A review of potential alpine newt (Ichthyosaura alpestris) impacts on native frogs in New Zealand

ABSTRACT

The alpine newt (Ichthyosaura alpestris) is an adaptable European amphibian occupying a range of habitats from low to high elevation within its native range. The Italian subspecies I. a. apuana, illegally introduced to New Zealand, has established geographically close to habitats of Archey’s frog (Leiopelma archeyi) and Hochstetter’s frog (L. hochstetteri). The newt’s introductions in Europe suggest competitive impacts on local amphibians were not severe, although as a disease and parasite vector its impact may have been underestimated in the past. The most identifiable risk to native frog populations in New Zealand is introduction of new diseases or parasites, although the newt may have wider ecological impact should it substantially invade native frog habitats. Alpine newts need eradicating from New Zealand as soon as possible, while the status of captive-held fire-bellied newts (Cynops spp.) also requires review, given their disease risk and their potential for release into the wild.

KEYWORDS:

• Biosecurity
• conservation
• invasive species
• Leiopelma
• New Zealand
• newts

Introduction

With the growing concern about declining amphibian populations it is easy to overlook the potential for some amphibians to become established and even invade new habitats once they are introduced into a new region (Wells 2007). Well-known examples of invasive amphibians are the American bullfrog (Rana catesbeiana) and the cane toad (Bufo marinus) (Lever 2003; Wells 2007).

Since humans arrived in New Zealand, much of the country’s biota has been transformed and continues to be altered, including the arrival of exotic species and their detrimental impact on endemic species through predation, competition and disease (Wilson 2004; Gibbs 2006; Trewick & Morgan-Richards 2014). Biological invasions, still occurring worldwide at an alarming rate, are widely acknowledged as threats to the integrity and functioning of ecosystems (Barlow & Goldson 2002; Poulin et al. 2011). Incursions of unwanted species into New Zealand continue despite biosecurity provisions at its borders (Kriticos et al. 2005; Maria 2014; Chapple et al. 2016); for example, agricultural pests or diseases, including varroa bee mite (Varroa jacobsoni), southern marsh mosquito (Ochlerotatus camptorhynchus), Queensland fruitfly (Bactrocera tryoni) and bacterial kiwifruit vine disease or Psa (Pseudomonas syringae pv actinidiae) (Auditor-General 2002; MPI 2014a, 2015a, 2015b, 2015c). The detrimental impact of introduced mammals on New Zealand’s endemic biota, particularly the avifauna, is also widely recognised (e.g. King 1984, 1990; Bell 1991; Russell & Clout 2005; Russell et al. 2008), while management steps to reduce their numbers are often debated, such as the use of the pesticide 1080 (Elliott & Suggate 2007; Wright 2011).

Incursion threats to New Zealand’s endemic herpetofauna, including its distinctive native frogs (Anura: Leiopelmatidae), have generally received less publicity, but nevertheless are recognised in conservation and scientific circles (e.g. Bell 2010; Kikillus et al. 2012; Bishop et al. 2013; Chapple et al. 2016). Here, I aim to review one such threat, the recently discovered incursion of the alpine newt Ichthyosaura alpestris (Amphibia: Urodela), summarising the newt’s biology and its potential impacts on endemic frogs, particularly Archey’s frog (Leiopelma archeyi) and Hochstetter’s frog (L. hochstetteri), and the efforts made to eradicate the species before further spread from its introduction site in New Zealand. The endemic leiopelmatid frogs are of considerable evolutionary significance and all four extant species are globally and nationally threatened (Roelants & Bossuyt 2005; Bell 2010; Newman et al. 2010, 2013; Bishop et al. 2013). This review has developed from an internal report (Bell 2015) requested by the Native Frog Recovery Group of the Department of Conservation (DOC).

Exotic herpetofauna in New Zealand

Apart from the alpine newt, a range of other exotic amphibians have been introduced to New Zealand. Three Australian Litoria species (family Hylidae) have successfully established in the wild (L. aurea, L. ewingii and L. raniformis) and are now widespread (Bell 1982a; Pickard & Towns 1988; King et al. 2009), although there have been declines (Bishop 1999). The green and golden bell frog (L. aurea) is known to prey upon Archey’s frog (Thurley & Bell 1994), while the smaller brown tree frog (L. ewingii) is slowly extending its range in the North Island, partly the result of further liberations (King et al. 2009). Should it eventually invade habitats of Archey’s and Hochstetter’s frogs, then the brown tree frog could have a negative ecological impact on them through competition and possibly predation of younger frogs, which are only 7–10 mm (snout–vent length) at metamorphosis (Bell 1978b; B Bell, pers. obs.).

