Neil Silverwood

Birds of play

Packs of kea are reliable entertainers in places such as Arthur’s Pass or Glacier Country, and new research is showing that kea are smarter and have more complex communication than previously thought. But large flocks in tourism hotspots conceal the fact that kea numbers are dramatically falling across the Southern Alps. Why is this? How can we reverse it? And what do we still have to learn about them?

Written by       Photographed by Neil Silverwood

Corey Mosen and photographer Neil Silverwood camped at the head of Rainbow Creek in Nelson Lakes National Park in search of kea fledglings—but their first expedition together yielded only empty nests.

Catching kea requires two things: patience and a hot-pink jandal. You can substitute something else for the jandal, but the patience is compulsory.

First, you’ll set up camp on the tops, above the bushline, where tussock and wind make the air blurry. You’ll have to rise before dawn to start your watch, because kea are crepuscular—they go foraging at the times when day and night merge. As you wait, huddled up against the chill, you may see a glow of light on the next ridge over, as someone else gets ready to survey.

You’ll wait, and listen. You don’t want to attract birds from several ridges over, because you’re trying to figure out if there are any kea in this particular spot. If there’s a resident bird, or pair, you’ll hear them before you see them, a long screech: ‘Keeeee-ah’. They’re named after this call—it’s what researchers refer to as a contact call, when one kea meets another. In human terms, it’s hello.

When you hear one, or glimpse a dark-green shape gliding across the valley, you’ll need to get its attention. Feel free to make a lot of noise—in fact, the more noise you make, the more interested in you the kea will be. Something brightly coloured pretty much always works. That’s where the jandal comes in, although if your tent is red or orange, you might not even need it. Kea-catching expert Corey Mosen uses whatever he has to hand—a bit of tape, a hat, a glove—to lure them into the range of his net-gun.

Each bird that’s caught receives leg bands, and some of them get a name, too, so they can be spotted and identified later. A couple of adult females get tiny backpack transmitters, so Mosen can find their nests and return the following year to see if the kea parents succeed in raising any chicks.

This is how Mosen spends his summers—catching kea. It gives him an idea of how many there are in a particular area, and how many new kea are being added to the population.

But that’s all it’ll be, an idea. No one knows how many kea there are, because their habitat extends along the whole spine of the South Island, from Waitutu in the south to Kaiteriteri in the north. Most populations aren’t monitored, but according to those that are, the number of birds has crashed.

The density of kea in Nelson Lakes National Park dropped 80 per cent between 1998 and 2011, across just one generation of birds. Nationwide, kea numbers are estimated to have fallen between 50 and 80 per cent over the past 36 years—that’s three generations of birds. Kea were added to the Department of Conservation’s list of nationally threatened species in 2012, and in 2017, their ranking on the International Union for Conservation of Nature’s Red List was changed from ‘vulnerable’ to ‘endangered’.

Over the last spring and summer, Corey Mosen kept an eye on two kea nests in Nelson Lakes National Park. The female kea are fitted with radio transmitters in order to make it easier to track them. In the St Arnaud Range, overlooking Lake Rotoiti, he stops to listen for a signal.

There may be as many as 7000, or as few as 2000 kea. Either way, it isn’t very many birds. There are more kiwi than there are kea.

“I think people assume that there’s gazillions up there,” says Mosen.

“They’re kind of in your face, and people don’t associate that with being endangered. But when you’re studying and looking for them in the mountains, they’re not actually that abundant.”

We used to think there were too many kea. For a century, the birds had a bounty on their beaks, endorsed by the government and prompted by ‘kea strikes’, when flocks of the birds attacked high-country sheep. A kea beak was worth as much as £1, or $120 now. Records show that at least 150,000 kea were killed between 1860 and 1971, when the bounty was abolished. They became fully protected in 1986.

Kea are resourceful and resilient. They’ve adapted to living in a range of environments, many of them extreme: peaks, bluffs, spurs, ridges, permanent snowfields, alpine meadows, tussock lands, rainforest and beech forest. They survived a century of being a target. Why aren’t they flourishing now?

[Chapter Break]

The helicopter can’t settle its skids on the bed of the Beautiful River in Kahurangi National Park, west of Nelson. The pilot holds it in place as Mosen and I jump out. In a moment the roar of it has gone, and Mosen checks his GPS. The first nest we’ll check is upstream a little further.

There’s one thing that clearly indicates whether kea are thriving or struggling: their nests. Kea nest in the same spots year after year, underground, in crevasses between rocks, holes between tree roots. Mosen is keeping an eye on nine nests throughout the two national parks on his doorstep, visiting them several times to check on how they’re doing—whether there are eggs, whether those eggs hatch, whether the fledglings make it to adulthood.

