McMurtrie, P.; K A. Edge, D. Crouchley, D. Gleeson, M.J. Willans, and A.J. Veale. Eradication of stoats ( Mustela erminea ) from Secretary Island, New Zealand
Eradication of stoats ( Mustela erminea ) from Secretary Island, New Zealand
P. McMurtrie 1, K-A. Edge 1, D. Crouchley 1, D. Gleeson 2, M. J. Willans 3, and A. J. Veale 4 1Department of Conservation, Te Anau Area Offi ce, PO Box 29, Lakefront Drive, Te Anau 0640, New Zealand.
INTRODUCTION Stoats ( Mustela erminea ) were fi rst introduced into The scale and experimental nature of this programme mainland New Zealand in the late 1880s in response to required that, in addition to determining whether the feral rabbit ( Oryctolagus cuniculus ) plagues that were outcome of these objectives is met, opportunities for destroying pasturelands and posing a serious threat to the learning must also be undertaken. Stoats captured after New Zealand economy. Stoats have had dramatic effects the initial eradication campaign could be used to determine on New Zealand’s naïve native animal species, many of the level of subsequent reinvasion to Secretary Island. which evolved without terrestrial predators (King 1984). Here we present all trapping data from 2005 to December Stoats are very mobile and are capable swimmers (Taylor 2009 and a preliminary analysis of stoat reinvasion using and Tilley 1984) and were observed by Richard Henry, molecular genetic techniques described by Gleeson et al. curator on Resolution Island in Fiordland, by 1900 (Hill (2010). A demographic study of the Secretary Island stoat and Hill 1987). The stoats probably invaded other remote population prior to trapping and after the initial knock islands in Fiordland, including Secretary Island, at around down is underway (A. Veale unpubl. data). Diet analysis the same time. on the original trapped population has been undertaken but Secretary Island is administered by the New Zealand is not included in this paper (E. Murphy pers. comm.). Department of Conservation. In 2004, it became the focus of a 10 year programme to eradicate stoats and red METHODS deer ( Cervus elaphus ) (Edge et al. 2011). Techniques for eradicating stoats from islands had been piloted Study area successfully on several smaller islands in Fiordland (Elliott Secretary Island (8140 ha; 1196 m), at the entrance et al . 2010). Stoats had also been trapped on 19 islands to Doubtful Sound on the western coastline of Fiordland ranging in size from 1 67 ha and within varying distances National Park (Fig. 1), is the second largest island on from the mainland over a four year period, in order to the Fiordland coast and the third highest island in New produce a predictive model of stoat reinvasion; of 46 stoats Zealand. The island is separated from the mainland portion captured, only one was caught on an island further than of Fiordland by Thompson Sound to the east (minimum 304 m offshore. Based on these results, Elliott et al . (2010) distance between the two is C. 950 m), and by Doubtful concluded that large islands like Secretary and Resolution Sound to the south (minimum distance to closest stoat would be suitable for stoat eradication attempts. Our paper population is two passages C. 170 m and 600 m via Bauza describes an approach to the eradication of stoats from Island which is largely stoat free). Secretary Island based on an operational plan with two key In 1963, Secretary Island was designated a ‘Special objectives (Golding et al. 2005): Area’ within Fiordland National Park by the New Zealand 1. To eradicate stoats from Secretary Island. The Government due to the island’s unmodifi ed vegetation plan defi ned ‘eradication’ as the complete removal of the and the real (or apparent) absence of introduced browsing resident stoat population and the establishment of a long or grazing animals including the brushtail possum term control and monitoring programme to manage re (Trichosurus vulpecula and red deer. Introduced rodents invasion. were also absent, making Secretary Island the largest inshore island in New Zealand free of such pests. Stoats 2. To achieve and maintain a zero density stoat were the only mammalian pests known to be present. Red population on Secretary Island so that indigenous species deer probably arrived in the late 1950s (Mark and Baylis currently existing on the island or introduced to the island 1975), but it took some time for a population to establish can thrive. (Crouchley et al . 2011).
