Trap tensions expose haphazard kea oversight
An internal battle at the Department of Conservation over Goodnature traps shows a lack of coordination and adhoc management over an endangered bird. David Williams reports.
The Department of Conservation is being accused of not having a clear plan to save an endangered alpine parrot – a claim supported by internal emails released under the Official Information Act.
Wanaka’s Paul van Klink – a DOC contractor who has monitored kea for the department after aerial 1080 drops – ruffled feathers last year by speaking out about the threat to kea posed by self-resetting traps made by conservation technology company Goodnature.
The fallout from Van Klink’s comments, first published in the Otago Daily Times in May last year, are captured in emails released to Newsroom. They paint a picture of a department that is unsure of the risks, or efficacy, of the traps it’s using, managers that make decisions while seemingly ignorant of trap trial results, and decisions being made about positioning traps seemingly with little scientific basis. The internal discussions also reveal tensions between the department and Goodnature – which put out a joint press release last week about a new parrot excluder that will be fitted to the A24 traps.
DOC’s Wellington-based director of threats, Amber Bill, says kea are managed on a site-by-site basis, often through its over-arching predator control programme Battle for our Birds, with a focus on key threats, like predators.
Van Klink, a former DOC staffer, says he’s happy that public pressure has resulted in the parrot excluder being developed. But he tells Newsroom he’s prepared to stop doing kea work, something he’s been involved with for years, because of DOC’s lack of leadership and in the absence of a clear plan. “I’d rather be working on projects that I really see are making a difference.”
DOC needs to make the parrots a priority, he says, and form a kea recovery group. Kea are going to become more and more at risk, Van Klink says, adding: “You’re not going to save them by just throwing 1080 out there.”
“There’s no management plan for [kea], there’s no one calling the shots, there’s different spokespeople, there’s no collaboration exactly on what we’re doing, what the priorities are, it’s very ad-hoc.”
The clever, ground-dwelling mountain parrots, peppered throughout the Southern Alps, used to be the enemy. The Government had a bounty on them for more than a century to protect sheep. But attitudes changed and the sociable and destructive birds have been fully protected since 1986. Despite that, they’re listed as nationally threatened, with only about 3000 to 5000 animals left.
Van Klink first aired his concerns at the Kea Conservation Trust’s so-called “Kea Konvention”, held at Arthur’s Pass, a known kea hotspot, in April last year. That morning, DOC director-general Lou Sanson thanked the trust for its work, adding: “You virtually know more than we do.”
The trust, a volunteer organisation, published a draft strategic plan for kea conservation, funded by DOC, in 2013. Two years later, DOC told Stuff it was working on a national plan to save kea, but nothing has emerged. In that story, Green MP Eugenie Sage said the lack of a recovery plan for kea was evidence “we're not taking protection seriously”. Sage is now Conservation Minister.
Claims of mismanagement of an endangered species adds pressure to a department already under scrutiny. There are accusations from within its ranks of a toxic culture and criticism that it has lost its way by embracing corporate management methods.
There are also questions about its management on issues like kauri dieback and the Mackenzie Basin. Newsroom revealed in August that advice to Conservation Minister Sage about the Mackenzie was doctored by a DOC manager, allegedly to cover up the department’s failures.
Set and forget
In the imperfect fight against bird predators, Goodnature’s A24 traps – bought in bulk for about $125 each – seem the perfect solution. Given a growing anti-1080 sentiment, it’s an advantage they’re not toxic. They fire multiple times, thanks to a gas cannister, and have a pump that automatically refreshes the lure. They’re also light, making them easy to deploy, and can be set and left for six months, leaving dead animals to scavengers.
For a budget-conscious organisation like DOC, not having to send qualified staff to check and clear traps every month is a god-send. Between 2007 and 2010, the department contributed $770,450 to Goodnature for the development of its self-resetting traps and accessories.
(Goodnature co-founder Robbie van Dam says the ambition for its traps isn’t necessarily to save money and time, but to broaden the area of pest control. For example, instead of checking traps set over 250,000ha every month, it might be possible to make annual checks over three million hectares.)
But that strength – to basically set and forget – can also be a weakness. DOC can’t know for sure, without physical checks or video monitoring, if kea are being maimed or killed. Concerns about traps killing “non-target species” have already led to the development of a weka excluder.
Kea are not to be underestimated. A DOC kea expert says: “They’re smart, strong, flexible, curious, and bloody persistent. I’d only go with something that’s bomb-proof.” That makes Goodnature’s job harder, obviously, to make a “bomb-proof” excluder that will still kill rats and stoats. (Rat clogging in A24 weka excluders is mentioned as a problem in several DOC emails. Van Dam says clogging is “theoretical” – “I’ve not seen that.”)