Several anurans were introduced to New Zealand but did not establish: Bufo bufo, B. calamita (?), Rana esculenta (?) and R. temporaria from Europe; and Litoria caerulea and L. adelaidensis from Australia (McCann 1961; Robb 1980; Bell 1982a, 1982b; King et al. 2009). The Agricultural Department imported six dozen L. caerulea for pest control from Sydney in 1897, augmented by a further consignment in 1899, liberating them into Hawke’s Bay, Wellington, Auckland, Whanganui and Nelson districts (McCann 1961; Robb 1980; Bell 1982a), but the species apparently failed to establish (Bell 1982a, 1982b; Thomson 1926). At least four L. adelaidensis from South West Australia were found calling at a New Plymouth lake in 1981 (Bell 1982a, 1982b), but again failed to establish.

In 1999, a human-induced incursion of the Australian eastern banjo frog (Lymnodynastes dumerilii; Anura: Myobatrachidae) occurred west of Auckland in an area where Hochstetter’s frog is found, but the population was exterminated (Whitaker & Bejakovich 2000; Gill et al. 2001). Other amphibian species have been found at ports or intercepted from imported cargoes; for example, cane toad, Asian black toad (Bufo melanostictus), South American hylids (B Bell, pers. obs.) and the Australian Litoria gracilenta (Bell 1982a, 1982b). There have also been importations of anurans for teaching, research and medical diagnosis, including the African clawed frog Xenopus laevis (family Pipidae) and the cane toad (King et al. 2009). Overall, some 200 species of amphibians and reptiles have been intercepted entering New Zealand (Chapple et al. 2016).

At least three urodeles are known to have been imported through the pet trade, the Mexican axolotl (Ambystoma mexicanum), the Japanese fire-bellied newt (Cynops pyrrhogaster) and the Chinese or oriental fire-bellied newt (Cynops orientalis) (Garthwaite 2000; Whitaker 2000; King et al. 2009; NatureWatch NZ 2015; New Zealand Herpetological Society 2015). Axolotls are frequently released into garden ponds, and Whitaker (2000) had several unverified reports of self-sustaining populations, notably in the Waikato region and river. A wild axolotl population was established in a farm dam near Tangoio, Hawke’s Bay, following a release of six animals in 1987, but was exterminated by DOC in the 1990s (Garthwaite 2000; King et al. 2009). Whitaker (2000, p. 230) noted that ‘other species (e.g. fire-bellied newt C. pyrrhogaster) are now widely distributed in captivity and breed regularly, suggesting it is only a matter of time before they establish’. Fire-bellied newts carry a potential disease and toxicity risk (Martel et al. 2013, 2014; Kolby et al. 2014; McEwan 2015), and potentially are regarded as a threat to native lizards in New Zealand (Romijn et al. 2012). Another urodele incursion has been recently reported in Australia. The European smooth newt (Lissotriton vulgaris) has established in the Melbourne area (Tingley & Woolnough 2014; Tingley et al. 2015). Newt incursions in both Australia and New Zealand, with no indigenous urodeles, differ from those in North America and Europe where newts and other urodeles already occur (Arnold et al. 1978; Wells 2007; McEwan 2015).

Apart from introduced frogs, New Zealand has experienced a successful incursion among its lizard fauna—the delicate or rainbow skink (Lampropholis delicata) from Australia has become locally abundant in the northern North Island in areas occupied by endemic lizards such as the copper skink (Oligosoma aenea) (Peace 2004; Jewell & Morris 2008; Chapple et al. 2016). Kikillus (2010) concluded that at least eight species of exotic reptile legally traded within New Zealand are predicted to be capable of surviving in a portion of the New Zealand environment and at least three species have the potential to breed successfully in warmer microclimates. She noted that public education and possible regulations imposed on the New Zealand exotic reptile trade may prevent introductions of these species into the local environment and still allow selected species to be enjoyed by the New Zealand public. Risks include disease transfer; for example, the prevalence of Salmonella in exotic reptiles is higher than that of native reptiles, though considerably lower than that of exotic reptiles overseas, with all serovars identified being previously reported in New Zealand (Kikillus et al. 2011). Kikillus (2010) recommended further research involving climatic tolerances and breeding potential (i.e. soil moisture content, juvenile survival, sex ratio and predicted climate change), while the methods developed may be applied to other species and other geographic regions, allowing investigation into the establishment risk of alien species including amphibians such as newts in New Zealand (Kikillus et al. 2010, 2011, 2012). This may help guide control and management efforts and help stem the tide of the growing problem of invasive species. The methodology included online trading tools for estimating propagule pressure via the pet-release pathway (Kikillus 2010; Kikillus et al. 2012).

