The first test
If we’re serious about Predator Free 2050—about ridding New Zealand of every last rat, stoat and possum—then Fiordland poses its sternest backcountry test. If we can pull it off here, we can pull it off in any of our national parks. What makes this land so difficult—and so important?
On July 25, 2016, then-Prime Minister John Key stood before a small crowd at Zealandia, a wildlife sanctuary in Wellington, and declared war.
“We have adopted the goal of a predator-free New Zealand. By 2050,” he assured guests, “every single part of New Zealand will be completely free of rats, stoats and possums.”
Looking at this part of the country, I’m not so sure. Fiordland National Park—12,600 square kilometres, barely scratched by road or track, suffering some of the planet’s foulest meteorology. It will be raining here for more than half the year—eight metres of rain disembogues annually over Milford Sound. Chill southerlies mean the forest is defeated at just 1000 metres of altitude: above is gaunt granite, some of the oldest rock in the land.
The single biggest problem with pests in this park is getting to them. Glaciers have carved a confusing coastline—is that a passage or a peninsula? An island or an isthmus? You’ll need a boat, a good one—or a helicopter, and lots of fuel. Any endeavour here has to plan for immensity, hindrance, contingency and the consent of that funereal firmament.
It occurs to me that if John Key was serious about ridding the country of that unholy trinity, the acid test—the proof of concept—looms right here in front of me. If we can do it in Fiordland, we can do it almost anywhere.
This isn’t going well. We got here about two o’clock, with the simple ambition of firing a lead sinker over a high tree branch with a slingshot. It’s now 4.30pm, and we’re running out of sinkers. Telemetry brought us here—me and Department of Conservation (DOC) rangers Emmanuel (Em) Oyston and Bex Jackson—to this one beech tree, of thousands in the Iris Burn, in the southern reach of the Kepler. The receiver is still chirping, insistent: the long-tailed bat we want is right up there.
It sounds easy in theory—fire a sinker, with fishing line attached, over your chosen branch, then pay out the line until the sinker grounds on the far side. Then, tie a light string to the fishing line and pull it back to where you stand. Then, braid a climbing rope to the string and pull it back the opposite way again. Now, you can climb up to your bat.
I’ve lost count of the times we’ve flunked step one. Bush protocol dictates that each person gets three goes at firing the sinker, and the slingshot has changed hands many times. No DOC manual explains how to shoot it precisely through a pandemonium of forest canopy, but at last, we succeed.
Maybe 25 metres up, a small, innocuous cleft marks, we think, the bat portal. Oyston wriggles into a full-body harness, then checks the gear rack round his middle: knife, silky saw, belay device, radio, receiver, antenna. He double-checks the anchor Jackson has set up around a neighbouring tree trunk, complete with redundant backups and escape loops.
Then he clips a mechanical ascender into the rope, slips his feet into a pair of cord loops, and starts hauling himself upwards, like an inchworm, standing up in the loops, then sliding the ascender to head height before repeating the move. He reaches the roost entrance, where the receiver endorses his labours with an affirmative chirp. Then he cuts away the obstructing brushwood with the silky saw before fixing some guy lines.
Suddenly, he stops work. Even over the sputter of the radio, he sounds unnerved.
“This branch is rotten. I’m coming down.”
All the same, the pair decide that the branch, decrepit as it is, may still do the job. Jackson begins assembling a harp trap—a light aluminium rectangle, maybe a metre across, with very fine monofilament droppers every 50 millimetres, just like its namesake.
A net it is not: there are no horizontal lines—if there’s too much monofilament, the bats’ echolocation will detect it. So there’s just enough to arrest their flight as they leave the roost. Then, they fall into a fabric catch bag hanging from the bottom of the trap, where sheets of clear polythene form an envelope. Long-tailed bats are good climbers, and will clamber up the sides of the catch bag, but the polythene stops them there.
After more tribulation—the branch was so rotten, the climbing rope ate into it—Oyston and Jackson haul the trap up to the mouth of the roost and fix it with guys.
At this latitude, at this time of year, bats don’t leave the roost until—and I wrote this down—9.54pm. An ultrasonic bat detector is broadcasting their squeaky pre-flight excitement. Then, against the gloaming, the first, tiny flicker from the roost. The bat evades the trap with a deft barrel-roll. The trap isn’t set quite square; nevertheless, when we lower it to the ground 20 minutes later, there are 37 long-tailed bats huddled together in the envelope, as though it was just another roost. I help to gently pick them from their cell and place them into gossamer drawstring bags. “They might wee all over you,” warns Oyston. And they do. And I am still delighted.