On the frontline
Fiordland has been the A24 frontline for DOC. The traps have been used there for nine years, with significant numbers deployed since 2014. In July last year, there were 2300 traps used across 7000ha, with no observations of a single kea fatality, injury or trap damage. Compare that to the current and proposed sites for 1080 drops – which, research shows, helps protect kea overall but also kill kea – which cover 320,000ha.
In Fiordland, there are at least seven confirmed kills in DOC traps, which have since been modified and strengthened, and one kea death in a Sentinel possum trap. (Van Klink wonders why the improved DOC200s from Fiordland aren’t being used in kea habitat across the country.)
Bill, the DOC threats boss, says there are now approximately 9500 A24s spread across public conservation land. In December 2016, the department decides to use A24s for predator control in the Haast Kiwi Sanctuary, to protect tokoeka, and several Canterbury high country valleys in the territory of orange-fronted parakeet, or kākāriki. Bill: “This decision was informed by a range of advice from scientists, technical specialists and field staff, including the department’s experience in Fiordland.”
When he spoke at last year’s Kea Konvention, Van Klink was worried about inadequate field testing of A24s and DOC’s lax monitoring. “I didn’t say A24s are no good,” Van Klink tells Newsroom. “I called DOC out for making decisions that didn’t do kea much good.”
He wasn’t alone. Forest & Bird chief executive Kevin Hague confirms his group repeatedly raised with DOC the threat to kea from A24s, calling for research to be done to reassure them the gains would outweigh the risks. Another source says some DOC staff expressed serious misgivings about A24s being used in kea country, but were ignored. Van Klink says he was thanked by some DOC staff for speaking out.
DOC’s internal emails show the subsequent ODT story prompts high-level debate by the likes of DOC’s principal scientist for threats Craig Gillies, Eastern South Island operations director Andy Roberts and principal science adviser Graeme Elliott. (DOC has redacted all names of senders and recipients, but Gillies and others can be identified by their job titles.)
It emerges that captive trials, conducted five months before the traps were rolled out in Haast and Canterbury, showed kea putting their heads in an A24 trap, fitted with a weka excluder, all the way to the trigger. The expert advice, in last year’s emails, is that the risk to kea outweighs the benefits of using A24s.
The tone by senior managers takes on greater urgency in July – two months after the Van Klink article appears – when photos and video footage emerges of a kea poking its head into an unset trap on the West Coast.
Roberts, the eastern South Island boss, is also the lead DOC director on kea matters. He emails several people, including deputy director-general of biodiversity Martin Kessick, on July 11 last year. He talks of the “emerging risk of kea non-target impacts”. Monitoring of kea by DOC is “minimal”, he writes – contradicting what he told the ODT in May. Threshold figures – the point at which to remove the traps – “have not been set”, Roberts says, and support from Goodnature is “patchy”. He adds: “But so has our involvement with them”.
(Van Dam says his company is happy to be examined and held to account about the effectiveness of its technology. “We are really proud of our technology and track record.”)
By the end of that week, Roberts recommends A24s at South Westland and Arthur’s Pass be disarmed and replaced with DOC traps. Smaller trials using video surveilliance should be started, he suggests, and Goodnature be contacted “urgently to solve kea interaction” risks.
The problem for DOC, is that, as another email notes, orange-fronted parakeet in certain Canterbury valleys “are going extinct”. “Effective stoat control at Arthur’s Pass matters now.”
Perhaps that prompts the change. On July 24, Roberts, circulates “interim advice”. No traps are to be set above 900m altitude in kea habitat. (A scientist later describes this as “silly in the extreme” – “the alpine parrot thing is worthy only of the Woman’s Weekly”.) A small sample of traps will be video monitored. Some traps will be deployed with weka excluders, others with “experimental kea excluders”. (One email says an additional clip is needed to secure a spring on the weka excluder that kea might be able to pull.)
Should kea impacts be detected, Roberts says, the use of A24s will be reviewed.
“Keas are very rare and killing them as a way of finding out whether the traps are a risk is not a good idea for kea.” – unnamed DOC senior scientist
Reports of problems pile up. The initial emails, in May, mention a kaka killed by an A24 at Boundary Stream, in the Hawke’s Bay. In September, a weka is killed on Kawau Island, in the Hauraki Gulf, a day after an A24, fitted with a weka excluder, is set. The email says: “The obvious conclusion is that weka can access A24 traps even when the weka excluder is attached.”
