Conservation status, threats and management options for the Open Bay Island skink (Oligosoma taumakae)

Abstract

The Open Bay Islands are a Māori-owned Wildlife Refuge located in South Westland (South Island) and support two of the rarest lizard species in New Zealand: the Open Bay Island skink and the Open Bay Island gecko. Both species have been assigned a threat status of Nationally Critical – the highest priority for conservation action. The terrestrial and diurnal habits of the skink make it particularly vulnerable to predation by wēkā, a flightless rail native to New Zealand but introduced to the Open Bay Islands c. 100 years ago. Here, we present results from a 5-day survey of the two main islands (14.7-ha Taumaka and 2.2-ha Popotai) and consider options for reducing the negative impacts of wēkā on skinks and other terrestrial fauna (wēkā removal, exclosure fencing and refuge supplementation; all subject to iwi approval). In total, 19 skinks were caught over 125 trap-days. The majority (16 individuals) were caught on Popotai Island, from which skinks had not been previously recorded. We conclude that removing wēkā from both islands is the only option guaranteed to increase skink abundance, with collateral benefits to other terrestrial fauna on the Open Bay Islands.

Keywords:

• Open Bay Islands
• conservation
• Oligosoma taumakae
• skink
• wēkā impacts

Introduction

New Zealand has a diverse endemic skink fauna currently comprising 33 extant described species plus several presumed new species that lack formal description (Hitchmough et al. 2007; Chapple et al. 2009). Skinks were recently proposed to have colonized New Zealand in the early Miocene (16–22.6 mya), arriving from New Caledonia via long-distance over-water dispersal (Chapple et al. 2009). Alternatively, skinks may have arrived from New Caledonia via presently-submerged land links (Towns et al. 1985). Following extensive and rapid speciation, New Zealand skinks came to occupy all major terrestrial habitats, with distinct open country, forest and coastal lineages (Chapple et al. 2009). Offshore islands in particular have become increasingly important sanctuaries for skinks, with many species now surviving only on islands free or cleared of introduced mammalian predators (Daugherty et al. 1994; Towns & Daugherty 1994; Towns & Ferreira 2001).

The Open Bay Islands (Taumaka and Popotai) in South Westland are home to two of the rarest lizard species in the New Zealand: the Open Bay Island skink (Oligosoma taumakae; Chapple & Patterson 2007) and the Open Bay Island gecko (Hoplodactylus sp. ‘Open Bay Islands gecko’; Hitchmough 1997; Hitchmough et al. 2007), an undescribed member of the forest gecko (H. granulatus) species complex. Both species were assigned a threat status of Nationally Critical (the highest priority for conservation action) under the national threat ranking system devised by the Department of Conservation (Hitchmough et al. 2007). This high threat ranking reflects their extreme rarity, relict distribution and on-going threat of predation by wēkā (Gallirallus australis), a native flightless rail that was introduced to the Open Bay Islands c. 100 years ago (Whitaker & Lyall 2004; Chapple & Patterson 2007).

Knowledge of the ecology and life history of the Open Bay Island skink is limited. Notably, the presence of the skink was not reported until 1986, despite earlier visits to the Open Bay Islands by numerous ecologists (Miskelly 1993). Lizard surveys conducted since have revealed only small numbers of animals (<10 skinks per survey), mostly from north-western Taumaka Island (Miskelly 1993; Tony Whitaker, Whitaker Consultants Limited, Motueka, pers. comm.). Here, we report on the findings of a recent (2008) lizard survey undertaken on the Open Bay Islands, including the discovery of a second population of skinks on Popotai Island, and outline options for managing the impacts of wēkā predation.

