New Zealand pest management: current and future challenges

Abstract

New Zealand is under increasing pressure from terrestrial and aquatic pests, weeds and diseases that threaten the country's ecosystems and economy. Ongoing improvement in existing pest management methodologies and novel approaches are required in response to public concerns about animal welfare, increasingly stringent trade requirements, abolition of groups of pesticides and resistance to existing pesticides as well as, possibly, biological control agents. Surveillance and pest monitoring are needed to increase the chances of early interception of invasive species or to confirm their eradication. Core capabilities in taxonomy, genomics, phenology, ecology, pest impacts, development of novel control tools and social science are required and must be maintained nationally. Given New Zealand's unique environment, the ecology of invasive pests cannot be presumed to be the same as that in their native ranges, yet currently many pests in New Zealand are managed with poor understanding of their bionomics and impacts. Failure to address these areas will have serious adverse impacts on New Zealand.

Keywords:

• biosecurity
• biological control
• biological invasions
• pesticide use
• public acceptability
• surveillance
• vertebrate toxic agents

Introduction and background

The considerations presented in this contribution were originally developed as an ‘Emerging Issues’ paper by the Royal Society of New Zealand (http://assets.royalsociety.org.nz/media/2014/03/Challenges-for-pest-management-in-NZ.pdf). That paper sought to link pest management issues across New Zealand's productive and indigenous ecosystems. Such analysis has rarely been undertaken. This contribution develops the paper further by providing an overview of New Zealand pest management to impart a fuller understanding of the nature and diversity of the challenges and issues. However, a comprehensive compendium of such considerations would require a book-length contribution and is therefore beyond the scope of this scientific paper. The species referred to are presented as examples to highlight generic pest management issues, rather than being a definitive listing.

Unusually, for an Organisation for Economic Cooperation and Development country, New Zealand's economy and reputation are closely linked to its agricultural lands, natural environment, and managed and wild aquatic ecosystems (Goldson 2011). Productive capacity and economic growth have flourished through the introduction of hundreds of economically important plants and animals and this has often been in the absence of their associated weeds, pests and diseases (e.g. Rahman 1982; Goldson 2011). In spite of this, all ecosystems remain continuously threatened by invasive organisms (Ministry of Agriculture and Fisheries 2003).

Within the productive ecosystems, reducing the impact of already-arrived invasive exotic species has often relied on pesticides, biological control and, occasionally, integrated pest management (IPM) and/or trapping regimes. Conversely, in the indigenous terrestrial environment, invertebrate, microorganism and weed pests often go uncontrolled. Environmentally damaging vertebrate pests are an exception to this and are often the subject of intensive control campaigns. Such initiatives are usually based on aerially applied toxins and/or ground-based trapping (Saunders & Norton 2001; Nugent & Morriss 2013). Vertebrate pest control is also vital for maintaining access to international markets by reducing the risk of transmission of bovine tuberculosis (Mycobacterium bovis Karlson and Lessel, [TB]) from vertebrate wildlife to livestock (e.g. Warburton & Livingstone 2014). An increasing threat is posed by invasive herpetofauna, for example, the still-dispersing rainbow skink Lampropholis delicata De Vis (Gill et al. 2001) and Litoria frogs (Litoria raniformis Keferstein); these species have been found to carry the amphibian fungal disease Batrachochytrium dendrobatidis Longcore, Pessier and D. K. Nichols (Moreno et al. 2011).

The New Zealand pest management environment is not stable and never will be, given perennial external and internal challenges. External factors, such as globalisation and free trade, threaten to accelerate the rate of pest incursions in spite of well-directed efforts by New Zealand to have an excellent biosecurity system (Ministry of Agriculture and Fisheries 2003). Dozens of severe and well-known exotic pest threats external to New Zealand, as well as lesser-known species, can become pests on arrival. Such new pests can inflict their own damage and can disrupt the management of existing pests. Compounding the immediate effects of such biosecurity failure are other exacerbating factors including land-use intensification (Goldson 2009), the effects of climate change (Ministry of Agriculture and Fisheries 2003), the decreasing social acceptability of some pest management solutions (e.g. Warburton 2012), ongoing imposition of phytosanitary requirements by offshore markets (e.g. Goldson & Gluckman 2014) and pesticide withdrawal (e.g. endosulfan) by the New Zealand Environmental Protection Authority (EPA). Other negative biological influences include the loss of pesticide efficacy through acquired resistance to pesticides (e.g. Bourdôt 1996) and observed possible decline in classical biological control through selection pressure on pest species (Goldson & Wratten et al. 2014). These issues are not confined to New Zealand, with the rate of implementation of eradication programmes against arthropods increasing exponentially around the globe (Tobin et al. 2014).

Pest impacts on New Zealand's ecosystems

The impacts of pests on New Zealand's ecosystems are diverse and do not segregate neatly into those affecting indigenous versus exotic ecosystems; rather, they span the productive and natural ecosystems.

Indigenous ecosystems

About 8.64 million ha of New Zealand's land area—about 33%—is protected, with most of that (about 8.2 million ha) under the stewardship of the Department of Conservation (DOC) and the remainder with Ngā Whenua Rāhui, regional councils and the QEII National Trust.

New Zealand's commitment to the 1993 Convention on Biological Diversity and the 2000 New Zealand Biodiversity Strategy calls for the protection of its natural ecosystems, flora and fauna. This is demanding, given the issues outlined above. As a result of habitat loss and predation, many indigenous species are threatened with some on the brink of extinction or confined to offshore islands. There have already been widespread extinctions; Worthy & Holdaway (2002) have shown that 40% of the avifauna has become extinct since the arrival of humans. Of notoriety here are invasive mammal predators, including rodents (ship rat [Rattus rattus L.], Norway rat [R. norvegicus Berkenhout], Pacific rat [R. exulans Peale], house mouse [Mus musculus L.]), possums (Trichosurus vulpecula Kerr), feral cats (Felis catus L.), hedgehogs (Erinaceus europaeus L.) and mustelids (stoats [Mustela erminea L.], ferrets [M. furo L.] and weasels [M. nivalis L.]) (King [2005] and references therein).

Exotic mammalian herbivores also impact on indigenous vegetation (e.g. Scroggie et al. 2012), as do insect herbivores with broad host ranges. For example, the critically endangered and iconic Cook's scurvy grass (Lepidium oleraceum G. Forst. ex Sparrm.) and similar species, are already under pressure from introduced armyworms, aphids and cabbage white butterfly (Pieris rapae L.); the latter also interacts with a white rust fungus, Albugo candida (Pers) Roussel (Hasenbank et al. 2011). Related to this, the DOC is now leading an eradication programme against an incursion of the ‘great white butterfly’ P. brassicae L. in Nelson (e.g. Phillips et al. 2013; Hiszczynska-Sawicka & Phillips 2014).

The destructiveness of invasive insects is also exemplified by the exotic social wasps Vespula germanica Fabricius and V. vulgaris L., particularly in beech forests (Beggs 2001). These aggressive species impact on indigenous invertebrates, avifauna and ecosystem services, as well as disrupting tourism, viticulture, forestry and recreation. Brockerhoff et al. (2010a) have reviewed the impacts of such exotic invertebrates, including herbivores, predators, parasitoids, decomposers and others, on native terrestrial species and ecosystems. The emergence of exotic infectious diseases of wildlife, such as avian malaria (Plasmodium relictum Russell), has also been linked to the spread of exotic invertebrates such as the mosquito Culex quinquefasciatus (Say) (Tompkins & Gleeson 2006) and the insidious appearance of kauri dieback (Phytophthora ‘taxon Agathis’ [PTA]) affecting kauri (Agathis australis (D.Don) Lindl. ex Loudon). This shows what can happen without effective detection and management of pest organisms (Krull et al. 2013). In general, pathogens of plants and animals are often extremely difficult to detect during an incursion. Overall, the DOC spends NZ$73.5 million p.a. on pest-led and site-led pest control and significant additional costs are incurred by regional authorities, private companies, farmers and citizens generally.