Pages 455-460 In: Veitch, C. R.; Clout, M. N. and Towns, D. R. (eds.). 2011. Island invasives: eradication and management. IUCN, Gland, Switzerland. 455 Island invasives: eradication and management
Stoat trapping on Secretary Island Full details of methods for the trapping programme on Secretary Island are provided in the Operational Plan (Golding et al . 2005). In brief, these involved the following techniques. A 108 km network of trap lines was established on Secretary Island from October 2004 to April 2005 along main ridge lines and spurs, habitat boundaries, waterways and traversable terrain (Fig. 2). Based on previous successful eradications of stoats from islands in Fiordland, we needed a minimum average density of one trap tunnel per 9 ha (Elliott et al. 2010.). We also needed to ensure that every stoat on the island would encounter a trap (Parkes 1990; Parkes et al . 2002). Home range estimates for stoats vary according to gender, season, and food availability, so it was important to determine the smallest likely home range in order to decide the maximum spacing between traps. Home ranges are smallest when prey, especially rodents, is easily available. For example, average home ranges were 93 (SE±7) ha for four male stoats and 69 (SE±8) ha for fi ve female stoats in a Fiordland beech ( Nothofagus sp.) forest when rodents were abundant (Murphy and Dowding 1995). Larger stoat home ranges were reported in areas where rodents are scarce, with estimates of 204 ha for males and 124 ha for females (Murphy and Dowding 1995), 223 ha for males and 94 ha females (Alterio 1998), and 210 ha for males and 89 ha females (Miller et al . 2001). Without similar home range information for stoats on Secretary Island, our trap network was based on the smaller home range sizes of Murphy and Dowding (1995), which meant Fig. 1 Location of Secretary Island within Fiordland that wherever possible traps should be no more than 700 National Park. m apart. However, due to the extremely rugged terrain on the island there were seven locations where this distance exceeded 700 m (see Fig. 2). A total of 945 tunnels each containing 2 Mark IV Fenn kill traps (DB Springs Ltd. Worcestershire, England) were placed at 135 m intervals along the trap lines and at 150 m intervals along the eastern coastline (accessible by boat), yielding an average tunnel density of 1 tunnel per 8.6 ha (Fig. 2). Two tunnel types were used to house traps: 300 wire mesh tunnels with wooden bases and 645 wooden tunnels with wire mesh ends. Wooden and wire mesh tunnels were placed in a repeated sequence along trap lines comprising one wire tunnel followed by two of wood. The variation in tunnel types was used to overcome any possibility that a few stoats were unwilling to enter either one type of tunnel. Previous stoat eradications in Fiordland used pre baiting whereby stoats were free to enter traps set with the safety catch on and with bait was left inside and outside the trap entrance. It is not possible to determine how crucial pre baiting has been to the success of these programmes. Pre baiting is relatively inexpensive and the amount of bait take observed during the pre baiting phase suggests it may reduce the time taken to achieve the initial knock down. Traps were pre baited twice on Secretary Island: 20 June 26 June and 5 July 11 July 2005. During pre baiting, each tunnel site was baited with one fresh hen’s egg and a piece of meat (ca 3 cm cube of beef, rabbit or venison) on the bait block between traps. An additional hen’s egg was also placed outside the trap tunnel on the ground and another approximately 1 m off the ground on a tree. Fig. 2 Secretary Island trap lines and mainland traps Stoat trapping began on Secretary Island from 20 buffered to 700m. 30 July 2005 using the pre baiting regime. Traps were
456 McMurtrie et al .: Eradication of stoats, Secretary Island
checked twice during this initial trapping period and were Mainland from July 2005 January 2008. Sub sampling only re set or re baited if required. Thereafter traps were from July 2005 and February 2006 was random, thereafter left set and baited, then serviced three times annually in all of the stoats captured were analysed for each of the November, February and between May and July. stated time periods. The intention is to include all of the stoats caught for all time periods for future analysis. The location, tunnel type, type of bait used, weight and sex of each stoat trapped were recorded and the carcass Tail tissue samples were dissected in the laboratory, frozen for future analyses of diet and aging using cementum where 50 mg of muscle tissue and caudal skin were