Our first night on Puangiangi, 31 April 2012, was frightening. It wasn’t the scary 12-volt electrics, which were way dodgier than anything William Heath Robinson might have concocted, or getting to grips with how the water might be directed to the inside tap, but rather the things that went bump in the night.

That day, we had arrived with the locksmith, gained entry and gotten to understand the infrastructure as best we could in the fading light, eaten and went to bed. Soon enough a lot of thumping and scrabbling came from the ceiling and walls of the house. This was frightening as it could have meant that Peter’s hard work in 1999 to eradicate rats had been undone. It turned out though, that this is possibly what walls and ceilings are meant to sound like on the mainland, if they were full of geckos.

Poking around the following day, we discovered that all the geckos were common geckos (now raukawa gecko, Woodworthia maculata), and that they liked the shed as well as the house. There was not a box or a nook that didn’t house a gecko, or several. Our tools and gear in the shed now need a regular clean lest they disappear beneath piles of sticky gecko droppings. They are also numerous in piles of corrugated iron outside, and Ross Webber’s various mussel-float constructions are popular too. It was a bit of a party trick to show visitors the hundreds of geckos gathered under a corrugated iron stack, but Lynn Adams and I got a shock when a robin showed up and began swallowing not-so-small geckos whole when we lifted the iron.They also get into our various rainwater drums scattered about the island to provide water for birds, and often could not climb out and drowned, until we worked out they needed a length of rope coming out the top of the drums to escape by. Dad was quite taken when he visited by the sight of little gecko faces peering over the edge of the spouting on the house, where they would gather at night probably to get a drink of the dew collecting from the roof. A look on tree trunks at night will show them on their way to catch insects, or take nectar and fruit. We’ve also seen them in the house in the daytime popping up from gaps near the windows to try to grab flies on the glass. At this time of year they are pretty torpid on the cool mornings, and they need careful rescuing if we are opening and closing doors and find them in the cracks. They wake up within a few seconds when given some warmth from a hand, and make a break for it. Peter noted only a few common gecko when he first visited Puangiangi, and the recovery since rat eradication has been explosive.

A sunny day early on soon also revealed glossy brown skink (Oligosoma zelandicum), and northern grass skink (Oligosoma polychroma). These common skinks certainly should be there, but we often remark that numbers seem quite low compared with what we have seen on similar islands, and are also a surprisingly low proportion of the known lizard population. Late summer, however, shows good numbers of young skinks in the grass. Perhaps the adults are more circumspect given the moderate numbers of kingfisher around, or maybe sunny skink habitat is slowly shrinking as the pasture becomes more rank and is taken over by shrublands.

Glossy brown skink. Note vertical markings around the jaw, and brown iris. Photo courtesy Ivan Rogers.


Northern grass skink. Note the strong brown and cream lateral lines, straw-coloured iris and blunt snout. Photo courtesy Ivan Rogers.

What else should be there? Forest gecko would have a similar ability to common gecko to persist in the face of rat predation, and would be expected to recover at the same rate, but so far we have not found any. They are on Maud Island/ Te Hoiere and some of the other Sounds islands, although they were last reported from Takapourewa in 1897. It’s only a few tens of thousands of years since Puangiangi would have been connected to the mainland, so I think they must have been there at some point, if not now.

Marlborough green gecko would be another possibility, although again we haven’t seen any yet. When Sue and I went to Auckland to visit Ross and Jean Webber in 2012, though, Jean said she had seen a green gecko by the island’s house where they sat in 2009-10 contemplating where the interim owner, and their money, might be. Jean is not a lizard expert and I wondered whether she might have seen a dull-greenish common gecko which are, er, common, but when pressed as to the type of green she confidently pointed to a tissue box with an apple-green graphic on the side, the colour of the Marlborough green. They are pretty elusive and who knows if we might corroborate her observation one sunny morning when walking past a kanuka and see one contrasted against the foliage. Other lizards that are nearby and might have a chance of remaining on Puangiangi are striped gecko and spotted and speckled skink. All of these lizards made it into our management plan as candidates for translocation if they were not found in the meantime.

In the 1980s when Peter visited Ross, the party was shown an old photo of a gecko which Ross had kept as a pet. The photo was most likely of a Duvaucel’s gecko. The nearest current population of this threatened species is on Kuru Pongi/ The Trios. Ross was clear in conversation with me that it was the only one of its type on Puangiangi and that he had collected it from a “stack”. Conscious that I might wear out my welcome with my incessant questioning, I wasn’t able to establish for sure that he might have meant, say, North Trio. His reticence might have been because, if he had been allowed pet geckos then, he certainly would not be now without a permit.

When present, Duvaucel’s stand out because of their size and are easy to find after dark when they come out to feed. A careful look around confirmed Ross’ comment that they were very unlikely to be on Puangiangi. I imagine their size and docility would have had them eaten by rats early on if they had been there historically.

Both for our ecological restoration and to assist the species, a Duvaucel’s translocation would make good sense. Peter’s translocation proposal was approved by both Ngati Koata and DOC, and we arranged to visit Kuru Pongi over several nights in early March 2014. Jim Williamson came out from Nelson with Brigadoon and he was to be the ferryman over however many trips it took to get the permitted number of animals. Jim kindly covered all his expenses for his trip and saved us a considerable sum in transport costs.

The weather wasn’t great on the evening of the 3rd, but we made the 7 km trip to Kuru Pongi anyway. Alison Cree in her 2014 book on tuatara quotes Hugo Schauinsland, a museum director from Bremen, who collected in the area with his wife and assisted by George Webber in December 1896 and January 1897, landing on Kuru Pongi- “an undertaking which…could almost have cost me my life”. We were to land on the western side in Jim’s inflatable but just before we were ready to go, a squall blew up. We didn’t want to go one better than Schauinsland and land on the rocks upside down and we were ready to head back to Puangiangi for the night. Luckily though, Jim found a little bay on the northeastern side that was sheltered enough to get to land and that appeared to give access to the top. Peter ferried the catching party ashore in the inflatable and Lynn Adams and Chris Birmingham from DOC, and I, got ourselves and our bins of gear up a tight scree onto the main ridge of the island before nightfall.

Gaining the ridge on Kuru Pongi. Squall completely evaporated, Puangiangi in the background. Photo courtesy Chris Birmingham.

The going on Kuru Pongi is very slow, with low, scratchy scrub. On the main ridge there was a hint of a flagged route from earlier parties, of which there have been few, and we found a flat site and sorted out our gear. Lynn gave me and Chris a refresher on how to tell a small Duvaucel’s from a large Common, and males from females (we wanted more females), and we began catching just before dark. Lynn withheld crucial habitat information and established a handy lead in the catch race by grabbing drowsy geckos from under rocks while Chris and I were concentrating on the trees. As it got properly dark we all started to contribute to the steadily growing pile of animals in individual calico bags in their fish bin. The scrub was so impenetrable that there was little point trying to move around much, and we did pretty well by standing still and listening for them moving, and watching for them on the trees. They could be distracted by our head lamps while an unseen hand gathered them up easily without any sudden lunges that might have harmed them or caused them to shed their tails. The island is also home to tuatara and burrowing petrels, and it was nice to see them on their island going about their nightly activities, or inactivities in the case of tuatara.

By midnight it was clear that we could catch the full complement in a single night and so we decided to keep going. By 3:00 we had 51 animals, all with their tails, back at base and Lynn checked each for over-all health, and identity given that I was catching too. We slept for a few hours and dawn revealed that Lynn had ended up uncomfortably close to a drop-off, which two of us found quite amusing. A bit of grunting and heaving got the boxes and us down to the beach and Jim and Peter picked us up at 7:00.

Being weighed mid-skin-shed, on the kitchen table.

Back on Puangiangi, Peter and Lynn weighed and measured each animal and marked them for future identification using the recommended toe-clipping method. Hugh Blank from Pure Science kindly donated the electronic balance we have for weighing animals. The proportion of females was just right at 70%, but with relatively few gravid, and all were in robust health when examined more closely in daylight. By early afternoon we were able to release them at the chosen site in the northeastern forest. We had chosen a place where there are some solid rock slabs with gecko-sized cracks. They all made for the cracks with little encouragement.

Jim Williamson releases Duvaucel’s, northeastern forest. Photo courtesy Chris Birmingham.

This was a good moment to reflect on the contribution the late Tony Whitaker made to this project. He gave advice to Peter just days before his untimely death, and the techniques we used, and getting the whole thing done in under 24 hours, owe much to his work on lizards over many years. The kohekohe canopy is reasonably high and we won’t stand much chance of seeing them in numbers for a while. We’ve looked for them twice in the intervening months and have seen a single one, sitting quietly on its rock after dark. I don’t want to disturb any of the animals unnecessarily and the geckos don’t benefit from us knowing their every movement, so checking for translocation success will be of the low-impact variety. The obvious threat to them will be if predators get on to the island, and it will remain important to keep weka numbers vanishingly low, although the northeastern forest is not especially favoured by weka as far as we can work out.

At the release site, northeastern forest

To my knowledge, this is the first translocation of any species from Kuru Pongi, possibly save for early “collecting” trips for tuatara where the animals were released into a garden or suchlike, or indeed Ross Webber’s pet gecko.

