The conservation status of New Zealand bats, 2009

Abstract

The New Zealand Threat Classification System (NZTCS) is a national system used to assess the risk of extinction faced by New Zealand plants, animals and fungi. The system is specifically designed to be relevant to New Zealand's unusual ecological and geographic conditions. We undertook a re-evaluation of the status of seven bat taxa based on our knowledge of New Zealand bats using revised NZTCS criteria. Five taxa were listed as Threatened or At Risk: one as Nationally Critical (long-tailed bat Chalinolobus tuberculatus ‘South Island’), one as Nationally Endangered (southern lesser short-tailed bat Mystacina tuberculata tuberculata), two as Nationally Vulnerable (long-tailed bat ‘North Island’ and northern lesser short-tailed bat M. t. aupourica) and one as Declining (central lesser short-tailed bat M. t. rhyacobia). One taxon was assessed as Data Deficient (greater short-tailed bat M. robusta) and one (little red flying fox Pteropus scapulatus) as Vagrant. We suspect declines result primarily from predation and competition from introduced mammals, habitat degradation, and disturbance.

Keywords:

• threat classification
• Mystacina
• Chalinolobus
• Pteropus
• bats
• threatened species

Introduction

In 2002, a process and criteria for assessing the threat status of New Zealand's flora and fauna (the New Zealand Threat Classification System—NZTCS) was published (Molloy et al. 2002). The NZTCS is a national system led by the Department of Conservation (DOC). It is a tool that uses objective criteria and information drawn from a wide range of experts to rigorously assess the risk of extinction faced by New Zealand's plants, animals and fungi. Each taxon is placed in a category that reflects the level of risk it faces. The system is specifically designed to be relevant to New Zealand's unusual ecological and geographic conditions, resulting in a large number of local and regional endemics, which are naturally rare rather than under immediate threat of extinction (Townsend et al. 2008). It is intended to complement the world view provided by the IUCN Red Lists (IUCN n.d.).

Hitchmough (2002) presented the results of applying the NZTCS across a range of New Zealand taxa. An update was undertaken in 2005 (Hitchmough et al. 2007), which documented changes in the threat status of taxa and added new ones to the list. The number of taxa listed as threatened increased from 2372 to 2788, and the number listed as Data Deficient increased from 2047 to 3031. These changes were mostly a result of improved knowledge and the assessment of taxa that had not been considered previously, but a small number of taxa changed category as a result of genuinely improved or deteriorated status.

NZTCS listing has no direct or automatic impact on the legal status or resourcing of work on threatened species. However, it provides vital information for processes such as applying legal protection to species via amendments to the Schedules to the New Zealand Wildlife Act (1953). Although threat status is only one of several criteria used in prioritisation of conservation efforts (Joseph et al. 2008), classifying taxa according to the risk of extinction they face allows prioritisation of work and resources to those species that need them most. The published lists also form a basis for national outcome monitoring to measure the impact that DOC and other agencies and community groups have on their management of natural heritage. These indicators are reported nationally and internationally (Ministry for the Environment 2007; Hoare et al. 2010).

In 2007, a review of the NZTCS resulted in a new manual for classifying New Zealand taxa according to their threat of extinction (Townsend et al. 2008). The fundamental approach was unchanged from Molloy et al. (2002), but there were changes to some categories and criteria, and to the recommended process for evaluation of taxa. As part of the implementation of this revised system, we re-evaluated the threat status of New Zealand bats (Chiroptera) in 2009. This paper reports the results of our assessments. This followed parallel re-evaluations of the vascular plant flora (de Lange et al. 2009), and bird (Miskelly et al. 2008), reptile (Hitchmough et al. 2010), amphibian (Newman et al. 2010), marine mammal (Baker et al. 2010) and marine invertebrate (Freeman et al. in press) faunas.

Three bat species, all of which are endemic, are known from New Zealand: the lesser short-tailed bat (Mystacina tuberculata), greater short-tailed bat (M. robusta) and long-tailed bat (Chalinolobus tuberculatus) (O'Donnell 2010). The lesser short-tailed bat and greater short-tailed bat belong to the endemic and mono-generic family Mystacinidae, whilst the long-tailed bat is a member of the large and widespread family Vespertilionidae and is endemic at the species level. There have also been occasional cases of vagrants arriving, but all except the little red flying fox (Pteropus scapulatus) have been accidental imports that arrived dead with freight (Daniel 1975; O'Donnell 1998, 2005).