We settle in. Jackson draws a squirming bat from a bag, then blows lightly on its belly, parting its fur, looking for plump nipples: “Lactating female.”
Then she stretches out one forewing, looking for a tiny ID tag, and reads off the numbers. Bats are scarce enough in these parts that Oyston knows the sex and provenance of each one in the roost.
Not that long ago, our knowledge of bats came down to whether they were there or not. Such is the level of fine detail that species monitoring has reached these days. This population wasn’t even surveyed until 2011, and by 2013, it was clear it was headed for oblivion if something wasn’t done. DOC laid out bait stations and rat traps.
“Without rat control,” says Oyston, “the population just flatlines. Then you get a mast, and it plummets. You can’t just relocate bats—once they’re gone, they’re gone.”
Short-tailed bats in the nearby Eglinton Valley have prospered, thanks to sustained trapping and poisoning.
“We know you can quadruple the population with rat control,” says Oyston, “but we had no budget for the Iris Burn.”
Fiordland National Park occupies 15 per cent of the conservation estate, but gets less than one per cent of DOC’s budget, so it fell to Treasury to stump up for the first Battle for Our Birds pest-control campaign in 2014, when the Kepler bats got some breathing space after an aerial 1080 drop. The population graph levelled off, then started to climb after another 1080 drop in 2016.
The female population here at Rocky Point has since doubled, which sounds encouraging until you learn that amounted to only 65 in 2017. We had held more than half of them in our hands.
“Out you get,” orders Andrew Smart, and the gaggle of tourists turn their cameras from a couple of panhandling kea to Oska, a six-year-old Czechoslovakian wire-haired pointer, as he jumps from the ute.
Oska’s fluoro-orange Conservation Dogs jacket brightens a rainy scene at Monkey Creek, on the Hollyford road. Smart, who is a biodiversity ranger and the manager of DOC’s whio recovery programme, knows that a few pairs of the blue ducks hang out nearby. By now, if the season’s been kind, they should have ducklings with them, and he wants to know how they got on.
We wade into the chill riffles. I totter on the slippery rocks, but Smart has been doing this so long, he doesn’t even need to watch his feet. He’s gazing upstream, canny to the abstruse difference between a wet rock and the steel blue of a whio. This is a man who can tell lichen and whio poo apart from 20 metres. Oska lifts his nose into the wind, and freezes, fixed on a clump of dead rushes.
“A pair nested in here last year,” says Smart, who teases apart the fronds to find fluffy down still inside. He digs gently beneath. “Sometimes, they bury their eggs if they’re going out for a bit.”
But it’s no nursery this year, and we walk on.
Oska, with professional detachment, is undistracted by a family of grey ducks as they flee up a side creek. Around a bend, the stream gets rowdy as it squeezes down a chute, and Smart calls off the search.
“They won’t take ducklings into water like that by choice.”
We retrace our steps to the ute and cross the road. Now, we’re wading the Upper Hollyford, thigh-deep and gelid. Smart knows the home patch of every pair up here, and upriver he sights a familiar pair. He sighs in relief at the three stone-grey ducklings in their wake. “They’re really vulnerable to stoats when they’re nesting,” he says, “especially the females, because they do all the incubating.”
Along reaches without stoat control, it’s not unusual for every last sitting female to be killed. Some whio populations around the country are now 70 per cent male.
Between 1999 and 2005, Smart trained surveillance cameras on whio nests along the Clinton and Arthur Rivers. Over the summer of 1999-2000, the beech trees masted, producing a crop of seeds that carpeted the forest floor with rat food. Sensing plenty, the stoats, too, bred to capacity, and the Milford was soon aflood with predators. At Smart’s study site, it took an average of nine days before newly laid whio eggs were plundered by stoats. Only one clutch of eggs hatched, and the entire brood was promptly killed the next day. Other times, he says, with a mix of loathing and admiration, “a stoat will use a nest as a larder, picking one off each day”.
Here in northern Fiordland, whio have enough to deal with as it is. In 2016, 9.2 metres of rain fell in Milford Sound, up the road—the wettest year in close to a century of records. Trickle becomes torrent in minutes, sweeping ducklings away and inundating the nests of whio, which don’t have the benefit of climate-change projections.
But predator control is giving them at least some respite. Four river systems, the Worsley, Clinton, Arthur and Cleddau, along with Sinbad Gully, below Mitre Peak, are lined with stoat traps. The aim is to grant sanctuary to at least 50 pairs of whio over these 65,000-plus hectares.