analysis of teeth. Sex was determined from the presence/ removed. DNA was then isolated, using a Bio Rad absence of a baculum bone, unless the specimen was badly AquaPure Genomic Tissue Kit (Cat# 732 6343) following degraded, in which case it was recorded as “unknown”. the manufacturer’s protocol, and re suspended in 100 l of Tissue or bone samples were taken from all stoats captured supplied buffer. for DNA analysis. All samples were genotyped using sixteen microsatellite In July 2006, all wire mesh tunnels were removed from loci developed from a range of mustelid species. Primers the island due to disturbance from native birds such as used were MER005, MER030, MER022, MER041, kaka ( Nestor meridionalis ), kea ( N. notabilis ), and weka MER009, and MER082 developed from M. erminea (Gallirallus a. australis ). Concurrently, each wooden (Fleming et al . 1999); MVI057 developed from M. vison tunnel was modifi ed to contain a single trap. In July 2007, (O’Connell et al . 1996); WE7 and WE8 from M. sibirica all remaining traps were replaced with single set stainless (Huang et al. 2007); MLUT27 and MLUT32 developed steel DOC 150 traps (CMI Springs Ltd. Wellington, NZ). from M. lutropola (Cabria et al . 2007); MA1 developed from Martes americana (Davis and Strobeck 1998); Managing reinvasion MEL1 and MEL4 developed from Meles meles (Bijlsma et al . 2000); RIO11 and RIO19 developed from Lantra A coastal trap line comprising 180 double set DOC 150 canadensis (Beheler et al. 2005). PCR amplifi cation and traps in wooden tunnels was established on the mainland genotyping followed Gleeson et al. (2010). along Thompson Sound and Pendulo Reach (Fig. 2). Stepping stone islands to the south and south east were DNA analysis already being trapped as part of the Fiordland Stoat Immigration Study (Elliott et al . 2010). Trapping on the For statistical purposes, the data were grouped into mainland (hereafter referred to as Mainland) commenced three ‘populations’: 1) Secretary Island residents (n=35) in March 2005 with a subsequent check during the initial consisting of 10 from the initial knockdown and 25 trapped knockdown on Secretary Island. Thereafter, traps were in February 2006 (these latter were mostly juveniles and serviced in November and February, which activated traps considered to be survivors from the initial knockdown); 2) with fresh bait immediately before juveniles left their natal all stoats trapped from February 2007 – June 2008 (n=20); den, and cleared the traps after most juveniles had dispersed and 3) all samples from the nearby mainland site (n=35) to establish new territories (King and Powell 2007). from July 2005 January 2008. Microsatellite genotypes were analysed using GenALEx Monitoring for stoats at low density v. 6.2 (Peakall and Smouse 2006) to generate observed and Tracking tunnels were not used to monitor stoat activity expected heterozygosities, allele frequency scores and on the island because their rigorous use required a very Hardy Weinberg equilibriums. Pairwise FST parameters large number to be set and serviced (Brown and Miller for each population pair were estimated according to Weir 1998; Choquenot et al. 2001; King et al. 2007).Given that and Cockerham (1984). The data were analysed using the the probability of a stoat entering a tracking tunnel and a Bayesian clustering method implemented in STRUCTURE kill trap tunnel is similar, we viewed dead stoats in traps ver 2.3 (Pritchard et al. 2000) to provide another estimate as preferable to stoat footprints in tracking tunnels. Based of pairwise FST parameters and to determine the number on the success of previous stoat eradication operations in of distinct genetic units ( K) in the dataset. This method Fiordland up to 2004, we assumed that kill traps would does not require prior knowledge of sampling localities provide good detectability for stoats at low density. Trained and assigns individuals into groups minimising deviations stoat indicator dogs were used on and off the tracks and from Hardy Weinberg proportions and genotypic linkage we also requested contract deer hunters to record their equilibrium. The admixture model with correlated allele observations of stoat sign. frequencies was chosen. Ten replicates were conducted for each run, consisting of a burn in period of 100,000 Molecular data MCMC (Markov Chain Monte Carlo) steps followed by Molecular DNA techniques have successfully identifi ed 10 6 iterations. The ∆ K method of Evanno et al . (2005) was survivors from invaders following island rat eradication applied and plots of the log posterior probability of the data programmes (Adbelkrim et al. 2007; Rollins et al. 2006; [ln P(D)] for each value of K examined. Russell et al . 