Materials and methods

Study site

The Open Bay Islands (43° 50′ S, 168° 53′ E) are located approximately 5 km off the coast near the mouth of the Okuru River in South Westland, and consist of two small islands (14.7-ha Taumaka [c. 630 m × 280 m] and 2.2-ha Popotai [c. 410 m × 180 m]; Information Management Unit, Department of Conservation) and several smaller barren islets and rocks (Burrows 1972). The islands are Māori-owned and designated as a Wildlife Refuge under the Wildlife Act 1953 (Hieatt & Newton in prep.). The faunal values of these islands include significant populations of New Zealand fur seal (Arctocephalus forsteri), Fiordland crested penguin (Eudyptes pachyrhynchus) and three Nationally Critical species not known from anywhere else – the Open Bay Island skink, Open Bay Island gecko and Open Bay Island leech (Hirudobdella antipodium) (Miller 1997; Whitaker & Lyall 2004; Chapple & Patterson 2007; Hitchmough et al. 2007; Hieatt & Newton in prep.). Rodents and other introduced mammalian predators are not known to have ever been present. Detailed descriptions of the birds, vegetation, soils and geology of the islands were given by Cockayne (1905), Mutch & McKellar (1954) and Burrows (1972).

Survey methods

All survey work was carried out from 10–14 March 2008. Methods used to locate and/or capture skinks were visual searches, pitfall trapping and intermittent use of a dog trained to indicate the presence of lizards. Although efforts were also made to search for Open Bay Island geckos, the rationale for focusing on the skink was that it is terrestrial and diurnal, and therefore presumably more vulnerable to predation by wēkā than the primarily arboreal and nocturnal gecko. Most of the interior of Taumaka Island is covered in 3–4 m-tall kiekie (Freycinetia banksii) that is largely impenetrable to human observers. Hence, the area that could effectively be searched and trapped for skinks was limited to the main track running along the length of the island, several smaller tracks accessing the north-western shoreline, the vegetated edge of this shoreline, and the area surrounding the hut and helicopter landing site. All previous skink captures were obtained from piles of wood and roofing materials stacked near the hut and from within the hut structure (Miskelly 1993; Chapple & Patterson 2007; Tony Whitaker pers. comm.). These stacks had been removed prior to our survey and could therefore not be checked. Wooden debris and corrugated iron remaining on the island was carefully lifted and checked for skinks.

On Taumaka Island, 31 pitfall traps (2 L paint tins; Resene Colourshop) were placed in accessible areas that appeared to provide good skink habitat (dry, sunny areas with dense cover nearby to provide refuge from wēkā predation). Traps were dug into the ground to leave their rims flush with the surface, and covered with plywood lids raised 1–2 cm off the ground by wooden spacers glued under the corners of each lid. This allowed skinks to enter traps whilst preventing heat stress and predation by wēkā. Steel pegs were used to secure lids in place. A small amount of soil and vegetation was added to each pitfall trap. Drainage holes were punched into the bottom of each trap. Pitfall traps were baited with canned pear (Whitaker 1967) and checked daily. Bait was replaced every 1–2 days.

The number of traps operated daily on Taumaka Island ranged from 22 to 31. Eight pitfall traps were uplifted on 12 March due to flooding following persistent rain, and re-installed the next day. On the morning of 13 March, nine pitfall traps were installed on Popotai Island, moved to new sites the following morning to increase coverage, and uplifted later that day. Pitfall traps on Popotai Island were deliberately positioned next to dense vine (Muehlenbeckia australis) thickets and/or bracken (Pteridium esculentum), as these areas appeared to offer the greatest refuge from wēkā predation. The total number of trap-days per island (i.e. number of traps×number of days they were operated) was 107 for Taumaka Island and 18 for Popotai Island. Total trapping effort (both islands combined) over the survey period was therefore 125 trap-days.

All skinks captured were measured (snout-vent length [SVL], vent-tail length [VTL], and the length of any tail regeneration if previously broken; all to the nearest mm using a clear plastic ruler), sexed (adults only; by hemipene eversion and/or examination of the cloacal region) and weighed to the nearest 0.1 g using digital (ATEC Multi-Function-250) scales. All skinks that had SVLs ≥ the smallest animal to which sex could be assigned based on the presence of hemipenes were considered adults. Because adult females were non-reproductive (lacking vitellogenic follicles or embryos) at the time of the survey, it was not possible to determine the minimum SVL at which females attain sexual maturity. Neonates (skinks born in the summer before our survey) were easily identified by their small body size and placental scarring on their abdomens. Skinks were given a temporary mark on the head with a non-toxic pen (Pilot xylene-free silver marker; Whitcoulls Limited) to allow for easy identification of recaptures, and released within 1 m of their capture site. All pitfall traps were removed on 14 March.