Weeds are a threat to one third of all of New Zealand's threatened plant species and could potentially degrade 7% of the conservation estate within a decade, corresponding to a loss of native biodiversity equivalent to NZ$1.3 billion (Williams & Timmins 2002). More than 300 weeds have been listed to be of conservation concern, including heather (Calluna vulgaris [L.] Hull), alligator weed (Alternanthera philoxeroides [Mart.] Griseb.), hawkweed (Hieracium L. spp.), gorse (Ulex europaeus L.), mistflower (Ageratina riparia [Regel] R.M. King and H. Rob), old man's beard (Clematis vitalba L.) and Scotch broom (Cytisus scoparius [L.] Link).

Pathogens found in New Zealand crops can also infect native biota in the same families as their cultivated hosts, often acting in concert with insect vectors such as aphids. For example, the endangered native plants Pachycladon Hook.f. spp. and Cook's scurvy grass can become infected with the turnip mosaic virus (Fletcher et al. 2011). Similarly, New Zealand's single rare and endangered cucurbit, Sicyos australis Endl., can be damaged by certain isolates of cucumber mosaic virus and the watermelon mosaic virus (Delmiglio & Pearson 2006).

To date, the long-term costs of loss of native biodiversity from indigenous ecosystems due to invasive organisms have not been comprehensively estimated, in part because native biota can be valued in diverse ways. Estimates of the value of biodiversity in indigenous ecosystems are rare and highly subjective. However, as an indication, Patterson & Cole (1999) put this at NZ$44 billion in 1994. Overall, it remains clear that a wide variety of potential complex and unpredictable types of impacts in indigenous ecosystems exist that can often depend on the presence and competency of vectors in novel associations.

Exotic ‘managed’ ecosystems

Forty-five percent of New Zealand's land area (12.1 million ha) is devoted to primary production, often comprising species-sparse and highly invasible ecosystems that are particularly prone to exotic pest build-up and damage (e.g. Goldson & Tomasetto et al. 2014). Pests, diseases and weeds have been estimated to cost the productive sector NZ$1.28 billion p.a. and this increases to NZ$2.45 billion p.a. when the downstream effects are factored in (Biosecurity New Zealand 2009). Numerous pests exist, but, by way of example, the bacterial disease Pseudomonas syringae pv. actinidiae (PSA) in kiwifruit is anticipated to cost between NZ$310 and NZ$410 million over 5 years (Greer & Saunders 2012). The potential annual impact of the clover root weevil (Sitona obsoletus Gmelin), should current biological control breakdown occur, has been estimated to cost the pastoral sector NZ$444 million p.a. (Saunders et al. 2013). Likewise, the costs and averted economic impacts from eradications of forest insect pests in New Zealand over 20 years have been substantial (Table 1).

Table 1 Costs to the Crown of invertebrate incursion responses in New Zealand, estimated economic impacts and estimated net benefits of eradication.

Organism, incursion response period and outcome of response	Incursion response cost to the Crown (approx. NZ$ million)	Estimated economic impact over a given number of years (NZ$ million, present value)	Approximate averted costs (i.e. economic impact less cost of eradication) (NZ$ million)	Sources
White-spotted tussock moth (Orgyia thyellina) (1996–98)—eradicated	12	25–177 (20 years)	13–165	Brockerhoff et al. (2010b)
Gum leaf skeletoniser (Uraba lugens) (1997–98—eradicated	4	101–142 (20 years)	97–138	Brockerhoff et al. (2010b)
Painted apple moth (Teia anartoides) (1999–2006)—eradicated	61	58–356 (20 years)	−3–295	Brockerhoff et al. (2010b), C. Baddeley, pers. comm. 2014
Fall webworm (Hyphantria cunea) (2003–06)—eradicated	7	19–83 (20 years)	12–76	Brockerhoff et al. (2010b)
Gum leaf skeletoniser (Uraba lugens) (2003)—eradication not attempted	120^a	101–142 (20 years)	−19–22^a	Brockerhoff et al. (2010b)
Hokkaido gypsy moth (Lymantria umbrosa) (2003–05)—eradicated	6	3–291^b (20 years)	−3–285	Brockerhoff et al. (2010b)
Didymo (Didymosphenia geminata) (2004–ongoing)—not eradicated	17	58 (8 years)	n/a	NZIER (2006), C. Baddeley, pers. comm. 2014
Red imported fire ant (Solenopsis invicta) (2001–09)—eradicated (three responses in Auckland and Hawke's Bay)	10	665 (23 years)	655	MAF (2001), C. Baddeley, pers. comm. 2014
Varroa mite (Varroa destructor) (2000–09)—not eradicated (North Island and South Island responses)	18	316 (35 years)	n/a	NZIER (2001), C. Baddeley, pers. comm. 2014
Clover root weevil (Sitona lepidus) (1996–ongoing)—not eradicated	10^b	1400–8100 (35 years)	n/a	NZIER (2005a), Baddeley (2007)
Southern salt marsh mosquito (Ochlerotatus camptorhynchus) (1999–2010)—eradicated (several responses in the North Island and South Island)	77	n/a	n/a	C. Baddeley, pers. comm. 2014
Didemnum sea squirt (Didemnum vexillum) (2006–09)—not eradicated	1	n/a	n/a	Ansell & Coates (2008)
Mediterranean fanworm (Sabella spallanzanii) (2008–09)	1.3	n/a	n/a	Biosecurity New Zealand (2010b), C. Baddeley, pers. comm. 2014

^aFor this particular detection, eradication was not attempted due to an unfavourable cost-benefit analysis; eradication cost estimate (see Brockerhoff et al. (2010b) and references therein for details).

^bInvestment by the Crown (from 1996 to 2007) for research and information/advice.

In 2008, the expenditure on pest control in the productive sector was estimated to be NZ$970 million p.a. (Biosecurity New Zealand 2009). Furthermore, practices in terrestrial pest management are not without externalities, such as contamination through pesticide residues, loss of biodiversity through non-target impacts, and damage to soil structure through repeated cultivation. In addition, the Ministry for Primary Industries currently spends more than NZ$50 million p.a. managing a vast array of biosecurity risks and these costs can escalate during major biosecurity responses. The estimated cost of response to a single invasive fruit fly (Bactrocera tryoni [Froggatt]) interception in Whangarei in January 2014 was about NZ$1 million (two were caught in total), while the costs of an established self-sustaining fruit fly population would be orders of magnitude higher, due to trade implications affecting NZ$3.6 billion worth of exported horticultural products p.a.

Pastoral weeds are estimated to cost the New Zealand economy NZ$1.2 billion p.a. in lost animal production and control costs (Bourdôt et al. 2007), and bovine tuberculosis could seriously damage New Zealand's reputation and access to high-value export markets worth around NZ$14 billion p.a. (Hutchings et al. 2013).

Interactions can be complex and far reaching, as in the case of varroa mite (Varroa destructor Anderson and Trueman) and associated vectored viruses, which have seriously affected wild honeybee populations (e.g. Todd et al. 2007). This, in turn, has affected pollination services for fruit and seed production throughout urban and rural areas and is constraining the production of unique manuka honey and other products supporting emerging industries.