Lizard diversity would be a good marker of ecosystem health, and we wanted to get some more professional advice on what if any other lizards are and should be there. An expert team of Lynn Adams, Sue Keall (Victoria University), Peter, Rod Hitchmough and Ivan Rogers (both DOC) visited for a week in mid-March 2015, equipped with binoculars with torches taped to them, Gee’s minnow traps and pitfall traps. They confirmed what we thought about the state of common gecko and the two widespread skink species.

One evening Lynn rang me and said they had found something significant, and demanded that I guess. I immediately suggested that they had found spotted skink (Oligosoma lineoocellatum), which turned out to be right and they all thought I was very brainy. Lynn doesn’t know that Peter and I had already considered that spotted skink ought to be there and it was probably just a matter of the people with time and the right identification skills showing up. My reputation won’t be punctured by putting this on the internet because that is a great place to keep secrets. After finding the first spotted skink, the team quickly found more and have concluded that it is at least moderately abundant.

Peter, Rod Hitchmough and Sue Keall search for skinks, Woolshed Bay. Photo courtesy Ivan Rogers.

The team did not find any forest gecko, Marlborough green gecko, striped gecko or speckled skink, but noted that ongoing survey effort would be needed to say they were not there for sure. In the meantime, they said we should go ahead and translocate striped gecko, a species found on Te Hoiere and Takapourewa and in need of insurance populations. Peter has completed the first stage of the permit application as of last month.

Title photo, common gecko in a corrugated iron stack. Courtesy Roy Grose.

Sooty Shearwater Update

In late April 2016 we checked the sooty colony again with the burrowscope. The sooty chicks would be fledging in early May and if there were healthy chicks present in late April, then they would probably have had their last parental feed and would be likely to fledge successfully.

We found 16 healthy sooty chicks, in various states of down coverage, but all looking good. Down coverage is a misleading method of aging chicks and a single wet night on the surface can turn a fluffball into a bird that looks ready to go.

Not only was 16 a vast improvement on last year’s presumed 3 fledges, it was from 20 active burrows (of 30 known burrows) at egg-laying time in late November. This is a high success rate and implies that all was well at the colony with respect to weka predation and any other adverse factors beyond our knowledge.

Weka numbers were a bit lower this year than last, but we did not see any evidence of successful weka predation last year either. I guess some at-sea event, such as food concentrations moving further south than expected, might have explained last year’s result.

Can we ease up on weka-catching therefore? Quite possibly with respect to the sooties, but not if we are hoping that any of the smaller seabirds like fluttering shearwater, fairy prion or diving petrel will establish.

Stating the Obvious

So, you go shopping for an island for conservation purposes and buy one. Why? Then what?

To answer the why,

I suppose we had gained some experience of how species and restoration projects went, and figured that we were capable of taking one on for ourselves.

As funders, we had been happy with the way DOC-led species projects on the public estate had gone, and we were learning something from the people who ran the projects. There were a few outstanding successes. Applications began to dry up though, perhaps because we had taken care of the ready-to-go projects and those remaining were too nebulous or expensive, or both, that they didn’t suit us. I tend to conclude though that DOC had had the stuffing knocked out of it by endless restructuring, and people were too busy polishing their resumes or reapplying for their own jobs to do anything. Whatever the reason, the money we had earmarked for conservation was going unspent.

We were also learning about ecological restoration projects, especially through our involvement as volunteers on publicly-owned Mana Island. The model of a community group working with the government did not suit me though. The volunteer workforce was looked on by some DOC staff as a group which could be used to tick the required community engagement box, but was otherwise just a nuisance to be kept at arm’s length. Conversely the volunteers ranged from brilliant to useless, and those who thought the project was a holiday and the DOC rangers were there to entertain them, annoyed the heck out of me.

Having established that I was too difficult and demanding to fit in anywhere else, and also for positive reasons, we started thinking about our own project. Although I am well qualified in one scientific field, it would be a mistake to think that I could branch out and offer anything in the threatened species area as, say, a captive-breeder, researcher or research manager. However, there were templates available for ecological restorations and we figured it was realistic to attempt one with the right team in place.

A mainland sanctuary project was considered and rejected pretty much straight away. Simple preservation of an area of forest through purchase and covenanting did not cut it. In the presence of predators, even mainland ecosystems covered in natural vegetation are little more than biodiversity deserts, despite looking pretty and attracting public attention such as the recent crowd-funded purchase of a beach in the upper South Island, laudable as a statement about land alienation but achieving little for conservation.

The next step up, the predator-reduced, enclosed sanctuaries, have costly chemical or wire fences. Expenses are quite often shifted onto the unsuspecting taxpayer or ratepayer without proper analysis of the value of the spend. In the absence of mainland New Zealand becoming pest-free, fenced sanctuaries will require centuries of commitment so that the gains will not be lost the minute interest wanes, funding dries up and the fence is breached. Any conservation project can also have its detractors, and I didn’t fancy dealing with sanctuary neighbours who might be unhelpful or hostile either. An island project was therefore top of my mind.

To state the obvious, islands far enough from the nearest pest-infested area already have a fence. It’s not impermeable, but we figured that provided a pest-free island’s tenure was impermeable, it would cost little to ensure that traplines, bait stations, tracking tunnels were monitored in perpetuity. The same thinking has many islands in State ownership at the forefront of conservation efforts as arks for threatened species, pending the mythical breakthrough on the mainland.

Having figured an island project might be the one for us, we knew there was a dearth of choice. Many worthwhile islands are already in public ownership, and many are in multiple undivided title based on traditional tenure but stymied from contributing economically or conservation-wise through the paralysis brought by having many owners. Most or all of the easily accessible, fee-simple private islands have become the playthings of the rich (local and foreign), with little or no conservation benefit being contemplated or implemented. The larger ones have usually been subdivided and therefore would be difficult to bring under effective management for conservation purposes.

I was aware of Puangiangi when Ross Webber put it on the market in 2004. It was still in play simply because Ross had kept hold of it for 47 years through waves of island-buying fever, and because the more intelligent well-heeled could see it was a nightmare to get to and build their mansion on. 2004 was not the right time for us though and it didn’t compete for dinnertime discussion on top of everything else in our lives. The government’s Nature Heritage Fund was interested and put up a proposal to buy it, which was ultimately rejected by the then Labour conservation minister. It therefore went the way of the others, in to the hands of a property-developer type who had no interest in conservation, largely  financed by Ross leaving money in (I can only hope without the benefit of professional advice).

Photo taken for 2004 sale finds it way into Air New Zealand’s in-flight magazine, in an article about Rangitoto near Auckland. Perhaps the backdrop of D’Urville rather than the Sky Tower might have assisted in getting it right.

Conservation got a second chance though when the interim owner fell over and Ross got the island back for resale. Ross just wanted the mortgagee-in-possession headache to go away, but still preferred the island to go to someone who would make good use of it. After there had been some shenanigans with another local property, he wanted to know we weren’t just a front for offshore buyers, and interestingly he did not think the Government would make the best possible custodian either. Peter knew Ross and gave us a good reference. Our 2012 tender was accepted and Ross was finally free of an island he had left 8 years previously at age 74.

This link contains a CNBC video (scroll down a fair way) which evidently passes for news, and says much about the mind-set around islands that we did not want to see continue for Puangiangi. (The numbers in the video are laughably wrong by the way.) It took plenty of effort but the island is now safely held by our Charitable Trust. It must be used for public benefit (conservation) according to the law and to our Trust Deed. Although we take plenty of selfish pride in having got it this far, the asset is strictly alienated from us and is not our personal property. Our intention is that our Trust will become a perpetual one, and we have also been obtaining advice from DOC’s legal people about the best overlying structure to put on the island to further ensure protection. Interestingly they are not currently recommending any of the Wildlife Sanctuary options and the like available under the Wildlife Act, but lean more to the flexible Protected Private Land agreement offered under the Reserves Act. Extending the existing QEII covenant is also an option.

About the third question most people ask about the island (after “how do you get there?” and “can I come too?”) is “how big is it?”. According to a 1991 report by Ian Atkinson and Rowley Taylor, who compiled the predator status of all islands bigger than about 5 Ha, Puangiangi at “69” Ha squeaks into the 100 biggest at about Number 93. They measured areas by cutting out mapped outlines of the islands and weighing them. My estimate using a more modern method was 63 Ha. That is about 2.5 times the size of Matiu-Somes, a fair bit smaller than Mana but a similar length, and within coo-ee of many of the islands with the highest value to conservation.

Then what?

Well, apart from stating the obvious that we had managed to secure Peter’s services and thereby make the project viable, we wanted to get some other experts out for an ecological survey, and to write a management plan. My previous article talks about the botanical survey, you will find one about seabirds and future articles will cover land birds, lizards, and invertebrates. Having got some basic survey information, and with reference to the late Tony Whitaker’s 2002 plan for restoration of the adjacent Wakaterepapanui, from which we liberally stole (the plan, not the island), we got to preparing the plan.