Methods

Taxonomic units assessed

In this paper we use ‘taxa’ to collectively refer to both described and undescribed taxa. Undescribed (a subset of ‘taxonomically indeterminate’; Townsend et al. 2008) are entities believed to be distinct but not yet furnished with a formal name.

We assessed seven bat taxa: greater short-tailed bat (one taxon), lesser short-tailed bat (three taxa), long-tailed bat (two taxa) and little red flying fox (one taxon). Lesser short-tailed bats are described as three sub-species (Hill & Daniel 1985), although significant variation in genetic diversity within each sub-species has been recorded (Lloyd 2003a, b). In 2002, Hitchmough (2002) ranked threat status of the three sub-species separately (Table 1). However, recent genetic studies of the conservation status of lesser short-tailed bats (Lloyd 2003a, b) led Hitchmough et al. (2007) to split two of the sub-species into four potential ‘Evolutionary Significant Units’, which led to eight taxa of concern being evaluated (Table 1). Lloyd (2003a, b) identified six highly divergent lesser short-tailed bat matrilines, which diverged simultaneously 680,000 to 930,000 years ago. This is a divergence time characteristic of many mammalian sub-species and several avian species, which is why these lineages were used during the previous ranking (Hitchmough et al 2007). In this re-evaluation, we chose not to use the matrilines. They occur sympatrically in many populations, but they cannot be identified without detailed genetic analysis and, consequently, cannot be managed as separate units. We therefore used the original taxonomy of Hill & Daniel (1985).

Table 1  Conservation status of New Zealand bats in the 2002 and 2005 assessments.

Current scientific name^1	Taxa assessed	Threat Status 2002	Threat Status 2005
Chalinolobus tuberculatus	1. Long-tailed bat ‘North Island’	Nationally Vulnerable	Nationally Vulnerable
C. tuberculatus	2. Long-tailed bat ‘South Island’	Nationally Endangered	Nationally Endangered
Mystacina robusta	3. Greater short-tailed bat	Data Deficient	Data Deficient
M. tuberculata aupourica	4. Northern lesser short-tailed bat	Nationally Endangered	Nationally Endangered
M. t. rhyacobia	5. Central lesser short-tailed bat	Range Restricted	5a. Eastern short-tailed bat Range Restricted^2 5b. North-western short-tailed bat Range Restricted
M. t. tuberculata	6. Southern lesser short-tailed bat	Nationally Endangered	6a. Southern North Island short-tailed bat Nationally Critical^2 6b. South Island short-tailed bat Nationally Endangered
Pteropus scapulatus	7. Little red flying fox	Not Assessed	Not Assessed
^1Taxonomy follows Hill & Daniel (1985) and Daniel (1975); ^2Assessed as different Evolutionary Significant Units (O'Donnell 2010)

Long-tailed bats are currently described as a single species. However, there appear to be significant differences in size (O'Donnell 2005), echolocation calls (Parsons 1997) and genetic diversity (Winnington 1999) between populations in the North and South Islands. It is not yet clear whether these differences warrant sub- specific status; therefore, the New Zealand Bat Recovery Group has recommended that the long-tailed bat is treated as two separate taxa as a precaution until the taxonomy of this species has been finalised (O'Donnell 2010).

Assessment procedure

The re-evaluation was undertaken during a 1-day meeting of experts (the authors) with a range of affiliations and experience on 14 December 2009. We incorporated information from the public and a broader pool of experts not directly involved in the listing process. A call for submissions on the threat status re-evaluation of bats was made via the New Zealand Department of Conservation website (http://www.doc.govt.nz/getting-involved/consultations/closed/new-listing-of-threatened-status-of-new-zealand-bats/) on 22 December 2008. Submissions closed on 10 April 2009. Eight submissions were received. The role of the expert panel members was to provide knowledge on their particular field of expertise at the threat classification list meeting, to answer queries on listing decisions reached, and to consult with peers to bring as much information as possible to the meeting (Townsend et al. 2008).

The panel placed taxa into threat categories based on the criteria provided by Townsend et al. (2008). This process was guided by submissions received, panel knowledge and reference to recent publications relating to taxonomic and population status.

The categories used in our evaluation (as defined by Townsend et al. 2008) were:

1. Extinct

2. Threatened (including Nationally Critical, Nationally Endangered and Nationally Vulnerable)

3. At Risk (including Declining, Recovering, Relict and Naturally Uncommon)

4. Not Threatened

5. Non-resident Native (including Coloniser, Migrant, and Vagrant)

6. Introduced and Naturalised

7. Data Deficient

The first four categories include taxa in order of their threat status (1 being most threatened, 4 being least threatened). Many Data Deficient taxa are also likely to be under threat.