Aquatic ecosystems

Pest species destabilise aquatic habitats and negatively modify water flow with consequences for drainage, irrigation, power generation and recreational activities. The incursion and outbreak of the freshwater diatom didymo (Didymosphenia geminata [Lyngb.] M. Schmidt) in South Island streams has epitomised this. In the marine environment, invasive species displace native species, modify coastal habitats and affect human health. They also pose threats to aquaculture, commercial fishing and other maritime industries, including recreational pastimes. The costs can be severe. Production losses to aquaculture from a single species of sea squirt (Styela clava Herdman) were estimated to be NZ$15 million p.a. in 2005 (NZIER 2005b). More recent estimates suggest that, if S. clava spreads to Marlborough (in the northeast of South Island), production losses over the next 8 years could amount to NZ$383 million (Deloitte 2011). Within the marine environment, expenditure on recent incursions (i.e. ‘defensive costs’) by biofouling organisms such as S. clava, the Whangamata sea squirt (Didemnum vexillum Kott) and the Mediterranean fanworm (Sabella spallanzanii [Gmelin]) has been estimated to be NZ$2.2 million, NZ$1 million and NZ$1 million, respectively (Ansell & Coates 2008; Biosecurity New Zealand 2010a). Notably, these estimates represent only one-off incursion response costs incurred by central government and/or industry. Such expenditure, to date, has been modest compared with expenditure on incursion responses and management of terrestrial pest outbreaks (see examples in Table 1) (Ministry of Health 2004; Minister for Biosecurity 2006; Brockerhoff et al. 2010b). Moreover, if the decision were made that eradication is not feasible, there are then inevitable additional costs related to ongoing management and pest impacts.

No estimates have been made of the impacts of these or other biofouling species on non-market values. However, a ‘willingness-to-pay’ study assessed the dollar value of marginal changes to indigenous marine biodiversity and other attributes of the coastal marine environment associated with a potential incursion by the European green crab (Carcinus maenas [L.]) (Bell & Yap 2008; Bell et al. 2008). Of the four attributes evaluated in the study (loss of shellfish species, loss of recreational shellfish take, loss of coastal vegetation and the inability of children to paddle at the water's edge), the loss of indigenous (shellfish) biodiversity was valued most by the respondents. Although focused on a single estuary, the study concluded that, if comparable impacts were experienced throughout New Zealand, the expected marginal loss to these non-market values could amount to between NZ$325 million and NZ$600 million (Bell & Yap 2008). Such challenges are immense for a country the size of New Zealand as it has the fourth-largest marine exclusive economic zone in the world.

New Zealand's pest management issues, options and challenges

As discussed above, pests cost New Zealand billions of dollars. New methods of pest suppression are needed and these can only be obtained through long-term research and deep biological and ecological understanding.

Outlined below is a general discussion of New Zealand's main pest management issues, options and challenges, along with some possible solutions. In considering invertebrate pest management it is notable that only half or less of New Zealand's insect fauna has been described (Emberson 1998). Moreover, New Zealand's agricultural and forestry ecosystems are unusual in that they comprise only relatively few introduced, economically useful plant (and animal) species. While such economic areas may comprise endemic invertebrate diversity, frequently this does not include key exotic pest-suppressing species. These ecosystems also provide numerous unfilled niches, making them exceptionally susceptible to severe pest outbreaks by species that often have little impact elsewhere (Goldson & Tomasetto et al. 2014). New Zealand's indigenous terrestrial ecosystems have higher biodiversity than the managed productive ecosystems but here, too, relatively few species are found such as parasitoids, generalist predators or spiders that can exert biotic resistance to invasive species.

Invertebrate pest impacts and management

In many intensively cultivated situations, and sometimes in combination with various forms of integrated pest management (e.g. van Emden & Peakall 1996), pesticides of various sorts have been used for many years to prevent pest damage. This has included the use of some ‘soft’ biopesticides such as those based on Bacillus thuringiensis israelensis combined with synthetic insect growth regulators used to eradicate the salt marsh mosquito (Ochlerotatus camptorhynchus [Thomson]) (e.g. O’Callaghan et al. 2003). Similarly, painted apple moth (Teia anartoides Walker) eradication in Auckland was based on aerial applications of B. thuringiensis var. kurstaki (Btk) (e.g. Suckling et al. 2007) (see below). Biopesticides have also been developed as control agents against the soil-dwelling New Zealand grass grub (Costelytra zealandica White) using commercial formulations of the bacterium Serratia marcescens Bizio (e.g. Jackson 2007). Effective biological control has been achieved in New Zealand's extensive ecosystems by classical methods, as exemplified by three braconid endoparasitoid species belonging to the genus Microctonus Wesmael, introduced to control three exotic Curculionidae that cause severe damage to forage legumes and grass species (see below). While very successful (e.g. Barlow & Goldson 1993; Goldson & Wratten et al. 2014), this work has, at the same time, raised concerns about biosafety of such controls in terms of the impacts on non-target indigenous species. Concurrent work was done on how to minimise such unfortunate consequences by the judicious choice of control agents (e.g. Barratt et al. 2010).

Notable success in plant breeding has been achieved through the use of obligate biotrophic endophytic fungi (Epichloë/Neotyphodium spp.) to confer resistance in Gramineae against a range of pest species (e.g. Johnson et al. 2013) and in breeding lucerne (Medicago sativa L.) for resistance to aphids (e.g. Barlow & Goldson 2002).

While there has been good progress in developing environmentally friendly pest management options, synthetic pesticides are still necessary in numerous situations. Of concern is that some of the older, standby, broad-spectrum pesticides are now being phased out, which could leave critical gaps in control strategies (see below).

Pathogenic microorganism impacts and management

Pathogenic microorganisms present severe threats to all aspects of New Zealand life. Some, such as rust, Melampsora medusae Thum. and Melampsora laricis-populina Kleb. (Van Kraayenoord et al. 1974), are windborne and therefore spread rapidly. Livestock diseases may be dispersed through stock movements, as agriculture intensifies. Management demands timely and effective interception of disease but this is problematic because early detection is difficult and symptoms may be non-specific and/or be delayed by latent periods. Inevitably, large, genetically similar monocultures such as Pinus radiata D.Don, kiwifruit, apple plantations and pastures all face risks of invasion and very high rates of disease spread.

Human pathogens aside, foot and mouth disease (Aphthae epizootica) is the best known and studied infectious vertebrate disease. As well as bovine TB (mentioned above), other threats to livestock health include Toxoplasma gondii Nicolle and Manceaux, a microscopic parasite vectored by domestic and feral cats and Theileria (Theileria orientalis Ikeda), a tick-borne parasite that causes anaemia in cattle. Other threats to indigenous wildlife include avian malaria, which causes ‘crusty bum’ (cloacitis) in kakapo (Strigops habroptilus [G.R. Gray]) and various forms of bird flu (Orthomyxoviridae) that present severe threats to human health.

Significantly, taxonomy, interpreted at the species level by regulatory agencies, struggles to account for strains that can vary in pathogenicity (McCann et al. 2013). This can lead to complacency if an organism (mild strain) is already in the country, or if there is poor information about virulent strains. PSA in kiwifruit (see above) is a classic example where an incursion of a virulent strain was initially indistinguishable from a benign strain already present in New Zealand (Chapman et al. 2012; McCann et al. 2013). Currently, increasing concern is emerging about Phytophthora strains causing diseases such as kauri dieback (see above).

The potential impact of rusts, such as myrtle rust (Uredo rangelii J. A. Simpson, K. Thomas and Grgur) on pohutakawa Metrosideros Banks ex Gaertn. trees (Morin et al. 2012), as well as various pine needle diseases (Dick et al. 2014), are also worrisome.