The island is part of the Cook Strait Ecological District- think wind and salt. Islands include: Takapouwera/ Stephens; the Rangitotos- Wakaterepapanui, Puangiangi, Tinui; Nga Kiore/ Jags; Kuru Pongi/ Trios; Nukuwaiata and Te Kakaho/ Chetwodes; Titi; Mana; and Kapiti. Add the northern tip of Rangitoto Ki Te Tonga/ D’Urville, Capes Lambert and Jackson, the exposed headlands of Arapawa and the Port Underwood coast, and the coastline from Wellington to Kapiti. The mainland areas and D’Urville and Arapawa carry pest mammals, but all of the other islands have either always been pest-free or have had pests eradicated. Occasional incursions have been reported on many of them. Puangiangi is important to the ecological restoration of the other Rangitoto Islands, and vice versa given how close they are to one another. Puangiangi’s nearest other neighbours are Takapouwera and Kuru Pongi, which to a greater or lesser extent are seabird islands. I’ve previously discussed the place of seabirds as keystone species and to state the obvious we were always going to see Puangiangi as a seabird island. What level of intervention to achieve this is desirable?

Level 1: doing nothing apart from keeping pests away


The island would be likely to re-vegetate rapidly, but forest development and health would be limited by grazing. Weed species such as pines and macrocarpas would spread out and might dominate except in those areas already with mature forest. New weeds might arrive and establish. The weka would probably prevent seabird establishment and they would limit growth of the reptile population. To state the obvious, we would have only the limited range of birds, lizards and invertebrates that were already there, plus one or two self-introductions. The outcome would be a modified Cook Strait ecosystem.

Level 2: Level 1 plus removing sheep, weka and weeds and keeping it that way

Sooty1aThis option would allow seabirds to establish, and the forest would over time regenerate to what it would have been, minus any lost taxa. The populations of the existing animals plus any that could get themselves there would increase as habitat improved. The outcome would be a fully functional seabird island typical of Cook Strait.

Level 3: Level 2 plus restoration and translocation

robinThe reintroduction of species not on the island but which should be, gives an ecosystem rather close to what it would have been before humans and pests came along. If threatened species were included in translocations, it would help their security by giving them an additional site.

Level 4: A refuge

IMG_0284_1xThe island could be made available to house acutely threatened species, possibly at some detriment to an ecological restoration intended to provide a typical assemblage of plants and animals.

Preferred Intervention

The island’s size, location, security of tenure and good likelihood of being able to stop it filling up with pests again, convinced us that the highest-cost option, Level 3, was too good an opportunity to miss. We were able to formulate our goal pretty easily and to state the obvious, we propose “to return Puangiangi to a state that most closely resembles a fully functioning, Cook Strait seabird-island ecosystem in pre-human times”.

An earlier article talked about our approach to restoration of the forests and shrublands, and to support this our plan includes how to reduce the risk of a fire, which would be the single biggest threat. Realistically we can’t fight a fire except perhaps around the house, but we have a total ban on open fires and smoking.

Also integral to the goal is having  a proper biosecurity plan and all visits involve bag and equipment checks to reduce the chance of assisting pests to arrive. Unfriendly signs discourage trespassers and mitigate the attendant risk of bringing pests ashore. The island’s title extends to the waterline and there is no foreshore reserve or Queen’s Chain, so landing is able to be prohibited.

As for species reintroductions, we have made lists of birds, lizards and invertebrates based on what is found or should be found on the other relatively unmodified islands in the area. Our thinking about seabirds is described in a separate article. Land bird species that were missing in 2012 and might have been expected to be there include tui, bellbird, brown creeper, tit, robin, kakariki, fernbird, kaka and saddleback, not all of which might be a good idea to reintroduce. Lizards initially identified as candidates included spotted skink, speckled skink, Duvaucel’s gecko, striped gecko, and Marborough green gecko. Tuatara and frogs might also be expected to have occurred there. Missing invertebrates could include Cook Strait giant weta and ngaio weevil.

I hope that doesn’t read like a shopping list to establish yet another zoo masquerading as an ecological restoration. The emphasis is on commoner species which would have been present on the island, not some hodge-podge of threatened species based on the “I want” attitude, to which DOC appears to acquiesce all too often.

It’s worth noting that our goal is also consistent with DOC’s objectives for the region, as expressed in their Nelson-Marlborough and Island management strategies. It also fits in with the Department’s aims for management of individual species.

Although we did not look seriously at Level 4, a refuge for acutely threatened species, the island would probably be able to support small numbers of, say, kiwi, takahe or orange-fronted parakeet. We figured however that there were better opportunities to provide a long-term secure site for such animals. For example, although several pairs of takahe would thrive on the pasture lands, we see the grass reverting to forest over the next few decades, thereby displacing takahe. We did see a role for Puangiangi as emergency accommodation though. We made DOC aware of this and indeed the recent discovery of mice on Maud/Te Hoiere meant the island hosted some refugees from there.

We think we can get rid of the weeds and do all of the species translocations in ten years. The other tasks, not requiring large one-off efforts but more steady background work like maintaining the predator-free state, attracting seabirds, catching weka, keeping up biosecurity and fire measures, engaging with the owners and administrators of the other islands in the group, maintaining the infrastructure on the island, reviewing the management plan and keeping records of progress, will carry on, to state the obvious.

Forests, Shrublands, Grass and Cliffs

When Peter, Sue and I visited Puangiangi to look at it before buying, we could see the expected forested areas and pasture from the boat, but the landing area was inauspicious. A few big pines, brush wattle, tree lucerne, kikuyu grass and a pohutukawa, weeds or out of their natural range, were near the track from the landing bay. On reaching the edge of the first forest remnant I saw it was eaten out underneath by sheep and contained, groan, what I thought were some olive trees.

Once I’d had a chance to think and observe properly, however, I realised they were adult fierce lancewood, Pseudopanax ferox, something I’d only seen as the distinct juvenile form in gardens. This rare plant was everywhere and the project started to look like not just a restoration but also one where special plants occurred and could be protected. Things continued to improve when we found the forest sheltering the house to be reasonably sound coastal broadleaf forest, with a canopy of kohekohe, hinau, titoki, kaikomako, nikau, P. ferox and another rarity, large-leaved milk tree, Streblus banksii.

Juvenile Pseudopanax ferox, with fivefinger and Coprosma propinqua, D’Urville as a backdrop

That day we also had a quick look at the farmed area, where rank grass was being slowly invaded by tauhinu, with not a broadleaved seedling in sight, and at a manuka remnant on one of the high points. A circuit in the boat before heading back to French Pass revealed several forest areas, regenerating scrub, shrublands, swathes of wharariki/flax (Phormium cookianum), cliffs and screes.

One of our earliest priorities after taking over was to do a proper ecological and botanical survey. Geoff Walls, Barbara Mitcalfe and Chris Horne came out in October 2012 for a two-day survey which turned into five when we were caught by a storm. Geoff and his family own land on D’Urville and he is a very experienced ecologist, also having visited Puangiangi as early as 1981. Marlborough District Council part-funded his visit so he could do a Significant Natural Area assessment for them. Sue and I have been privileged to spend time on field trips with Wellington Botanical Society éminences grises Chris and Barbara. The three did the most comprehensive botanical survey yet on the island, putting up with the general topography, rain, having to crawl along ridgelines as the storm hit, and still having the energy to be witty and intelligent company back at the house. Barbara was undeterred despite being high up the waiting list for a knee replacement and Chris’ supply of filthy limericks was inexhaustible.

Chris and Barbara (note ever-present hand lens), October 2012

The team compiled a list of some 138 indigenous vascular plants and made a vegetation map. The island’s ecosystems comprise:

  • Coastal broadleafed forest, 15%;
  • Regenerating coastal broadleafed forest, 10%;
  • Mixed coastal shrublands with wharariki, 20%;
  • Pasture with tauhinu, 27%;
  • Kanuka and Manuka shrublands, 2%;
  • Salt turfs, 1%;
  • Cliffs and scarps with scattered vegetation, 25%.

(This is my slight simplification of the information in Geoff’s report and I have also  tweaked the area percentages to make them add up to 100 once again.)

The Broadleafed Forest

This forest is typical of many in the Cook Strait Ecological District and Wellington residents familiar with Otari/Wilton’s Bush would feel at home with the mix of common trees given above for the forest around the house (Telecom Bush). The predominant tree is kohekohe. Soon after possum and rat control began at Otari, Sue and I began finding strange black fruits on the ground. I had never seen a fallen kohekohe fruit, as possums ate nearly all the flowers and rats got any fruits that did develop. The kohekohe are a real feature of Otari now pest numbers are way down. Puangiangi has never had possums and no longer has rats, and it is such a pleasure to see the kohekohe trunks in early winter, covered with white flowers, to be followed by copious fruits in summer. The multi-stemmed trees have great character, hanging on on the steep and sometimes mobile slopes, their canopy shorn off by salty gales.