Everyone gets a GPS, and a radio. Then they have to show Lindsay Wilson, a biodiversity ranger of long experience, that they can use them. We are, after all, about as disconnected from civilisation as it’s possible to get in nearshore New Zealand. Here in Preservation Inlet, human presence is mostly echoes and relics—Puysegur Point lighthouse, the Golden Site mine, the country’s first whaling station, and the ill-conceived Tarawera Smelter at Isthmus Sound.
We stuff packs with raincoats, of course, and first-aid kits, dayglo triangles, rolls of pink tape, tracking tunnels, ink pads, then Goodnature A24 self-resetting traps and long-life chocolate lures. Another legacy of failed human judgement—stoats and rats—still prowls the dozens of islands that lie, like a fleet at anchor, across this reach.
The boat’s contingent—DOC staff, volunteers and sponsors—splits into pairs, each assigned a different island. Wilson hands out maps, etched with crosses plotted into the GPSs. They spell doom: lines of new-tech traps that disgorge their own victims, re-set their mechanisms and keep themselves primed with fresh lure.
It’ll be six months before Wilson has to think about these traps again, and in a place like Preservation Inlet, autonomy saves him a fortune. The biggest cost is coming all this way. This boat, the Southern Winds, barely testing her anchor in this perfect mooring, is worth “the thick end of $3 million” alone.
The boat’s mate, Peter Young, has plied these waters since he was 16, first as a deckhand, then a crayfisherman, most recently as a contract skipper.
“There are probably 350 islands in Dusky [Sound], and a thousand rocks you can hit,” he says.
In Young’s head is a vast 3D chart: “Radar’s just a fuse away from failure.”
With deft helmsmanship, he drops Wilson and me onto the rocks of the main island in the Cording group, where we continue setting out a trap line we started yesterday evening.
Already, rat corpses lie beneath two of the A24s we installed, but even Wilson, who’s been doing this for decades, can’t tell whether they’re Norway rats or “shippies”. He folds the tail forwards against the body—a ship rat’s tail, when measured so, should extend beyond the head, but these don’t. Their owners are huge, so Wilson records them as Norways, along with the place and time of death. Then he pops the cadavers into a plastic bag for DNA analysis.
This is important. The rats’ DNA will tell him whether they’re part of a resident population on this island or have swum here from somewhere else. That provenance is the difference, for DOC, between biosecurity and pest control.
We push on, over a carpet of ferns, weaving between solemn stands of tōtara, miro, and rātā in the first roseate blush of December. Everywhere, throngs of miro seedlings are racing for the light, which means there are no deer on here right now. But they were here not long ago: hundreds of saplings are bare stalks for the first two metres, green buds only now starting to reappear. Above deer height, they’re lush.
The going varies between an easy amble and a maddening thrash through jumbles of windthrow and supplejack. I mark our way with ties of pink tape, and every 50 metres, Wilson nails a trap to a tree trunk, screws in the carbon-dioxide cartridge that will arm it for another six months, then inserts a small tub of longlife lure. I nail a pink marker triangle above while he jabs the coordinates into a GPS. We put two traps into every hectare, 51 in all, close enough to ensure a rat can’t help but bump into one.
That evening, senior ranger Colin Bishop examines our catch.
“We keep underestimating rodents,” he tells me. “We’ve had good rat eradications in the past, and put measures in place to deal with any swimmers. That’s worked well for maybe 10 years, then all of a sudden, we find an established population on the island.”
Why does that happen?
“Trapping basically selects for critters that know how to avoid your systems. They learn. You might check a trap and find a rat paw in it. Well, there’s a good chance that rat’s going to survive, and no way is he gonna stick his head into any more traps. And he’s on your island.”
A bewitching waft of bacon and eggs gets me out of my bunk and into Tuesday, which has dawned squally and sullen. We don’t need the radio to tell us about the 35 knots of sou’west wind—Peter Young calls it a “bad hair day”. Late December, and I’m wearing three layers.
I get the job of checking stoat traps along the coastline of two of the Small Craft Harbour Islands, and as Young eases the inflatable into a cove, white-fronted terns scull upwind, their bills full of baitfish.
It’s a steep scramble through dracophyllum, but I can follow nominal tracks left by penguins. The last trap has the dated remains of a stoat in it, which means the creature had paddled at least 800 metres across the narrows of Cunaris Sound, or maybe more than two kilometres from further shores. In either case, it was a routine paddle for this accomplished swimmer. We still don’t know just how far a stoat can swim, but in 2010, one was found on Kapiti Island, separated by five kilometres of turbulent sea from the Wellington mainland. In a 2013 study, a female stoat swam against a moderate current in a flume tank for nearly two hours without rest—the equivalent of dog-paddling 1.8 kilometres.