2010). To be useful, the technique requires Assignment tests were carried out to determine the most measurable genetic differentiation between sample probable origin of the individuals captured after the initial populations. We used molecular analysis to determine eradication operation using GENECLASS 2.0 (Piry et al . the frequency of immigration by stoats to Secretary Island 2004). The likelihood of the multilocus genotype of each following the initial knockdown. individual being assigned to the resident Secretary Island DNA extraction and microsatellite amplifi cation population or the Mainland population was calculated in order to identify putative residual individuals or migrants. Of 189 stoats caught, 89 were used in the genetic Ten thousand MCMC simulations per population were run analysis. Fifty four stoats were from Secretary Island, using the Lh/Lmax likelihood computation (Paetkau et al . including 10 from July 2005, 25 from February 2006, 5 2004). An individual was considered to be a disperser if from February 2007, 6 from May 2007, 1 from June 2007, the L /L P value was below 0.01. 1 from January 2008, and 7 from June 2008. Thirty four h max samples were obtained from stoats trapped on the adjacent
457 Island invasives: eradication and management
Table 1 Sex of stoats caught on Secretary Island between Table 2 Summary statistics for stoats from Secretary
July 2005 and December 2009. Island and Mainland. N = Sample Size; N A= mean number of alleles per locus; N PA = number of private alleles with Time period Male Female Unknown frequency > 0.05; H O = observed heterozygosity; H E = July 2005 34 56 5 expected heterozygosity. Nov 2005 2 5 2 Location Year N NA NPA HO HE Feb 2006 13 28 3 Secretary I. 2005 06 35 4.06 0 0.498 0.539 July 2006 0 1 5 Secretary I. post Nov 2006 0 0 0 eradication 2007 08 20 4.69 5 0.491 0.579 Feb 2007 3 3 0 Mainland 2005 08 34 5.06 2 0.471 0.572 May 2007 2 3 1 Nov 2007 1 0 0 Feb 2008 0 2 0 June 2008 3 4 1 RESULTS Dec 2008 0 0 0 Secretary Island stoat captures Feb 2009 1 1 7 Prebaiting and Trapping May 2009 0 3 1 Bait was taken from 95% and 99% of all trap tunnels Dec 2009 0 0 0 during the fi rst and second pre baiting periods respectively. Following the knockdown in July 2005 <10 stoats have been caught in each trapping period (Fig. 3) mostly in autumn and early winter. The sex ratio of captures was approximately 2 females for every male (Table 1). Stoat captures were generally well spread across the island with highest numbers in the west and north. Molecular analysis No signifi cant linkage disequilibrium was detected between loci, so all loci included in the analysis were considered independent. The mean number of alleles per population (Table 2) ranged from 4.06 for the original Secretary Island population, 4.69 for the post eradication Secretary Island population, through to 5.06 for the nearby Mainland population. There were no alleles of frequency > 0.05 restricted only in the original Secretary Island population, while there were fi ve alleles found only in the post eradication Secretary Island population, and two alleles restricted to the Mainland population. Ten alleles shared between post eradication Secretary Island and the Fig. 3 Stoat captures on Secretary Island from July 2005 Mainland that were not found in the original Secretary to December 2009. Solid bars denote the period from Island population. which trapped animals were analysed using molecular DNA techniques. Arrows indicate stoat plague events FST values between populations were relatively low, on the adjacent mainland driven by beech ( Nothofagus indicating little population structuring. Pairwise estimates sp.) masting events in the preceding autumn causing an were lowest between the post eradication population and increase in rodent numbers. the mainland (0.006), and highest between the original Secretary Island population and the mainland (0.03). The STRUCTURE analysis showed only slight differences between average loglikelihood estimates across different population scenarios ranging from K=1 to K=5. The best scenario revealed from plotting these estimates was K=2. The proportion of membership ( q) of each group to the two inferred clusters (Secretary Island vs Mainland) (Fig. 4) shows group 2 (post eradication Secretary Island) individuals being an admixture of both sources. GENECLASS identifi ed four individuals from the post eradication Secretary Island population as fi rst generation immigrants from the mainland, while three individuals were assigned to the original Secretary Island population
(Lh/Lmax P < 0.01).