The NZTCS offers a number of criteria (in status and trend sub-sets) for assessing each taxon under each threat category. All taxa were classified using both status and trend criteria. Status criteria (total number of mature individuals, or total number of populations, or number of mature individuals in the largest population, or area of occupancy of the total population) were generally considered first, followed by an evaluation of the trend criteria (ongoing or predicted population trend, due to existing threats, measured either by population size or area of occupancy). A series of ‘qualifiers’ (e.g. Data Poor, One Location, Partial Decline, Conservation Dependent, Range Restricted) was also available to enable additional information on each taxon to be captured and considered (Townsend et al. 2008). We defined area of occupancy as the cumulative home range (ha) of maternity roosts within a population. Generation time is defined as the average age of breeding bats in any one breeding season (IUCN 2001).

Results

Seven New Zealand bat taxa were assessed (Table 2), with five classed as Threatened or At Risk: one as Nationally Critical (long-tailed bat ‘South Island’), one as Nationally Endangered (southern lesser short-tailed bat), two as Nationally Vulnerable (long-tailed bat ‘North Island’ and northern lesser short-tailed bat) and one as Declining (central lesser short-tailed bat). One taxon was assessed as Data Deficient (greater short-tailed bat) and one (little red flying fox) as Vagrant (Table 2). No taxa were considered Not Threatened and there are no Introduced and Naturalised taxa in New Zealand.

Table 2  Conservation status and population parameter estimates and qualifiers for assessing conservation status of New Zealand bats, 2009.

Current scientific name^1	Common name of taxon	Threat Status 2009	Estimated number of viable populations	Estimated total population size	Estimated population decline over three generations (%)	Qualifiers (see text for definitions)	Sources of population estimates
Chalinolobus tuberculatus	Long-tailed bat ‘North Island’	Nationally Vulnerable	>15	20,000−100,000^2	50–70	Data Poor	Gillingham 1996; Alexander 2001; O'Donnell 2002b, 2005; Pryde et al. 2006; Borkin & Parsons 2009; Dekrout 2009; Borkin & Parsons 2010a
C. tuberculatus	Long-tailed bat ‘South Island’	Nationally Critical	10	<5000	>90		Griffiths 1996; O'Donnell 2000a,c; Sedgeley & O'Donnell 2004; Pryde et al. 2005
Mystacina robusta	Greater short-tailed bat	Data Deficient	1	Unknown	Unknown	One Location	O'Donnell 1999; Lloyd 2005
M. tuberculata aupourica	Northern lesser short-tailed bat	Nationally Vulnerable	2	c. 5000	Stable	Partial Decline, One Location, Conservation Dependent	Arkins et al. 1999; Lloyd 2003a,b, 2005; B. Barr, N. Miller, I. Petrove (pers. comm.)
M. t. rhyacobia	Central lesser short-tailed bat	Declining	8	c. 40,000	30	Data Poor, Range Restricted	Lloyd 2003a,b, 2005
M. t. tuberculata	Southern lesser short-tailed bat	Nationally Endangered	3	<4,000	>60	Partial Decline, Conservation Dependent, Range Restricted	O'Donnell et al. 1999; Sedgeley & Anderson 2000; Sedgeley 2001, 2003; Lloyd 2005; C. O'Donnell unpubl. data;
Pteropus scapulatus	Little red flying fox	Vagrant	0	0	0		Daniel 1975
^1Taxonomy follows Hill & Daniel (1985) and Daniel (1975); ^2Category from Townsend et al. 2008

Discussion

Threats

A wide range of threats to the continued viability of bat populations has been identified. The geographic range and numbers of bats have declined significantly and in many areas declines are continuing (e.g. O'Donnell 2000a; Lloyd 2003a, 2005; Pryde et al. 2005, 2006). Declines result from a combination of threats, primarily predation and competition from introduced mammalian pests, habitat degradation, disturbance and, possibly, disease (see below).