Results

Taumaka Island

A total of three skinks were caught over five days on Taumaka Island, ranging in size from 38–96 mm SVL (Table 1). Two skinks were caught in a pitfall trap set on the edge of a clearing (the helicopter landing site) at the north-western end of the island. The third skink was caught inside the hut, where it was intercepted while climbing up the diagonal timber framing below the sink. The dog did not detect any skinks on Taumaka Island.

Table 1  Capture data for Open Bay Island skinks (Oligosoma taumakae) caught on Taumaka Island, 10–14 March 2008. SVL, snout–vent length; VTL, vent–tail length; tail regen, length of regenerating tail if previously broken; n/a, data not available.

Date	Age	Sex (adults only)	SVL (mm)	VTL (mm)	Tail regen (mm)	Weight (g)
11/3/08	Adult	Male	88	76	0.5	15.6
13/3/08	Neonate	–	38	47	–	n/a
14/3/08	Adult	Female	96	54	n/a	n/a

Popotai Island

A total of 16 skinks (four adults, one sub-adult, eight juveniles and three neonates) were captured in pitfall traps over the two days that they were set. The dog indicated that skinks were present at several sites within the area that was sampled by pitfall trapping. Multiple captures were obtained in several pitfall traps, including one trap that caught five skinks (one adult female, two juveniles and two neonates). Skinks ranged in size from 41–88 mm SVL (Table 2). There appeared to be four discrete size (SVL) cohorts (Fig. 1), possibly indicating that female Open Bay Island skinks breed for the first time when c. 3 years old.

Fig. 1  Size-class distribution of Open Bay Island skinks (Oligosoma taumakae) caught on Taumaka and Popotai Islands. The four apparent size-classes may correspond to four age classes, in which case maturity would be reached once skinks are c. 3 years old. Long-term data from individually-marked animals will be needed to fully reveal the life-history traits of this little-known species.

Table 2  Capture data for Open Bay Island skinks (Oligosoma taumakae) caught on Popotai Island, 13–14 March 2008. SVL, snout–vent length; VTL, vent–tail length; tail regen, length of regenerating tail if previously broken; n/a, data not available.

Date	Age	Sex (adults only)	SVL (mm)	VTL (mm)	Tail regen (mm)	Weight (g)
13/3/08	Adult	Female	88	92	52	12.4
13/3/08	Juvenile	–	56	73	–	3.6
13/3/08	Juvenile	–	57	76	–	4.2
13/3/08	Juvenile	–	62	84	–	5.1
14/3/08	Adult	Male	87	54	31	14.3
14/3/08	Juvenile	–	58	61	0.5	4.3
14/3/08	Juvenile	–	59	84	–	4.6
14/3/08	Subadult	Female?	74	98	–	9.8
14/3/08	Juvenile	–	59	75	–	4.4
14/3/08	Adult	Female	81	91	–	12.3
14/3/08	Juvenile	–	57	76	–	4.0
14/3/08	Neonate	–	41	48	–	1.2
14/3/08	Juvenile	–	62	81	–	4.9
14/3/08	Neonate	–	41	51	–	1.3
14/3/08	Adult	Male	82	54	46	10.9
14/3/08	Neonate	–	42	53	–	1.4

Observations on other fauna

On Taumaka Island, six Open Bay Island geckos were found during nocturnal searches by spotlight. The largest gecko captured was an adult female with an SVL of 77 mm. She was visibly gravid with two embryos and appeared to be in the mid-to-late stages of pregnancy. No geckos were found on Popotai Island during a 2.5-h night search conducted by three observers. Wēkā were conspicuous on both islands. Popotai Island appeared to support a population of 6–12 adult wēkā (no chicks were seen). One wēkā was seen to be regularly patrolling a roughly circular patch of prostrate vine (M. australis) that was c. 1 m across with a small log sticking out of it. An adult-sized skink was observed basking on this log just before the arrival of the bird. While the observer was waiting for the skink to re-emerge so that it could be photographed, the wēkā returned five times over a 1.5 h period, each time circling and intensely staring at the vegetation, and on one occasion hopping onto the log to check the vegetation directly underneath it (Ian Rasmussen, Haast pers. comm.).