L-R: Kohekohe flowers, June; carpeting the ground; fruit, December


In the northeastern forest, which has stabilised screes, boulders and vertical slabs and must be near the slope limit to have vegetation cover, many of the kohekohe have stopped piles of rolling rocks on the uphill sides of their trunks. They have marched up to gain the ridge in a couple of places and hang on in the 150 km/h gales, making little oases of relative calm. They share the northeastern forest canopy mainly with akiraho (Olearia paniculata), a particularly indestructible giant tree daisy, and fivefinger (Pseudopanax arboreus), the only other Pseudopanax apart from P. ferox on the island, which is a little surprising given the species diversity elsewhere in the district. There are also some karaka, ngaio, Streblus, mahoe, tree broom (Carmichaelia odorata), wharangi, puka (Griselinia lucida) and kohuhu in the canopy. The puka are very large and terrestrial, and their grooved trunks/roots snaking up and down the rocks are a highlight. The kohuhu (Pittosporum tenuifolium) is another Cook Strait special, with its salt-resistant leaves being much more leathery than in the mainland form.

Under the canopy and in open spots, kiekie forms swathes which slow progress for those intrepid enough to explore the forest. The rockier areas have gardens of rengarenga lily, which of course also extend to the open cliffs. Lara Shepherd at Te Papa has been doing a DNA analysis of rengarenga from around the country, and a specimen from Puangiangi has the same provenance as others from the northern South Island area, likely deriving from early Maori plantings of material brought from the North. There are a few lianes, with Metrosideros perforata adding some colour to proceedings. The forest floor is populated with a typical array of ferns- Asplenium, Blechnum, Microsorum, Polystichum, Adiantum,and one of Sue’s favourites, Lastreopsis velutina. Also on some of the slabs is the little Peperomia urvilleana.

Kohekohe forest without an understory save for the unpalatable Polystichum (photo courtesy Geoff Walls), and vigorous regeneration in the prolonged absence of sheep (photo courtesy Roy Grose)

The other main areas of mature forest are Telecom Bush and a large area at the south of the island. Both are on slightly gentler ground, and Telecom Bush in particular was more affected by sheep. Now the sheep have gone, there is slow but gratifying regeneration of the understory. These forests have all of the canopy trees of the northeastern forest and add hinau, nikau, tree ferns and a few tawa. Peter and I have had a fair amount of time letting gravity take us round the southern forest, looking for sooty burrows, and we were pleased to find the tiny group of adult tawa that eluded the botanists. A brief survey in 1996 by Nelson Botanical Society had however recorded a seedling tawa, so we had been hopeful of finding its parents. Summer 2016 brings several tawa seedlings to the forest with regular though light rain helping them hang on. One has even popped up in Telecom Bush and we think the kereru may also be bringing the seeds from Tinui where there is an extensive grove.

L: Nikau inflorescence coming out, last year’s fruit behind; R: Metrosideros perforata; both photos January.

Apart from the areas of mature forest, there is substantial regeneration afoot, especially between the southern  forest and the central pasture lands, and on the southwestern slopes of the island high point down to Boatshed Bay (part of the area covenanted with QEII National Trust by Ross Webber in 1999 and retired from grazing then). These areas are dominated by fivefinger, akiraho, ngaio, mahoe, kawakawa, kohuhu. P. ferox, Coprosma propinqua, C. robusta and the hybrid between the two coprosmas. C. propinqua, tree broom, mahoe, mingimingi and tauhinu play host to the mistletoe Ileostylus micranthus, which is quite uncommon on the mainland because of selective possum browsing among other insults. It has also colonised tree lucerne, introduced around the house and slowly spreading elsewhere. On the non-native, leguminous host the Ileostylus can reach 3 m across in a lush, dark green as opposed to the often more “natural” pale yellow-green on the native hosts. Most of the forest trees are beginning to make an appearance, with the first kohekohe having just shown up in the covenanted area. The regenerating forest is extraordinary in its diversity and productivity, the abundant fruiting of the fivefinger being of special note to us and I expect the birds. It has come back quite quickly as the accompanying photos from 1981 and 2012 show.

Looking south to north, 1981 (above) and 2016 (below). Broadleaf forest regenerating at right. At far left, pasture finally retired in 2013 reverts to tauhinu. 1981 photo courtesy Geoff Walls.

Barbara and Chris are strongly of the view that this regeneration neither needs nor wants human assistance. Their philosophy is perhaps derived from the project at Hinewai, where nature is left to take its course no matter how long or short that might be, and they certainly don’t recommend the more extreme replanting interventions which Chris recently likened in a Letter To The Editor to little more than “gardening”. I share their view for the Puangiangi project, where unassisted regeneration is doing very nicely thank you. We might however temper this with a little bit of moving of kohekohe seedlings around the edges of Telecom Bush, which have been planted with shelter trees native and exotic and gardened in the past, and which might benefit from a bit of evening up of the species present.

The Stable Shrublands and Salt Turfs


The northern end of Puangiangi and some of the other ridgelines are a delight. Here the winds are too extreme for the broadleaved forest to take over any time soon, and the land is covered with a waist-high biodiversity treasure trove.

Clockwise from top left: Cook Strait kowhai flower, August; kowhai in situ (photo courtesy Geoff Walls); Melicytus sp.

The Cook Strait endemics shine, with three little mahoe relatives to the fore, Melicytus aff. obovatus “Cook Strait”, M. aff. novaezelandiae “Maritime” and a hybrid between them (identifications courtesy of Peter de Lange from specimens provided by Barbara). Here we also find P. ferox, Coprosma propinqua, taupata, ngaio, prickly mingimingi, wharariki, and clumps of dark green which we were excited to find was Cook Strait kowhai. It not only grows in the tangle of the more mature shrublands, but dominates some of the cliffs of harder rock down to within 20 m of the sea. Last winter I took the kayak out in the late afternoon and I could look up and see the kowhai flowers glowing yellow on the cliffs. If I had to pick one botanical standout on the island it would be these shrublands with their thriving populations of rare Strait endemics.

At and around spot height 120 m in the middle of the island is an area of manuka. Outliers have colonised the surrounding pasture on the western side down to near sea level. This part of the island is of gentler contour, being merely steep rather than precipitous. It would once have held the broadleaved forest we see elsewhere today, perhaps supplemented by a few missing trees such as matai and miro which occur on some other islands in the area. The forest cover might have helped the relatively large catchment there to have held water nearly permanently instead of ephemerally as today. The number of young manuka popping up and the relative absence of competing broadleaf seedlings (which were eaten by the sheep) might mean that the area of manuka shrubland expands and dominates the central third of the island for many decades. We need only look across the channel to Tinui to see how effectively manuka can come back on retired pasture.

Tinui from summit of Puangiangi, 1981 (above) and 2016 (below). Tinui was stocked and periodically fired until maybe 1960s, now clothed in manuka. Note succession from tauhinu to fivefinger and P. ferox on Puangiangi. 1981 photo courtesy Geoff Walls.

An area of kanuka dominates the 128 m summit of Mt Mistaken above the southern broadleaf forest and is spreading down the western slopes. Some of the kanuka at the very edge of the forest are large and old, and their copious litter makes great robin territory. The lower plants on the very exposed areas are festooned with the rare mistletoe Korthalsella salicornioides, its stems blending very well with the kanuka foliage. I was aware that it might be around, but still managed to walk past the main colony about eight times before Chris and Barbara spotted it on their first outing. It’s likely that kanuka will be the dominant species here for a very long time.

Kanuka spreads out, Mt Mistaken. Korthalsella salicorniodes at right.

Another rare plant found in the shrublands and open spots is wind grass, Anamanthele lessoniana. It was also obviously a favourite of Ross’, as he planted it around the house for low shelter and ornament, with its lovely showing of the wind through its blades and its seed-bearing tillers up to 2 m long. It’s fair to say that Puangiangi is also a national stronghold for Anamanthele.

Salt turf assemblage. Photo courtesy Geoff Walls.

At the southwestern corner of the island is an area of salt turf relatively unmodified by sheep. If Ross Webber or earlier farmers had ever had cattle it might have been lost. Here in the spray zone is a big area of silver tussock, native iceplant, glasswort, Samolus repens, Selliera radicans, Senecio lautus, and the two native daphnes Pimelea prostrata and P. urvilleana. These plants also occur sporadically around the shore and on open ridges. A monocultural salt turf of the little buttercup Ranunculus acaulis is at Woolshed Bay, by an archaeological site containing argillite, human-transported stones, charcoal and small bones. This area is more degraded by heavier grazing.

Cliffs and Scarps

The eastern screes and cliffs. Photo courtesy Geoff Walls

Most of Puangiangi slopes very steeply away into the sea. Although Ross told me he used to muster one of the very steep eastern faces on foot, we confine our interaction to views through binoculars or at the end of a rope drilling and poisoning some of the more easily accessible coniferous weeds.

The mobile scarps of scree and softer rock may once have held a forest of sorts, but at the moment they are limited to a few patches of grey scrub. From below, the largest scree looks devastated, the shoreline littered with boulders which appear recently arrived and a more or less constant stain of sand and mud in the sea. On a gentler bit, an island of forest hangs on- or has colonised, I don’t know which.

Wharariki in the salt exposure zone

Where slope, soil or aspect allow, the steeplands in the spray zone have extensive fields of low vegetation, dominated by wharariki (Phormium cookianum), silver tussock and with patches of rengarenga lily and linen flax (Linum monogynum). The latter two add a splash of white during flowering in spring. The shrubland species described earlier also make inroads. There are a few areas of hard rock cliffs, sometimes hosting rengarenga and Cook Strait kowhai.