Farmers introduced stoats to New Zealand in the late 1880s to control rabbits, but stoats set about decimating native birds instead. Pioneer conservationist Richard Henry, curator of Mauikatau/Resolution Island, was the first to underestimate their swimming skills. A dejected entry in his diary from August 1900 notes they had reached the sanctuary he thought he’d made for kiwi and kākāpō.
Ninety-nine years passed before stoats were first eradicated from a nearshore island—Te Kakahu o Tamatea/Chalky Island, not far from here. Encouraged, DOC set out to rid them from other, larger Fiordland islands: Pukenui/Anchor Island (1130 hectares) in 2001 and Te Puka-Hereka/Coal Island (1163 hectares) in 2005.
Records from four years of trapping on 19 islands suggested stoats were reluctant to swim more than 300 metres. They lied. Massive trapping efforts on Secretary and Resolution failed, and DNA testing duly showed that most summers, a dozen or so stoats strike out from the mainland in search of richer hunting—even more in mast seasons. While small islands are rarely breached, stoats seem to be emboldened by longer coastlines that forgive any miscalculation of strong currents.
For now, stoats are an ineluctable presence, but Lindsay Wilson thinks he can work with that, if he can just keep their numbers low enough. Barely clinging to Resolution Island by the merest filament of land, Five Fingers Peninsula is his test bench. In August 2017, he oversaw the transfer of 138 tīeke, South Island saddlebacks, to Five Fingers. Trusting and terrestrial, tīeke are the most hapless of stoat fodder. While no stoats had been trapped near the test site in years, Wilson knew they were still present on Resolution.
It was a contentious move, even in his own office. Some felt he was sending the birds to their deaths, but he insists it was worth it.
“We needed to see whether these birds could cope with stoats in low numbers,” he says.
If they could, he says, it would buy him more options. “It could change our thinking about what’s possible on inshore islands.”
And a notion that would make Richard Henry beam: “It means we might be able to move kākāpō back to Resolution.”
So far, the Five Fingers tīeke are holding on. A January survey found 27 birds, and saw one ‘jack bird’, an unbanded juvenile, which meant they’d already begun breeding. Another search in March found three more youngsters.
So great is the loss of biodiversity in New Zealand that, when you get a glimpse of its true richness, it takes you a while to recognise it. As we step ashore onto Chalky Island, we hear bird calls coming from the riotous rātā forest above that I can’t immediately identify— tīeke, mohua, kākāriki. A falcon scuds overhead. And kererū; big flocks of them. Wilson and I start to wend our way—carefully, because there are seabird burrows everywhere—up through coastal scrub and into a bushy myriad: rimu, matai, rātā, kāmahi, hutu. The sun spangles in the
We check a line of stoat traps, which should be empty. Chalky was rid of stoats in 1999. It does not suffer from having rats, mice, or anything else, which is why, as Wilson checks one trap, I can count seven tīeke, four mohua, a brown creeper, a tūī, and at least four robins. I give up counting the bellbirds.
Higher up, in the tussocks that line Divide Track, Wilson points out a bowl, a flattened round in the grass. It’s the imprint of a kākāpō. There are males here, retired from the breeding programme on Whenua Hou to the south. Not far from the hut, we find feathers on the track, and nearby, nipped stumps of tussock. A kākāpō was feeding here not long before.
This is what it’s all for. The leaden packs full of traps, and the drenching, groundhog days of checking and rebaiting them. The dehydrated dinners. The difficult boat landings. The fundraising, the footslogging. It’s why Em Oyston dangles from trees in the dark. Why Andrew Smart puts up with cold, wet feet.
But clearing Chalky was, in eradication terms, easy: the plan to purge the mainland of pests will be many orders of magnitude harder. All the same, people who know about this stuff are taking it seriously.
One evening, Colin Bishop spreads a map of southern Fiordland over the transom of the Southern Winds. Big swathes of the park are delineated in different colours—hundreds of thousands of hectares each, but they all meet at some choke point in the topography: the narrows of an inlet, the trunk of a peninsula, a pass between two lakes.
These are hypothetical front lines, a succession of beachheads that Bishop thinks could mark the path of a rolling maul, from the south coast up—550,000 hectares in all. Thinking big brings some equal-sized challenges, he says, but the payoff is huge, too.
“When I first started working in Waitutu, we were controlling possums over 2500 hectares by ground control, and it was pretty hopeless, to be honest. The contractors were doing a great job, but possums were re-invading really quickly. We had to go back every second year, all for an 80 per cent knockdown at best, which is what you average with ground control. Then we started doing aerial 1080.”