Discussion

Seabird islands without introduced mammalian predators and wēkā are capable of supporting exceptionally abundant skink populations. For example, population densities of the spotted skink (O. lineoocellatum; a similar-sized species to the Open Bay Island skink) range from 938–4482 individuals ha⁻¹ on Motunau Island off the North Canterbury coast (Lettink et al. unpubl. data) and exceed 3000 individuals ha⁻¹ on Takapourewa/Stephens Island in the Cook Strait (Stephens 2004). In contrast, we caught only 19 skinks from two islands totaling 16.9 ha in size: a minimum density of <2 individuals ha⁻¹. Although low, our trapping effort (125 trap-days) was 1.7 times greater than that used to estimate the density of spotted skinks on Motunau Island (75 trap-days; Lettink et al. unpubl. data).

The most significant survey result was the discovery of Open Bay Island skinks on Popotai Island, from which skinks had not been recorded prior to our survey. While the existence of a second population of skinks strengthens the prospects for the long-term survival of the species, skinks on Popotai Island appear to be restricted to a small area of habitat (<1 ha) that offers some degree of protection from wēkā predation. Because wēkā are both predators of lizards and competitors for their food source (predominantly invertebrates and some fruit), the introduction of wēkā onto the Open Bay Islands in the early 1900s is likely to have caused an immediate and substantial reduction in skink abundance, followed by a less marked decline in the intervening decades, with skinks now persisting only in areas that provide substantial refuge from predation. Whether their current density is sufficient to ensure long-term viability of the species is questionable. Unfortunately, the apparent density of skinks on Taumaka Island is now so low that monitoring to determine population trends may no longer be possible.

Impacts of wēkā on other fauna on the Open Bay Islands

Wēkā are opportunistic and omnivorous foragers, taking a variety of prey items including invertebrates, lizards, and the eggs and chicks of ground-nesting birds (Thomas 1982; Beauchamp 1987; St Clair & St Clair 1992; Beauchamp et al. 1999; Imber et al. 2003; Harper 2006 ,2007). We encountered only one beetle during our night searches for geckos on Taumaka Island, which is unusual for an offshore island without mammalian predators (pers. obs.). Other researchers have similarly commented on the paucity of the ground-dwelling fauna (e.g. Stirling & Johns 1969). In a 1989 study of the breeding biology of Fiordland crested penguins on Taumaka Island, predation by wēkā was responsible for the loss of 38% of failed eggs and 20% of chicks (St Clair & St Clair 1992). Chicks up to 620 g and 17 days old were killed by wēkā, apparently by repeated blows to the head (St Clair & St Clair 1992). Wēkā have been observed killing sooty shearwater (Puffinus griseus) chicks in a similar manner, eating just the head and discarding the body, on the southern Titi Islands near Stewart Island/Rakiura (Harper 2006). On Taumaka Island, St Clair & St Clair (1992) regularly saw wēkā entering burrows used by sooty shearwaters and Southern blue penguins (Eudyptula minor minor), and found the remains of at least seven fairy prions (Pachyptila turtur) in and around wēkā lairs. Wēkā on Taumaka Island have also been seen taking spotted shag (Stictocarbo punctatus punctatus) eggs (Stirling & Johns 1969) and tree wētā (Hemideina thoracica) from the leaf bases of kiekie (St Clair & St Clair 1992).

Overall, it appears that wēkā have a substantial, negative impact on the terrestrial fauna of the Open Bay Islands, leading Stirling & Johns (1969)) to conclude that: ‘In our opinion, the Wekas are modifying the natural fauna of the island to an extent that would justify their removal. 'Similarly, St Clair & St Clair (1992, p.24) suggested that wēkā removal ‘may be an ideal and relatively inexpensive solution’ to restore native fauna populations. The Department of Conservation has also recommended that wēkā be removed from the Open Bay Islands (Whitaker & Lyall 2004). Clearly, the negative impacts of wēkā on the terrestrial fauna of the Open Bay Islands must be carefully weighed against the cultural and conservation value of retaining wēkā populations on these islands. Genetic (mitochondrial DNA) analysis has suggested multiple (South and North Island) origins of wēkā released on Taumaka Island (Lambert 1999). The conservation value of genetic hybrids is questionable.

