Before a single kiwi was released into the hills above New Zealand’s capital, there were traps. Around 4,600 of them by the count reported a few years into the work, strung in a grid across 24,000 hectares of sheep paddocks, pine blocks, mountain bike trails and gorse on the western edge of Wellington, running roughly from the coast at Pariwhero Red Rocks north to Porirua. About 2,000 of those traps already existed, put out over years by community trapping groups who were there long before anyone mentioned birds. The Capital Kiwi Project added the other 2,600 or so from 2018 onwards, and then did the part nobody writes about, which is walking the lines and checking them, in winter, up farm tracks, in the wind, again and again, for four years, before there was any kiwi in the landscape to show for it.
That is the actual story here. The birds are the ending.
Paul Ward founded the project and has led it since. The idea he started with was close to absurd on paper. Kiwi recovery in New Zealand mostly happens in places built for it: fenced sanctuaries, offshore islands, forest blocks with no roads and no dogs and no people. Ward wanted to put them into a working landscape on the doorstep of a capital city, across land owned by farmers and councils and trusts and individuals, crossed by public tracks and used every weekend by cyclists and walkers and their pets. Nobody puts kiwi there. Kiwi had not lived in those hills for something in the order of 150 years, long enough that no specific last sighting seems to be recorded anywhere, and long enough that the usual phrasing is simply more than a century.
The reason it had to begin with traps rather than birds is a piece of arithmetic that governs everything about kiwi conservation. A kiwi chick hatches, is abandoned by its parents almost immediately, and has to feed itself and grow from a few hundred grams up to around a kilogram before it can survive a stoat. Below that weight it is prey. Above it, more or less, it is not. Stoats are small, fast, numerous and extremely good at finding a ground bird in a burrow. In unmanaged North Island forest, the proportion of chicks that reach stoat-proof weight is low enough that a population shrinks steadily even while adults are still calling at night. Release adult kiwi into a landscape full of stoats and you have not restored anything. You have delivered a food supply with a delay built in.
So the first thing Ward had to do was not get a permit for birds. It was to persuade a lot of people to let him put steel traps on their land and keep coming back to them. That is the part of the decade that reads as nothing and was in practice the whole project. Much of the 24,000 hectares is private, mostly farmland, some of it held by families who have farmed it for generations and have heard plenty of enthusiastic plans from town. Some is public reserve with its own rules. The project’s framing of its own structure is a tripod standing on relationships with iwi and Maori, with the communities in the area, and with the landowners, and Ward has been blunt that if you take any one of those three legs away the whole thing tips over. That is not a slogan when the trap grid physically has to cross hundreds of boundaries.
The trapping itself is unglamorous in a way that matters, because it never stops. A trap network is not an achievement, it is a standing cost. Stoats reinvade. They come in from the edges, from the pine, from the farmland beyond the grid. A line that is not checked is worse than no line, because it gives you a number you believe and a landscape that is quietly refilling. Peter Kirkman, the project’s head of operations, has had to run this as a logistics problem rather than a wildlife one: lines that get walked on a schedule, traps that get rebaited, data that gets entered, volunteers who keep turning up in bad weather for years on end without a kiwi to look at.
Only once that was running did the paperwork for birds become possible, and the paperwork is more specific than most people assume. Moving a protected native species in New Zealand requires an authority issued under the Wildlife Act 1953, applied for through a formal translocation proposal. The Department of Conservation’s own guidance for community groups sets out what that proposal has to contain: the conservation justification, an assessment of both the source site and the release site, disease management, the design of the transfer itself, a monitoring programme and a plan for managing the animals after release. Consultation with iwi at both the source and the release site is not optional, and the department sends the proposal out to affected iwi and other parties for comment as part of processing it. The process is budgeted at a minimum of 70 days and a maximum of 90. Approvals come with conditions attached, covering collection, transport, release, welfare, what must be monitored, how it must be reported, and what happens if a bird dies, including necropsy. A transfer report is normally due within two months of the transfer being completed. Where monitoring is ongoing, a monitoring report has to be produced every year.
That last detail is the one that gives the trapping its teeth. The permit does not just allow birds to be moved. It sets out what the project has to prove afterwards, in writing, annually, to the department. Capital Kiwi’s permit carried a target for the proportion of monitored chicks reaching stoat-proof weight, set at 30 percent. The underlying ecological threshold is lower still: roughly one chick in five surviving to that weight is the point at which a wild kiwi population can sustain itself rather than drain away. Before the first releases, the project’s own stated expectation was that it would beat one in five, based on what its monitoring of the trap network was telling it. Of the first twenty chicks fitted with transmitters, six survivors would have met the mark.