Claire Gaze sets off for the butcher’s, January 2013. Photo courtesy Peter Gaze

The central third of the island has gentler slopes and more shelter and it’s logical to see the influences of past farming lingering there the longest. When we took over we were nonplussed to find a flock of some 20 sheep, left to their own devices for some years. It was obvious from a restoration point of view that they had to go, but having established that they weren’t suffering through disease burdens, lack of water or sweltering in their enormous fleeces (which they were shedding), we decided to shoot them at the rate we could eat them. Sue’s family and our friend Jan from Toronto would be among the few to say they have had a Christmas dinner of the now-extinct Puangiangi sheep. Some of the older rams took a bit of cooking but I think it showed respect not to waste them. I also sent up a choicer lamb cut to Ross in Auckland.

By 2012 the pasture had become rank even with the sheep that were there. Tauhinu was all but the only coloniser, the sheep leaving it alone in favour of any broadleaf seedlings that came up. I bought from Television New Zealand some archival footage of a visit one of their journalists paid Ross as he was preparing to leave in 2004. Even at that stage he had the journalist out grubbing tauhinu.

Pasture at Woolshed Bay reverting to tauhinu from 2012 (above) to 2016 (below), in the face of extreme winds funneling up the bay. Note demise of macrocarpa at right.

By the time we arrived, the tauhinu was extensive but we could still walk anywhere in near enough a straight line. Now it is all but impenetrable in many places, manuka is spreading from the high point and broadleaf saplings are showing in the shelter of the tauhinu. We recently cleared some grass and tree lucerne at the edge of Telecom Bush and were surprised to find healthy kohekohe seedlings under the rank grass.


At the start of this article I implied an initial negative impression of the island’s botany, formed by a weedy landing but since fixed by exploration and experts. Really though, we are pretty lucky having only the weeds we do. There is no gorse, Darwin’s barberry, hakea, climbing asparagus, old man’s beard, and so on. The ones we do have are isolated and controllable, and this influenced our decision to totally eradicate them apart from tree lucerne, pasture weeds (which will eventually die out as the forest grows) and probably kikuyu grass which is stabilising land at the two beaches for the time being.

Macrocarpas colonise the eastern cliffs, 2012. If left they would eventually take out the Cook strait kowhai in the central background. Photo courtesy Peter Gaze.

A shelter belt of macrocarpa protected the country’s smallest woolshed (3 x 2 m) and its yards, dating we think from the King-Turners’ 1929-1957 tenure. The shelter belt itself was benign but seed was blowing on to the adjacent rock pavements and cliffs, and some sizeable trees had taken hold. Lest the island end up like coastal northern California, we engaged Marlborough Sounds Restoration Trust to spray the macrocarpas from the air with metsulfuron methyl, as they have been doing successfully with pines in the Sounds for some time now. We drilled and poisoned the accessible trees, one of which got its own back by blowing over and flattening the woolshed. Some of the trees are not dead yet after two visits from the helicopter and there are plenty of seedlings coming up, but we will persist. The house area held several large pines, presumably for shelter and firewood, and they have also been poisoned. Many have blown over now and some took out the access track when they did so. We have cut a hundred or so saplings from the covenant area, where the seeds appear to blow and take hold, and we are seeing only a very few now.

A single pohutukawa, presumably an ornamental planting, and an isolated pocket of European broom have also been dealt with, apparently without any ongoing seedling germination. An infestation of brush wattle around the house was cut down. The main area of wattle has now been taken over by regenerating shrubs and it is too shady there for any wattle seedlings. Further down-slope, the long-lived seeds mean regular attention is required to pull out seedlings, a thousand at a time if this month is anything to go by.


All said and done, this is a botanical restoration that is proceeding very quickly with minimal effort from people, and it is reasonable to think that Puangiangi will be weed-free, and have at least developing forest over the former pasture lands, in my lifetime. The broadleaf forest will have a resilient structure with a diverse understory, and the island will become more widely known as a national stronghold for its astonishing range of rare endemics.




Fifteen years ago I was very ignorant indeed about seabirds. When Sue and I helped with our first of I think 11 chick translocations, the first thing I learnt was that seabirds were not all just “seagulls’. Eventually I came to learn from the likes of Colin Miskelly and Helen Gummer that seabirds really are and ought to be pre-eminent in the New Zealand story.

New Zealand is still the centre for seabird endemism: 38 species breed nowhere else and 86 species breed here in total, out of 360 species worldwide. Populations-wise, however, they would once have been a much more dominant feature, before they were all but extirpated from the mainland by predators such as rats, stoats and pigs. Now we think of seabirds as perhaps belonging on rock stacks and little islands, despite a few happy mainland exceptions such as Hutton’s shearwater. I suppose this is yet another example of the shifting baseline idea expounded so well by George Monbiot in his book Feral: in our lifetimes we have thought of seabirds as breeding on islands, sometimes scarce; therefore it was always thus. In the recent past though, there would have been clouds of all sorts of seabirds, circling mainland high points, their nesting and burrowing and bringing food in from the sea and being eaten and dying and decomposing, having the biggest effect of anything on the ecology of the mainland.

We can see what it would once have been like by looking at their island strongholds today- vegetation modified by fertility changes, shrubs destabilised by burrowing, a profusion of life making use of their burrows and their leftovers. It’s no accident that there’s 300 mm of fertile topsoil across the formerly farmed piece of peneplain that is Mana Island, making it easy digging for today’s revegetation teams. A hundred thousand years of seabirds can do that. Ross Webber wasn’t confronted with barren rock on Puangiangi when he established his vege garden and orchard, rather a deep, free-draining, moisture-retentive soil.

Now, I see twenty-something Miss Canterbury on Petone beach (she is a red-billed gull banded red-black at her natal site in Kaikoura on the day of a big Ranfurly shield challenge by a naturalist with a sense of humour), resplendent in perfect plumage on a sunny winter’s day, standing aloof from the lesser of her species who eat chips, and reflect that even “common red-bills” have moved into the Vulnerable category now they have so few breeding sites left. And I’m well aware that there aren’t any petrel colonies on the windy hills around my home in Wellington, the last one possibly having gone about 60 years ago.

The present mainland colonies, natural and translocated, are often reliant on expensive predator management, including fences. We have an opportunity with the Puangiangi project to intervene inexpensively to create another seabird island, with its natural fence.

Puangiangi As a Seabird Island

There’s no doubt that Puangiangi would once have been a smaller version of Takapouwera/ Stephens Island, which even today holds two million fairy prions to name but one species. It was almost not even a discussion point to make seabirds the focus of our restoration efforts, but that decision does come at a cost and does limit what else we can do. Take weka (please): let’s say there were 500 weka on Puangiangi twenty thousand years ago, and 5 million of a mixed mega-colony of fairy prions, fluttering shearwaters, diving petrels, white-faced storm petrels, flesh-footed shearwaters and sooty shearwaters, with sundry outcroppings of half-a-dozen shag species, gannets, red-bills, black-billed gulls, terns, mollymawks, little blue penguins. The 500 weka would be very well-fed through grabbing unattended eggs, nestlings, weakened adults, but they wouldn’t make a huge difference to a population of 5 million birds. Contrast that with the situation in 2012. I estimated the population of weka on Puangiangi as 20 (that was a gross underestimate as it turned out, but 20 will suffice here). Let’s say we were able to re-establish  a colony of 20 pairs of sooty shearwater. If we left the weka there, it’s likely they would get many or all of the chicks each season. Weka are inquisitive with a good sense of smell and are constantly patrolling for new opportunities. Such a seabird colony would not go unnoticed.

Weka at sooty burrow, April 2014. It can just be seen at the end emerging with a piece of down.

If the burrow-nesting petrels lose an egg or a chick, there is no time to re-nest, and of course they can only afford the energy input of one per clutch. Sooties take several years before reaching breeding age also. Therefore, more or less any weka predation would send a nascent colony spiralling towards functional extinction, with only a few pairs of unsuccessfully breeding adults hanging on for any time. So if we were to have a go at re-creating Puangiangi as a seabird island, the weka would have to go. Weka are worthy of protection in their own right and ironically we have funded a project to re-introduce weka to an area of the upper south, so it’s not actually a no-brainer to choose seabirds over weka. However, there are more areas where weka are safe than there are islands where seabirds are safe, so we chose seabirds. Peter was able to secure a permit to catch weka and relocate them to the mainland, so the project was under way from our point of view.


Burrow-nesting seabirds were holding on on Puangiangi and on Tinui’s southern satellite Takawhero into the 1980s. Ross Webber was well aware of them, and indeed had found a chick and kept it at his house for a short time, trying to feed it on bacon scraps. Visiting field workers, including Peter, had found remnant sooty colonies in the southern and north-eastern forests and Peter brought out a predator trap for Ross to deploy in the southern colony. We found the remnants of the trap during May 2012. But, as the rats slowly got on top of Ross, they slowly got on top of the sooties. It’s highly predictable that sooties would be the last burrow-nester there: they are bigger and feistier than the fairy prions, fluttering shearwaters and diving petrels that are generally more common in Puangiangi’s latitude, and would be a match for a rat or a weka coming down a burrow, leading with its head. That protection would extend to incubated eggs, but after the egg hatches and the adults leave the nestling for longer and longer periods to go as far as the Polar Front to gather food, then the insidious numbers game of predation begins. That game was over perhaps a hundred years ago or more for the smaller and comparatively defenceless petrels and prions, but sooties hung on longest, before being wiped out or driven away.