Managing wēkā impacts on reptile populations

To date, two management strategies have been used to protect sensitive reptile populations from wēkā predation: wēkā removal and the creation of wēkā-free areas by exclosure fencing. For instance, wēkā were removed from Long Island in the Marlborough Sounds specifically to protect tuatara (Pete Gaze, Department of Conservation, Nelson, pers. comm.). In areas where the removal of wēkā is not desirable, fencing is being used in an attempt to limit their impact on lizard populations, for example in Zealandia Sanctuary near Wellington (http://www.visitzealandia.com/Site/Zealandia_Home/Inside/Our_Wildlife/Birds/Weka.aspx/) and on Ulva Island in Paterson Inlet on Stewart Island/Rakiura. On Ulva Island, a study is currently underway to determine whether translocated common skinks (Oligosoma polychroma) are able to establish in the presence of wēkā (Goodman et al. 2006). Skinks were released at two sites in 2005 and 2006, one of which was surrounded by a 50 × 50 m weka exclosure. Translocations of threatened lizard species may follow if common skinks are able to establish on the island (Andrea Goodman, St Arnaud, pers. comm.).

The appropriateness and success of wēkā removal will depend on a number of factors, including the value of retaining wēkā balanced against their impacts on other biodiversity, management costs and proximity to the nearest wēkā population (to prevent or limit re-invasion). Ensuring that wēkā are safeguarded elsewhere (e.g. Henderson & Goodman 2007) is a key consideration for any wēkā removal operation (Pete Gaze pers. comm.).

Management options

Below, we consider six potential management options for the Open Bay Island skink, and their likely consequences. The options are listed in the order recommended by the authors:

1. Removal of wēkā from Taumaka and Popotai Islands

This option will have maximum benefits for the Open Bay Island skink, Open Bay Island gecko, Open Bay Island leech, ground-nesting birds and terrestrial invertebrates. Future translocations of other species that are now rare or absent on the mainland would also become an option, thereby increasing the restoration potential of the islands. We consider this to be the only option that is guaranteed to increase the abundance of Open Bay Island skinks, and the best option for biodiversity conservation on the islands.

2. Removal of wēkā from one of the islands

Wēkā are capable swimmers with strong homing instincts, as shown by a banded wēkā that was taken from Maud Island in the Marlborough Sounds and released on the mainland, only to be found back on Maud Island 13 days later after swimming a distance of at least 700 m (Bill Cash, Department of Conservation, Picton, pers. comm.). Hence, removing wēkā from one island whilst leaving them on the other is, at best, only a temporary solution with temporary benefits. This option may be feasible for Popotai Island (given its smaller size and relative ease of wēkā removal compared with Taumaka) if funding is made available to regularly revisit the island to check for and remove any wēkā present. To prevent rapid recolonization, any wēkā taken off Popotai Island should not be released on Taumaka Island. Because the islands are only c. 200 m apart (a distance that is easily crossed by wēkā), recolonisation events may be constant. Due to uncertainties over funding availability and the frequency of recolonisation by wēkā, the effectiveness of the option remains unknown and cannot be guaranteed.

3. Control of wēkā by exclosure fencing on one or both islands

This is predicted to benefit only the Open Bay Island skink and other terrestrial fauna present within the enclosed area for as long as the fence remains intact. However, this option is fraught with difficulties, including: (1) logistical and financial constraints, particularly on Taumaka Island, where the area known to be used by skinks cannot easily be fenced (if at all) because of the steep terrain and flanking cliff edge; (2) the need for ongoing surveillance to ensure that seals and burrowing seabirds do not breach the fence; (3) entrapment of seabirds within the fenced area (fairy prions and sooty shearwaters may not be able to clear the fence during take-off); (4) potential impacts of a fence on the seal population and of seals on a fence, and; (5) the need to maintain lizard habitat inside the fence if this is altered by the exclusion of seals and other wildlife. Exclosure fencing could be used to protect part or all the area occupied by skinks on Popotai Island; however, all the above issues would have to be addressed.