The alpine newt incursion in New Zealand

The alpine newt incursion into New Zealand was first reported in July 2013 after three newts were detected on private property in the south Coromandel region (Van Winkel 2013; Whitaker 2013). Network analysis of mitochondrial DNA haplotypes identified the newt source population as I. a. apuana from Tuscany, Italy (Figure 1; Arntzen et al. 2016). This smaller subspecies resembles the nominate subspecies I. a. alpestris and is often captive-bred overseas (Staniszewski 1995).

Figure 1. Male Italian alpine newt Ichthyosaura alpestris apuana, the subspecies introduced to New Zealand. Photograph: MPI photo taken by AH Whitaker.

A rapid assessment, followed by delimiting surveys, confirmed the presence of a breeding population, and preliminary risk assessments considered the newt to be a significant threat to nearby populations of indigenous frogs (as vectors of the pathogenic chytrid fungus). Anecdotal and biological evidence suggested that the incursion was potentially more than a decade old. As part of the eradication response, the Ministry for Primary Industries (MPI) declared the species an ‘Unwanted Organism’ under the Biosecurity Act 1993, making it an offence to transport, breed, sell or display the newt. Failure to do so can result in 5 years’ imprisonment or a fine up to $100,000, or both. An ‘Unwanted Organism’ is any organism that a chief technical officer of MPI believes is capable or potentially capable of causing unwanted harm to any natural and physical resources or human health (MPI 1998).

By 2015 it was evident that this eradication had been challenged by a range of issues: the detectability of individual newts; the reproductive and behavioural biology of the newts; complex habitat features; a requirement for novel and innovative search and trapping techniques; investigation and compliance issues; land access; public interest; project confidentiality; and resource limitations (Van Winkel et al. 2015). From the onset, MPI led a collaborative approach with DOC, specialist consultants and contractors, regional council and university research associates. This response approach allowed the eradication process to progress significantly over the response’s first year in practice (Van Winkel et al. 2015).

An assessment of the potential impacts of the alpine newt on New Zealand freshwater ecosystems concluded that the newt could potentially invade a wide range of regions and freshwater habitats, such as lakes, ponds, ditches and slow-flowing streams, and interact with indigenous freshwater ecosystems in a number of ways—as predators, competitors and prey (McEwan 2015). Knowledge from the alpine newt’s native range, together with evidence from other introductions, generally suggested that impacts in freshwater ecosystems may not be severe. However, aspects of particular concern included potential predatory and competitive interactions with threatened non-migratory fishes, and possibly the behaviour of tetrodotoxin (sometimes referred to as tarichatoxin) in New Zealand food webs, a potent neurotoxin reported from western Taricha newts of North America (Wakely et al. 1966; Brodie et al. 1974; Marks & Doyle 2015). As a result of the latest situation report DOC stepped up its involvement with the newt incursion in 2015, proactively sharing expertise to better support MPI’s lead to ensure the best conservation outcome (DOC 2015).

An overview of the biology of the alpine newt

Geographical distribution

The alpine newt is more widely distributed in Europe than its name suggests, occurring in sizeable colonies in some lowland areas; although the largest populations are in hilly or mountainous districts, the species occurring from near sea level to 2500–3000 m (Steward 1969; Arnold et al. 1978). The newt occurs from extreme western Russia westwards to northern and eastern France, and from southern Denmark south to northern Italy and central Greece, with an isolated population in northwestern Spain (Arnold et al. 1978; Griffiths 1996).

Introduced populations of alpine newts have been reported in different parts of mainland Europe (Sillero et al. 2014; Arntzen et al. 2016), and it has been liberated on some Dutch islands (E Heiss, pers. comm. cited in McEwan 2015). The alpine newt is one of several exotic amphibians to have naturalised in Britain, where three native newts occur (Bell 1978a; Griffiths 1987, 1996; Bell & Bell 1995; Beebee & Griffiths 2000; Inns 2009; Lever 2009; Beebee 2013; Weller 2013).