DISCUSSION Fig. 4 STRUCTURE bar plot of estimation of the membership coefficient (Q) for each individual stoat for the three groups Unlike previous eradications on smaller islands in for K = 2. Each individual is represented by a thin vertical Fiordland, not all stoats were removed from Secretary line, showing degree of admixture. Black lines separate Island within the fi rst year of trapping. Our results indicate individuals from each of different population groups based that the stoat population is now being maintained at a very that are labelled below the figure. low number and, as a result of immigration and breeding by
458 McMurtrie et al .: Eradication of stoats, Secretary Island
residual resident animals, there has been no further decline Low population productivity on Secretary Island This latter fi nding may be related to the island’s size. strengthens the chances of eradication, which thus remains Many eradication programmes against mobile carnivorous a key objective. A harsh winter, further refi nements with predators have taken several years to reach completion. the existing trapping programme or new technologies may Examples include cats ( Felis catus ) (Bester et al. 2000; hasten removal of the residual population. Moreover, Veitch 2001; Algar et al . 2002), stoats (Crouchley 1994), stoat numbers have remained suffi ciently low on Secretary and mink ( Neovision vison ) (MacDonald and Harrington Island to achieve anticipated conservation outcomes such 2003). as the reintroduction of several species of threatened birds (Wickes and Edge 2009) . Monitoring species particularly Recent home range estimates obtained for stoats on vulnerable to stoats will be crucial in order to establish Resolution Island (home range diameter C. 486 m; Clayton a stoat density threshold for future reintroductions, et al. 2011) indicate that the decision to space trap lines at such as tieke/ South Island saddleback ( Philesturnus C. a distance no greater than 1400 m apart may have been an carunculatus ) proposed for 2015. The challenge is to detect over estimate of resident stoat home range. On Secretary stoats at extremely low densities without establishing a Island, a few stoats may have retained very small home prohibitively expensive monitoring programme. ranges, despite the signifi cant population reduction, and have therefore failed to encounter a trap. Since female stoats have smaller home ranges than males they may GENERAL CONCLUSIONS be less likely to encounter a trap. Nonetheless, twice as Our programme was based on applying techniques many females as males were trapped. Alternatively, some developed on smaller islands over a much larger area. stoats may avoid entering a trap tunnel either for extended Although we planned to put all animals at risk of capture, periods of time, or even in perpetuity, as was the case this appears not to have been achieved, probably due to on Maud Island, New Zealand (Crouchley 1994). On a broader range of habitat types than anticipated in the Secretary Island, stoat tracks were twice recorded in snow Secretary Island landscape. We also assumed that the level along ridgelines with traps present, which indicated trap of reinvasion would be lower than preliminary genetic avoidance. Based on Maud Island experiences, continued results have indicated. The experimental nature of this trapping can eventually eliminate stoats that have avoided programme has opened the door for testing new ground traps for periods of up to several years. in the fi eld of island eradications and challenging some Genetic data revealed enough variability across all of the previously held views of what should and should loci to show some degree of differentiation between the not be attempted (see Edge et al . 