4. Refuge supplementation

This option would involve increasing the number and/or quality of refuges available to Open Bay Island skinks, either by providing skinks with additional natural or semi-natural cover (e.g. vine thickets and rock piles) or by using artificial retreats (e.g. stacks of roofing materials). ‘Natural’ habitat modification could include replacing some areas of kiekie with mixed plantings of sedges, vines and bracken, although this may be difficult to achieve in the presence of burrowing seabirds, wēkā and seals. If artificial retreats are used, they would have to be reinforced in some way or fenced to keep seals and wēkā out. This strategy is risky because wēkā may learn to preferentially hunt in the vicinity of artificial structures if skinks are regularly seen there. In the only New Zealand study to have formally tested the effects of refuge supplementation on lizards, the addition of (Onduline) artificial retreats failed to increase survival of McCann's skinks (Oligosoma maccanni) inhabiting coastal duneland in Canterbury (Lettink et al. 2010). Worldwide, we are aware ofonly one study that has demonstrated an increase in reptile abundance following refuge supplementation. This study was on, the pygmy blue tongue lizard (Tiliqua adelaidensis) of Southern Australia, which lives in burrows (Souter et al. 2004). This method should therefore not be relied on as the only means of conserving the Open Bay Island skink. In general, refuge supplementation is most likely to benefit populations and species that are limited by a lack of natural retreat sites, rather than by other factors such as disease, food availability or predation (Brady et al. 2000).

5. Do nothing

This option may lead to the extinction of the Open Bay Island skink on Taumaka Island in the short-to-medium term. Population trends on Popotai Island are unknown, but skinks remain vulnerable to extinction because wēkā are present and there is only a small area (<1 ha) of suitable skink habitat.

6. Captive breeding and/or translocation of skinks to a predator-free island

Captive breeding was raised as a temporary option for Open Bay Island skinks by Hieatt & Newton (in prep.). It is listed here, but not considered to be a viable option at present due to the costs of setting up a captive breeding facility, the lack of a predator-free location for the eventual release of captive-bred animals, and the impacts of taking more skinks away from an already-rare source population.

Concluding remarks

We consider wēkā removal (Option 1) to be the only feasible option for reducing the impacts of wēkā predation on the Open Bay Island skink and other terrestrial fauna of the Open Bay Islands. Of the six options considered, it is the only management action that provides a permanent solution to the greatest threat that is currently limiting the abundance of Open Bay Island skinks and other affected fauna. In the absence of wēkā, skinks would be expected to reach high local densities wherever suitable habitat is present. Although female reptiles from New Zealand generally exhibit K-selected life-histories (i.e. delayed maturity, low annual reproductive output and extreme longevity [Cree 1994]), rapid response to management has been documented in some species. For example, the population of Fiordland skinks (O. acrinasum) released on Hawea Island in Breaksea Sound (Fiordland) following eradication of the Norway rat (Rattus norvegicus) increased by 400 percent in less than 5 years (Thomas & Whitaker 1994).

Finally, removal of wēkā from the Open Bay Islands would be inexpensive and easily achieved compared with the expensive and ongoing management solutions needed at mainland sites. Apart from the Open Bay Islands, there is only one other site that has two Nationally Critical lizard species: Macraes Flat in north-eastern Otago. At that site, populations of grand (O. grande) and Otago (O. otagense) skinks inhabit schist tors in tussock grasslands, where they are under threat from predation and habitat loss (Tocher 2006). It was recently demonstrated that it is possible to reverse population declines, using either intensive pest mammal control conducted over large (up to 2000 ha) areas or by using large-scale (11–22 ha) predator exclosures (Nathan Whitmore, Department of Conservation, Dunedin, pers. comm.). Compared with the intensive management currently undertaken at Macraes Flat, relatively little would be required to improve the threat status of the Open Bay Island skink, with collateral benefits to other resident fauna.

Acknowledgements

Field work was conducted with approval from the Department of Conservation (DOC) and Taumaka me Popotai Trust. We wish to thank the owners of Taumaka me Popotai for permission to visit the Open Bay Islands, Tim Shaw and Ian Rasmussen for excellent company and invaluable assistance in the field, Tony Whitaker for his field observations, Andrea Goodman for providing wēkā and skink management information, James Scott for helicopter transport and DOC South Westland/Weheka Area for funding the survey. Alison Cree, Jo Hoare, David Chapple, Tim Shaw, Grant Norbury, John Lyall, Rod Hitchmough and James Reardon provided constructive comments that improved this manuscript.