott and Walker were supposed to be looking for a completely different bird.

This was back in 1989. Elliott and Walker were biologists—Walker obsessed with giant snails and Elliott having spent several years studying banded rails, little birds that live in wetlands at the top of the South Island, where they are extremely well camouflaged. DOC offered the couple a spot on a trip to the subantarctic to search for another rail, the Auckland Island rail—a bird that was missing, presumed dead.

A trip to the subantarctic? They couldn’t sign up quickly enough. Elliott remembers watching the archipelago loom up out of the horizon, fortified by vertical cliffs, like a palisade of stone.

This is going to be hard work, he thought. It wasn’t.

“We found the little rail in about five minutes, and it was really quite common,” he says. “And we were thinking, ‘What else can we do that’s useful while we’re on the island?’”

They looked around. There were a lot of albatrosses. “At the time, it had just become news that the albatrosses were being slaughtered in large numbers by the tuna longline fisheries. So we thought, ‘Well, let’s count them.’”

An Antipodean albatross dives for fish scraps off the coast of Great Island, one of the Three Kings Islands northwest of Cape Reinga.

In Walker’s mind was a study she’d just read about wandering albatrosses on an island in the Indian Ocean, how their numbers had halved during the 1980s. What if the same thing was happening here?

When Walker and Elliott got home, they realised their spontaneous albatross count was “bollocks” and required more forethought than they’d given it. So they got started on the forethought.

There’s a science to counting birds. You don’t count all of them, because there are usually too many. You count a small sample, but then you have to make sure that the number you end up with can be stretched out to account for the entire population. Fortunately, statistics is one of Elliott’s other obsessions. They designed a mark-recapture study, where they’d put bands on the legs of a certain number of birds, then return year after year to see who was still alive, who’d paired up, and who’d started a family.

The tricky thing about Gibson’s albatrosses is that they don’t return to land every year, so the count has to allow for who’s still overseas. Young albatrosses spend several years foraging at sea before returning to the island where they hatched. Pairs meet up every two years to build a nest and lay an egg. Sabbatical birds, they’re called, the ones taking a year off from raising a chick.

The first official albatross census, Summer 1, took place from 1990-1991, and afterwards, Walker and Elliott knew this was how they wanted to spend their summers for the rest of their lives.

The juvenile albatross at left was doing the same thing when it was hooked by a Japanese tuna longliner east of Tasmania. The man cradling the bird—to demonstrate its size—is a crew member of an Australian patrol vessel.

Adams Island, where the albatrosses lived, was bursting with life. It was how New Zealand was supposed to be. Mostly, though, they just loved the company of the albatrosses—birds the size of children. “I’ve worked on lots of silly little birds and their behaviour is always fleeting, you know, they flit and they move quickly,” says Elliott. “It’s all over in a millisecond.”

But albatrosses’ gestures are amplified because the birds are so big. Their expressions and reactions are legible to a human onlooker. Elliott and Walker learned to spot mates greeting each other after two years apart fishing at sea, and young albatrosses on their first visit back to the island.

“When they turn up, they don’t walk very well, because they haven’t walked on land for five years,” says Elliott. “So you see these birds staggering around and you think, ‘Wow, that’s a youngster.’

And then you see them interact with other birds. They all stand back and they look slightly frightened. Like school kids. They look uncomfortable. And you can spot it.”

As well as designing the population study, Elliott and Walker established protocols for etiquette. They’d always greet an albatross before approaching. Perhaps they’d have a little chat with it. “‘Ooh, you’ve got an egg’, that kind of thing. It doesn’t sound very scientific, but whatever.”

[Chapter Break]

From Summer 2 to Summer 15, Elliott and Walker’s annual count produced good news: the number of Gibson’s albatrosses was on the up-and-up.

Case closed. The birds were doing great. DOC and Fisheries New Zealand, who had been paying for the study, stopped funding it after Summer 15. The pair went back to the subantarctic the following year out of curiosity, and in Summer 16, the Gibson’s population plummeted. The number of female birds dropped to a record low, fewer than there had been in Summer 1. Now Walker and Elliott were truly stuck. They had to find out whether this was a blip or a trend.

And so they called in favours. They hitchhiked on scientific vessels and with tourist expeditions. They cobbled together funding and support from other places: television productions, non-profits, a retired British Merchant Navy captain who loved birds. They cut a deal with another group of scientists studying petrels on the islands. And in the years when nothing came through, they started spending their own money.