The first releases happened in late 2022, in the Makara hills west of the city. Sixty-three birds went out in that first phase, with a departmental permit in hand for close to two hundred more. The birds came from places that had spare capacity because they had already done this work: Sanctuary Mountain Maungatautari in the Waikato, Otorohanga Kiwi House, and the Taranaki Kohanga Kiwi at Rotokare. Each transfer was arranged as a gift from one iwi to another rather than as a shipment. The request is made formally, as a tono. The handover happens at a powhiri, a formal welcome, which is the point at which responsibility for the birds passes to the receiving community. Ten birds moved from Rotokare in the Taranaki region in 2024, gifted by Ngati Tupaia and received by Taranaki Whanui ki Te Upoko o te Ika at a ceremony at Te Papa in Wellington, with Rotokare’s own conservation manager Fiona Gordon and staff from the Taranaki Kiwi Trust seeing them off. Birds from Maungatautari came from Ngati KorokI Kahukura. Ngati Toa and Taranaki Whanui are named by the project as having been essential throughout.
Rawiri Walsh works on the project as its kaimanaaki kiwi, which in plain terms is the person responsible for the care and guardianship of the birds, and as its liaison with iwi. He is of Taranaki Whanui ki Te Upoko o Te Ika, Nga Rauru Kitahi, Ngati Raukawa ki te Tonga and Te Ati Haunui-a-Paparangi descent. The role is both practical and procedural: it is the person who handles the birds and the person who makes sure the tono and the powhiri and the relationships behind them are done properly rather than decoratively. It is worth being precise about what the iwi role actually was, because it is easy to flatten. Mana whenua were not a blessing added to a conservation plan. Consultation at source and release was a legal condition of the permit, the birds moved as gifts between specific iwi who had to agree to give and to receive them, and the transfer of responsibility at the powhiri is the mechanism by which the receiving community takes the birds on. At the Maungatautari end, Poto Davis, a kuia of Ngati Koroki Kahukura, was among those seeing birds leave. In Makara, Ted Smith, an elder of the local community, has put the obligation about as directly as it can be put: if a people named after an animal cannot look after that animal, that people has earned a less flattering name.
Then the numbers came in, and they were not what anyone had budgeted for. Against a permit target of 30 percent of monitored chicks reaching stoat-proof weight, and a private hope somewhere in the mid to high forties, the project recorded more than 90 percent. Over three years following the 2023 season, only two monitored chicks were lost, one to a predator and one to natural causes. Chicks were hitting the kilogram mark in around four months, where the historical expectation is six to eight. That difference is not a quirk. A chick that spends four months in the vulnerable window instead of seven spends roughly half as long being findable, and a landscape where the stoat density is low enough to let them grow that fast is a landscape where the trapping has worked.
By the time the project reached its permitted total it had moved around 250 birds. The final translocation was staged at Parliament, with a powhiri at the Beehive in late April of this year, which was a deliberate piece of theatre and a fair one: a bird that had been absent from the hills around the capital for a century and a half, handed over in the building where the country’s conservation law is made. Ward’s summary of that evening was that it was ten years of work arriving in a single moment.
The more interesting number is the one nobody collected. The estimated population in those hills is now above 300, which is more than were ever released. The difference, somewhere between 60 and 70 birds on a conservative estimate, is chicks that hatched in the hills, were never caught, never fitted with a transmitter, and are not in any dataset. They are known about the way wild animals are known about: from tracking tunnels, trail cameras, acoustic recorders run through pattern-matching software, DNA taken from trapped stoats, and residents ringing in to say something screamed outside the house at two in the morning.
In September the last transmitters came off. The monitored cohort stopped being monitored. Wellington’s kiwi became, in the project’s own phrase, properly free range.
Which is where Ward’s worry sits now, and it is not the worry he started with. Ten years ago the question hanging over the project was whether it was being too ambitious. The question he names now is complacency. It is an unusual thing for a conservation project to say out loud at the moment of its success, and it is accurate, because nothing about the underlying situation has changed. The 24,000 hectares are still full of stoats trying to come back in. The trap lines still have to be walked by people who are no longer walking them towards a milestone. A good season is a good season and nothing more; the chicks counted this year do not insure the chicks of the next one, and a population that is growing because the trapping is working will shrink again, quietly and on the same arithmetic, if the trapping stops working.
And there is a threat the traps were never going to solve. Stoats kill chicks. The thing that kills adult kiwi near a city is dogs. The project has stated it plainly: uncontrolled dogs are the single biggest danger to adult kiwi in any new release area, and even a small pet dog is capable of killing one. A kiwi has no keel, no flight muscle and no way of getting off the ground, and a dog does not have to be aggressive or hungry to kill one, only curious. Ferrets and other large mustelids are a danger to adults too. But the dog problem is the one that cannot be trapped away, because it is not a wildlife problem at all. It is a problem about several hundred thousand people, their weekends, their vehicles, their dogs, and whether they bother. Putting kiwi next to a city was always a bet that a city could be asked to change its habits, and that bet does not get settled in a decade.
So the ending of this is not the bird at Parliament. It is the 60 or 70 young kiwi in the hills behind Makara that nobody has counted, nobody has handled and nobody is tracking, growing up in a landscape that works for them only as long as several hundred volunteers keep checking several thousand traps that catch, most weeks, nothing at all.