It was the weekend, and Jason Howard was at the library in Durham, North Carolina, flipping through children’s books to find something to read to his six-year-old daughter. A cover with a close-up photograph of a strange bird made him pause. He pulled the book off the shelf and came face to face with the downy cheeks and chubby beak of a green owl. No—a parrot.
Howard’s job involves tracing the evolution of human speech, so to him, parrots are among the most interesting birds in the world. Few living species can learn to make new sounds, and parrots are one of only three groups of birds which can learn vocalisations.
Howard, a neuroscientist, was working on the B10K project, an international endeavour to sequence the genomes of all 10,000 bird species from around the world—but he’d never heard of this one.
Later, reading Kakapo Rescue to his daughter, he was astounded by just how weird this parrot was.
Kākāpō are the heaviest of parrots, the rarest of parrots, the only ones that can’t fly, and the only ones that select their mates through a showing-off contest called a ‘lek’, which takes place only in the summers when podocarp trees bear fruit, a fickle but predictable event that happens every two years if you’re lucky and every four years if you’re not.
Kākāpō are probably one of the longest-lived birds—no one has been counting long enough to tell—and they are also ridiculously, undeniably cute, their round faces large and innocent, fixed in a permanent expression of hopefulness, or perhaps bafflement. There are just over 150 left, roaming free on three islands off the coast of New Zealand.
It’s debatable whether Howard or his daughter was more transfixed. By the last page, he’d realised that he could help the bird, or at least record its strangeness for posterity, by fast-tracking the sequencing of its genome. “It was one of those things which just became obvious—I’ve got to do this,” he says.
Howard works in Erich Jarvis’s laboratory at Duke University, one of the institutions taking part in the B10K project. Its ultimate goal is to sequence the genomes of one representative from every bird species, and it was working its way down the biological tree: first, one bird from every order, then family, then genera. Genomes are useful for studying phylogeny and for identifying susceptibility to disease—and the more genomes you have to compare with each other, the better. (So far, the study has shown that the same genes responsible for birds singing also relate to human speech, among other discoveries.)
While the kākāpō was in line to be sequenced towards the end of the project as the only representative of its genera, Howard could see obvious benefits in allowing it to jump the queue. The bird was already in a precarious position, and obtaining its DNA would be time-consuming. Transferring blood samples of endangered species across international borders is an exercise in paperwork and patience, and Howard didn’t know how long it would take to obtain those samples from a rare parrot on the edge of the world. So he settled in to wait.
Whenua Hou is a fleck in the sea off the left shoulder of Stewart Island, as though a tiny piece of it detached one day and floated away. Its longest beach, Sealers Bay, unfurls along the northern coast, and its single hut lies just inland, squatting in the lee of the dunes beside a stream dyed red by leaf tannins.
Whenua Hou means ‘new land’, and it does feel like a piece of freshly minted New Zealand. An incessant symphony of songbirds plays throughout the daylight hours, and in the late afternoon, fingers of golden light reach down through the forest.
By night, the music changes: the crashing of mottled petrels and hoiho through the bush, the high-pitched white noise of squeaking bats, and on the rare summer, the booms of male kākāpō seeking a mate—a breathy, low clarinet note that pulses all night long.
The island is like a second chance: Aotearoa, before we got our hands on it.
So it’s closely guarded. Arrival and departure is by air, bound to low tide and the vagaries of Foveaux Strait’s weather. The beach doubles as the runway. There’s no entry except on kākāpō business.
Sealers Bay Hut sits at the top of a spiderweb of tracks that splay out across the island, past old nesting sites and the remains of bowls where males once boomed.
The hut is lavish by tramping standards—two bunkrooms, a shower, a walk-in pantry, a big scarred table where staff and volunteers eat together, and a little office that is the centre of radio communications for the island and the hub of an extensive kākāpō spying operation. During breeding seasons, there’s little that a kākāpō can do without someone in the hut knowing about it.
The kākāpō’s main obstacle on the road to survival is itself—a problem that’s replicated in each one of its cells. Kākāpō DNA isn’t different enough from bird to bird, because the population passed through a genetic bottleneck in the early 90s, dropping to an all-time low of about 50, and now, too many of them are too closely related. Inbreeding results in infertility and susceptibility to disease—around 40 per cent of the eggs kākāpō lay are infertile, for instance, while most birds have never produced offspring.
Almost all kākāpō are descended from a small group of ‘founders’ transplanted from Stewart Island in the 1970s and 80s. But three are the offspring of Richard Henry, one of a group of males caught in Fiordland in the 1970s, and the only one of them to breed—a godsend to the population’s genetic diversity. He fathered three chicks—Gulliver, Sinbad and Kuia—who are now at the top of the list of kākāpō that the rangers hope will reproduce.