Complicating matters was the fact that Walker and Elliott had started a second albatross study back in Summer 4. This one counted Antipodean albatrosses, which live on their own set of subantarctic islands, the Antipodes. They’re the closest relatives of Gibson’s albatrosses: not similar enough to be the same species, and not different enough to be a different one. They’re more like cousins: same grandparents, different upbringing. Darker feathers, different habits.

Walker and Elliott called in reinforcements every summer to survey birds, and they’d switch islands themselves. Sometimes they’d study Antipodeans, sometimes Gibson’s.

Richard Robinson encountered these Antipodean albatrosses socialising on an empty nest, “making a hell of a lot of noise”. The albatross on the right is part of the mark-recapture study. He received his leg band as a chick in 2002, and lived at sea for several years before returning to Antipodes Island in 2010. He was spotted courting a female albatross in 2012 and 2013, and the pair made it official in 2014, raising a chick together. But his mate hasn’t been seen again since, and he’s been hanging around for close to a decade now without a partner.

The Antipodean albatross population crashed at around the same time. From Summer 17 to Summer 22, the number of birds on both islands just kept dropping. To this day, Walker and Elliott still don’t know what changed in Summer 16. “We just knew that those big albatrosses were copping it,” says Elliott.

Albatrosses spend years at a time at sea, flying in immense zig-zags, smelling the air, waiting for the scent of a meal: an airborne chemical that signals there are squid or fish near the surface.

Tuna-fishing boats bait their lines with squid and lay them on the surface of the water, where they stretch from horizon to horizon, carrying thousands of hooks. This is called surface longlining, and the “long” part of it is no joke—the lines can stretch for up to 100 kilometres. Albatrosses can detect a longline from about 30 kilometres away; tracking devices show them turning mid-flight to head straight for the squid buffet.

Hooks are designed to catch things, and they don’t discriminate between birds and fish. Albatrosses caught on hooks usually drown, dragged underwater by the line as it sinks.

If one half of an albatross couple vanishes, it can take the other years to find a new mate. Females may pair up again after a couple of years, but a young Antipodean male could spend a decade looking for a new partner, and an older male? “It’s decades if at all,” says Walker.  If one parent doesn’t return from a foraging flight, its chick dies, because working alone, the remaining bird can’t keep both itself and its chick from starving.

Young albatrosses don’t start a family until they’re at least eight years old, and even then, pairs raise only one chick every two years. That means the population can’t rebound quickly from any shocks. In fact, it hasn’t rebounded at all from Summer 16, as there are too many factors going against it. Walker and Elliott don’t know all of them, but one factor is obvious.

[Chapter Break]

There are some pretty good ways to avoid catching albatrosses, if only we used them.

Laying a longline in the water is called the soak, and so much depends on when the soak happens: setting it at dawn catches more birds than if it is done at night.

Fishers can scare birds away with a line of streamers that flies above the line of hooks. They can get the hooks out of the way quickly by weighting the longline, or by shielding the hooks with covers that retract when the hooks are 10 metres down.

Night setting; bird-scaring line; weighted hooks. Every seabird conservation advocacy organisation in the Pacific region recommends surface longliners do all three of these things. New Zealand law requires vessels to do two of them. In reality, it’s hard to know how many fishers are doing any of them. New Zealand doesn’t police its surface longline fleet very often, but trusts it to operate in good faith and self-report any seabirds it catches.

Because very few longline vessels have observers—in one New Zealand fishing area, there are no observers—it’s impossible to obtain independent proof of whether these safety measures are being used, or of how many seabirds are being caught.

Getting onto Antipodes Island is a logistical challenge—packing down for the season, Rexer-Huber sends gear by flying fox to a waiting vessel.
Every couple of days, Parker and Rexer-Huber check in on specific nests, mapped above, in order to record whether eggs have hatched or failed.

Research also shows that self-reported data is not trustworthy—without external monitoring. A Fisheries New Zealand study found the number of bycatch reports doubled when cameras were present on bottom longline vessels in 2016. In Australia, bycatch reports increased eightfold after the addition of cameras to longline vessels in 2015.

Installing cameras on vessels may solve the lack-of-observers problem. The government announced cameras would be introduced to all New Zealand surface longliners in October 2018, but this move has been postponed multiple times; as this magazine went to print, cameras were scheduled for November 2023.