In the marine environment, microorganisms and pathogens also threaten. The oyster herpes virus (OsHV-1) μvar caused severe mortality in farmed Pacific oysters Crassostrea gigas (Thunberg) in 2010, reducing stocks by 60%–80%, at an annual cost of around NZ$26 million to the industry (Barratt-Boyes 2012). Again, the impacts of diseases are compounded by slow recognition and inadequate response; this has, on occasion, greatly impacted on fish and shrimp industries (Walker & Winton 2010).

Weed impacts and management

As with invertebrate pests, New Zealand's managed ecosystems are prone to invasion by weeds. Indigenous ecosystems are more resilient but, when modified, they too can be overwhelmed by plant invaders (for instance, wilding pines) that are better adapted to disturbance or fire.

Garden escapes and live plant imports by gardeners and garden centres are a perennial source of new weeds. Plants present in gardens, but yet to establish in the country's broader ecosystems, form a huge pool of potential ‘sleeper’ weeds. For reasons that are yet to be fully understood, many weeds exhibit extended time-lags for as much as 100 years between their establishment and spread. This indicates that many future weeds may already be in New Zealand (Aikio et al. 2010). Generally, the risk posed by the 25,000 exotic plant species already present in New Zealand remains poorly understood and research is needed to identify the most likely future weeds among them (Diez et al. 2012). Seeds can also be transported into New Zealand by deliberate mail order and as accidental hitch-hikers on clothing and luggage. The hybridisation of New Zealand native plant species with exotic species, such that their whakapapa is compromised, is of particular concern for species with Māori heritage values.

Good progress has been made in New Zealand with the biological control of some weeds, for example, the suppression of mistflower (Ageratina riparia Regel) in natural ecosystems, and ragwort (Jacobaea vulgaris Gaertn.) and St John's wort (Hypericum perforatum L.) in the productive sector (Fowler et al. 2010; Paynter et al. 2012; Suckling 2013). Worldwide, there have been few issues with non-target effects on plants from modern weed biological control programmes (Suckling & Sforza 2014).

Vertebrate pest impact and management

Only two native land mammals are found in New Zealand (both bats). This is the result of 80 million years of geographical isolation. In contrast, 32 species of mammal and 35 bird species have established since the arrival of humans (King 2005), largely through deliberate introduction. New Zealand's native fauna is particularly vulnerable to the impacts of predation by mammalian pests. Rats, mice, mustelids, hares (Lepus europaeus Pallas), rabbits (Oryctolagus cuniculus L.), hedgehogs, possums, wild pigs (Sus scrofa L.) and feral cats all present serious threats. As a result, and to allow native species to recover, strenuous efforts have been made to create vertebrate pest-free offshore islands and fenced mainland sanctuaries (Norbury et al. 2014). Invasive mammals have now been eradicated from many islands (Veitch et al. 2011) and as a result, the country is regarded as a world leader in such eradications (Bellingham et al. 2010). While this has been largely achieved to date on small islands and fenced sanctuaries, it is now changing. Norway rats were eradicated from the 11,000 ha Campbell Island in 2002 (McLelland 2011) and all invasive mammals have been removed from Rangitoto/Motutapu Islands (3700 ha) in the Hauraki Gulf. Currently, programmes are under way to eradicate stoats and red deer (Cervus elaphus L.) from Secretary and Resolution Islands in Fiordland (8100 and 20,800 ha, respectively) (Edge et al. 2011), although reinvasion remains a challenge through pests’ ability to swim.

In general, New Zealand's pragmatic attitude towards the eradication of vertebrate pests from ever-larger areas is providing impetus for future challenges; in particular, the eradication or large-scale suppression of mammals on large inhabited islands (Glen et al. 2013a) as well as on the mainland (Glen et al. 2013b). A long-term vision for this is encapsulated in the ‘Predator-Free New Zealand’ initiative that seeks, in part through widespread public involvement, to achieve complete eradication or large-scale suppression of mustelids, possums and rats (http://predatorfreenz.org/).

Emerging issues for vertebrate pest management include the need for cost-effective, humane suppression at very large scales and a requirement for public support for mammal pest control or eradication (Cowan & Warburton 2011), especially where this involves toxins (see below). Vertebrate pest management differs from that of invertebrate pests and weeds, in that it has an expanding range of tools in the toolbox (Eason et al. 2014).

Freshwater pests and their impact and management

More than 200 freshwater plant and animal species have been introduced to New Zealand and have become established (Champion et al. 2012). Pests include microorganisms such as didymo (Didymosphenia geminata (Lyngbye) M. Schmidt), a wide range of invertebrates and a number of pest fish species, some of which continue to be illegally spread for recreational fishing.

The connectivity of aquatic environments and the flow of water through landscapes create special challenges for the control and management of aquatic pests; detection in these ecosystems is expensive and control options are limited. As in other ecosystems, early detection of unwanted species increases the feasibility of rapid responses to eradicate or contain spread. Environmental DNA (eDNA) can be effective as a detection tool in freshwater environments, as has been demonstrated in the Great Lakes system of the USA during the invasion of two species of Asian carp (Jerde et al. 2011).

Marine pests and their impact and management

New Zealand is highly dependent on maritime trade, with thousands of vessels visiting every year. In 2010, a review of non-indigenous and cryptogenic marine species revealed more than 330 non-indigenous species recorded in New Zealand's marine environments, with just over half of these (178) known to be established here (Kospartov et al. 2008). Another 350 species, whose geographic origins are unknown, were also found to be present in New Zealand's waters (Kospartov et al. 2008). Marine pests include microorganisms, large kelp species and a wide range of invertebrates such as crabs, tubeworms and sea squirts. Most have arrived by either attaching to the submerged surfaces of vessels and/or marine structures (biofouling) or in the ballast water carried by ships. More than 80% of goods transported worldwide are shipped by sea and the volumes being moved globally are increasing (Asariotis et al. 2010). It is estimated that c. 3–5 billion tonnes of ballast water are transported p.a. (Asariotis et al. 2010; GloBallast 2014). Pest species are also transported on fishing and marine farming equipment (e.g. Forrest & Blakemore 2006), as well as being spread through introduced aquarium material or introduced deliberately (e.g. Forrest & Blakemore 2006; Morrisey et al. 2011).

Current pest management tools, approaches and strategies

In keeping with many other jurisdictions, a number of tools of varying effectiveness and social acceptability can be identified for the eradication or suppression of weeds, pests and diseases.

Pesticides

Pesticides, in one form or another, have been used in all major pest control or eradication programmes since the 1800s. Initially, programmes against invertebrates used environmentally hazardous and/or persistent materials that are no longer acceptable, including insecticides such as nicotine sulphate and organochlorines such as dichlorodiphenyltrichloroethane (DDT) (see below). Similarly, programmes against weeds in the early 1900s used explosive sodium chlorate to control ragwort and other weeds.

Across the spectrum, from new incursions to the mitigation of existing pest threats, strategic planning still requires the use of pesticide-based tools on an appropriate scale and at a suitable frequency. To be effective, this requires a sound knowledge of the biology of each pest species and the pesticide's impacts, as well as proper integration of strategies and tactics.

As already discussed, the trend now is away from broad-spectrum and persistent products towards more selective agents, for example the toxin para-aminopropiophenone (PAPP) for the control of stoats and feral cats (Eason et al. 2014) and species-selective toxicants (Rennison et al. 2013). Some new insecticides are very short-lived (e.g. biological insecticides that contain living organisms or the toxins produced by them). This serves to improve their environmental safety but often calls for repeated applications. Such products may operate through direct contact (e.g. aerosols) or require ingestion while feeding (e.g. Btk). A few others are taken up systemically through the vascular system of plants (e.g. the neonicotinoids). Insect growth regulators (IGRs) and ecdysone represent other agonists that disrupt pest development by acting on their endocrine systems, but these are often environmentally persistent and require careful timing to be effective.