Meanwhile, other kākāpō have succeeded wildly in passing on their genes. A lothario named Blades has inexplicable appeal to the females of Whenua Hou. This isn’t necessarily a bad thing, because kākāpō are polygamous—both sexes mate multiple times, with more than one partner—and the more partners a female kākāpō has, the more likely her eggs are to be fertile.
When I arrive at Sealers Bay Hut on a clear May morning, it’s the living-room freezer that catches my eye. It’s been turned into a whiteboard, with a list of all the females that have nested this season, and after their names, egg shapes for each one they’ve laid. A smiley face drawn inside an egg means it’s fertile, and a little beak and legs appended means it hatched. Chicks that died are crossed out with an X, but so far, the surviving ones far outnumber them.
There’s another list of chicks on a corkboard above one of the sofas, and this one is decorated with gold, silver and bronze stars. This past summer was the biggest kākāpō breeding season in the programme’s history, and rangers have to prioritise chicks. Without knowing who any of the chicks’ fathers are until paternity testing is completed later in the year—all kākāpō dads are deadbeats, and don’t participate in childrearing—it’s a little bit of a guessing game. “Quite a few of our chicks could possibly be offspring of Gulliver or Sinbad, but we don’t know yet, so we’re treating them as potentially important,” says the recovery programme’s lead scientist, Andrew Digby. “They might just be from Blades, who has mated with the most females.”
It doesn’t help that the kākāpō family tree is very foggy in places. University of Otago conservation geneticist Bruce Robertson can identify ‘relatedness’ from snippets of kākāpō DNA, but it’s impossible to tell if a pair are cousins, for instance, or parent and child, or siblings.
So when Robertson heard of Jason Howard’s plan to sequence the kākāpō genome, he was elated—this was the tool he’d been dreaming of for more than two decades.
It fell to him to choose a kākāpō that would represent the species, and after some deliberation, he selected Jane, who had been found on Stewart Island in 1989. “She’s one of the founding kākāpō, but she also has a limp, a gammy leg or something like that, so she probably won’t ever have a possibility of contributing to the gene pool,” he says. “So I thought this was one way that Jane could contribute.”
The short pieces of DNA that Robertson had been working with up until this point were the equivalent of a few pages torn out of a book. Acquiring Jane’s genome would be like having an entire library to browse. He’d be able to investigate issues such as susceptibility to disease; he’d be able to figure out how old the kākāpō were, or perhaps pinpoint the problems they were having with fertility.
But acquiring blood samples from Jane turned out to be the easy part for Howard—he was having trouble finding the right genome-sequencing sponsor. The Jarvis lab had just sequenced the budgie genome at a cost of US$100,000, and it didn’t have that kind of cash sitting around for another parrot genome that wasn’t top priority. “We had a couple of companies that said they would sequence it for us, but they were these newly developed companies, and I tried them out, and they didn’t really work that well,” he says. But Howard had noticed that sequencing technology was improving, and the price of it was falling, at a breakneck rate.
So he settled in, once again, to wait. He watched BBC documentaries on the kākāpō with his daughter, the funny parrot now a frequent topic of dinner-time conversation. “Most of my work is neurogenomics and it’s kind of hard for my kids to relate to it,” he says. “But we would talk about this kākāpō project—should we try to help save the species, how we could do that.”
The kākāpō provided a helpful window into evolution, he added, prompting all kinds of questions: Why can’t it fly? Why does it make nests in the ground instead of in trees? What does nocturnal mean? “It was something I could talk to my daughter about and she could understand.”
The sun set less than an hour ago, but it’s already completely dark in the forest on Whenua Hou. On the tops of the island, the squeaking of bats dies away and the wind whistles through the trees. Tumeke the kākāpō left her nest a few minutes ago to go foraging for fruit—setting off a doorbell ding-dong in Sealers Bay Hut—and I’m tagging along with recovery programme manager Deidre Vercoe to check on Tumeke’s chick, a two-month-old that hasn’t been visited in five days.
We’re passing through Hoki’s territory, and stop to check her feeding station, a trough of pellets suspended above a set of scales. The lid unlocks only at the frequency of the radio-transmitter backpack that Hoki wears; it’s important to maintain the females’ weight between 1.5 and 1.7 kilograms, otherwise they disproportionately produce male chicks.
“Oh, hello,” says Vercoe in surprise, and all of a sudden there Hoki is in the centre of the track, frozen in the narrow beam of light from our headlamps.
Vercoe fumbles for something in her pack, and Hoki takes one slow-motion step forward, placing one claw down, then the other, to accept an almond from Vercoe’s fingertips.