Soon after, a senior scientist blasts his Wellington boss, saying he’s convinced DOC is killing kea. It’s “foolhardy”, he writes, for DOC to keep using the traps and measuring the rate of kea deaths to subsequently decide whether they’re safe. “Keas are very rare and killing them as a way of finding out whether the traps are a risk is not a good idea for kea.”
Another scientist, talking about robins in orange-fronted parakeet country, rages: “If you followed the advice we’ve given, you wouldn’t be using A24s without weka excluders, and until the traps with weka excluders had been shown to be effective, you wouldn’t be using them at all.”
In November, footage emerges of kea interacting with A24s on Resolution Island in Fiordland. There’s “considerable investigation” but no confirmed fatalities. A decomposed parrot carcass, of unknown species, is found about two metres from an A24 trap. The cause of death is “speculative”.
There’s also evidence of kea interaction with A24s in the southern branch of the Hurunui – Canterbury’s orange-fronted parakeet territory.
The risks fray the nerves of DOC specialists tasked with protecting kākāpō, takahē and short-tailed bats. At one point Roberts exclaims: “I’m up to my neck in A24 issues!”
However, the threat would have been clearer, to kea at least, if DOC bosses had taken notice of early A24 trials.
In November 2016, a month before it’s decided A24s can go into Haast and Canterbury valleys, a Rangiora-based ranger writes about a camera trial with captive kea. (The email is forwarded to another DOC staffer on December 6 last year.)
In the trial, at Christchurch’s Willowbank Wildlife Park, white tissue is inserted in the bait compartment. To remove the tissue, kea need to push past the trigger mechanism. The email says: “Kea accomplished this within 13 mins of the A24 set up. Note this is with kea who have been exposed to unbaited A24 in the preceding trial.”
Another email, written by a science adviser in the threats unit, emerges from July 2016. It’s headed: “Is there an urgent need for adding A24s to the Waimakariri predator trapping network?” The adviser talks of unquantified risks to kea and weka, and notes that most traps in Canterbury beech forest valleys don’t catch mustelids (ferrets, stoats and weasels) and there’s low incidence of multiple captures. “Therefore, based on this information, there does not appear to be an urgent need for adding A24s to the Waimakariri predator trapping network.” However, the adviser says A24s might be useful “as a back-up” for controlling ship rats.
Fast-forward 13 months, to December last year, and DOC managers still have unanswered questions. Are kea getting killed by A24s? Are the weka excluders working for kea? Footage of kea interaction with the traps is put together for analysis.
DOC staff visit Goodnature’s headquarters last December. An email sent afterwards confirms a parrot excluder is being developed. (Van Dam tells Newsroom the company had been working on a kākāpō excluder long before this issue was publicised and this prototype was adapted for kea.) The excluder is meant to be available for testing in May of this year – but is actually completed in August.
In late 2017, field staff at Haast and Waimakariri are given two options with the already-deployed A24s – fit weka excluders to protect kea or disarm the traps.
Anti-Goodnature or pro-kea?
Arguments within the department reached boiling point with talk of an “anti-Goodnature trap campaign”.
Te Anau’s principal biodiversity ranger Lindsay Wilson had told his bosses the level of concern about kea deaths “may be disproportionate to the risk”. On Saturday, August 19 last year, Wilson writes again, saying A24s seem to pose a lower risk than that by other possum, stoat and rat control tools. “I am baffled by the anti-Goodnature trap campaign that seems to be raging with a wealth of anecdote, misinformation, perception and innuendo and little fact.” He adds: “I am optimistic these traps will be modified to all but eliminate the risk to weka and kea.”
Three days later, a Nelson-based principal advisory scientist hits back, saying he’s not part of an anti-Goodnature trap campaign, but part of a pro-kea campaign. He’d like to see A24s working safely and well. “I, too, am sick of anecdote, misinformation, political pressure and little fact.” He adds: “I think it is pretty disgusting that we have run unsafe traps in the past and caught a lot of kea, but that is not an excuse for continuing to do that with A24 traps.”
The tensions aren’t just internal. On November 1, a Goodnature director responds to a DOC staffer’s query about work on its kea excluder, involving DOC scientists. The terse response says: “Your involvement introduces unwanted complexity and compromises this work and its people”.
(A DOC email from January this year says the relationship between DOC and Goodnature was “quite broken”.)
The public face is of two organisations working happily together. But in truth the Wellington company, founded in 2005, has gradually moved away from the department. When Goodnature was just three blokes in a shed, 90 percent of what it produced went to the conservation estate. These days it make 2000 units a week, and less than 10 percent of its business is required by DOC. The company’s products – all made in New Zealand – are distributed in 19 countries.