Introduced stoats (Mustela erminea), rats (Rattus spp.), possums (Trichosurus vulpecula) and cats (Felis catus) all prey on, or attempt to prey on, New Zealand bats (Dwyer 1962; Daniel & Williams 1984; O'Donnell 2000a, b; Lloyd 2005; Pryde et al. 2005; J. Scrimgeour, pers. comm.). Bats are vulnerable to predators throughout the year; in summer when they congregate in large colonies to give birth and rear young (O'Donnell & Sedgeley 1999), and during winter when they may remain inactive (in torpor) for long periods within roosts (P. Webb, Clarke Webb Ecology Ltd, pers. comm.). The precise impacts of predation are uncertain. The disappearance of greater short-tailed bats from mainland New Zealand coincided with the spread of the Pacific rat or kiore (R. exulans) (Worthy 1997), and this species of bat subsequently became extinct during the 1960s when ship rats (R. rattus) arrived on its last known refuge, Big South Cape Island off Stewart Island/Rakiura (Daniel 1990). On the mainland, significant declines in the survival of long-tailed bats from three colonies in the Eglinton Valley were strongly correlated with irruptions of both stoats and ship rats (Pryde et al. 2005).

There has been speculation that both introduced mammals and native and introduced bird species may compete with lesser short-tailed bats for food sources. Many of the types of fruit eaten by lesser short-tailed bats are also eaten by many bird species, rodents and possums (Daniel 1976). Molloy (1995) suggested introduced wasps, which feed on nectar, fruit and insects, may compete with lesser short-tailed bats. Ecroyd (1995) demonstrated that ship rats and possums all fed on significant amounts of nectar from flowers of wood rose/pua o te reinga, Dactylanthus taylorii, making it unavailable to lesser short-tailed bats. Possums cause serious damage to native ecosystems by selective browsing; they also feed on buds, flowers and fruits of a wide variety of native trees and shrubs and have also caused mortality of native trees (Cowan & Waddington 1990). By doing this they are modifying bat habitat and, possibly, competing for some foods with lesser short-tailed bats.

Although any pest control programmes that target possums, rodents, stoats or feral cats are likely to be of benefit to bat populations, there are some risks associated with operations that use toxins. Lesser short-tailed bats are potentially susceptible to toxins because of their broad diet and habit of feeding on the ground (Arkins et al. 1999; Daniel 1976, 1990; Eason & Spurr 1995; Lloyd & McQueen 2000, 2002). One lesser short-tailed bat has been found dead on cyanide bait, which is used for possum control (Lloyd 2005). At least 125 lesser short-tailed bats died following a diphacinone operation at Pureora in 2008 (D. Smith, DOC, pers. comm.). However, losses of short-tailed bats following this operation were unlikely to be typical because in this operation the bait matrix was in the form of a paste and was presented in bait bags stapled to tree trunks in the open and close to the ground, rather than concealed in bait stations. Uneaten bait from those bags that were disturbed by rats or possums could have fallen to the ground, perhaps increasing the chances of bats feeding on the baits or on ground-feeding invertebrates that had fed on the baits. Previous work has shown that cereal baits, which are typically used, generally appear to be unattractive to bats and no primary or secondary mortality has been recorded following their use with 1080, brodifacoum and pindone toxins (Lloyd 1994; Lloyd & McQueen 2002; Sedgeley & Anderson 2000; C. O'Donnell unpubl. data), despite there being instances where bats in captivity have shown interest in them (Beath et al. 2004). However, we are concerned about continuing use of diphacinone in areas inhabited by short-tailed bats, as bats are known to be susceptible to diphacinone (Eason & Wickstrom 2001). Diphacinone is used to control vampire bats (Desmodus rotundus) by systemic treatment of livestock on which vampire bats feed (Thompson et al. 1972; Schmidt & Badger 1979). When bats feed on blood from cattle within 72 hours of the cattle receiving an injection of 1 µg g^-1 of diphacinone, the bats receive a lethal dose. Considerably higher levels of diphacinone than those that kill vampire bats are reported from tree wētā (Hemideina crassidens) feeding on diphacinone baits (Fisher et al. 2007). There are several reports of wētā and other ground-dwelling invertebrates feeding on toxic baits during pest control operations (Fisher et al. 2007). It therefore seems reasonable to assume that short-tailed bats will be vulnerable to secondary poisoning during rodent control operations with diphacinone.