Life-history stages

Fertilisation in the alpine newt is internal, and adults are mostly terrestrial but resort to water in spring and early summer for breeding, with elaborate courtship displays. The males produce spermatophores which are picked up by the female with the cloaca and stored until fertilisation in her spermatheca. In the Italian subspecies I. a. apuana breeding activity begins in the spring and is interrupted by the winter. Thus eggs are laid at least twice a year and the two larval cohorts live together, indicating adaptations to unpredictable aquatic environments in a Mediterranean dry climate (Andreone & Dore 1992). The eggs—usually laid singly and attached to plants, sticks or stones—hatch out into gill-bearing tadpoles, which grow first the front and then the back legs and, finally, following atrophy of the gills, emerge from the water as lung-breathing adults (Steward 1969). Around 150 eggs are laid by each female newt, hatching out in 2–4 weeks, depending on temperature; generally, larvae metamorphose some 3 months after hatching (Steward 1969).

Alpine newts can overwinter in Europe before metamorphosing, with a notable tendency towards paedogenesis or neoteny, particularly in some subspecies (Steward 1969; Arnold et al. 1978; Ballasina 1984; Djorovic & Kalezic 2000; Sotiropoulos 2001; Denoël 2002; Caudata Culture 2003; Recuero et al. 2014). Adult female alpine newts reach 10–11 cm in length (Beebee & Griffiths 2000), sometimes up to 12.7 cm, adult males being smaller at about 9 cm in length (Steward 1969). After larval metamorphosis into miniature versions of the adults, the young newts then spend much of their time living a terrestrial existence, hiding by day in vegetation, or in sites like rock crevices or under logs (Beebee 2013; B Bell, pers. obs.). Attainment of sexual maturity is linked to temperature, taking 1.5–3 years, while longevity is at least 20 years (Staniszewski 1995; Beebee & Griffiths 2000).

Habits and habitats

In many respects the alpine newt appears to be more adaptable than most other European newt species, entering and leaving the water more often outside the breeding season (Steward 1969). Staniszewski (1995) noted that the species is very hardy, capable of tolerating a wide range of waters including frozen, stagnant, polluted ponds, and that it is the most aquatic true newt, with the Italian subspecies actively spending much (but not all) of its adult life in water. In the north of its range it occurs in a variety of habitats, even including shallow, open water in lowlands, but is commoner in cold, almost plantless ponds in woods and in pools, lakes and slow-flowing streams in mountain regions, or complexes of small ruts on muddy forest tracks (Arnold et al. 1978; Morrison 1994; Kopecký et al. 2010). In the southern part of its range sparsely vegetated mountain lakes and barren areas are favoured (Kwet 2009). Its habitat range, derived from habitat pictograms given by Ballasina (1984), is compared with habitat ranges of Archey’s frog and Hochstetter’s frog in Table 1.

Table 1. Habitats occupied by the alpine newt, Archey’s frog and Hochstetter’s frog. Black—major habitat; grey—other occupied habitat; white—absent, or occasionally present. Habitats for the alpine newt are primarily based on Ballasina (1984).

When on land, it is typically found in cool, moist places, even surviving at least short periods below freezing point. In the south of its range it is more strictly montane and can occur above 2500 m (Steward 1969; Arnold et al. 1978). On land the alpine newt is nocturnal and crepuscular, typically emerging early in the evening at the onset of dusk (Steward 1969). While relatively aquatic (Staniszewski 1995), the alpine newt nevertheless appears to wander about more freely than other newt species, with specimens being found far from water or under isolated stones in open fields, where other species of newt would rarely be encountered. It is likely to have the capacity to move into forests and creek systems in New Zealand, as in Europe it can be found in small watercourses, brooks and running water, as well as in deciduous and pine forest, hills and mountains (Ballasina 1984; Table 1).

Alpine newts introduced to France can disperse up to 2.5 km to the breeding pond, or from the point of introduction (Mathieu 2005). In Shropshire, UK, alpine newts were released into garden ponds, and the dispersal rates were limited to 70 m in 15–23 years (Bell & Bell 1995). In Europe, hibernation takes place on land, beginning in September–October and ending in February (or earlier in the south) to May. The variation in timing and duration of the active period depends on latitude and altitude. In highlands, the activity period is shorter (Kuzmin 1999), while New Zealand’s milder climate suggests a shorter period of winter inactivity compared with much of Europe.