And it’s growing. Right now, Van Dam says it has about 35 staff. “We’ll probably be about 50 about this time next year.”
In February of this year, the video assessment for DOC into kea interaction with A24s is concluded. “The kea risk from operational deployment is higher than previously thought,” a February 19 email says. As an interim measure, A24s in kea habitat are to be disarmed or have weka excluders fitted, the email says.
(DOC confirms A24s above 900m at Haast were disarmed in November last year, with the remainder disarmed in March. The timing in Canterbury – where only 400 of the scheduled 750 were deployed – is unclear right now, “due to unavailability of staff”.)
While one part of DOC is sounding the alarm, a few months later results of a field study land that are glowing of A24s.
A 200ha project at Harts Hill, just off the Kepler Track, near Te Anau, tested the mechanical reliability of 467 A24 traps. They’re found to be very reliable. What’s more interesting, perhaps, are the trapping numbers.
An April 16 email says: “What [redacted] has found to date is the CO2 available in the A24 traps at this trap layout density was enough to reduce a beech mast/plague event population of rats from 68% pre-treatment to 0%.” Control was sustained through the next 18 months of cannister replacements. The emailer adds: “As well as reducing the rat population within the project area, other pests including stoats and mice were observed killed by the A24s without exhausting the available CO2.”
DOC threat boss Bill tells Newsroom that A24s have been shown to be an effective tool for rat and stoat control, with the potential to reduce the long-term costs of the department’s trap networks in difficult terrain. But in July, she raises concerns with Goodnature about “the lack of investment into research that proves its ability to perform” when fitted with weka excluders. (Van Dam says the excluder, developed in 2012, was tested across five different sites. “The testing demonstrated similar kill rates and we were satisfied with the efficacy of our excluder on traps.”)
Bill’s concerns originated from an email questioning the efficacy of A24s being used along the Heaphy Track, in the upper South Island. The emailer accused Goodnature of shying away from investment and using “DOC’s reputation and name to promote their product”.
(Van Dam says Goodnature invests a “huge amount” in R&D. He describes DOC as a taskmaster, challenging Goodnature to constantly improve its traps. “It has to be robust enough to last in the worst places of New Zealand, it has to be as cheap as chips and it has to be absolutely infallible all the time.” Clearly there’s a lift in reputation in being associated with DOC. Van Dam: “When we turn up anywhere else on the planet they just look at what we’ve done and are absolutely gobsmacked.”)
Goodnature confirms in August that its new excluder prototype – DOC staff call it a “shroud” – is ready for testing. An efficacy trial is discussed by DOC staff. The biggest question, according to one emailer on August 29, is whether fitting the new excluder affects the catch-rate of rats and stoats.
Van Dam says its new parrot excluder has been on Codfish Island and tested in takahē pens and in captive facilities for kea around New Zealand. There has also been target species testing, which will be strengthened by field tests involving a large-scale deployment of traps – some with excluders, some without. “We’ll set up, say, a network of 300 or 400 traps and we will compare the compare the capture efficacy and continue to understand whether the capture effectiveness is modified by the inclusion of an excluder.”
“They’re a very, very important tool in the arsenal and I want to see them continue to be further developed and made even better.” – Kevin Hague
Kea Conservation Trust co-founder Tamsin Orr-Walker says her main concern with A24s is the speed with which the traps have been rolled out – and questions about the robustness of the testing. “I’m not saying they’re not good, because obviously having a trap that kills multiple times is fantastic in terms of targeting predators. But kea are a particularly difficult case and more research, I think, needs to go into these. Trials for parrot excluders have apparently been going on for a while; we’ll be interested in seeing the research behind that as well.”
Zero Invasive Predators, or ZIP, is a research company established in 2015 to test pest control techniques in the hope of keeping large mainland areas predator-free. ZIP’s innovation director Phil Bell says it has done a small amount of trapping with A24s, fitted with weka excluders, at its site on the West Coast, in the Perth Valley. “We’ve experienced no concerns at this point with them.”
The traps are catching pests, he says. “We’ve got no evidence of kea interacting with the traps let alone the excluder. In that regard, I guess you could argue it’s working as well.”
As a Green MP, Kevin Hague announced a $4 million pest control package in 2010 to trial Goodnature’s first trap, named The Henry – after Richard Henry, who’s known as the founding father of New Zealand conservation. Now Forest & Bird’s boss, he says his organisation supports the development of new trap technology. He uses A24s at his property, south-east of Greymouth. “They’re a very, very important tool in the arsenal and I want to see them continue to be further developed and made even better.”