Habitat loss, while significant in the past, is less of a contemporary threat. Early records note the disturbance of large colonies of long-tailed bats while European colonists were burning and felling trees for timber, and clearing land for agriculture or for mining (Buller 1892; Cheeseman 1893). Dwyer (1960, 1962) concluded that the decrease in the area of distribution of bats was correlated with the removal of indigenous forest during the 1800s and the failure of either bat species to survive in open country or urban areas. Disappearance of long-tailed bats from coastal and lowland regions in areas such as Canterbury, Otago and Southland coincided with the loss of forest cover (Hutton & Drummond 1904; Dwyer 1960). In the 1900s, logging generally targeted a significant proportion of trees preferred by bats for roosting (Sedgeley & O'Donnell 1999a, b). Logging of indigenous forests on public-owned land has now ceased (Griffiths 2002). The logging that does occur on private land must now operate under ‘Sustainable Management Plans’ which are required to demonstrate that safeguards are in place to ensure that bat populations are not threatened and sufficient large trees are always available to allow bat population survival (O'Donnell 2010). However, long-tailed bats have been recorded in commercial plantation forests, a highly dynamic habitat that has recently been subject to extensive logging to enable the land to be converted to pasture (Borkin & Parsons 2010a, Borkin & Parsons 2010b; Ministry of Agriculture and Forestry 2010). At a local level, fragmentation and incremental habitat loss continue in bat habitats because of the conversion of indigenous shrubland to exotic pine plantations, logging of roosts for firewood, the increased intensification in agriculture and the clearance of trees on road verges (Sedgeley & O'Donnell 2004; O'Donnell 2010).

Risk of disease is largely unknown. There is one instance of lesser-short-tailed bats contracting an unknown skin disease following translocation to Kapiti Island (Ruffell & Parsons 2009). Eight bats captured 232 days after release had scabbing on the distal end of their pinnae, while an additional two animals had severe swelling at the basal end. One bat also had balding around the groin; another had balding on the back. All eight bats were treated with Ivermectin and the antibiotic Enrofloxacin. Treatment with antibiotics temporarily halted secondary infections, but for most diseased individuals, amputation of the external pinnae was required to resolve the problem. The disease would probably cause a high mortality rate in a wild population of free-living short-tailed bats.

Long-tailed bat

Long-tailed bat ‘South Island’

Long-tailed bats have been recorded widely through the South Island but are now estimated to have a small population size (Table 2). There are no populations of South Island long-tailed bats on pest-free islands. Principal research on this taxon has occurred at only two sites: in the Eglinton Valley in Fiordland and the Geraldine area in South Canterbury. Capture-mark-recapture studies have been used to calculate survival and growth rates of these populations. Rapid declines have been predicted at a rate of 5% per annum in the Eglinton Valley (Pryde et al. 2005) and 9% per annum in the Geraldine area (Pryde et al. 2006). Past declines in the Eglinton Valley have coincided with irruptions of rats and mustelids, whereas those in Geraldine have coincided with predator visitations to roosts as well as accelerated rates of habitat loss (O'Donnell 2000b; Sedgeley & O'Donnell 2004). Further declines of a similar order have been recorded in three colonies in recent years in the Eglinton Valley coinciding with irruptions of mammalian predators (2006−2009, C. O'Donnell unpubl. data). Area of occupancy was estimated at >100 ha but less than < 1000 ha, based on ranges of individual breeding females being in the order of 20−123 ha (O'Donnell 2001) and collective colony roosting ranges, including non-breeding roosts, of 426−1391 ha (O'Donnell 2000c).

After extrapolating these rates of decline to the taxon, we classified long-tailed bat ‘South Island’ as ‘Nationally Critical’ based on Criterion C of the classification: that their population exhibited an ongoing trend or predicted decline of >70% in the total population due to existing threats taken over the next 10 years or three generations, whichever is longer (Townsend et al. 2008). At the rates of decline that have been measured, we estimate the decline will be in the order of >90% over the next three generations, based on a conservative estimate that a generation is 12 years. Small, echolocating bats are thought to be long lived (frequently >30 years, maximum age 38 years; Khritankov & Ovodov 2001; Wilkinson & South 2002). Most female long-tailed bats breed for the first time at 2−3 years old (1−4 y, O'Donnell 2002a), and the oldest known banded female is at least 20 years old (C. O'Donnell unpubl. data).