Meanwhile, in the past fishing year, the government checked just three surface longliners (there are around 30 in operation) and none of them was following the rules. They were doing one, using bird-scaring lines, but not using them properly: one line was ripped off, and the vessel continued fishing; another crew couldn’t get the streamer line to fly above the longline, which means it doesn’t scare birds in the right place; a third line was too short. Two of the vessels didn’t report bycatch properly; two of the vessels caught large numbers of seabirds.

On photographer Richard Robinson’s November 2022 visit to Antipodes Island, he barely saw adult albatrosses—just chicks in nests. “They’re ginormous, big, fluffy things,” he says. “They’re all hungry and they’re all crying out for food from their parents.” The parents were away foraging at sea. If anything happens to an albatross parent, its chick dies on the nest, as at right. Raising a chick is so energy-intensive that parents take a year off to recuperate before having another go.

There are other things that fishers can do to prevent seabird bycatch, like being careful about where and when they throw fish guts overboard. Each vessel has its own bycatch reduction plan detailing actions it has agreed to take. Not one of the three vessels was following its plan. In the previous two years, when more vessels were audited, only a third of surface longliners were following their plans. Sadly, fishers doing the right thing are being obscured by those who aren’t.

Fisheries New Zealand is aiming for zero bycatch, but the surface longline fleet hasn’t gotten any closer to this goal over the past two decades. The total number of seabirds caught every year by the fishery has been about the same since 2005—give or take 700 birds—despite fisheries liaison officers, education, bycatch reduction plans, working groups, and the requirement to have the 10 Golden Rules of responsible fishing somewhere on board. Over that same time period, Antipodean albatrosses and 12 other seabird species have been shifted into a higher category of endangeredness—they’re more at risk than before.

There were some 1070 drift longliners from 23 countries fishing within the area of this map in 2021. Vessels flagged to China (380) and Taiwan (286) outnumbered all others. The boats’ tracks describe straight lines on their way to high-seas fishing areas. Once there, they make rapid course changes as they set longlines and then retrieve them. There’s a relatively small overlap between the longliners’ range and the albatrosses’ range. But the density of fishing activity essentially creates a 12,000-kilometre-long net right across the Pacific. During any given year, longlines and their albatross-attracting scent cover at least 40 per cent of the world’s oceans—fishing is one of the most spread-out food production systems that we have, covering four times as much area as agriculture. This graphic is based on 1.5 million points of automated tracking data pinged by fishing boats—data which are monitored by Global Fishing Watch. These have been overlaid with the tracks of 36 albatrosses tagged on Antipodes Island in 2021. Male birds typically forage further south than female birds—they’re physically larger and better able to withstand strong winds at these latitudes. Females travel further north, into calmer seas but more dangerous waters, and come into contact with fishing boats more often. Walker and Elliott estimate that an extra 700 male and 1600 female Antipodean albatrosses are killed every year—on top of birds that die of natural causes. On Antipodes Island, there is now a severe population imbalance, with at least twice as many males as females. As that disparity has increased, male Antipodean albatrosses have begun courting and pairing up with each other.

Antipodean and Gibson’s albatrosses nest only in New Zealand’s subantarctic, but this is a problem that goes beyond our territorial limits. An albatross crosses the boundary of a country’s exclusive economic zone without getting a stamp in its passport, and it doesn’t stop to check what flag a fishing vessel is flying.

Where are the birds going, and what’s happening to them there?

[Chapter Break]

The tracking device was the size of two matchboxes end-on-end, maybe a little thinner—the first tracker small enough and cheap enough to transmit an albatross’s locations on the wing.

“So I’m going to put it on you,” Elliott told the seated albatross. It was Summer 29, and this was a two-person job—one person to stop the bird from flying away, another to attach the tracker. The device went right in the middle of the bird’s back, as close to its centre of gravity as possible. Walker wrapped some tape around the feathers to keep it there. By the time the albatross wriggled away, Elliott’s back hurt. This was like gardening, he thought, having to bend over for long periods at a time.

The bird was clearly displeased with her new accessory, but Walker and Elliott knew the tracker would fall off before she returned to the island. In the meantime, they hoped it would tell them what was happening to Antipodean albatrosses at sea.

And it did.

Sixty-three Antipodeans got trackers, allowing Walker and Elliott to keep an eye on their travels from home in Nelson.

They watched as female albatrosses ventured further north than anyone expected—which meant the birds’ travels overlapped even more with territory frequented by offshore tuna longliners. The birds’ foraging patterns, they realised, reflected how fish species were moving due to warming waters. In other words, the birds were mapping climate change.