2011). Molecular DNA mainland and original Secretary Island population, although tools have been invaluable in enabling managers to better understand what has happened on the island since the FST values were relatively low. Differentiation between these groupings was supported by the STRUCTURE campaign began. analysis which showed the data to be effectively split into two groupings. Evidence for immigration amongst the ACKNOWLEDGEMENTS remaining stoats captured on Secretary Island after the initial year was from allelic differences (new alleles appearing) The programme to eradicate stoats from Secretary and from the assignment test using GENECLASS, which Island has been achieved by government funding to the identifi ed four fi rst generation immigrants. There were also New Zealand Department of Conservation. The molecular three individuals from that group which were assigned to work was supported by funds from the Foundation for the residential island population, while the remainder were Research, Science and Technology. Advice throughout unable to be assigned to either group, so were most likely the planning and implementation of this programme has admixtures from both. been given by members of the Island Eradication Advisory Group: K. Broome, A. Cox, R. Empson, P. Cromarty, The level of immigration detected from July 2005 and I. McFadden. We wish to thank the many staff and to June 2008 was higher than predicted by Elliott et volunteers that have given their time to this programme, in al . (2010) possibly due to beech masting in 2006 and a particular D. Hamilton, P. Kirkman and the skippers of the subsequent rodent and stoat plague on the mainland in Department of Conservation vessel: B. Walker, B. Hawke, Fiordland. During mast years, it is likely that there will M. Peychers, and P. Young. We also wish to acknowledge be higher numbers of juvenile stoats dispersing from the the support of the local Te Anau – Manapouri communities mainland to inshore islands, such as Secretary Island. In including tourist operators working in Doubtful Sound who February 2007, one stoat was caught on Seymour Island have given their support to this programme. E. Murphy to the south of Secretary Island, the fi rst in seven years of provided useful comments on an earlier version of this trapping. Another stoat was seen on Anchor Island, which manuscript. We thank our reviewers Kim King and Sugoto had been free of stoats since 2001. Roy for their helpful comments and improvements to this manuscript. Further genetic work to include all of the stoats captured on the island since 2005 should help to refi ne the estimate for immigration. Molecular tools will also REFERENCES be used to determine the relatedness among individuals, Adbelkrim J.; Pascal M. and Samadi S. 2007. Establishing causes of thereby providing an estimate of population productivity; eradication failure based on genetics: Case study of ship rat eradication the absence of rodents on Secretary Island may mean in Ste. Anne Archipelago. Conservation Biology 21 : 719 730. that female litter size is reduced, which would explain Algar D. A.; Burbidge A. A. and Angus G. J. 2002. Cat eradication on why the number of stoats caught in summer on Secretary Hermite Island, Montebello islands, Western Australia. In: Veitch C. R. and M. N. Clout (eds). Turning the tide: the eradication of invasive Island is not higher. King et al. (2003) demonstrated the species , pp. 14 18. IUCN SSC Invasive Species Specialist Group. signifi cance of rodents driving population productivity in IUCN Gland, Switzerland and Cambridge, UK. four beech forest sites in Fiordland. A shortage of rodents Alterio, N. 1998. Spring home range, spatial organisation and activity can lead to increased mortality of embryos and young in of stoats Mustelia erminea in a South Island Nothofagus forest, New the den, while adult females remain healthy. Zealand. Ecography 21 : 18 24.