Trichoderma Persoon species are ubiquitous saprophytic fungi (Klein & Eveleigh 1998). The genus has been widely investigated for pathogen biological control as the species are easy to culture on a range of inexpensive substrates. A number of Trichoderma isolates have been used against a variety of aerial and soil-borne plant pathogens (Harman & Bjorkman 1998; Mukherjee et al. 2013). For example, Hypocrea atroviridis Dodd, Lieckf. & Samuels isolate C52 has been commercialised as a biocontrol agent for onion white rot (Sclerotium cepivorum Berk.) and T. harzianum Rifai isolate T39 has been found to be able to control grey mould Botryotinia fuckeliana (de Bary) Whetzel in grapes (Elad 1994).

Such products tend to be very selective and their generally low environmental impacts make them attractive for use in sensitive ecosystems.

In spite of such progress, an emerging problem is that, while more species-specific ‘green’ pesticides are being developed and used, the big agrichemical companies do not see New Zealand's small market as sufficient to warrant registration of such chemicals. Paradoxically therefore, it would seem that New Zealand's range of available invertebrate pesticides is likely to continue to comprise those broad-spectrum products that permit importation in sufficient quantities to provide an economic return to the importers. One exception to this, however, is the development of species-selective toxicants for rodents, which are a global problem, particularly in food safety and security.

For some insect species, attract-and-kill methods that combine a pheromone or other attractant with a toxicant and a carrier material are available. The aim is to control the target species by attracting it to large droplets of a formulation that causes mortality shortly after contact (e.g. Brockerhoff & Suckling 1999; El-Sayed et al. 2009). This is more acceptable than many other methodologies, in part because few species other than the target pest are likely to contact the toxin. Furthermore, the insecticide is able to be applied at much lower rates than in systems such as generalised spray application. However, one limitation of this technique is that it only works where a strong attractant component occurs in the pest species’ life history. This has become apparent during global fruit fly eradications (Suckling et al. 2015) when male annihilation of several Bactrocera spp. has been achieved using a combination of a powerful para-pheromone (methyl eugenol) and insecticide-treated blocks. Conversely, there is a complete absence of this tactic against medfly (Ceratitis capitata L.), for which no equivalent male attractant is known.

Biological control

Biological control is the control of a species through the introduction of a natural enemy. Examples include augmenting native herbivores in the case of weeds, introducing insect predators or parasites (Atalah et al. 2013) or using biopesticides (e.g. Jackson 2007).

International results from ‘classical’ biological control involving the release of natural enemies for the suppression of weeds and invasive invertebrates have been mixed, with only about 10% of releases since 1880 considered successful worldwide (Gurr & Wratten 2000). However, in the past 20–30 years there has been considerable success in New Zealand. As mentioned above, effective biological control of three very major curculionid pests of the country's main pasture species (ryegrass/clover) has been achieved via the introduction of three exotic species of the braconid endoparasitoid species belonging to the genus Microctonus Wesmael. These were M. aethiopoides Loan released against the lucerne weevil Sitona discoideus Gyllenhal in 1982 (Stufkens et al. 1987), M. hyperodae Loan released against the Argentine stem weevil (Kuschel) in 1991 (Goldson et al. 1993) and the clover root weevil S. obsoletus Gmelin released in 2006 (Gerard et al. 2006). Host parasitism rates by these wasp species reached up to 90% shortly after their release (Goldson et al. 1998; Barker & Addison 2006; Gerard et al. 2011) and have demonstrably reduced damage (e.g. Barlow & Goldson 2002). Similarly, biological control of target weeds has, in some cases, reduced the problem by 50%–83% (Fowler et al. 2000, 2010; Paynter et al. 2010). In general, however, the large number of current and future weed species presents a challenge for biological control programmes that rely on host-specific agents.

Research is needed to improve the success rates of biocontrol agents through analysis of ‘what worked in the past’ and how biological control can be better integrated with grazing management, chemical and physical control. For example, the efficacy of existing biocontrol agents may be enhanced through ecosystem manipulation (Wratten et al. 2012). Inundative methods that involve suppression of pests by mass rearing and release of natural enemies remain underdeveloped in New Zealand, due to the very small market size.

Plant breeding

A significant non-pesticidal response to pests is via plant breeding (see above). Exploiting the naturally occurring obligate Epichloë endophyte to control pests of ryegrass (Lolium spp.) and tall fescue (Schedonorus arundinaceus [Schreb.] Dumort) in New Zealand pastures has been highly successful (Johnson et al. 2013). As discussed, there has also been success in breeding lucerne (M. sativa) for resistance to aphids, and resistance in cereals to striped rust Puccinia striiformis Westend (e.g. Cromey 1992) although loss of resistance can be a problem.

A major programme is under way to identify varieties of kiwifruit (Actinidia deliciosa [A. Chev.] C.F. Liang and A.R. Ferguson) with resistance/and or tolerance to PSA (e.g. Cheng 2014). Such traditional breeding programmes have been accelerated by marker-assisted selection, which allows the selection of genes associated with particular traits. Transgenic modification remains a controversial area in New Zealand and, significantly, its use is still a barrier for some of this country's international trading partners. Possibly, a better understanding of how transgenic technology works, its benefits and risks combined with rapid technological advances, may result in acceptance of this sort of technology in New Zealand and its markets.

In general, plant resistance can be particularly useful in pasture or broad-acre cropping environments where the size of the area, cost and environmental impact often preclude the widespread use of pesticides. Further, forage production is not destined for export which obviates any concern about cosmetic damage or the presence of adventive invertebrate species. This situation is quite different from that in high-value crops such as fruit and vegetables where, even in the presence of effective plant resistance, some cosmetic damage and/or biological contamination (e.g. mites) can lead to outright product rejection. In such situations, despite the effectiveness of IPM, often significant disinfestation procedures are still required prior to product export.

While plant breeding for pest resistance has been successful, there are two corollaries: the first is that, in New Zealand, plant breeding has tended to focus on crop yield and quality (e.g. apples and pine trees) rather than pest resistance; the second is that, in all cases, there remains the omnipresent threat of possible new pest incursions that are well able to completely disrupt existing pest management approaches (see below).

Physical methods

In some circumstances, various physical interventions against vertebrate pests, such as mass trapping (Warburton et al. 2008), shooting (Choquenot et al. 1999) or the manual removal of pest species, have been used to reduce pest populations (Yamanaka 2007).

Shooting is used to control large species such as red deer, pigs and goats, either by ground-based hunters or from helicopters (Parkes et al. 2010; Barron et al. 2011). Manual removal of weeds in managed and natural ecosystems also works when population densities are low. This approach, called rogueing, is valuable when targeting specific weeds that pose a singular threat such as the recent incursion of black grass (Alopecurus myosuroides Hud) as a result of seed spillage in Canterbury. Fences (Scofield et al. 2011) combined with grazing and manipulation of plant competition will emerge should reliance on herbicides decline and/or herbicide resistance become more widespread.

Physical treatment tools have also been successfully trialled in the marine environment (e.g. smothering pests with geotextile fabric, polyethylene or dredge spoil; Coutts & Forrest 2007), as well as heat treatment (e.g. Wooton et al. 2004). Research into control of biofouling on vessels as well as in marine aquaculture (Fitridge et al. 2012) may yield useful outcomes that will help to reduce the global spread of species.