Some of the trackers stopped short, and when they did, Walker and Elliott overlaid the track with the transmissions of fishing vessels operating in the area. Eight albatross trackers merged with the paths of fishing vessels and stopped transmitting. One track joined the path of a fishing vessel and then continued, its path duplicating the path of the boat, all the way to port in Pagopago, American Samoa.

The consequences of all this became very clear in Summer 33.

[Chapter Break]

The Antipodes Islands look as though they’re sinking back into the sea, their tussock flat-tops tilted, capsizing on a geological timescale. It was Graham Parker and Kalinka Rexer-Huber’s first time counting albatrosses there, and they weren’t prepared for all the violence.

Parker and Rexer-Huber spend a lot of time living on remote specks in the ocean. They’ve worked with wandering albatrosses on Gough Island, a dot in the middle of the southern Atlantic, and southern royal albatrosses on Campbell Island, in New Zealand’s subantarctic. They’d been subbing in for Walker and Elliott, counting Gibson’s albatrosses, since Summer 26. But this was different.

“It was quite striking, the level of aggression,” says Rexer-Huber. Male albatrosses—she calls them “blokes”—arguing, fighting. Ganging up on females as soon as they landed on the island, six blokes at a time, all trying to mate with one bird. “We’ve not seen that before to that degree.”

So many females had been killed at sea that there was now a serious population imbalance. Three blokes for every female.

In Summer 29, a wildlife documentary crew filmed a male-male pair of Antipodean albatrosses that had set up a nest together, and the clip went viral, as though it was an example of progressive albatross politics. (The nest was empty; they weren’t foster fathers.)

Not every albatross finds a mate, says Rexer-Huber; that’s normal. “There’s still plenty of males sitting around on their lonesome waiting for someone who doesn’t come back, or potentially, who haven’t had a match for many years. They just sit around for several months; they’re there all the time. But that’s different to when those same lonesome blokes are really actively harassing everybody.”

Parker and Rexer-Huber plan tomorrow’s route from the hut on Antipodes Island. “Every single day, you look at the weather and go, ‘Oh, that aspect might be a little bit more friendly given where it’s howling from today’,” says Rexer-Huber.
Looking closely at this albatross chick’s beak reveals its large nostrils; the birds have unusually big olfactory bulbs, the part of the brain responsible for interpreting scent. They navigate by smell as well as sight.

Meanwhile, Parker, a former commercial fisher, was in discussions about a longline fishery off the coast of Otago which is at high risk of catching Antipodean albatrosses. A collision between a container ship and a longliner in New Zealand waters in 2021 had led to observers being withdrawn from some fishing vessels, including all the Otago surface longliners. As a result, it’s not clear how birds and fishers are getting on there—which makes it hard to come up with better ways of preventing bycatch.

He was Zooming into these meetings from his hut on Antipodes Island, then he’d leave the hut, walk up the hill, and watch the effects of albatross bycatch in action: the noise, the aggression, the strange behaviour. “On a personal level it was quite difficult.”

Once most of the birds had paired up, and the ambiguity about who was getting together with who was settled, the fights diminished and the mobs dispersed—to Parker and Rexer-Huber’s immense relief.

It’s a tricky thing to talk about. They don’t want anyone to think, Oh, well, it’s the birds’ fault, they’re beating each other up. Rather, human actions are now shaping albatross behaviour.

[Chapter Break]

The albatross study raises an existential problem in conservation science. This data is very useful for figuring out how the world is changing, but there are no incentives for collecting it. The timeframe is just too long.

The average monitoring study might be funded for three or four years—enough to catch a species in an up phase, or a down phase.

“Those short studies, you actually can’t get the true story,” says Elliott. “Things like climate change and the depletion of global fisheries are things that play out on really long timescales. You don’t see anything in three or four years.”

Yet no institution, public or private, could guarantee the support to keep the count going.

“You’d never get a 30-year project out of a government without somebody being a champion for it,” says Elliott. “These wandering albatrosses—if Kath and I hadn’t been so keen, it would have fallen over years ago.”