Mosen, sandy-haired and cheerful, is the kind of person who can figure out a solution to a problem using whatever’s in his pocket at the time. He started volunteering for kea work in 2007, taking time off from his job as a builder in Whanganui to spend long days tracking birds in Aoraki/Mt Cook and Mt Aspiring National Parks. He stuck around doing kea work, first unpaid, then barely paid, for long enough that the Department of Conservation (DOC) and the Kea Conservation Trust (KCT) hired him. These days, he spends more time tracking, catching and monitoring kea in the wild than anyone else.

I quickly learn why it’s going to take us three days to visit four nests. We walk downriver—from the Beautiful to the Karamea, then the Leslie—and we stop whenever the GPS shows Mosen that a nest is near, and the only direction left to go is up.

Corey Mosen trained Ajax, a border-collie cross, to track kea scent, making it easier to find their nests, which are usually located in small, natural caves. When kea chicks begin to fledge, their feathers transform from fluffy grey to smooth green.

We climb loose, loamy slopes, soft to the step. The beech forest on the banks of the Leslie has been thoroughly grazed by deer, the understorey gone except for large bushes of horopito. Dracophyllum fronds crunch underfoot, and their trunks, bent like elbows, are useful handles for pulling myself up as we climb. And climb. And climb. Kea nests are never above the bushline, but they prefer the heights of the forest.

“People ask me what I do all day,” says Mosen, “and I tell them I walk up stairs over and over again.”

Mosen eventually stops at the foot of a tree whose roots are wrapped around several large rocks. It forms a kind of promontory in the forest, leaning out over the edge of the slope, beech leaves falling like confetti into the wide green void below us. The roots are covered in tufts of white down, and one rotting log has been shredded into fine dust.

This nest belongs to Marie, says Mosen, peering into a cavity between the rocks. She and her mate were caught during Mosen’s phase of naming kea after scientists, so he called them Marie and Pierre, after the Curies. But it turned out there was already a Pierre in the database, so the male was renamed Megatron.

Most of the 1600 birds in the kea database have been named by Mosen. Thinking of names is easier if you have a theme going, he says, like pharaohs, X-men, scientists, or outer space.

There’s the sound of muffled flapping, and then Mosen hands me a wriggling bird. As soon as I grab its legs it goes still, gazing up at me placidly from an eye ringed with yellow—the sign of a juvenile. Mosen measures the length of its beak and the size of its head.

“It’s a male,” he says.

I have the sense of something looking at me, but it takes me a while to notice the parrot silhouette on a branch, just a couple of metres away.

“That’s dad,” says Mosen.

Megatron doesn’t approach, but watches us from the shadows, shifting from one leg to the other, as Mosen weighs the fledgling and bands its legs—white Z on black.

The young kea blinks at me. He smells like pure warmth, like the breeze that means it’s finally summer. I place him back in the crevice between the rocks and he shuffles out of sight.

“What shall we name the juvenile?” asks Mosen, as we plunge down the slope to the Leslie.

“We should start a new theme,” I suggest. There’s a long walk ahead of us—plenty of time to think of a good one.

They receive a health check and their legs are banded so that their survival can be monitored.

A juvenile kea is a hopeful sign. In Kahurangi, only two per cent of nests added a new kea to the population between 2009 and 2014. Back then, there was little to knock back stoats or possums in the region, but after predator control began in 2014, nest success climbed to 50 per cent the following year.

Kea meet untimely ends in a number of ways: poisoning by chocolate, or lead, or 1080; setting off traps intended for predators; being struck by cars; getting into accidents with snow-groomers, power lines, rubbish bins; being attacked by mammals, especially while on their nests.

One of the biggest threats to kea is an all-too-familiar one: mammalian predators. The population crash is thought to be largely the work of stoats—nests have trail cameras installed within them, and the footage shows kea parents fighting off intruders. It also reveals which predators tend to win.

“When you put a camera on the kea, you see they’re getting harassed all the time—the level of interference and harassment and predation by possums, rats, stoats, cats, it’s quite scary,” says Josh Kemp, who manages kea for DOC.

“It’s amazing that any of the nests make it through, when you see what goes on. But if you remove that, they live for ages, and they’re really good breeders and parents. They don’t have any food issues in the wild—there’s plenty of natural food and they don’t have any genetic depression or inbreeding problems. We’re in a good position. If we can’t save them, we’re really dumb.”

Part of the problem, says Kemp, is that only about a third of kea habitat is predator-controlled.