Fertility control

Fertility control of vertebrates has been attempted (Duckworth et al. 2006, 2010) but it is still beset with challenges (see below). Conversely, mating disruption of insects in orchards is used in New Zealand to control moth species. Such disruption is based on synthetic pheromone formulations that can control up to four pest species with a single dispenser, thereby providing pesticide-free, residue-free, high-value apples for export. This benign technology can be applied either aerially or from the ground. Currently, around 30% of the New Zealand pipfruit industry is using a mating disruption system, resulting in a total saving on pesticides of c. NZ$670,000 p.a. As a result, use of pesticide ingredients classified as ‘extremely’ and ‘highly hazardous’ to human health have been discontinued completely. Mating disruption-based IPM programmes also conserve ecosystem services, such as higher functional biodiversity effective against secondary pests (Suckling et al. 1999). Again, unfortunately, such methodology is restricted to those species that show strong pheromonally based reproductive behaviour.

Sterile insect technique (SIT)

Internationally, SIT plays a significant role in containment and eradication programmes for certain pests, especially fruit flies (e.g. Klassen & Curtis 2005). Inundative releases of sterile insects over wide areas can reduce the fertility of a field population; however, for SIT to be used as an operational technique during eradication, several criteria must be met. For example, treated areas must be large enough and/or isolated enough so that the effect of immigration from surrounding areas is minimal (Barclay et al. 2011).

The use of SIT as a control tactic has many advantages, including species specificity and compatibility with the use of other control tactics, as well as effectiveness in low-density pest populations. This is valuable during the endgame of an eradication. New Zealand's only operational case of SIT use involved the painted apple moth in Auckland, where there was an overflooding release of parental males carrying inherited sterility (Suckling et al. 2007). Such inherited sterility offered the significant advantage in that any mating with wild females led to hundreds of sterile progeny in the F1 generation. The technique was also parameterised on light brown apple moth for use in California (Kean et al. 2011) and field success has been demonstrated in New Zealand (Stringer et al. 2013).

Pressures on existing pest management tools

In spite of progress, pressures on currently available pest management tools remain a major policy and technical challenge. As discussed, existing methodologies, especially the use of pesticides, are being questioned by both the public and the market and, at the same time, existing pest management regimes can be obliterated by biosecurity failure. For example, New Zealand's internationally acclaimed IPM suppression of the cosmopolitan pest Helicoverpa armigera (Hübner) in tomatoes (Cameron et al. 2006; Walker et al. 2010) collapsed completely after the arrival of the potato-tomato psyllid Bactericera cockerelli (Sulc). The psyllid forced growers to go back to calendar-based spraying that often includes the use of toxic organophosphate pesticides. With this, the threat of insecticide resistance also emerges when control is marginal (Page-Weir et al. 2011) and the introduction of a classical biological control agent for the psyllid (if one is found) may well pose inherent risks to native insects, leaving the problem and control unresolved.

Obsolescence and discontinuation of various classes of pesticides

New Zealand pest management is being challenged by the declining range of pesticides available.

The use of persistent organochlorine pesticides (chlordane, DDT, lindane, aldrin, dieldrin, endrin, heptachlor, mirex, hexachlorobenzene, toxaphene, etc.) has been progressively restricted by a succession of legislative measures, such that, by the mid-1970s, their use had effectively ceased in agriculture and horticulture (MfE 2006). Moreover, their widespread presence in sheep dips has created legacy issues at some locations (Roberts et al. 1996). Pesticides acceptable for use have been restricted further, largely based on the Stockholm Convention on Persistent Organic Pollutants. This convention became international law in May 2004, was ratified by New Zealand in September 2004 and entered into force on 23 December 2004. The agreement is focused on the abolition or strict control of the use of persistent organic pollutants that do not break down readily in the environment, are capable of long-range transport, bioaccumulate in human and animal tissue (and biomagnify in food chains) and/or pose the risk of causing adverse effects to human health and the environment.

As a result, by 2013 the EPA had discontinued its permission for the manufacture of, or importation of, the following pesticides: benomyl, carbofuran, carbosulfan, dichlofenthion, ethion, famphur, isazofos, omethoate, phoxim, methyl parathion, endosulfan and pyrazophos. In addition, other active ingredients and/or insecticides containing these actives have time-limited approvals as follows: diazinon (1 July 2028), fenamiphos, methamidophos, prothiofos and terbufos (1 July 2023), fenitrothion and phorate (1 July 2016). The EPA also decreed that, as well as the default controls already in place on these approvals, additional controls should be applied to all insecticides that remain in use. Such requirements include setting application parameters such as maximum application rates and frequencies, restricting the method of application, such as prohibiting aerial application of some substances, and restricting indoor application to automated methods. Further measures are label statements to indicate that the substance is an organophosphate or carbamate, label warnings of risks to bees and stated crop re-entry intervals. It is now also necessary for users of organophosphate and carbamates to hold an approved handler certification (e.g. EPA 2013).

In spite of such measures, no doubt synthetic pesticides and their use in general will remain under close scrutiny. A current example of this is a suspicion that neonicotinoids are implicated in bee colony collapse disorder (e.g. Whitehorn et al. 2012).

Evolved resistance

Restriction imposed on the use of pesticides is being compounded by cases of evolved resistance by a wide range of organisms, from fungi to weeds. Thus, formerly useful and chemically based strategies are now no longer effective.

This has led to the development of New Zealand resistance management strategies (http://resistance.nzpps.org/) that are vital to ensure the sustainability of current regimes. This often amounts to the alternating use of different mode-of-action synthetic pesticides that can be combined with various non-chemical inputs. New Zealand agriculture benefits from having few herbicide-resistant weeds, but should the recent appearance of glyphosate resistance become more widespread, then the annual cost to the arable sector could be as high as NZ$25–50 million p.a. (Groundworks Ltd 2012). Moreover, wider distribution of herbicide-resistant giant buttercup could cost New Zealand dairy farmers NZ$750 million p.a. in lost milk solids revenue (Bourdôt & Saville 2010).

As mentioned above, Goldson et al. (2014a, 2014b) have suggested that New Zealand's simple broad-acre ecosystems, with their relative lack of biological complexity, may risk the loss of efficacy of parasitoid control agents due to selection pressure for host resistance. The workers noted this could particularly be the case when highly effective parasitoids are parthenogenetic but their hosts reproduce sexually. Such circumstances can be seen to cause an unequal ‘evolutionary arms-race’ whereby the host is able to evolve, while the parasitoids are much less able to do so. This could be the case for both L. bonariensis and S. obsoletus biological control. Indeed, the once effective biological suppression of L. bonariensis now seems to be considerably diminished (Goldson et al. 2014a, 2014b).

Resistance to fungicides is exemplified by the loss of efficacy against apple black spot (Fusicladium pomi [Fr.] Lind) of the currently used demethylation inhibitor fungicides, myclobutanil and penconazole (Beresford et al. 2013). The evolution of genetic resistance in Norway rats to anticoagulant poisons has caused great concern in Europe (Pelz et al. 2005) and anthelmintic resistance by gut nematodes is also a significant issue for New Zealand livestock managers (e.g. Leathwick et al. 2001).

Animal welfare

The welfare impacts of vertebrate pest control are increasingly scrutinised in New Zealand (Warburton et al. 2010) and this needs to be taken into account when assessing the overall costs and benefits of pest control operations (Littin 2010). Attention has therefore been given to improving and assessing the relative humaneness of traps (Warburton et al. 2008) and poisons (O’Connor et al. 2007; Sharp & Saunders 2011) (see below). Internationally, New Zealand is at the forefront of incorporating ethical and welfare principles (Warburton & Norton 2009) but, if the country's use of a wide range of pest control tools is to be maintained, vigilance will be required in terms of adhering to international trends in this area.