In Britain, introduced alpine newts have, thus far, mainly turned up in gardens and park ponds, but are widely distributed through multiple liberations (Bell & Bell 1995; Beebee & Griffiths 2000; Beebee 2013). Lever (2009) concluded that as both a congener of and a competitor with native newt species, the alpine newt has the potential to cause ecological perturbation in Britain, despite its coexistence with other newt species in continental Europe. In both Sunderland and near Brighton they thrive in ponds with the three-spined stickleback (Gasterosteus aculeatus), a small pugnacious fish (Beebee & Griffiths 2000), but in Shropshire (Bell & Bell 1995) no fish were observed at any of 13 ponds where newts were breeding, although five of nine ponds with only adult newts had fish in them (carp Cyprinus carpio, goldfish Carassius auratus, rudd Scardinius erythropthalmus or three-spined stickleback).

Food and foraging

Overall, the alpine newt is regarded as a generalist predator that exhibits ‘extraordinary’ behavioural plasticity in its prey-capture behaviour according to Heiss et al. (2013). Beebee & Griffiths (2000) noted that locally introduced alpine newts were mainly terrestrial during the summer months in Britain, emerging at night to hunt invertebrates. In their aquatic phase, adults fed avidly on frog tadpoles, but avoided toad larvae (Beebee & Griffiths 2000). Steward (1969) described the food of adult alpine newts as small crustaceans, insects and their larvae, worms and slugs, noting that, in water, adults prefer to find most of their food on or in the mud at the bottom, although they also hunt active or free-swimming prey. In Europe, instances have also been reported of alpine newts eating frog eggs and those of their own or other species of newt (Steward 1969).

The feeding habits of their aquatic larvae are similar to other European newts: after the eggs hatch, the minute larvae feed on infusoria, but as they later become larger and more mobile, they actively pursue small creatures such as water fleas Daphnia (Steward 1969). By the time full larval size is reached, they can tackle insect larvae and small worms, or ‘almost anything they can catch and swallow’ (Steward 1969, p. 96). Further information on feeding in the Italian subspecies that occurs in New Zealand is given by Fasola & Canova (1992). They tabulated the proportion and volume of prey items taken by alpine newt larvae, neotenics, juveniles and adults, finding marked seasonality in prey taken. The proportion of prey eaten differed from studies in northern Europe; for example, the diet of adult Italian alpine newts was mostly Crustacea and Collembola, but elsewhere it was mainly Crustacea, Diptera and Ephemeroptera (Braña et al. 1986; Joly & Giacoma 1992). MPI (2013, 2014b) concluded that alpine newts are voracious predators, also adding that they feed on a variety of things, including amphibian and fish eggs and larvae, aquatic invertebrates and small vertebrates, and that ‘this makes them a high risk to New Zealand’s already endangered native frogs, which would be placed under further threat from newts’ (MPI 2014b, p. 1).

Diseases and pathology

Proportionally, amphibians are more threatened than either birds or mammals at a global level (Stuart et al. 2004; Hoffmann et al. 2010), and disease has been identified as a key agent in the ‘enigmatic decline’ of many amphibians, even where suitable habitat remains (Stuart et al. 2004). Newts can harbour pathogens and parasites that could potentially spread to endangered endemic amphibians (MPI 2013, 2014b). The alpine newt is probably the most successful introduced species of urodele in Britain (Wisniewski 1989), Wilkinson et al. (2015) finding that the species was of high concern and high research priority due to its risk of spreading disease, especially chytridiomycosis. It is not clear what might trigger its uncontrolled expansion in the UK, potentially at the expense of native fauna (Bond & Haycock 2008).

The amphibian chytrid fungus Batrachochytrium dendrobatidis (Bd), is reported to have infected alpine newts in Europe (Spitzen-van der Sluijs et al. 2014), and occurs in the area of origin of New Zealand alpine newts along the Italian peninsula (Zampiglia et al. 2013). The alpine newt was identified as an asymptomatic vector in the spread of Bd in introduced populations in Britain (Arntzen et al. 2009), while in New Zealand initial screening showed > 70% to be infected (J Laycock, pers. comm. cited in Arntzen et al. 2016). Laycock et al. (2015) point to the potential value of screening amphibian intruders, such as alpine newts, for Bd in the MPI collection formerly held by AH (Tony) Whitaker.