“We need to be getting up over half if we want to keep the kea across the South Island, or we’re going to lose them in large chunks. Our aim is to retain the whole species range, but they’re really marginal in the Kaikōuras, and between Nelson Lakes and Lewis Pass. There are large tracts where you can go for multiple days without hearing a kea—in contrast to South Westland, where you’ll hear a kea within half an hour.”

[Chapter Break]

I wake up to the sound of kea shouting their names into the dawn. The sun has just lifted above the Matukituki Valley, turning the southern face of Mount Aspiring pink, and the slanted light is thick and syrupy. On the tops, above the valley, a gang of kea are fossicking in the tussock. Here, the Cascade Saddle ends in a sheer cliff, plunging 1500 metres to the river flats below.

I crouch and four of the kea surround me, sidling close enough for me to see the bright-yellow rings around their eyes. Kea roaming in packs are usually juveniles—they have a long adolescence before they find mates, and they spend it exploring the world and, sometimes, making trouble.

But these kea are more interested in each other than in me. One lets out a squeaky-toy cry and tries to kick another, and they tussle for a few moments, the tropical orange underneath their wings flashing in the light, as though they’re ablaze within. Then one bird leaps off the edge, turning a long loop above the valley, and the others watch it from the edge before they, too, dive off.

Amalia Bastos shows Neo an orange token. Next, she shows him a black one, seen below her left hand. When she conceals both tokens in her fists, Neo must indicate which one contains the black token. When he does, he’s rewarded with food. Kea are quick to learn, says Bastos: “It’s just a matter of figuring out whether they’re picking things up in a really dumb way of ‘All right, this gives me food’, or a more clever way of actually thinking about it.”
Just outside the township of Arthur’s Pass, Death’s Corner is one of the country’s most popular kea-spotting sites. Any kea heard or sighted can be reported at, a citizen-science project aimed at building the national picture of where kea are and how the population is getting on. If a bird has leg bands, it probably has a name, too.

Most animals play only when they’re young, but kea never stop. They even have a special call which announces to any other kea within earshot that it’s time to be silly. Ximena Nelson, a University of Canterbury biologist specialising in animal behaviour, has named it “the warble”, one of seven kea calls she has identified. When she plays it to a room full of people, they burst out laughing. When she plays it to kea, they start to muck around, chasing each other, throwing things, performing acrobatics, leaping on their backs so that other kea can jump on them. Even a kea on its own, no other birds in sight, will start playing at the sound of a warble.

“Kea are really unusual in terms of their play behaviour,” says Nelson. “Play is usually associated with young mammals. Human adults play, but we’re really weird—in all other mammals it’s associated with juvenile stages. The only other adults that reliably play are kea.”

Many animals have specific calls, but most of the ones described so far are alarm calls—chickens, for instance, have different calls for aerial predators or ground predators. There are calls for food and calls for mating. Other than kea, no animal has been shown to have a play call.

Kea certainly have more than seven calls, says Nelson, but it’s going to be tricky to figure out their meanings. Most calls are used in many different contexts, and they vary geographically, as though different kea populations have their own dialects.

The yellow colour that marks the eye, lower mandible and cere of a juvenile kea fades to black by the time the bird is three or four years old. “Juveniles form into little bunches and wander round wreaking havoc,” says Josh Kemp. “Once they mature and find a mate and pair up, they’re usually found in the same section of mountains pretty much every day.”

“I think there’s a lot more to it,” she says.  “I’ve heard calls in really isolated places that I’ve never heard before—that are totally different.”

It also means that kea have a certain level of intelligence, because associating specific actions with sounds requires a decent amount of cognitive power. On top of that, there are two other clues to kea intelligence, says Alex Taylor, who leads the University of Auckland’s Animal Minds lab—their sociability and their ingenuity.

“The key hypothesis for how humans got so smart is that humans have to have a really good understanding of their tools, but also have to work with each other. You get an elevated level of intelligence in a species that is both highly social and good at problem-solving. Kea might have something similar going on because they’re highly social and figure out how to get food from hard-to-reach places —though mainly with their beaks.”

[Chapter Break]

The kea sitting on my shoulder, Loki, pushes his beak against my face, sifts through my hair, then pinches the skin of my cheek, examining me one texture at a time. Then he pokes his beak through my jumper. In a moment, he’s made a hole in my sleeve, and he becomes completely absorbed in enlarging it.

There’s a thump on the back of my neck as another bird lands—it’s like being hit by an errant basketball—and I feel the cool touch of its beak on my ear. Parrots have a small patch of skin covered in tiny hairs at the top of their beaks—it’s called a cere—which they use to sense what’s around them. To find out what something is, they touch it with their beak, the way we might reach out a fingertip. After that, they grab it.