Intensification

Most of New Zealand agriculture is intensifying in the interests of improved productivity. With this, production tends to be towards the technical upper limit of what can be produced. Very often, such systems have a reduced capacity to absorb the impact of pests without significant losses in yield (e.g. Goldson 2009).

Biosecurity failure

As discussed, the impacts of biosecurity failure can be catastrophic to production, not only because of the impact of the pest itself, but also because of the ability to disrupt finely tuned, minimal-impact pest management regimes.

Seedling establishment

In broad-acre agriculture, pest damage often results in the inevitable need to recultivate. As well as already-mentioned impacts on soil structure and energy use, it is the emerging seedlings that are most prone to pest impacts and these can be difficult to protect except by the use of systemic pesticides, often applied as seed coatings.

Advances and development of new tools and strategies for pest management

Pest management approaches and technologies require ongoing basic and applied research to maintain pest suppression using publicly acceptable methods. Improvement and innovation are required to offset the trends discussed above.

Irrespective, synthetic pesticides are likely to continue to have a role for the foreseeable future, but escalating costs of development and registration, as well as increasing demand for sustainability, will limit their availability in New Zealand. For vertebrate pests, new toxic agents continue to be developed in this country (e.g. Eason et al. 2013).

Given such circumstances, work must continue, unabated, in the pursuit of pest management solutions. It is, therefore, useful to consider current thinking, approaches and scientific progress; this is outlined below.

Ongoing development of surveillance and pest management strategies

Prioritisation of surveillance effort remains an abiding challenge for pest management, as early detection is essential for eradication (Brockerhoff et al. 2010b; Tobin et al. 2014). This is a requirement for both pre- and post-border biosecurity, because pest interception and eradication obviate inevitable damage and the need for varyingly successful management strategies. Importantly, though, lack of detection does not indicate pest absence; unless a pest species is conspicuous it can be very difficult to find the first few invaders, reinvaders or the last few survivors (Parkes & Nugent 2011). This can be particularly difficult in dealing with incursions of microorganisms, such as plant pathogens, and for marine species. Prioritisation of surveillance is further fraught because of the unpredictable behaviours shown by threat species across New Zealand's ecosystems (see above). Risk assessment approaches definitely bring benefits, but these are not foolproof and depend on prior knowledge (Hulme 2012). Notable advances have been made in prioritising species for attention based life-history traits to heighten probability of discovery and successful control. This approach identified Japanese honeysuckle (Lonicera japonica Thunb.) as problematic in view of the very high risk of accelerated spread when different global clones mingle and then set seed (Paynter et al. 2012).

When eradication cannot be achieved, informed ecological approaches to pest management are required that extend well beyond those historical pest management approaches that are often based on killing organisms in sufficient numbers to reduce damage. Advances must now aim to measure responses in terms of desired outcomes (e.g. economic productivity or biodiversity response; Clayton & Cowan 2010). A plant example of this is Nassella tussock (Nassella trichotoma [Nees] Hack. ex Arechav.). This grass species, of low digestibility to livestock, invaded indigenous tussock grasslands in the eastern parts of Marlborough and Canterbury from the 1860s onwards. By the early 1900s, near monocultures of the weed had developed. Since the middle of the 20th century, the reinvading plants have been removed annually by digging them out in regionally coordinated management programmes. Such effort resulted in densities that no longer reduce live-weight gains of sheep and other grazing animals (Bourdôt & Saville 2009). However, whether such continuous management of the weed is worthwhile has been a topic of intense debate among both farmers and scientists. To inform this debate, the potential range of Nassella tussock in New Zealand was estimated using a climate model combined with data on Nassella's global distribution. Regional authorities can now target those sites for surveillance that are most at risk of invasion. Similar models can be found, based on extensive data obtained from international and local weed or pest distributions, that are now enabling territorial authorities to target surveillance operations to best effect (e.g. Worner et al. 2012; Narouei Khandan et al. 2013; Senay et al. 2013). Linked to this, Bayesian inference is being increasingly used to assess uncertainties and to assist managers with their interpretation of surveillance data (Ramsey et al. 2011; Morrisey et al. 2012). Information technology is also starting to permit pest management based on ‘real time’ control (e.g. Guarnieri et al. 2011).

In response to the increasing need for additional vertebrate pest management technologies, research continues on the use of biological control methods for rabbits (specifically, rabbit haemorrhagic disease) (Scroggie et al. 2012), but not on any other vertebrate pests.

In general, new technologies are being employed to help with insect pest management efforts. For example, pheromones and other odorants are being tested to uncover the presence of invaders (Brockerhoff et al. 2006; El-Sayed 2013). This technology provides information on geographical distribution and an ability to monitor population density, which are central to eradication and pest management programmes. Sex pheromones are now available for about 30 horticultural and other insect pests (Suckling et al. 2012).

Other technological approaches are being developed through the New Zealand Better Border Biosecurity Collaboration (B3–Better Border Biosecurity 2014). These include the use of DNA barcoding for rapid pest species identification (Armstrong & Ball 2005); the assessment of next generation sequence-based technology for the identification of important biosecurity threats in multi-taxa formats (such as the mix of pathogens found in a plant tissue sample or the range of insect species caught in a trap); linking of taxonomic identification of a pathogen to its relative pathogenicity; and the use of trace elements and stable isotopes to infer natal origins of exotic threat populations (Holder et al. 2014).

Within waterways, techniques such as environmental DNA sampling, as mentioned above, can improve aquatic monitoring and surveillance, although further research is needed to improve the efficacy of such methods (Wood et al. 2013).

All of these sophisticated approaches are requiring more trained local and central government staff (e.g. regional council and fisheries officers) to assist in translating such science-intensive technologies and methodologies for pest eradication and management.

Fertility-based biological control methods

The growing ethical concern about lethal methods for the control of larger mammalian pests requires innovation. The pursuit of fertility-based control measures is one approach for dealing with such concerns. Although the application of such methods has not been extensive, significant progress was made with the fertility control of possums before research was halted (Duckworth et al. 2006; Cui et al. 2010). Other recent advances have included the recognition of the potential use of the Vaccinia virus as a delivery mechanism for fertility control agents (Cross et al. 2011). This has been augmented by research into the ‘Trojan female’ technique (Tompkins et al. 2013b). The method involves exploiting natural maternally inherited mitochondrial DNA mutations that affect male but not female reproductive fitness. ‘Trojan females’ carrying such mutations, and their female descendants, produce ‘sterile male’ equivalents under natural conditions over multiple generations (Gemmell et al. 2013). Similarly, with insects such as mosquitoes, research is being undertaken to develop transgenic strains of males that result in high levels of mortality among the progeny under certain conditions (Massonnet-Bruneel et al. 2013). Similar self-disseminating fertility control of small pest mammals such as rodents has yet to be developed (Warburton 2012).

There remains a concerted effort to continue to develop fertility control of marine biofouling pests (Willis & Woods 2011) and sterile-male approaches in invasive fish such as carp, again using the ‘Trojan female’ technique are being pursued (Teem et al. 2011). Moreover, work is continuing into the use of a species of native sea urchin (Evechinus chloroticus Valenciennes) for the biocontrol of biofouling organisms (Atalah et al. 2013) and on the development of koi (escaped domesticated common carp [Cyprinus carpio L.]) herpes virus for managing the pest in lakes and streams.

The use of pheromones and other attractants for attracting beneficial biological control insects into an infested vicinity offers some promising leads in areas such as weed and insect classical biological control (Suckling 2009), as well as for some invertebrate species in aquatic environments, including paddle crabs (Ovalipes catharus White) where electrophysiology is being used to provide insights from brain recordings.