Long-tailed bat ‘North Island’

Long-tailed bats have been recorded widely throughout the North Island and also occur on Little Barrier and Kapiti Islands (O'Donnell 2005; Borkin & Parsons 2010a), both now free of exotic mammalian pests. Number of populations and total population size is unknown, but we suspect there are >15 populations (Table 2). There has been only one study of survivorship (Pryde et al. 2006). In this study, at Grand Canyon Cave in the King Country, it was uncertain whether the population was stable or declining slowly, but the population was large (>700 bats, O'Donnell 2002b). The panel considered that given the general weight of evidence about the impacts of introduced mammalian predators and habitat loss on long-tailed bats in the South Island, similar mechanisms of decline are likely to be operating in the north. Most populations monitored recently appear to be relatively small (based on average emergence counts from roosts of generally fewer than 15 individuals; Alexander 2001; Dekrout 2009, K. Borkin, University of Auckland, pers. comm.) except on the East Coast of the North Island where the average emergence was 86 bats (Gillingham 1996). Therefore we classified long-tailed bat ‘North Island’ as Nationally Vulnerable because they are thought to have a large population but a high ongoing or predicted decline. We assessed the taxon as falling into the trend description under Criterion E1: we predict an ongoing decline of 50–70% in the total population over the next three generations. However, the population is still likely to be markedly larger than the South Island taxon (>20,000 mature individuals) because there are reports of long-tailed bats from much larger areas of the North Island. We considered area of occupancy to be in the order of >1000 ha based on two radio-tracking studies in the King Country, which recorded roosting ranges of >100 ha per colony (C. O'Donnell unpubl. data).

Greater short-tailed bat

We were unable to evaluate the threat classification for greater short-tailed bats because of a lack of information on their status. We therefore designated them as Data Deficient. Until recently, greater short-tailed bats have been considered extinct because there have been no confirmed sightings since 1967 (Daniel 1990). However, unusual mystacinid-like echolocation calls were recorded using bat detectors on Putauhina Island, approximately 1 km from Big South Cape (Taukihepa) Island in 1999 (O'Donnell 1999). The last confirmed sighting of this species was on Big South Cape Island, which led to speculation that greater short-tailed bats might still be extant (O'Donnell 1999). Since then there have been a number of reports of bats being seen on both Putauhina and Big South Cape Islands (P. Schofield, M. Charteris, P. McClelland, Canterbury Museum and DOC, pers. comm.). A recent (spring 2009) 3-day visit to these islands failed to detect any evidence of greater short-tailed bats, although the weather was poor and the trip brief (C. O'Donnell unpubl. data).

Lesser short-tailed bat

Northern lesser short-tailed bat

The northern short-tailed bat is known from areas of indigenous forest in Northland and from Hauturu/Little Barrier Island (Lloyd 2005). Population size and trend has not been studied except on Hauturu/Little Barrier Island, but lesser short-tailed bats are thought to be abundant there and number in the thousands (Table 2; Arkins et al. 1999; B. Lloyd, S. Parsons unpubl. data). Sightings have been rare in mainland Northland despite numerous surveys using bat detectors and it is thought that, compared with the 1970s, the taxon is in serious decline there (Daniel 1979; Lloyd 2005; B. Barr, N. Miller, I. Petrove, DOC, pers. comm.). There have been no confirmed reports of lesser short-tailed bats from the last known Northland population at Omahuta-Puketi since 2002. Hauturu/Little Barrier Island has been free of introduced mammalian predators since 2004 and the population of lesser short-tailed bats there is thought to be abundant and secure (R. Griffiths, DOC, pers. comm.). Area of occupancy is unknown but is probably >1000 ha based on Hauturu/Little Barrier Island being 3052 ha in size. Therefore, we classified northern short-tailed bats as Nationally Vulnerable based on our prediction that the overall population is stable (±10%) and is predicted to remain stable over the next 10 years or three generations, whichever is longer. We estimate the population to still be small (in the order of 5000 mature individuals; Criterion B1 of the classification). Qualifiers to this classification are that the secure population only occurs in one location, and that a partial decline is occurring, with the taxon declining over the majority of its range, but secure in one location. Retaining this status is conservation dependent, relying on Little Barrier Island remaining predator free and island quarantine procedures being effectively maintained.