Loki is busy grabbing. With his long, curved beak, he’s manipulating the fabric of my jumper as though he’s trying to figure out what it is, how it works, what’s on the other side of it.

Amalia Bastos is laughing, and for the first time I notice the tiny holes covering the shoulders of her polo shirt. She’s been working with these kea for two years—there are 13 here at Willowbank Wildlife Reserve in Christchurch, most of them named after famous villains. She wants to figure out how they think.

Most testing of the birds so far, says Bastos, has been binary—can they do a particular activity or not? For her PhD at the Animal Minds lab, she’s studying how kea do those activities, what kind of cognitive processes they use to solve problems.

“The lab wants to know how much they actually understand of what they’re doing—or whether they’re just putting their beaks into something and seeing if it breaks,” she says.

Bastos shows me some of the basic training she’s been doing. When she conceals black and orange tokens in her hands, the birds remember which of her fists contains the black token, even if she crosses her wrists. The kea touch a yellow circle painted on a block of concrete. They flip over another block of concrete using a piece of wooden dowel as a lever. And they’ll open small wooden boxes by tugging on a string attached to the door. These are all preliminary skills, says Bastos, which she’ll use in experiments targeting the birds’ understanding of the laws of physics. Can they predict the trajectory of an object that they can’t see? If something is hidden, do they forget that it’s there? Do they recognise that objects fall in a straight line? Do they make logical inferences about the world, or do they simply learn by association?

I glance down to see a kea gnawing on the sole of my sneaker. It looks up at me with an expression of, “Well, what?” then swaggers off down the path. Its head feathers are puffed up, giving it a rakish, nonchalant quality.

“There’s something about the texture of shoes that’s really fabulous,” says Bastos. “They love shoes.”

Each kea has a particular area of the aviary where only they train—it helps prevent the others from barging in on experiments. While one kea is working, several others hang around, watching, and occasionally offering very loud commentary. When Bastos and I move to another area of the aviary, the birds follow us. As they walk, they sway from side to side. It’s the gait of drunks, or toddlers.

On the whole, kea have pretty good attention spans, provided they’re being entertained.

“If the task is really new, they will work on it for 30 minutes, no problem,” says Bastos. “If they’ve done it a hundred times before, you might get 15 trials out of them and then they’re gone. They’re like, ‘Nah, I’ve done this before’.

“I try to vary what I’m doing with them, that’s why I have so much stuff going on at the same time. You do need to keep up the motivation so they keep working with you.”


The birds recognise Bastos, too. Each year, she spends three months with kea before moving on to the other species she is studying for her PhD—dogs and New Caledonian crows. When she returns, the birds crowd around her and shout.

New Caledonian crows provide an interesting point of comparison. The only bird species shown to make their own tools, and the only non-human species to use hooks, they’re one of the most-studied birds in terms of intelligence. Kea can be trained, intensively, to use tools, but they don’t regularly use them in the wild.

“I mean, in a way, they don’t need to,” says Bastos. “You don’t evolve something you don’t need. If you have a beak like that, why would you need a tool? Because you can just rip into anything you want.”

Rather, one of the kea’s big advantages may be in collaboration. A study by Taylor’s former PhD student Megan Heaney found kea would wait up to 65 seconds for another kea to help them with a task; so far, the only other species shown to do this are elephants, great apes and dogs. (The dogs required a lot of training, and weren’t very good at waiting.)

New Caledonian crows don’t appear to have the same kind of social intelligence—they don’t show evidence of understanding collaboration, for instance—and they’re also more fearful, so it takes them a lot longer to become accustomed to something new.

Kea fearlessness, says Bastos, probably makes them seem smarter than they are.

“Most of their amazing skill comes from their social understanding, but also the fact they will just try anything—they’re so brave,” she says. “For a bird species, it’s pretty rare.”

Comparing intelligence between species is difficult, says Bastos, because so many factors play into it. But she and Taylor are optimistic about what’s still to be learned about kea.

“If I’m interacting with a New Caledonian crow versus a kea, it’s the kea that really wows me,” says Taylor. “If they get a taste for new things, then they just love new things way more than any other species. You walk in with a pen in your shirt; the New Caledonian crow is never going to fly over, land on your shoulder, and take the pen out.”

[Chapter Break]

Kea with a taste for new things also create problems. If kea learn that humans are a source of food, that’s when the real trouble begins.

When kea are being a pain, says Andrea Goodman from the KCT, it’s often because they’ve been fed by humans at some point. Most of the time it’s unintentional—crusts thrown out the bulldozer window, cat food on the porch, veggie scraps chucked out behind the hut. Once kea learn to associate food with an area, they’ll keep visiting, waiting for more.