Similar studies run entirely on people’s enthusiasm, and mostly people named Graeme. Graeme Taylor has been working on grey-faced petrels around Auckland (“for centuries”, says Elliott); Graeme Loh has been monitoring fairy prions nesting in the cliffs of the Otago Peninsula since 2000. Graham Parker and Kalinka Rexer-Huber have been counting albatrosses for more than a decade. Jim Mills, who started monitoring red-billed gulls in 1964 and tracked the birds through a population plunge and recovery, continues his population count to this day. Mills lives in upstate New York and travels to Kaikōura every year. One of his project’s outputs was a paper in Science in 2011 that calculated the amount of fish needed in the ocean to sustain seabird populations. It assembled data from seven ecosystems and 14 bird species around the world, finding that within each ecosystem, there was a tipping point: if the amount of fish and krill dropped by more than two-thirds, then seabird decline was inevitable. It sounds obvious, but studies like this also provide the kind of solid evidence that feeds legislative change.


Still: “A long-term monitoring project isn’t very exciting,” says Parker. “And then the data are kind of simple. So people in universities aren’t necessarily super excited about long-term datasets either. There’s more exciting things you can do.”

“It’s just slow returns as opposed to quick returns,” says Rexer-Huber. “And, really, if you’ve got students doing PhDs, for example, you need data that can be providing interesting results quickly.”

Graham Parker and Kalinka Rexer-Huber take feather samples from a 23-year-old male Antipodean albatross. The bird found a mate the previous summer after 14 years of failed courting attempts.

The study also hints at the fate of other seabirds living in less-friendly parts of the subantarctic. Salvin’s albatrosses nest on the Bounty Islands and the Snares Islands, which are “both very difficult places to spend long periods of time”, says Rexer-Huber. White-capped albatrosses are considered less endangered than Gibson’s and Antipodeans, but we don’t have a fix on how they’re doing. They were counted from the air until recently, a type of survey which can have deceptive results.

Both species are frequently killed by New Zealand fishing vessels. If we had good-quality information about them, worries Rexer-Huber, maybe we’d discover they were in more trouble than we’d realised.

[Chapter Break]

Internationally, the number of albatrosses caught by fishing boats has decreased over time, which looks like an improvement. In reality, says Stephanie Borrelle from BirdLife International, there are simply fewer albatrosses to catch.

If nothing changes, Antipodean albatrosses will become extinct within her lifetime. She’s frustrated: the problem is fixable, and the solutions pretty simple. “It’s really just about getting people on board. They’ve been faffing about for 25 or 30 years and the measures are still the same.”

Borrelle acts as an observer at meetings of international fisheries organisations which set rules for fishing nations around the Pacific. There isn’t a lot of interest from other nations in solving the bycatch problem, she says. Lately, tuna stocks have been increasing, so the industry plans to increase fishing efforts, which will increase bycatch in turn. That said, one piece of recent positive news has come from the same industry: a group of tuna-fishing companies called SeaBOS, which includes a couple of large global suppliers, are adopting the three seabird-safety measures for their fleets.New Zealand’s attempts to advocate for seabirds have met with little success. “It’s hard to spend political capital on seabirds when we don’t have much of it anyway,” says Borrelle. It’s also awkward for New Zealand to advocate for safety measures that it doesn’t use itself.

Young albatrosses on Antipodes Island are staying single longer than ever before. After hatching, they forage at sea for three to six years before returning to their home island. Some birds pair up and start a family at the age of eight; most do so around age 12. “But it’s all been mucked up for years now,” says Kath Walker, “with some not breeding till their early 20s.”

Meanwhile, on Auckland Island, as on Antipodes Island, the albatross population has never recovered from the crash of Summer 16. The number of nests is still half of what it was before the crash. The number of birds surviving from year to year is lower, and the number of chicks raised to adulthood is fewer. The population isn’t recovering fast enough to replace the birds who are disappearing at sea.

Still, Elliott and Walker are preparing for Summer 34. It takes half a year to write the report about the previous summer and half a year to get ready for the next summer, and that’s all a bit tedious. But when Walker and Elliott get to Adams Island, they’ll see everyone again. The old males, albatrosses who have been there since Summer 1, will turn their beaks away wearily when Walker checks up on them, the message clear: You again? They’re probably 50 years old by now, those original birds. The new mums will look pretty relieved after they’ve managed to lay an egg, and Walker and Elliott will congratulate them effusively. The young birds, back on their first visit to the island, will mistake the humans for albatrosses and attempt to dance with them. (Walker and Elliott won’t admit to it, but they will dance right back.)

The fledging chicks will squeal-whistle happily in the weeks before they first take flight. There will just be fewer of all of them than before.

“You go back each year to see who’s alive and who’s dead,” says Elliott. “We just end up feeling like voyeurs of extinction.”

We are grateful to Live Ocean Foundation which contributed to the production of this feature.