Hoki curls her claw around the nut, holds it to her beak to nibble as though it’s an ice-cream cone, and blinks slowly.
She’s close enough that I can see the fine white down of her face, the dark whiskers sprouting either side of her beak and nostrils, and the tiny green-rimmed feathers scalloping her head, like tropical-print chainmail. And I can smell her, a sweet fragrance like a hint of old perfume lingering on a collar or a cuff. To other kākāpō, with their super-powered olfactory glands, she must smell as bright as her feathers look to me. Vercoe told me earlier that booming males radiate a powerful sweetness, like lighthouses of scent in the dark.
Hoki takes another almond and scampers up the trunk of the mānuka tree next to me, using her beak and claws for purchase, until she’s at my eye level, moving her head side to side to peer at me from different angles, or looking back at me over her shoulder, like a model trying out different poses in front of a camera.
She peels off a strip of mānuka bark and chews it laconically, holding my gaze all the while. I have the feeling of being very keenly assessed.
All of a sudden, a mottled petrel bellyflops through the bush at our feet, swimming with its wings through the underbrush, and Hoki swiftly scales the mānuka until she’s over our heads, her white face gazing down owlishly. Then she climbs above the canopy and is lost to the sky.
We’ve encountered Hoki on a good day, says Vercoe: “You never know what you’re going to get with her.”
Hoki is the original celebrity kākāpō—in 1992, she was the first chick successfully raised by humans, when the rimu crop failed on Whenua Hou and chicks began to starve. There’s a picture book about her, too, but unlike her brother Sirocco, she sometimes disdains rather than embraces human company.
Last year, everything fell into place for Jason Howard and the kākāpō.
In February, a ranger climbed 10 of Whenua Hou’s rimu to count the ratio of fruit buds to leaf tips. More than 10 per cent fruit, and kākāpō will breed the following summer. It was 24 per cent, which forecast an unprecedented breeding season, and it meant action stations for the Kākāpō Recovery Programme. By October, Sealers Bay Hut was home to food delivery volunteers, nest babysitters, and the Sperm Team, a bunch of would-be kākāpō matchmakers.
A hemisphere away, Howard discovered a new genome-sequencing company. Pacific Biosciences not only had a reputable new sequencing technology, but an executive, Jonas Korlach, who once lived in New Zealand. Korlach knew what a kākāpō was and why it needed help, and agreed to collaborate on the project, helping to reduce the cost.
“It turned out to be good that we waited,” says Howard, “because where the technology is right now, we would have to do it all over again if we’d done it two or three years ago. We hit the sweet spot; the time was just right.”
Meanwhile, in California, a software engineer named David Iorns had just discovered conservation genetics. A New Zealander, he’d relocated to the US to work for Science Exchange, a company that matches up researchers with labs to carry out the tests they need.
Iorns had never paid much attention to science, but he quickly got hooked—especially on the idea of genome sequencing, which he saw could prove useful for many of New Zealand’s bottlenecked species.
So he promptly set up a non-profit, The Genetic Rescue Foundation, and launched a website. “Within three days, I was on the phone with Andrew Digby,” he says, “and I thought, ‘Wow, this is going much better than expected’.”
Digby wanted to sequence the genome of every living kākāpō. It had never been done before, for any species in the world, and for kākāpō the impact could be significant.
It could reveal whether or not there was a genetic basis for two of the kākāpō’s biggest problems, infertility and disease. It could show if the Fiordland and Stewart Island kākāpō had local adaptations, and how much of their genetic diversity had been lost. It could fill in the blanks on the family tree, and could allow the ages of older kākāpō to be calculated. And those are only the things that Digby can think of. “It’ll give us answers to questions we haven’t even asked yet,” he says.
Iorns was thrilled. He set about fundraising, and created a crowdfunding page on experiment.com. Donations poured in from around the world—expat New Zealanders, parrot-lovers who were keen to see the world’s rarest parrot survive, people who’d been captivated by Sirocco, the kākāpō whose antics made the species famous in Stephen Fry’s documentary Last Chance to See.
“I managed to raise on behalf of the foundation US$25,000 in private donations—that funded the first 40,” says Iorns. “We also did a crowdfunding project and we raised another US$20,000, which gets us pretty close to doing 80.”
The project was named Kākāpō 125, after the number of kākāpō at the time—the fact that two of them have since died is a poignant illustration of the bird’s vulnerability—and Iorns is still fundraising to sequence the remaining kākāpō, as well as a collection of museum specimens.
In December, the first kākāpō were booming on Whenua Hou, while partway round the world, Jane’s DNA was finally undergoing sequencing.