DOC’s threats boss Bill says the new parrot excluders will be fitted to A24s, including those used by DOC-funded community groups, next month. “DOC is committed to reducing the risk posed by predator control methods, including traps, to threatened species as kea.”
Some might argue that DOC has been thrown into a panic over a perceived threat to kea, without evidence of an A24 kea kill. But Goodnature can’t have it both ways. If there’s an argument that a lack of rat and stoat carcasses isn’t evidence its traps aren’t working, then it can’t say a lack of recorded kea deaths means its traps are safe.
As Wanaka’s Van Klink puts it, DOC seems to be pursuing “pest control at all costs”.
After spending more than a year worrying about the threat to kea, Goodnature has found itself plunged into an A24 crisis over another legally protected bird.
A Takaka-based senior ranger wrote to DOC management last month to say four weka had been killed by A24s fitted with weka excluders. The bait in the traps – set in Gouland Downs, part of the upper South Island’s Kahurangi National Park – had been changed to a prototype stoat lure. “The traps are obviously not performing that well, both in their efficacy and also for excluding weka as non-targets.”
(Despite this, Bill tells Newsroom on October 10 that weka excluders were being used at Gouland Downs “to prevent impacts on weka”. She adds: “These excluders are also considered to minimise risks to kea.”)
Video footage also emerges from Totaranui, in the Abel Tasman National Park, “clearly shows a weka going all the way through an excluder and eating some bait”. The emailer writes: “I think this is a national issue as DOC and some community groups are already using these traps and excluders to trap stoats and to protect weka.”
On September 28, a senior DOC ranger writes to Goodnature. “I am interested to see the testing and evidence that Goodnature would have collected when the ‘weka excluder’ was developed,” they write. “We purchased these excluders in good faith that they would exclude weka.” That ranger’s team drops all other work so the 688 traps at Gouland Downs can be fitted with a makeshift extension (using pipe and cable ties) to the excluder.
Days later, two Goodnature staff are sent to Gouland Downs to help attach the makeshift extensions.
“They seem to put more pressure when we use this new bait.” – Robbie Van Dam
Goodnature director Van Dam tells Newsroom the new bait elicited a more significant interaction from weka. “Now we’ve been provided with evidence, we’ll communicate with people that they should consider the type of bait that they will apply.”
But isn’t there a fundamental problem here? The weka excluder didn’t exclude weka. Doesn’t that mean it doesn’t work?
Van Dam argues the problem is the bait. Previously the excluder was excluding weka consistently, he says. “They seem to put more pressure when we use this new bait.”
In his combative style, Van Dam points out that DOC200 traps have killed weka in the Abel Tasman. Goodnature’s being held to a higher standard, he appears to be saying.
Once the cause of the Gouland Downs incident is established the company will act accordingly, Van Dam says – though he doubts a product recall will be necessary.
(Senior biodiversity ranger Hans Stoffregen, of DOC’s Golden Bay office, says as far as he knows no further weka have been killed. The next scheduled check of Gouland Downs is in two months. “We are reasonably confident that the extension will prevent weka from getting into the traps and being harmed or killed,” Stoffregen says.)
“Killing species that we’re committed to protecting is the last thing that Goodnature wants,” Van Dam says. “We feel this deeply, just exactly the same as everyone else. We’re just trying to do our best.”
In conservation, it seems, nature demands that our best continually gets better.
At the Kea Konvention, where Van Klink first raised the issue of A24s, DOC assured the Kea Conservation Trust it was working with iwi to create a recovery plan for kea. Trust co-founder Orr-Walker tells Newsroom: “They’re in consultation at the moment with iwi to actually set up a recovery plan for kea, which is great.”
Yes, she says, DOC hasn’t had the focus on kea it should have had. But she adds: “I feel that DOC now recognise that there has been this lack of work and support for kea up until relatively recently and they’re very keen to rectify that.”
The trust gets huge support from the department. Kea expert Josh Kemp, a DOC staffer, is its science adviser and it gets funding from the department, sometimes over multiple years, to do the work it hasn’t been able to.
Whatever those reasons, it’s clear that it’s not just DOC battling for our birds, these days.
Orr-Walker: “We have a species that was pretty-much reviled for 100 years and now we’ve got communities that are really stepping up to get birds back. To see that is just incredibly heartening and I think probably the reason why the species won’t become extinct.”
Can you help our journalists uncover the facts?
Newsroom is committed to giving our journalists the time they need to uncover, investigate, and fact-check tough stories. Reader donations are critical to buying our team the time they need to produce high-quality independent journalism.
If you can help us, please donate today.