Central lesser short-tailed bat

This taxon is present in indigenous forests of the central North Island, particularly the Urewera Ranges, Whirinaki Forest, Kaimanawa Ranges, Mt Ruapehu, Pureora Forest and northern Taranaki (Lloyd 2005). Short-tailed bats have been reported also commuting and foraging within central North Island Pinus radiata plantations (Borkin & Parsons 2010b). However, reported roosts within these forests are limited to long-dead native trees (Borkin & Parsons 2010b). The total population was estimated as being in the order of 40,000 individuals in eight populations in 2002 (Table 2; Lloyd 2003a, b). A small population in the Tararua Ranges, previously grouped with southern lesser short-tailed bats, has been re-assigned to this taxon on the basis of recent genetic analyses (Lloyd 2003a). We classified central short-tailed bats as Declining. Declining taxa do not qualify as Threatened according to Townsend et al.'s (2008) criteria because they are buffered by a large total population size and/or a slower decline rate. However, if the declining trends continue, such taxa may be listed as Threatened in the future. We classified this taxon according to Criterion B1: the total population size is 20,000–100,000 mature individuals, but there is an ongoing or predicted decline of 10–50% in the total population or area of occupancy due to existing threats, taken over the next 10 years or three generations, whichever is longer. We estimated that the population may be declining at a rate in the order of 1% per annum (>30% in three generations). Area of occupancy is unknown but is probably >1000 ha. Predation by introduced mammals is a major concern, because cats, stoats and rats have been recorded visiting roosts (Lloyd 2005; J. Scrimgeour, pers. comm.). In a recent example of the potential impacts of predators, a feral cat caught and killed a minimum of 102 bats over seven days at a single roost site near Ohakune (J. Scrimgeour, pers. comm.). Furthermore, there are no populations of central lesser short-tailed bats secure on predator free offshore islands. Other threats include habitat loss through logging in some areas, disease and potential secondary poisoning from use of pest control toxins (Lloyd & McQueen 2000; Griffiths 2002; Ruffell & Parsons 2009).

Southern lesser short-tailed bat

This taxon is only known from three locations: Oparara in north Westland, Eglinton Valley in Fiordland and Codfish Island/Whenua Hou near Stewart Island/Rakiura (O'Donnell et al. 1999; Lloyd 2005). The Codfish Island population numbers c. 2000 individuals (Sedgeley & Anderson 2000) and is likely to be secure following the eradication of the last introduced mammalian pests in 1999. The Eglinton valley population is estimated at c. 1800 individuals (C. O'Donnell unpubl. data), but the population is vulnerable to predator irruptions. Forty percent of tagged short-tailed bats alive prior to a rat population irruption in 2006 (following heavy masting of beech Nothofagus spp.) were recorded the following summer (2007), compared with the 76% of tagged bats that were recorded the following year when predator numbers were low (J. Sedgeley & C. O'Donnell unpubl. data). This was similar to c. 30% mortality in long-tailed bats in the same area in years when predator numbers are high (Pryde et al. 2005; C. O'Donnell & M. Pryde unpubl. data.). Area of occupancy of maternity roosts is 17 ha in the Eglinton Valley (Christie 2003) but on Codfish Island is >500 ha (C. O'Donnell unpubl. data). Short-tailed bats are also thought to have declined significantly in the Oparara area since their discovery there in 1996, and there have been few sightings of them in recent years, despite numerous intensive surveys (Greaves 2004; Lloyd 2010; C. Woods, B. Chalmers, R. McClellan, DOC, pers. comm.).

We classified southern lesser short-tailed bats as Nationally Endangered. We estimate the total population size is <5000 mature individuals and there is an ongoing or predicted decline of 50–70% in the total population due to existing threats, taken over the next 10 years or three generations, whichever is longer (Table 2). The major threat is predation by introduced mammalian predators. In comparison with northern short-tailed bats, a much higher proportion of the population (c. 50% cf. <10%) is known to be under significant threat.

Little red flying fox

Little red flying fox, which is an Australian species, was assessed as a Vagrant based on there being one confirmed record in Hamilton in 1927−1929, at a time when there were unusually large migrations of this species into south-eastern Australia (Daniel 1975).

Adequacy of the assessment

Overall, there is some uncertainty about the conservation status of New Zealand's bat taxa because detailed data on trends of all populations are unavailable and it is not known whether all threats operate in all populations and at the same rates. Studies from the South Island provide most of the detailed analyses of rates of decline (Pryde et al. 2005, 2006; C. O'Donnell unpubl. data). We are uncertain if the same factors causing decline in the South Island are also operating in the North Island. We have made the assumption that causal factors are similar across most mainland populations, but acknowledge uncertainty about rates of decline. Potentially, the conservation status of taxa in the North Island may be worse than our classification suggests. Predation by introduced mammals is a major concern in the North Island and we may have underestimated its impact on these populations. The worsening of conservation status in both South Island bat taxa (see below) is associated with increased knowledge of these populations. This suggests that, given the same level of study, North Island bat populations would also likely undergo a change in status, given they are subject to the same threats. In addition, it is not known how present forestry operations may be affecting the health and long-term viability of bat populations within plantation forests (Borkin & Parsons 2010a).