In the coal mine of Waitutu Forest, kākā make excellent canaries. Because they nest inside hollow trees, they’re pitifully vulnerable to stoats. And because females do all the incubating, they suffer a highly skewed death rate—in 2005, male kākā outnumbered females by six to one, and surveys saw not a single youngster.
In the decade since Bishop began his aerial 1080 campaign, female kākā have increased nearly fourfold, and juvenile kākā by a factor of 20.
“The benefit of doing broad-scale control like that is just huge—the re-invasion rate really slows down.”
This map, then, is part experience, part ambition: “We could build on that, go bigger again, and work towards these natural barriers—the coast and the sounds—and really increase the bang we get for our buck.”
Bishop might be describing the modus operandi of Zero Invasive Predators (ZIP), a public-private pest-control research and development start-up that’s bench-testing the tools and techniques we’ll need if we’re serious about going predator-free. The ZIP model, ‘remove and protect’, is about picking target zones between natural barriers such as fast-flowing rivers, eradicating pests inside them, then using those barriers to thwart re-invasion. ZIP chief executive Al Bramley, an engineer by trade, has spent the past three years figuring out how to do pest control at high resolution, using technology to add precision to a task that, historically, has simply worked on percentages.
So far, the remove-and-protect strategy has worked. On Taranaki Mounga, a pest-eradication and species-reintroduction project covering a vast area of Taranaki, ZIP got “a couple of thousand” rats over 1600 hectares down to 12, with one drop of 1080, then down to zero with ground baiting. Then, Bramley ramped it up. In the Jackson-Arawhata catchment in South Westland, he got the same result over 2300 hectares.
“People don’t appreciate how intensive our detection network is,” says Bramley. “At Taranaki Mounga, we put two detection devices into every rat’s home range. And then we added cameras, chew cards—everything we could think of. Then we watched, for 50 days.”
Bishop’s plan is intoxicating, but pest-control experts rarely allow themselves to yearn. Possums will be a doddle: they’ve barely established in western Fiordland, and they’re slow breeders, but the idea that you might kill every last rat in here—and, critically, stop its neighbours moving back in—may yet be an ambition too far for Lindsay Wilson.
“ZIP’s pushing the boundaries, testing new approaches, and it’s great work and a great ambition. But I’d be very surprised if 1080 took out every rat. We know we can get them with anticoagulants—we’ve done it many times—but to do it with 1080 would be a big shift. You’d expect some of those rats to get a sub-lethal dose, and not die.
“We regularly get zero tracking with rats, but we know they’re still out there. Time will tell.”
Bramley acknowledges that rats are a challenge, “largely because they can re-establish much faster than possums. But it’s not impossible—we’ve done it before on lots of islands, so why not the mainland? I’m hopeful we can completely remove predators from New Zealand, and keep them out, across the whole country, in the near future.”
At least the same climate that makes Wilson’s life hard makes Fiordland tough going for rats, too.
“A rat plague in Fiordland beech forest is just normal background noise in the Urewera,” he tells me. “Up there, 70 or 80 per cent tracking is normal. Even if you knock them hard, they’re back in six to eight months. Down here, if you get the timing right, it can be really effective—you might only have to do six months of control every four or five years.”
Fiordland will be one of Predator Free 2050’s toughest gigs, but there’s an ace in the hole: Fiordlanders themselves.
Social licence is almost a given. Very few people live inside the national park, but those who do are already doing pest control of their own. More broadly, DOC enjoys such wide public support in Fiordland that commercial and private sponsorship, services-in-kind and volunteerism boost its annual budget here by around 60 per cent.
For more than a century, our biodiversity has been huddled on a few safe islands, lifeboats from the wreck.
Aotearoa itself is under foreign occupation, but finally, we’re turning our gaze back to the mainland and daring to set plans for its liberation. Fiordland may be the acid test of technology, but it will be tenacity that carries the day—Fiordlanders themselves who will prove it’s possible.
Aboard this trip are two sponsors from the Coal Island Trust, here to set out and service trap lines, and Mike, an orchardist who wishes to remain anonymous. His company paid for the A24 traps we set out on the Small Craft Harbour Islands. He won’t publicly disclose the money he donates to this effort every year, insisting that “it’s small bikkies compared to what the big sponsors put in”, but he does disclose it off the record, and it’s nothing like small bikkies. And he’s been helping out for years.
“We’ll continue to do that into the future,” he says, “because this is a long game.”