Takawhero, centre left, from Puangiangi. Sooties present in 1980s, current status unknown.

Ross was pretty clear in conversation with Sue and me that he considered the sooties to be present to this day, but Peter and I looked for the old southern colony during what would be peak fledging week in May 2012. We should have found the old burrows to be clean, with squirts of bright white droppings away from the entrances, clean beaten ground from all the comings and goings and gardening, and a distinctive smell to the burrows. Instead we found cobwebby burrows, drifts of leaves blocking others, penguins in some others and weka following us around. We were to be starting from scratch, but of course the island was at least rat-free thanks to the efforts of 1999 described in another article.

So, let battle commence. It’s an important job every trip to catch as many weka as we can fit into the pen Dad built for them in the shed. Once in the pen, they get to forage for insects and worms in the leaf litter we bring in, and eat specially prepared meals and our leftovers- cheese, pasta, eggs, blue cod frames. In reality they mainly occupy themselves knocking over their water, trying to escape and staring balefully at us when they jump on to the top of the sacking we have hung as little refuges for the more timid weka to hide from their more aggressive brethren. We try to limit catching to near the end of a trip so we hold them for as short a time as possible, and they are largely very resilient. To start with, we transported them in cardboard or Corflute carry boxes, but after a near miss with one in its box almost being blown over the side of Aston’s barge, and the fact they just would not settle in the boxes, scrabbling around the whole trip into Garne and Saville Scenic Reserve, Claire made up some heavy canvas bags with tie tops. Like most birds, they settle well in the dark of the bags, even travelling two-up without incident if we run out of bags. At the release point they are tipped out of the bags and they orient themselves and run off impressively. We removed the 100th weka early in December 2015 and several times have thought we have gotten the last one. I’m still aware of a very few as I write though.

At the start of this article, I said we have helped with some chick translocations. These are intricate processes developed here by our endangered seabird scientists (ambiguity deliberate).  Almost every individual of each species of burrow-nesting seabird returns to nest in its natal colony, often coming back after however many years they spend at sea before beginning to breed, to within a few metres of where they were born. This fixing on their natal colony is thought to be learned during the nights (1-20 or more depending on the species) they emerge from their burrows to exercise wings and shed down and weight before they fledge. Only through over-population at a colony (unlikely these days) or in rare individuals with inclinations for wandering beyond the norm, or through some not-yet-understood at-sea interaction with others, will dispersal occur. The simple act of grabbing an adult and putting it in a hole somewhere else will in no way over-ride this loyalty to the natal site. People have tried- the adults fly back. However, moving chicks before they develop site loyalty, and feeding them until they fledge, might induce those birds to return as newly minted breeders to the new colony.

Chick translocations have been done with many species of petrels in New Zealand, many successfully. People like Brian Bell and Graeme Taylor practised on common species first, refining techniques so that endangered species like taiko and Chatham petrel could be tackled with justified optimism. Realising that canned sardines (whizzed up with water and fed by syringe) were a transportable, non-perishable means of approximating the chicks’ natural diet, expanded the type and remoteness of site which could be considered for the technique. People with attention to detail, like Helen Gummer, came to tailor the feeding regime individual by individual- not just sticking to a species template- by careful weighing and  measuring in the field as each project ran its course. Building up the numbers of trained people, such as Sue, means that chick translocation is now mainstream thinking for ecological restoration projects where common petrels are the target, and for work with the endangered ones too.

Sue feeds Chatham petrel chick, Sweetwater Conservation Covenant.

The typical three-year cycle of translocations is costing around $60,000 plus whatever notional price one wants to put on the volunteer chick-feeding labour. The maximum number of chicks that can be handled at a new site is about 240, with a target return rate of about 30% after however many years it takes for the species to reach breeding age. An analysis of the completed translocations around the country, however, indicates that many sites are going to need top-ups to give the new populations a statistically good chance of becoming self-sustaining. Funders and community groups should be thinking of a decent six-figure sum and bulk volunteer wrangling for a completed translocation project.  Puangiangi’s remoteness and difficulty of access and getting around suggests that chick translocation would be marginally achievable at best. Also, the community engagement plus of having large numbers of volunteers involved is not relevant for us. We have initially plumped for an alternative:

Speaker systems to attract seabirds are now a reality around the country. We’re up to our seventh I think, installed on the Kermadecs, Wakaterepapanui in the Rangitoto group, and now Puangiangi. A solar panel charges a hefty battery and, at night or in the daytime depending on the species being targeted, the solid-state electronics inside a waterproof case play the tracks on a sound card through 350-watt outdoor speakers. The systems now cost under $2500 and they require little, although not zero, maintenance. If they work, they attract adults, obviating waiting out the 2-8 years for translocated chicks to return as breeders.  We thought that Puangiangi would be a good candidate for the speaker method, what with a sooty colony not long gone with perhaps a few adults still alive that were born there, and large colonies of fairy prions to the north on Takapourewa, some of which might be enticed south.

The first site chosen was obviously to be near the bigger of the old sooty colonies, and we put up the speakers at the upper edge of the southern forest. It was a long, hot walk to carry in the heavy and awkward gear from the shed and we were grateful for the help of Roy Grose and Chris Birmingham from DOC, who were visiting us to catch up with Peter and to eyeball me. Sue and I were a bit worried that good intentions by DOC to assist us with the project might become subsumed by other priorities over time, but these immensely capable people gave me great confidence from the beginning of this project. Chris was at the time working nearby on Maud Island but has regrettably left for Te Anau to a dream takahe job along with his partner Linda Kilduff and new baby. Roy is sticking it out through the endless restructuring of DOC and still works in the region albeit with ever larger responsibilities.

Peter and I put in some starter burrows in the grass around the speakers and turned the system on, the sound card carrying recordings of sooty shearwater, fluttering shearwater, fairy prion and diving petrel. We gave visiting botanist Barbara Mitcalfe a demonstration and she thought it an awful cacophony, but it seemed to work for the birds. We also rigged up a trail camera and hooked it up to the cellular network through a big Yagi aerial fixed high up in a nearby akiraho. At 1 AM on 3 December 2012, a month after the system went in, I got the first photo (at the head of this article) sent to my phone back in Wellington, of a sooty shearwater on the ground.

The site saw quite a lot of activity. During March-April 2013 for example, the camera recorded birds on the ground on 15 of 56 nights monitored. We identified sooty shearwater, fluttering shearwater (two on the ground simultaneously several times) and probably diving petrel.

Trail camera stills showing sooty shearwater, fluttering shearwater and diving petrel at the speaker site, 2012-13

One strange observation was a morepork/ruru having a go at a small petrel on the ground next to the speaker system. Did we encourage unusual behaviour or are ruru expected to be predators at seabird colonies?

Ruru attacks petrel on the ground, April 2013.

On the strength of these results we got another two sound systems. One was installed at lower altitude than the active one, at the edge of a tussock/ salt turf area which looks ideal for smaller petrels to dig burrows. The best we managed in two years there was some feathers and, separately, half a diving petrel. There was no sign at all of burrow construction and we have since shifted the speakers to above Boatshed Bay and put in a gannet tape.

A third system has been put in at the north end, within earshot of any of the millions of petrels on Takapourewa that might fly a few km south when  foraging. Cameras showed a good rate of visitation by fluttering shearwater, and we promptly added some shearwater decoys to help things along. The initial fluttering shearwater model was beautifully made for free by Richard de Hamel and then a two-piece fibreglass mould taken from that. Peter spent wet evenings in the shed on Puangiangi making resin shearwater halves, which were then glued together and painted. They look great, but to date neither they nor the speakers have enticed anything to occupy burrows at the site.

Peter installs the decoys at northern speakers, January 2014.

In early December 2013 we returned to the original speaker site and found breeding sooty shearwater. There was a newly-refurbished look to the old natural burrows, and we found a couple of occupying birds at great cost to Peter’s arm. We installed some trail cameras on likely burrows.

A pair of sooties at their burrow entrance. Egg-laying season December 2013.

In January we went back with a burrowscope, courtesy of Nicky Nelson at Victoria University, and found 16 incubating birds.

After the chick hatches, the parents alternate between incubating and heading to sea to get food for themselves and their chick.

Adult sooty arrives presumably with food, March 2014.

As it grows, the chick is left by itself for longer periods. At the colony we recorded the longest time between feeds as 10 days.

First emergence, April 2014. This chick obviously survived the weka intrusion shown near the beginning of this article.

A few minutes later- had enough. Chick at maximum weight, fortunately.

The number of active burrows dwindled away during summer and autumn though, and three chicks got to the stage where they came to the surface at night-time, exercised their wings, shed down, lost weight and eventually took off in early May.

Exercising wings on a rainy night, April 2014. Most down has been shed. (The date on the camera was wrong).