459 Island invasives: eradication and management
Beheler, A. S.; Fike, J. A.; Dharmarajan, G.; Rhodes, O. E. and Serfass, King, C. M.; White, P. C. L.; Purdey, D. C. and Lawrence, B. 2003. T. L. 2005. Ten new microsatellite loci for North American river otters Matching productivity to resource availability in a small predator, the (Lontra canadensis ) and their utility in related mustelids. Molecular stoat ( Mustela erminea ). Canadian Journal of Zoology 81 : 662 669. Ecology Notes 5: 602 604. MacDonald, D. W. and Harrington, L. A. 2003. The American mink Bester, M.N.; Bloomer, J. P.; Bartlett, P. A.; Muller, D. D.; van Rooyen, (Neovison vision ): the triumph and tragedy of adaption out of context. M. and Buchner, H. 2000. Final eradication of feral cats from sub New Zealand Journal of Zoology 30 : 421 441. Antarctic Marion Island, southern Indian Ocean. South African Journal of Wildlife Research 30 : 53 7. Mark, A. F. and Baylis, G. T. S. 1975. Impact of deer on Secretary Island, Fiordland, New Zealand. Proceedings of the New Zealand Ecological Bijlsma, R.; Van De Vliet, M.; Pertoldi, C.; Van Apeldoorn, R. C. and Van Society 22 : 19 24 . De Zande, L. 2000. Microsatellite primers from the Eurasian badger, Meles meles . Molecular Ecology 9: 2155 2234. Miller, C.; Elliot, M. and Alterio, N. 2001. Home range of stoats (Mustela erminea ) in podocarp forest, south Westland, New Zealand: Brown, J. A. and Miller, C. J. 1998. Monitoring stoat ( Mustela erminea ) Implications for control strategy. Wildlife Research 28 : 165 172. control operations: power analysis and design. Science for Conservation Publication : 96. New Zealand Department of Conservation, Murphy, E. C. and Dowding, J. E. 1995. Ecology of the stoat in Nothofagus Wellington. forest: home range habitat use and diet at different stages of beech mast cycle. New Zealand Journal of Ecology 19 : 97 109. Cabria, M.T.; Gonzalez, E.G.; Gomez Moliner, B.J. and Zardoya, R. 2007. Microsatellite markers for the endangered European mink O’Connell, M.; Wright, J. M. and Farid, A. 1996. Development of (Mustela lutreola ) and closely related mustelids. Molecular Ecology PCR primers for nine polymorphic American mink Mustela vison Notes 7 (6) : 1185 1188. microsatellite loci. Molecular Ecology 5(2) : 311 312. Choquenot, D.; Ruscoe, W.C. and Murphy, E. 2001. Colonisation of new Paetkau, D.; Slade, R.; Burden, M. and Estoup, A. 2004. Direct, real time areas by stoats: time to establishment and requirements for detection. estimation of migration rate using assignment methods: a simulation New Zealand Journal of Ecology 25 : 83 88. based exploration of accuracy and power. Molecular Ecology 1: 55 65. Clayton, R.I.; Anderson, D.; Byrom, A., Edge, K A.; McMurtrie, P. M.; Peakall, R. and Smouse, P.E. 2006. GENALEX 6: genetic analysis in Veale, A. and Torr, N. 2011. Using genetic analysis and trapping data to Excel. Population genetic software for teaching and research. Molecular model the probability of persistence of feral stoats ( Mustela erminea ) Ecology Notes 6: 288 295. on Resolution Island, New Zealand. In: Veitch, C. R.; Clout, M. N. and Piry, S.; Alapetite, A.; Cornuet, J M.; Paetkau, D.; Baudouin, L. and Towns, D. R. (eds.). Island invasives: eradication and management , pp. Estoup, A. 2004. GeneClass2: A Software for Genetic Assignment and 413 417. IUCN, Gland, Switzerland. First Generation Migrant Detection. Journal of Heredity 95 : 536 539. Crouchley, D. 1994. Stoat control on Maud Island 1982 1993. Ecological Parkes, J. P. 1990. Eradication of feral goats on islands and habitat Management 2 : 39 45. islands. Journal of the Royal Society of New Zealand 20 : 297 304. Crouchley, D.; Nugent, G. and Edge, K A. 2011. Eradication of red deer Parkes, J. P.; Macdonald, N. and Leaman, G. 2002. An attempt to (Cervus elaphus ) from Anchor