In New Zealand frogs, Bd was first recorded in the southern bell frog near Christchurch in 1999–2000 (Waldman et al. 2001). It was first recorded in Archey’s frog in the Coromandel Peninsula in 2001 and at Whareorino in 2006 (Bell et al. 2004; Shaw et al. 2008). It has not been recorded in wild populations of Hochstetter’s frog (Bell 2010; Shaw et al. 2013). Bell et al. (2004) suggested that the progressive northward declines of Archey’s frog populations in Coromandel over 1996–2001 might indicate a spreading pathogen, with an average spread rate of about 12 km per year. If the principal causal factor for those declines was Bd, this suggests that the pathogen was in Coromandel by late 1996, with the first recorded decline near Tapu, even though chytridiomycosis was not identified in New Zealand until 1999–2000 (Waldman et al. 2001) and not in Coromandel until 2001 (Bell et al. 2004). There remains the possibility that the frog’s decline may even have coincided with the nearby introduction of the alpine newt, although this requires closer investigation, particularly as Bd also occurs in introduced Litoria frog populations (Waldman et al. 2001; Shaw et al. 2013), including green and golden bell frogs local to the Coromandel area (Bell 1982a). Reports across New Zealand suggest that populations of introduced frogs (Litoria spp.) crashed over 1993–1995 (Bishop 1999). In the Coromandel region, declines of green and golden bell frogs were reported over 1997–2000 (P Thomson in litt.; Bell et al. 2004). As noted by Arntzen et al. (2016), leiopelmatid frogs now have low susceptibility (Ohmer et al. 2013), and Hochstetter’s frog may show resistance to Bd (Moreno et al. 2011), but there are many different strains that could have varied effects (Herbert et al. 2011).

Batrachochytrium dendrobatidis was the only species within the entire phylum of the Chytridiomycota known to parasitise vertebrate hosts until Martel et al. (2013, 2014) reported the discovery of a second highly divergent chytrid pathogen, B. salamandrivorans sp. nov. (Bsal), that causes lethal skin infections in salamanders. This fungus (Bsal) has resulted in steep declines in salamander and newt populations, including the alpine newt, in northwestern Europe. East Asian salamanders were susceptible but able to limit the infection. Further, the fungus was detected in a more than 150-year-old museum specimen of the Japanese sword-tailed newt (Cynops ensicauda). This suggests it had originally emerged and co-evolved with salamanders in East Asia, forming its natural reservoir, and was introduced to Europe rather recently through the trade of species such as fire-bellied newts (C. orientalis and C. pyrrhogaster) as pets, which are also kept as pets in New Zealand. Martel et al. (2014) reported that frogs and caecilians were immune to Bsal, so it may pose no risk to leiopelmatid frogs in New Zealand. Since the discovery of Bsal and its likely origins were reported, the world’s amphibian conservationists have gone into full alert (Stark 2015), with the US Fish and Wildlife Service introducing restrictions on importation and interstate transportation of salamanders (US Department of Interior 2016; Zimmer 2016) and, in the UK, Cunningham et al. (2015) urging that measures to minimise the likelihood of further imports of Bsal should be put in place.

Ranaviral disease is a further threat to amphibians (Daszak et al. 1999). Alpine newts from the Cantabrian Mountains, Spain, have been recently reported to be experiencing high mortalities and population declines after infection with a ranavirus (Price et al. 2014; Arntzen et al. 2016). North et al. (2015) point out that ranaviruses are causing mass amphibian die-offs in North America, Europe and Asia, and have been implicated in the decline of European common frog (R. temporaria) populations in the UK. Despite this, we have very little understanding of the environmental drivers of its occurrence and prevalence, and its likely impact on leiopelmatid frogs in New Zealand. Using a long-term (1992–2000) dataset of public reports of amphibian mortalities, North et al. (2015) concluded that, in Britain, links between occurrence, prevalence, pond characteristics and land management practices provided useful management implications for reducing the impacts of ranavirus in the wild, revealing the influence of biotic and abiotic drivers of the disease, with many of these abiotic characteristics being anthropogenic. The disease prevalence in the common frog increased with increasing frog population density, presence of fish and wild newts, increasing pond depth and the use of garden chemicals. The presence of an alternative host reduced prevalence, potentially indicating a dilution effect. Ranavirosis occurrence was associated with the presence of common toads (B. bufo), an urban setting and the use of fish care products (North et al. 2015).

Invasive amphibians may also bring to a new environment a range of other diseases or parasites that in turn might be a risk to native species. These include other fungal infections (e.g. Oodinium), a range of bacterial infections (e.g. Aeromonas, Pseudomonas, Mycobacterium, Columnaris) and parasites (e.g. the protozoan parasite Opalina and endoparasites such as trematodes, nematodes and tapeworms). External parasites are relatively scarce in amphibians (Staniszewski 1995).