Her genome would act as a reference, meaning that sequencing the remaining kākāpō genomes would be much cheaper and faster. The difficult part of genome sequencing is in assembling it in the right order (see sidebar on page 53), and Jane’s genome would act as a scaffold, or a map, against which other kākāpō genomes could be pieced together. “Basically, what it cost him [Jason Howard] to produce the reference genome will probably cost the same for us to do 100-plus additional genome sequences,” says Iorns.
Near the peak of Whenua Hou, the forest gives way to low scrub, tussock and the bare contorted branches of mānuka. It’s high enough to watch the rain clouds sweeping over the strait to Sealers Bay, but the Southland coast is too far away to see. This is Sinbad’s territory—he’s named after a Fiordland gully where many kākāpō were found—and in breeding season, he fashions his bowl up here, along with a couple of other males.
I glance back down at Sinbad, but he seems to have dissolved into the air. There was a kākāpō here a moment ago, chewing. Then he lifts a claw to take a step forward and the shape of his body coalesces right in front of me.
The colours of his back feathers are perfectly calibrated to the carpet of mosses, lichens and dead twigs: a mix of every shade of green, with mottled black striations passing through them.
Sinbad has different features from the average kākāpō—his eyes are rounder, beadier, making him look particularly inquisitive.
He has spent several summers booming in his bowl—a gentle, distant drumbeat—but he hasn’t been popular with the ladies, so the Kākāpō Recovery Programme has been trying to help him out behind the scenes.
Each kākāpō’s movements are tracked via a backpack transmitter, which contains a small drop of mercury. This indicates what kind of movement the kākāpō is engaged in—from high-energy mating to low-energy nesting. If a male is recorded as mating, a radio receiver within his backpack switches on, detecting the female’s unique frequency, and checking the female’s activity log to ensure she was the partner. When a kākāpō next wanders past a hilltop proximity sensor called an Errol, its activity is ‘read’ and the information sent via satellite to an email address. Every morning, Kākāpō Recovery Progamme staff awake to a delivery of gossip: Who has mated the previous night, and with whom.
If a female is deemed to have chosen an unsuitable partner—too closely related, too young, already overrepresented in the gene pool—the Sperm Team track down a better mate for her. Speed is of the essence. The team has up to five days to artificially inseminate a female after she has mated, but kākāpō have a habit of roosting at the tops of trees, which poses a challenge to sperm collection.
Artificial insemination is also used to improve fertility; if a female has mated only once, she may be artificially inseminated in an attempt to increase the likelihood of her eggs being fertile.
When Jane’s genome landed in Bruce Robertson’s inbox on January 22, 2016, it was as though all his Christmases had come at once. The genome looked like a series of peaks on a rainbow-coloured line graph, each spike colour-coded to one of the bases—blue for C, black for G, green for A, orange for T. And it went on for miles: There were 1668 DNA pieces, or contigs, each up to 29 million bases long.
Robertson’s task now was to figure out what it all means.
For starters, he looked at previously completed bird genomes such as the chicken and zebra finch, isolated genes known to control certain attributes, and looked for the same sequence within the kākāpō genome. “We’ve been looking at genes from different species that are associated with spermatogenesis, and you can just ‘Google’ genes—blast-search them against the kākāpō genome. We can look at whether kākāpō have that gene, whether it’s functional, what does it look like, and what sort of genetic diversity is present.”
As this issue of New Zealand Geographic went to press, Jason Howard was completing a final round of sequencing on Jane’s genome, while Robertson was gearing up to start work on comparing the first 40 kākāpō genomes upon their completion in September. This initial batch includes most of the founders, as well as Sinbad, Gulliver, Kuia and their dead father, Richard Henry.
Importantly, the genome data will also be publicly available online to non-profit researchers. “There are many potential end-users of the genome, and we’ve had quite a lot of interest already,” says Andrew Digby.
Jason Howard is particularly interested in where kākāpō fit into the evolutionary tree of life. “There has not yet been a conclusive study of kākāpō being able to imitate human speech,” he says. “We think that the kākāpō, as far as parrots are concerned, is potentially the basal, the bottom level, of parrots.”
Unlike Jane’s DNA, genetic material from the remaining kākāpō won’t be crossing the Pacific. At Digby’s insistence, these will be sequenced at home, by a new lab in Dunedin, New Zealand Genomics, a collaboration between Otago, Massey and Auckland universities.
There, a computer whirs through an algorithm, comparing thousands of base pairs with Jane’s, sorting pieces of an ancient puzzle. What they contain will tell us a lot about the present circumstances of the world’s strangest parrot, and may force us to reconsider its past, too. These sequences will also be our best hope of a future where the number of nesting females no longer fits on the freezer door of Whenua Hou.