A spate of kea killed by motorists along State Highway 73 near Arthur’s Pass led to the introduction of kea signs in 2017. Arthur’s Pass is located in kea habitat, and the local community, with help from the Kea Conservation Trust, is working to reduce the risks to kea in the area: removing lead flashing from buildings, making power cables chew-proof, and asking the public not to feed the birds.
‘Don’t feed the kea—it turns them into beggars’, can be read on signs at kea sites around the South Island. Feeding kea makes them more likely to hang around roads and areas of human habitation, which increases the risks to them from vehicles, from lead poisoning, and from human food: so far, two kea are known to have died from chocolate poisoning.

Goodman is a diplomat of sorts, tasked with resolving human-kea conflict around the top of the South Island. She helps people protect their shoes and solar panels, and finds ways for forestry companies to live alongside birds with a propensity for wreaking $20,000 of damage in a night.

It’s usually flocks of juvenile kea causing problems, she says. They move away from where they hatched, and they’re not old enough to pair up and mate, so they form groups and take off to see the world.

“You know how kids put everything in their mouth? It’s kind of the same. It’s exploring, and testing, and trying things out.”

Except the things that juveniles try out are electrical wiring, bicycle seats, spa-pool covers and gumboots.

Kea diets vary depending on where they live, but they’re polyphagous and frugivorous: in other words, they’ll eat anything, but they love fruit. They dine out on at least 200 native plant species—leaves, shoots, seeds, flowers, roots and bulbs—as well as invertebrates, other birds, and mammals. Their cere allows them to smell roots beneath the soil, so often they’ll dig for food.

“Their bill is a pick shape,” says Kemp, “which you use for mining in the ground, trying to find little nuggets of gold under the soil, which is what they’re doing—they’re digging little excavations in the soil, and hoping to find orchid roots and insect larvae.”

An unusually high proportion of New Zealand’s alpine plants are fleshy fruited, and recently it was shown that kea have an important role in their seed dispersal—about 12 per cent of alpine plants rely on kea to reach new places. This is because alpine ecosystems act like islands, with few species crossing between them. But kea fly long distances, swallow seeds whole, and have a digestive system slow enough to last a trip between alpine zones.

Biologist Laura Young learned that it generally takes 120 to 140 minutes for a seed to pass from the beak to the other end, a measurement she made during her PhD research on the role of kea in seed dispersal. (It involved feeding a lot of blueberries to the kea at Willowbank and timing how long it took for the birds to produce blue poo.)

Young’s research means that kea can add another unusual attribute to their collection:  they’re the only parrots in the world known to help disperse seeds.

[Chapter Break]

I’m turning kea names over in my mind as Mosen and I walk out down the Leslie River. We’ve just visited the last nest, scrambling up the side of a sodden bluff on a ladder of tree roots.

Of the four nests we visited, one female kea was incubating infertile eggs, and another nest cavity was vacant—no one was home when we called in. One nest contained the young male Mosen banded, and the last had three fledglings tucked deep within a rocky crevasse.

It’s the start of a good summer for kea. By the end of the season, chicks have fledged in six of the nine nests Mosen monitors in Kahurangi and Nelson Lakes national parks. In January, Mosen and a crew of KCT volunteers survey 26 sites in the Stuart Mountains in Fiordland, catching and banding a record 44 kea, and putting transmitters on two adult females to monitor them throughout the next breeding season.

Kea in the wild generally forage for plant matter, such as rata flowers, while kea near human habitation tend to forage for lead, which tastes sweet. Lead poisoning poses a serious threat to kea, damaging their immune system and cognitive function, and leading to organ failure. Birds living near huts or townships are frequently found to have high lead levels, prompting the KCT to create a network of vets and wildlife parks which treat and rehabilitate lead-poisoned kea.

Then, during the month-long Wapiti Bugle in autumn, hunters in the same area of Fiordland note down all kea they see and hear. It’s something they’ve been doing for the past three years, but in 2018, they include any band details they see, helping to build the picture of the newly surveyed population.

Back in Kahurangi, I’m trying to think of a name that encapsulates kea omnivorousness, the fact that they’re good at teaching each other about things—but especially about what food is, and how to find it. A name that commemorates a great New Zealander, too.

“Let’s call him Holst,” I say, finally. “After Alison.”

We might call ourselves after kiwi, but I think kea embody our most defining quality: keenness. I can’t help wondering whether our national willingness to wade right in, to figure things out on the spot, came from the glint in the eye of a bird—one which, upon seeing something new, walks right up to it and grabs hold.

More by

More by Neil Silverwood