Potential impacts of the alpine newt on native frogs

In New Zealand, the introduced alpine newt may pose a threat to Archey’s frog and Hochstetter’s frog in various ways, potentially affecting a range of leiopelmatid life-history stages (Table 2). Of concern is its discovery ca. 2 km from Hochstetter’s frog habitat and ca. 7 km from Archey’s frog habitat in the Coromandel region (R Burns, pers. comm.), particularly given its potential risk of vectoring of disease. The newt could possibly have further detrimental impacts as a competitor, a predator and, perhaps, as a toxic prey item. It is an attractive species (Figure 1): Beebe & Griffiths (2000, p. 198) refer to it as ‘undoubtedly one of Europe’s most attractive amphibians’, with Morris (2014, p. 5) aptly noting that, in Cornwall, it was ‘a classic case of lovely species, wrong location’. Given this, there remains some risk that the alpine newt might still be illicitly spread in New Zealand, despite its classification by MPI as an ‘Unwanted Organism’; indeed, this may have already taken place.

Table 2. Possible risks to Archey’s frog and Hochstetter’s frog from the alpine newt. Hatchlings are hatched larvae prior to metamorphosis, young frogs are those in their first three years following metamorphosis.

Potential agent	Development stage	Direct risk to Archey’s frog	Direct risk to Hochstetter’s frog	References
Batrachochytrium dendrobatidis (chytrid)	Eggs	?	No	Waldman et al. (2001)
	Hatchlings	Yes	No	Bell et al. (2004)
	Young frogs	Yes	No	Ohmer et al. (2013)
	Older frogs	Yes	No	Shaw et al. (2013)
Batrachochytrium salamandrivorans (chytrid)	Eggs	No	No	Martel et al. (2013)
	Hatchlings	No	No	Martel et al. (2014)
	Young frogs	No	No
	Older frogs	No	No
Ranaviral disease	Eggs	?	?	Daszak et al. (1999)
	Hatchlings	Potential	Potential	Price et al. (2014)
	Young frogs	Potential	Potential	North et al. (2015)
	Older frogs	Potential	Potential
Parasites	Eggs	?	?
	Hatchlings	Potential	Potential
	Young frogs	Potential	Potential
	Older frogs	Potential	Potential
Predation	Eggs	?	?	McEwan (2015)
	Hatchlings	Potential	Potential	Arntzen et al. (2009)
	Young frogs	Potential	Potential
	Older frogs	No	No
Competition	Eggs	No	No	McEwan (2015)
	Hatchlings	No	No	Arntzen et al. (2009)
	Young frogs	Potential	Potential
	Older frogs	?	?
Toxicity	Eggs	No	No	Wakely et al. (1966)
	Hatchlings	No	No	McEwan (2015)
	Young frogs	No?	No?
	Older frogs	?	?

McEwan (2015) records that alpine newts were dubbed ‘masters of change’ by Heiss et al. (2013), due to the high levels of behavioural plasticity they displayed—a quality which is hypothesised to be an important pre-requisite for the invasion of new environments (Stayton 2011). She noted that they have already changed their phase regime in response to the New Zealand environment and that potential further adaptations include changes to prey species, habitat types and toxicity, and that the impacts of invasive species can take time to become apparent. For example, the introduced green and golden bell frog was originally thought to have little or no interaction with endemic frog species (Bell 1982a), but further investigation revealed it to be a predator in some situations, being observed well away from breeding ponds at 800 m in Whareorino forest where both Archey’s frog and Hochstetter’s frog occur (Thurley & Bell 1994). Some alpine newts might disperse similarly, though they would probably be potential predators of only young Leiopelma (Table 2).

Arntzen et al. (2016) argued that the effect of competition and predation on Archey’s frog is likely to be minor, because Archey’s frogs are terrestrial breeders, inhabiting cooler moist native forest habitat at 100–1000 m. above sea level, whereas the alpine newt usually breeds in ponds and, more rarely, in running water (Breuil & Parent 1987). Currently, the newts appear to be restricted to around the lowland pastoral site of introduction, as occurred in Britain (e.g. Bell & Bell 1995). The newt has the greatest potential to spread across such open habitats where ditches, pools and ponds are frequent, rather than into Leiopelma habitats where still waters for breeding are few, although they do use small seeps and pools (Steward 1969). Some newts might therefore potentially move up creeks and seepages occupied by Hochstetter’s frog, or even to more terrestrial sites at higher elevation where Archey’s frog mostly occurs (Stephenson & Stephenson 1957; Bell 1978b, 2010; McLennan 1985), and could breed there if small ponds or pools are present. Whether the newt would invade these habitats in any numbers remains uncertain, however, and the risk seems low, particularly for Archey’s frog as Arntzen et al. (2016) suggest. Nevertheless, a precautionary approach is required as the alpine newt is an adaptable species (Heiss et al. 2013; McEwan 2015), and could possibly invade new types of habitat once introduced into a new region (Wells 2007).