Helen Gummer points out that this might be the first successful example of acoustic attraction of sooty shearwater which have gone on to (re-)found a breeding colony. The colony looked forlorn during 2012, and we first noticed adult sooties in December 2012, a month after the speaker system was installed. The burrows were obviously active the following season (beaten earth, smell, clear burrows and so on) and I don’t think I would have been so stupid as to miss such signs in 2012.  In possible support of this claim, we also found an old sooty burrow grouping elsewhere on the island away from any speakers, and there has been no activity there, despite weka pressure all but vanishing island-wide.

The 2014-15 season gave a similar low fledging rate from about as many nesting attempts. We can only speculate on the reason- inexperienced breeding birds seems likely, as they should be the ones pulled in by the speaker system. An undetected predatory animal on the island would surely have left a mess, which we did not ever see. At-sea factors may have also influenced the fledging rate; poor results have been seen elsewhere when possible changes in sea temperatures, say, have caused food supplies to dwindle or move to different areas.

As the 2015-16 season starts we can report 16 birds on eggs, from a total of 28 GPS-mapped burrows which have held birds at some point over the last three years.

We plan on disturbing the colony as little as possible and are not intending to band the chicks or the adults. We do want to know the fledging rates each year and to get an idea of the population trend. I guess we are also not completely certain that they are all sooties. Many of the larger petrels seen at sea nearby are flesh-footed shearwater, which breed at the same time, so a mixed colony may be possible. Either will do just fine. They are easily told apart but burrowscoping is limited to confirming that a sooty-sized petrel and not a penguin is in the hole.

Other Seabirds

Red-billed gulls live in the area seasonally, feeding on schools of small fish and concentrations of crustaceans. Once we saw a little bay on the eastern side with red-bills all but walking over one another to feed on some crustaceans which had filled the bay. Red-bills don’t breed on Puangiangi and we don’t have any idea which breeding colony they come from. They have contributed to the nesting success of the resident pair of karearea though. Their 2014 nest was littered with the beaks and feet of red-bills, the only pieces they didn’t seem to have eaten. Red-bills are not far short of the size of the smaller male karearea and if they had caught them close to the sea, it must have been a feat to carry them to the nest site at about 60 m altitude.

Black-backed gulls are quite common. A pair of reef heron is often seen, as are white-fronted tern, the occasional Caspian, and various shags such as king, little, pied and spotted. There is a thriving breeding colony of spotted shag, spread out from the northern tip of the island and along the eastern side, nesting adults and young being seen from the sea in crevices in the cliffs.

Little penguins seem to be common, and they are encountered at the sooty colony and all around the island. They also nest under various man-made structures such as Ross’ house and one of his old pottery kilns, which is housed in a converted mussel float.

None of these birds seems to need human interference, but we are trying to establish a breeding colony of gannets. Their regional breeding colonies are thought to be full and thriving and they are often seen fishing in the area. A grouping of generic fibreglass boobies, sourced commercially and painted up in Australasian gannet colours by Peter, has been set up above Boatshed Bay, on what we hope is an attractive, clear site. A speaker system broadcasts the sounds of a gannet colony from dawn to dusk, and indeed we did seem to have some early success. Peter was checking the speakers and I was looking down from the spur, frantically signalling to Peter that he should look behind him. A gannet was there and unconcerned at being a metre behind Peter. We took it as a sign that the decoys were good ones, on two counts. A gannet was also filmed staying overnight but there has been no nesting so far.

Not a bad likeness. Spot the real one.



In August 1903 eminent naturalist Henry Travers wrote to the Colonial Secretary Sir Joseph Ward about “Reserves for Scenery and Native Birds, etc”. Travers had seen first-hand the formerly abundant birdlife on  Takapourewa (Stephens Island). But “a year or so later cats were introduced; I was there a few weeks ago, and there is not a native land bird left on the island in consequence of the depredation of the cats”. Of course the loss of birds included the extinction of the Stephens Island wren.

Takapourewa is in the middle distance to the north of Puangiangi and the Rangitoto group. Henry Travers went on to say “I also venture to suggest, that a number of the Tuatara lizards (from Takapourewa) be placed on the two Rangitoto islets on which there is bush; this would prevent any possible chance of their becoming extinct in this portion of the Colony”.

Document courtesy Dr Colin Miskelly, Curator of Terrestrial Vertebrates, Te Papa

Travers showed insight into predator-prey interactions at the turn of the 20th Century or earlier, and advocated removal of pigs, goats and cats from Kapiti before it was used as a bird sanctuary. His suggestion for the Rangitotos to be a tuatara sanctuary might have been forlorn though, as they were certainly already carrying rats.

Rodents and mustelids and possums took a long time to be recognised as pre-eminent causes of decline and extinction of New Zealand’s native birds, lizards and invertebrates. Elsewhere in the world habitat loss is often key. There, birds and mammals for example have co-existed for millions of years and the birds have evolved to cope with mammalian predation. Not so on our small piece of Gondwana, yet even today those only peripherally interested in conservation, and those with agendas that it suits, stick to the notion that things will be all right if only the forests are protected from destruction.

Ninety-five years after Travers speculated that the Rangitotos might become a sanctuary for tuatara, Nicky Nelson from Victoria University and Peter Gaze were on Takapourewa. Nicky was doing a PhD on the induction of eggs from gravid tuatara and their incubation back in the lab, to better understand the role of temperature in determining the sex of the hatchlings.

Nicky and Peter discussed how the 400 young tuatara resulting from Nicky’s work could best be used to establish a new wild population. She pointed (quite possibly from the same spot as Henry Travers) to Wakaterepapanui, the northernmost of the Rangitotos some 10 km to the south, remarking on what ideal habitat it would be. The problem was that Ecology Division of DSIR had found that Norway rats were present on all three islands and Tim Markwell (Victoria University) had more recently confirmed the presence of kiore on Wakaterepapanui. These rats would prey on tuatara eggs and young as well as out-competing them for invertebrate food.

Norway rats and kiore do not co-exist easily and it is tempting to think that kiore were once widespread on all three islands, but that Norway rats were introduced to Tinui after European colonisation and gradually swam to the northern islands, replacing kiore in the process.  Were the few surviving kiore on Wakaterepapanui evidence of more recent colonisation there by the Norway rat? Possums and ship rats were fortunately absent, and there have been only two records of single stoats from Puangiangi. Stoats, present on D’Urville and within theoretical swimming distance, are efficient predators of Norway rats, and the presence of the latter therefore indicated that stoat invasion was rare and that they had not bred on the islands.

If the potential of Wakaterepapanui to house tuatara was to be realised then the rats had to go, and from all three islands to prevent re-invasion one to the other. This was feasible given the wins already had with aerial brodifacoum by the late 90s, but it was relatively expensive. The consent of three different landowners would also be needed. Wakaterepapanui is publically owned and managed by the Department of Conservation. Puangiangi in the middle was owned and occupied by Ross Webber and the southernmost, Tinui and its islet Takawhero, had a multitude of Maori owners but was by then uninhabited.

Victoria University took the initiative and got a grant from the South Pacific Conservation and Development Fund, which used the money obtained from the French government after they had bombed the Rainbow Warrior and killed a crewman. With finances covered it now fell to Peter and his colleagues at DOC to obtain consent for the operation. Discussions with many of the individual owners of Tinui went well– the eradication of rats being consistent with a management plan previously prepared by Tony Whitaker for the owners of the island. It was not possible to meet with every owner but a senior kaumatua of Ngati Koata deemed that there was sufficient endorsement. Ross Webber was also keen to see the end of rats on his island. He had been trapping and shooting them, without measurable effect, and he was sick and tired of them getting into his stores. A complication was that the aerial application of brodifacoum was licensed for stock-free islands only and Ross still had a good flock of romneys on Puangiangi.

By early July 1999 a 50-page Assessment of Environmental Effects had been prepared which resulted in DOC approval for the operation, the Medical Officer of Health had been notified and the Marlborough District Council had given a non-notified resource consent. The bait was ordered. In August all of Ross’ sheep were barged to a neighbouring farm on D’Urville Island. This was not without some difficulty despite the services of shepherd Lionel Roughton and his dogs: these sheep just weren’t used to being rounded up with dogs, Ross’ last one having died some time previously. As the sheep needed to stay off the island until all bait had degraded, Peter arranged with the Rai Valley butcher to get a meat pack for Ross onto the weekly mail boat.

Ross Webber’s sheep head for D’Urville August 1999. Peter encourages lead sheep on board at the end of a sturdy rope.

After these months of preparation the actual operation, organised by Mike Aviss and DOC staff from Picton, was completed in a day – 24 August 1999. There was no non-toxic pre-feed. The bait [Pestoff 20R as 6 g pellets with a concentration of 0.02 g/kg brodifacoum] was flown from a barge moored at Tinui and applied in sixty-metre-wide swathes at 15 kg/ha. Bait was spread by hand on Takawhero to the southeast of Tinui.

Ready to spread the brodifacoum, 24 August 1999

There is good reason to believe the eradication was successful. Trapping soon after the operation failed to detect rats on any of the islands. The juvenile tuatara were then duly moved to Wakaterepapanui, along with green gecko and giant weta.

The helicopter finds a place to land on Wakaterepapanui; tuatara being unloaded. Photo courtesy Jo Monks.