and Secretary Island, Fiordland, New eradicate feral goats from Lord Howe Island. In: Veitch C.R. and Clout Zealand. In: Veitch, C. R.; Clout, M. N. and Towns, D. R. (eds.). Island M.N. (eds.). Turning the tide: the eradication of invasive species , pp. invasives: eradication and management , pp. 422 425. IUCN, Gland, 233 239 . IUCN SSC Invasive Species Specialist Group. IUCN Gland, Switzerland. Switzerland and Cambridge, UK. Davis, C. S. and Strobeck, C. 1998. Isolation, variability, and cross Pritchard, J. K.; Stephens, M. and Donnelly, P. 2000. Inference of species amplifi cation of polymorphic microsatellite loci in the family Population Structure Using Multilocus Genotype Data. Genetics 155 : Mustelidae. Molecular Ecology 7: 1776 1778. 945 959. Edge, K A.; Crouchley, D.; McMurtrie, P.; Willans, M. J. and Byrom, Rollins, L. A.; Woolnough, A. P. and Sherwin, W. B. 2006. Population A. 2011 Eradicating stoats ( Mustela erminea ) and red deer ( Cervus genetic tools for pest management: a review. Wildlife Research 33 : 251 elaphus ) off islands in Fiordland. In: Veitch, C. R.; Clout, M. N. and 261. Towns, D. R. (eds.). Island invasives: eradication and management , pp. 166 171. IUCN, Gland, Switzerland. Russell, J. C.; Miller, S. D.; Harper, G. A.; MacInnes. H. E.; Wylie, M. J and Fewtser, R. M. 2010. Survivors or reinvaders? Using genetic Elliott, G.; Willians, M.; Edmonds, H. and Crouchley, D. 2010. Stoat assignment to identify invasive species following eradication. Biological invasion, eradication and re invasion in islands in Fiordland. New Invasions 12 (6) : 1747 1757 . Zealand Journal of Zoology 37: 1 12. Taylor, R. H. and Tilley, J. A. V. 1984. Stoats ( Mustela erminea ) on Evanno, G.; Regnaut, S. and Goudet, J. 2005. Detecting the number of Adele and Fisherman Islands, Abel Tasman National Park, and other clusters of individuals using the software STRUCTURE: a simulation offshore islands in New Zealand. New Zealand Journal of Ecology 7: Study. Molecular Ecology 14 : 2611 2620. 139 145. Fleming, M.A.; Ostrander, E.A. and Cook, J.A.1999. Microsatellite Veitch, C.R. 2001. The eradication of feral cats ( Felis catus ) from Little markers for American mink ( Mustela vison ) and ermine ( Mustela Barrier Island, New Zealand. New Zealand Journal of Zoology 28 : erminea ). Molecular Ecology 8(8) : 1352 1354. 1 12. Gleeson, D.M; Byrom, A.E. and Howitt, R.L.J. 2010. Non invasive Weir, B.S. and Cockerham, C.C. 1984. Estimating F statistics for the methods for genotyping of stoats ( Mustela erminea ) in New Zealand: analysis of population structure. Evolution 38 : 1358 1370. potential for fi eld applications. Short Communication. New Zealand Journal of Ecology 34 (3) : 356 359. Wickes, C. and Edge K A. 2009. Secretary and Resolution Restoration Plan. Unpublished report, New Zealand Department of Conservation, Golding, C.; McMurtrie, P.; Edge, K A. and Willans, M. 2005. Secretary Te Anau*. Island Operational Plan. Part A – Stoat Eradication. Unpublished report, New Zealand Department of Conservation, Te Anau*. * Unpublished reports and operational plans available at: http://www. doc.govt.nz/conservation/land-and-freshwater/australasias-top-25- Hill, S. and Hill, J. 1987. Richard Henry of Resolution Island . John restoration-projects/fi ordland-islands-restoration/ McIndoe Ltd. Huang, C. C.; Lin R. C.; Li, S. H. and Lee, L. L. 2007. Characterisation of polymorphic tetranucleotide microsatellite loci from the Siberian weasel ( Mustela sibirica ). Molecular Ecology Notes 7 : 483 485. King, C. M. 1984. Immigrant killers: introduced predators and the conservation of birds in New Zealand. Oxford University Press, Auckland. King, C. M. and Powell, R. A. 2007. The natural history of weasels and stoats: ecology, behaviour and management . Second Edition. Oxford University Press.
460