If the alpine newt were to invade native frog habitats, it could gain access to retreat and breeding sites of both native frog species, being a small and elongate urodele. After hatching, Hochstetter’s frog larvae complete development in seepage waters near the oviposition site (Bell 1985; Beauchamp et al. 2010), and potentially could be preyed upon by newts there. The male Archey’s frog guards the young after they hatch (Bell 1985), offering possible protection from any predatory newt. Whether egg clusters of either species would be vulnerable to newt predation is uncertain, but unlikely. Smaller juvenile stages of both native frog species could be potential prey if newts were present (Table 2), although leiopelmatids have defense mechanisms that might render them unpalatable (Bell 2010), but they are eaten by bell frogs (Thurley & Bell 1994).

In turn, alpine newts might be at risk to native frog predation. The Maud Island frog (L. pakeka) will occasionally eat prey as large as an adult Wellington tree wētā (Hemideina crassidens) (B Bell, pers. obs.), so adult newts could be taken as well as younger size classes, at least by Hochstetter’s frog, which is larger than Archey’s frog (Turbott 1942; Bell 1978b). All European salamandrid species, including newts, are furnished with skin glands, which secrete an irritating or poisonous liquid of varying efficacy (Steward 1969). An important consideration in the context of newts as prey is the fact that, like other salamandrids, alpine newts produce tetrodotoxin (Wakely et al. 1966; Yotsu-Yamashita et al. 2007; McEwan 2015). The alpine newt is considered only mildly toxic, however, compared to fire-bellied newts already in the pet trade in New Zealand (see Tsuruda et al. 2001; Mochida et al. 2013). Tetrodotoxin could be harmful to New Zealand species which consume alpine newts and it accumulates in newt predators (William et al. 2004; McEwan 2015), and could possibly become more toxic in New Zealand (McEwan 2015). In Australia, the toxic impacts of the invasive smooth newt could potentially affect a wide range of native taxa there (Tingley et al. 2015).

Options for alpine newts introduced into Shropshire, UK, included their eradication before they could spread further, thereby avoiding any future ecological disruption that the newt might cause to native amphibian communities (Bell & Bell 1995). Since then they have been recognised as more of a problem, as potential carriers of diseases—two pathogenic chytrid fungus species (Bd and Bsal), as well as ranaviral disease, which have contributed to declines of amphibians in Europe and elsewhere (Berger et al. 1998; Stuart et al. 2004; Hoffmann et al. 2010; Martel et al. 2013, 2014; Zampiglia et al. 2013; Price et al. 2014; Winchester et al. 2015; Arntzen et al. 2016). For the security of native frog populations in New Zealand, the alpine newt incursion into the wild poses some risk (Table 2), indeed it may already have had some biosecurity impact as a disease vector. Future research could involve climate and habitat matching models in order to target New Zealand regions in which the newt might survive, as recommended for exotic reptiles in New Zealand (Kikillus 2010; Kikillus et al. 2010).

Conclusions

The establishment of the alpine newt in New Zealand represents the arrival of a new amphibian order, akin to the arrival of the smooth newt in Australia (Tingley et al. 2015). The alpine newt has the potential to detrimentally impact native frogs, especially as a carrier of disease. Japanese and Oriental fire-bellied newts, kept as pets in New Zealand, also carry a potential disease risk. It is easy to overlook the potential for amphibians to invade new habitats once introduced into a new region (Wells 2007), and the alpine newt could, given time, potentially invade native frog habitats. As a competitor, a predator and a slightly toxic prey species, it may present further risks, though these appear low at present (Arntzen et al. 2016). To eradicate the Coromandel population as soon as possible is therefore prudent and necessary, since it occurs geographically close to native frog habitats. The status of fire-bellied newts held as pets in New Zealand also needs review, given their potential to carry disease and the possibility of their accidental or illicit release into the wild.

Acknowledgements

I thank Paul Bell, Kerry Brown (DOC), Rhys Burns (DOC), Amanda Haig (DOC) and the late Tony Whitaker for providing material, including the photograph. I miss the opportunity of further discussing these issues with Tony, with whom I had frequent contact after the alpine newt incursion was first notified in 2013. I also thank Kelly Hare (Victoria University of Wellington), Jaap Knegtmans (MPI) and other reviewers for suggesting constructive changes.

Disclosure statement

No potential conflict of interest was reported by the author.