Two other areas of uncertainty in assigning conservation status to bats are of concern to us. The first concern is that all bat taxa in New Zealand are particularly vulnerable to predation and other catastrophes because of the colonial nature of their roosts. The most vulnerable life history stage is when bats are gathered in their maternity roosts. Both long-tailed bats and lesser short-tailed bats breed in trees, and colonies of both taxa can occupy as few as 20−30 primary trees, regardless of the size of their roosting ranges (O'Donnell 2001; Christie 2003; Lloyd 2005). If a predator attacks a maternity colony, or there is a disease outbreak, a large number of bats can be killed at one time (Pryde et al. 2005; J. Scrimgeour, pers. comm.). Although we considered area of occupancy of maternity roosting ranges in our assessments, and area of occupancy was small relative to the large foraging ranges of bats (O'Donnell et al. 1999; O'Donnell 2001; Christie 2003), our use of this criterion may still under estimate their vulnerability to decline.

Our second concern is that the safe status of single populations on offshore islands led us to assign threat categories that did not fully reflect the vulnerability of taxa to extinction. This concern has been noted for other animal groups (e.g. reptiles, Hitchmough et al. 2010). This anomaly shows in the rankings for northern and southern lesser short-tailed bats. Both taxa have secure offshore island populations. However, while the mainland population of the southern taxon now has some security because it is the focus of major conservation management to reverse declines, the mainland population of the northern taxon, which is not presently managed, is sliding towards extinction. The ranking anomaly results from c. 50% of the population of southern lesser short-tailed bats still being under threat of extinction if management is ineffective. In comparison, we judged the mainland portion of the northern population (the most threatened portion) to be small; while the island population (which now represents >95% of the total population) is relatively safe, which, according to NZTCS criteria, makes the taxon comparatively secure.

Changes in status since the last evaluation

The threat categories of three New Zealand bat taxa have changed since 2004 because of genuine changes in their conservation status. The status of two taxa has worsened. An accelerated rate of population decline has been detected in long-tailed bat ‘South Island’ based on further and more precise measures of survival obtained during study of bat populations subject to a series of predator irruptions in the Eglinton Valley, and continued declines in the managed population in the Geraldine area (Pryde et al. 2005, 2006). The status of southern lesser short-tailed bats also declined because of an accelerated rate of population decline. This decline went undetected earlier because we were unable to monitor any populations accurately. However, since 2006 we have microchipped lesser short-tailed bats in the Eglinton Valley and monitored their response to fluctuations in rat density (C. O'Donnell unpubl data). Declining population trends similar to those of long-tailed bats ‘South Island’ have been observed. In addition, the continued failure to re-find the population of southern lesser short-tailed bats in the Oparara catchment (Lloyd 2010) has reinforced concern about the continued survival of this population.

The conservation status of one bat taxon – northern lesser short-tailed bat – improved overall as a result of conservation management action (the eradication of rats from Little Barrier Island). However, the status of the same taxon on the mainland is still declining towards likely imminent extinction, with no recent reports of bats from Omahuta-Puketi, the location of the only known mainland Northland population.

Future prognosis

Although New Zealand bats are under considerable threat, the Department of Conservation has an active recovery programme for them and is entering a phase where attempts will be made to reverse declines. Over the last 15 years, a wide range of threats have been identified and intensive ecological studies have been undertaken in an effort to elucidate threats and improve knowledge to aid the development and implementation of conservation management techniques. Following this initial intensive research phase, active conservation management measures are now beginning to be implemented. These are described by O'Donnell (2010) and include: using legal mechanisms for protection; general advocacy and education; developing community-based conservation initiatives; control of introduced predators at key sites; active protection of roosts sites; protection of aquatic and terrestrial foraging habitats; development of a variety of habitat restoration techniques; and translocations to habitats free of introduced mammalian predators. Recently, declines in both long-tailed and lesser short-tailed bat populations have been reversed when targeted rat control has been undertaken in Fiordland beech forest (C. O'Donnell unpubl. data). However, there are still serious concerns about whether declines can be reversed in areas where bats are now rare, threats are unmanaged, and in taxa where no secure island populations exist.

Acknowledgements

We thank Ben Barr, Richard Griffiths, Jonathon Miles, Nigel Miller, Oliver Overdyk, Irene Petrove, Rory Renwick, Jessica Scrimgeour, Dave Smith and Thelma Wilson for their input into the relisting process and Jo Hoare, Don Newman and Lynette Clelland for constructive comments on the manuscript.