When Puangiangi came on the market in 2012, we were not sure it had stayed free of stoats and rats, the two most likely invaders. We also didn’t know about the other islands as Wakaterepapanui is seldom visited and the bait stations on Tinui are not checked. Sue and I were really only considering Puangiangi amongst competing options for our next project because of the predator eradication, and I guess we would have walked away if our pre-purchase visit had shown they were back. Sue remarked that it really is just a useless, though beautiful, piece of rock if it can’t be kept all but free of pests.

When we landed that March, the proxy tracking tunnel on the firm golden sand at Boatshed Bay was free of little mammalian footprints and there was no sign around the buildings (rats would want for shelter more than they would want for food on a place like Puangiangi). Common gecko, which Peter remembered in only small numbers during the 1980s, were everywhere; darkling beetles and giant tree weta (at instars you just don’t see on the mainland) likewise. Our first night in the house in May was frightening though. The scratching and thumping of animals in the walls and ceiling made us think for a while that it must be riddled with rats, until we found it was a favourite gecko residence. Indeed when we opened up some walls for plumbing fittings, 200 mm of accumulated shed gecko skins and droppings were piled up on many of the dwangs. And there were no mummified rats anywhere.

In the future, stoats may occasionally swim from D’Urville, but we hope the swift currents will mean that any arrivals are single (and not pregnant). Any re-invasion of rats would likely be from boats. It was a calculated risk to install moorings on safety grounds, and we can only hope that no-one runs a line to the land when they are moored. There will never be a jetty and we will continue to wade ashore. Trespassers do beach boats and kayaks and this creates considerable risk. The No Landing signs should make it obvious what is required and locals have been asked to watch for illegal landings and to intervene as required. Dogs being let ashore from boats would also be a disaster as they would find some of the ground birds there irresistible.

After the project got under way in 2012, Peter set out a line of DOC200 traps at landing sites and in outbuildings, baited with Erayz and checked monthly. We have since extended the network to 17, with good coverage along the ridgelines and tracks which the animals would prefer to use. If we were never to do anything else on the island, keeping the traps intact and serviced would be the job that does get done.

The Telecom Hut

Just to the east of Ross Webber’s house is a little building made of lightweight metal panels, tied down with wire and with a locked door. At the rear is a telephone pole with climbing spikes, giving access to an old aerial and a solar panel from the 1980s. On our first visit to the island after purchasing it in the mortgagee sale, we brought out a locksmith who had driven over to French Pass from Nelson to meet us. He quickly gained access to the house and shed and installed new deadlocks, since beefed up with various other devices. We also encouraged him to cut a key for the little hut and I gather he got permission from the appropriate people as presumably the equipment inside was not owned by us.

Inside was a pile of grumpy geckos, several hundred metres of old cable, a deep-cycle battery and solar controller, an auxiliary telephone handset, a “Country Set” for receiving and transmitting over a microwave link to the mainland, and several old inspection forms filled out and left. On our pre-purchase visit we had been able to spot that there was a telephone in the house, by jumping up and grabbing brief looks through the salt-covered kitchen window (funny, we have never had the windows covered in salt since). I’d arranged with Telecom to get a subscription to start on Settlement Day, which was slightly complicated by the previous account having not been paid for some time. We were delighted to discover that the phone worked perfectly, if a little scratchily on the incoming side. People back on the mainland can hear very clearly however.

The Telecom hut is next to one of the kohekohe forest remnants, the only piece in the central western part of the island. We instantly dubbed the forest Telecom Bush, which also alludes to a remnant back in the Makara area of Wellington called Post Office Bush, which I have had the pleasure to visit with Wellington Botanical Society. Telecom Bush was good sheep shelter, and although no longer under pressure from the sheep, is not yet getting its understory back. It is however in the territory of one of the original robin pairs, and Peter found one of the first nests after translocation had been built in Telecom Bush.

Ross told us that he had had the telephone connected in the early 1980s, just before Telecom was privatised. For the cost of a city telephone connection, Ross got special equipment installed in one of the more remote places in New Zealand. I gather they may have given him some decommissioned lead-acid batteries and a solar panel too. (When we arrived there were two generations of solar panel on the shed roof, supplying electric light to part of the house. The 80’s-vintage one was the same as on the telephone pole but the connections had long rotted out. A more recent panel was charging the old batteries in the porch, and there was a little post-Ross electric water pump also to supply running cold water to the house. We had no idea how to maintain the system but it did limp through to the time we did a major electrical upgrade.)

Not everyone comes out on the right side of a deal with a major corporate like Telecom but I guess it was their remit to provide telephones regardless of whether an individual connection was economic for them. Imagine how useful it would have been for Ross to be able to keep up with family and friends, and to order supplies by telephone after 25 years of not having one, not to mention how much it would help if Ross were to suffer an injury or illness that needed outside intervention.

Gerard Hindmarsh wrote in one of his lovely articles about Ross that the phone was hidden in an alcove in the wall behind a calendar, lest passing yachties might sneak in an make toll calls at Ross’ expense. Ross says that story is “bloody nonsense”. There is a framed message board on the kitchen wall which might well have such an alcove behind it, but from the way the phone is installed over on the opposite wall it’s pretty obvious it has been there forever, with nothing which might have given cover from the freeloading yachting fraternity. Indeed Ross enquired whether we found useful his pencil drawing, on the wall next to the phone, of the set-up and location of buried cabling from the house through to the Telecom hut. Regrettably the drawing is most probably under a layer of lurid purple paint courtesy of the intermediate owner.

The line has non-existent bandwidth and would not support even dial-up internet, but we greatly value the security and connectivity it gives. We have since found that the island has reasonable Vodafone cell coverage, at least on the open headlands with a view to the tower at Bulwer. Telecom’s tower is a bit further away but we can sometimes text on the Telecom network from the high points. The corrugated iron roof of the house cuts the cell signal inside to zero, but we have bought a Yagi aerial and clamped it to a tall piece of pipe up behind the house, pointing it at Bulwer and occasionally trimming trees from the line of sight and making sure the aerial doesn’t spin round in the gales. My iPhone can be connected inside and so in addition to the “land” line we now also have good cell telephony and satisfactory internet (courtesy of the hot spot feature) in the warm and dry.

We were checking the equipment in the Telecom hut one day in 2013 and found a new inspection report dated that February. We had had no idea that a technician would come out regularly to check the equipment over. I was quite concerned about possible implications for biosecurity and so penned a lengthy note about the new status of the island and left it in the hut with a request to phone me. Next February I got a very friendly phone call from John the technician, who explained that he came out once a year by helicopter (when the bill is paid at least). I was able to tell him the latest of Ross, with whom he had got on well, and we established that there were no concerns about dogs, or rats hiding in equipment, as the tech serviced the public estate conservation islands as well and was well up with the special requirements for gear checking and so on.

(The tech works for a company which was spun off from either Telecom or one of the electricity companies. I lose track of all that and am tempted to refer to everyone as either The Post and Telegraph Office or the Electric Power and Gas Board, but shall use Telecom in this article and forget either their previous or subsequent incarnations.) Whilst it is hilarious that we receive a $1500-plus helicopter visit once a year to check our fifty-dollars-a-month phone line, I did say to John that he might want to coordinate with us and share a boat to the island. He explained that they used to go by boat but with the helicopter they can do eight visits a day compared with two previously.

The phone line was dead one visit in Spring 2014 and after experimentation Peter and I figured it might be a cable fault somewhere between the Telecom hut and the house. Out of natural love and affection for Telecom we thought we should have a go at fixing it rather than contact Faults and assert that the $4.95 monthly line maintenance fee might be spent on another helicopter trip. I fronted at the electrical wholesaler back in town, asked a bunch of dumb questions which were helpfully answered by the sales clerk and a customer, and returned to the island with 63 metres of the latest gel-filled cable, a new handset, connection box and the like. We hooked up the new cable in a temporary manner and from first principles were able to work out what the blinking lights on the Country Set meant, and that all the existing equipment was indeed OK apart from the old cable. At that point I realised that my pacing out of a metre needed recalibration and we were 4 metres short on new cable. Fortunately we were able to shake the geckos out of the enormous coil of spare cable in the hut- which I had clean forgotten about- and spend a hot afternoon putting it in a new trench.

This winter the phone gave trouble again and it was clear that the battery was not charging enough to work the transmitting and receiving equipment. This did mean contacting Faults and I crafted a careful email, explaining that the fault was in a remote location, was related to the solar panel and/or battery, and that there was no point sending me a standard reply inviting me to walk around the house, check all the extensions, unplug the ADSL line filters and whatever else they say to do before they will send out a technician. They duly sent me a reply inviting me to walk around the house, check all the extensions, unplug the ADSL line filters and whatever else they say to do before they will send out a technician. Fully expecting this I had copied the email to John the technician and eventually he or I was able to get the help desk to actually read my message and assign him to the job. We now have had another helicopter visit and sport a shiny new battery and a working phone once again.

I’m really happy to have and use this somewhat anachronistic piece of infrastructure. It is important to our safety and it reminds us of Ross’ time on Puangiangi and what a step forward it must have been for him to get the phone on. These days the kereru like to perch on the aerial of the Telecom hut, the guy wires moan in the wind, the geckos are largely undisturbed inside, and the old Country Set does its job.