Improved use of existing toxins to combat vertebrate pests

As well as developing various sterility-based suppression approaches for mammalian pests, research continues into refining existing vertebrate control devices and toxic agents (VTAs) to increase their effectiveness against possums, stoats, rabbits, feral cats and rodents (Warburton et al. 2008; Eason et al. 2010). Recent innovations include a self-resetting trap (the ‘GoodNature’ trap: www.goodnature.co.nz) and species-selective, multi-dose devices that spray toxins such as para-aminopropiophenone (PAPP) or sodium nitrite directly on to the fur of pests such as possums or stoats; these toxins are subsequently ingested during grooming. Such tools hold promise for use in less accessible areas. The pre-feeding of non-toxic baits (Nugent et al. 2011) is now enabling spatially targeted control of mammalian pests (Porphyre et al. 2013). Other research, based on statistical analysis, has shown the circumstances where several single-capture kill-traps deployed at one location may be more cost effective than multiple-capture traps (Warburton & Gormley 2013).

Such improvement in the use of VTAs means that the same kill rates can be achieved with a fraction of the amount of toxic bait used in the 1970s–90s (Fisher et al. 2011; Nugent & Morriss 2013). The ability to predict the timing of irruptions of rodents has also improved in recent years due to better knowledge of the mast seeding that causes resource-based pulses of pests (Kelly et al. 2013). This is now being combined with better understanding of how trophic interactions among mammalian pests may influence control efficacy (Ruscoe et al. 2011; Tompkins et al. 2013a).

The use of toxin synergists and/or analgesics are speeding up pest animals’ metabolism and leading to more rapid death with reduced suffering; this is leading to improved acceptability of toxin use (Morgan et al. 2013). Food-based lures are also being used to attract pests to control devices and pheromone-based lures may increase the effective search area of control devices (Linklater et al. 2013).

Non-target risks have required a shift from the use of broad-spectrum VTAs to species-selective VTAs (Cowled et al. 2008). New Zealand is leading the way in this area, with the development of the first new rodenticide with enhanced efficacy for the genus Rattus Fischer von Waldheim since the formulation of brodifacoum 25 years ago (Hopkins 2013; Rennison et al. 2013). So far, research and development efforts have been restricted to rats, but the concept is extendable to a range of small mammal pests such as rabbits, possums and mice.

A key issue that urgently needs to be addressed with the development of new vertebrate pest control tools is the long timeframe between development of the product, its registration for field use and availability for widespread application (sometimes it can take 20 years; Eason et al. 2010).

Identification of new weed control agents

Microbially based biopesticides continue to show promise for the control of weeds although they lack the reliability and effectiveness of their chemical counterparts; work is continuing on improving them and their cost effectiveness (e.g. their shelf life) (Glare et al. 2012).

The identification and development of additional classical weed biological control solutions is a promising area because of its strategic alignment with demands for residue-free production for export. However, arthropod biological control agents are difficult to deploy and the next big challenge is to achieve a higher rate of success at a landscape scale. This will require detailed long-term research into what predisposes control agents to succeed. Certainly, classical biological control of weeds appears to create fewer challenges than that for invertebrates. The lack of negative impact (Suckling & Sforza 2014) is in part, because some candidate weeds have no New Zealand relatives and non-target impacts are not an issue.

The options to manage aquatic weeds are limited. Again, emphasis must be put on the prevention of spread, based on surveillance for early detection. Until 2005, only two herbicides were registered for use in aquatic areas and these proved ineffective against a range of high-impact species including Spartina anglica C.E. Hubb and S. alterniflora Loisel. (spartina), Alternanthera philoxeroides (alligator weed) and Hydrilla verticillata (L.f.) Royle (water thyme). As a result of extensive research, three additional herbicides are now available for aquatic use and two other highly selective herbicides have been permitted under previous authority provided by the Pesticides Act (1979). Through this work, the impact of these four invasive aquatic species has been curtailed. Use of such herbicides under aquatic situations has recently been reassessed by the EPA as suitable for ‘hard to control’ weed species in the marginal edges of aquatic systems. New options for the biocontrol of aquatic weeds are being pursued. These initiatives have expanded the aquatic weed control toolbox, providing management agencies and energy companies with effective selective control options and permitting improved biosecurity options.

General comment on New Zealand pest management

Controversy arising from differing human values and perceptions

Millions of pest species are killed each year using different control programmes and, as mentioned above, this can generate strong, negative public reactions (Warburton 2012). Indeed, even classifying certain animals as pests is controversial. Some species (e.g. deer and trout) are considered to be pests by some but as a resource by others (Figgins & Holland 2012). Proposed control of species such as cats generates a great deal of heated debate (http://garethsworld.com/catstogo/); while domestic cats are companion animals, domestic and feral cats are also major predators in New Zealand ecosystems (van Heezik et al. 2010). The ‘cat debate’ often flares up in the media and the ensuing furore results in greater awareness of the potential impacts of such species on native fauna. Aerial delivery of toxins such as sodium fluoroacetate (1080) still occasions vehement protest (Green & Rohan 2012) and while the risks and benefits of 1080 have been well documented, opposition to aerial poisoning persists (PCE 2011; Veltman & Westbrooke 2011).

Unexpected consequences of vertebrate pest control

Vertebrate pest management, in particular, is complex and multidimensional, especially when considering control of multiple interacting vertebrate pests in a whole-of-ecosystem context (Tompkins & Veltman 2006). Perverse or unexpected outcomes can arise when the removal of one pest results in an increase in another (Ruscoe et al. 2011), or when pest suppression occurs in tandem with other drivers of global change such as land use change (Norbury et al. 2014) or climate change (Tompkins et al. 2013a).

Varying perceptions of wilderness and biodiversity

In New Zealand, the ‘conservation estate’ (i.e. wilderness) is clearly distinguished from farmlands. This is in contrast to the UK where the British public regard farms and surrounding areas as (effectively) national parks and centres of biodiversity. This delineation creates a different dynamic between users of farms and conservation areas in New Zealand when it comes to pest management. There is also a contrast in New Zealand attitudes towards the use of chemical treatments in the marine environment as opposed to productive terrestrial ecosystems. For example, few mussel farmers use chemicals to control biofouling in New Zealand, whereas in other countries, chemicals have been used to eradicate or manage marine pests such as the black-striped mussel (Mytilopsis sallei Récluz) in Darwin (Bax et al. 2002) and ‘killer algae’ (Caulerpa taxifolia [Vahl] C.Agardh) in southern California (Anderson 2005).

Organisation and coordination of pest management responses in New Zealand

The allocation of resources between pest management or biosecurity surveillance and response activities is always a challenge. An agreed set of priorities across the various levels of government may include issues such as who provides monitoring, where this should be carried out and why. Further, after the discovery of an incursion, decisions must be made at short notice, often with incomplete data. This calls for the pooling of New Zealand pest management expertise and judgement.

Some contingency plans do exist. For example, clear agreement has been reached on what to do if there is an outbreak of foot and mouth disease (A. epizootica) or if new PSA infections of kiwifruit are detected; here, strong economic drivers to respond come into play. Conversely, action required when dealing with natural ecosystems is less clear and, consequently, responses can vary. An example of this is the marine Mediterranean fanworm (S. spallanzanii), a major pest of shorelines, which can reduce native biodiversity and has significant, indirect effects on nutrient cycling in marine environments. It was detected in Lyttelton Harbour in Christchurch in 2008 and in Waitemata Harbour in Auckland in 2009. Soon after its discovery in Lyttelton, an attempt was made to eliminate it and its density was reduced to less than 3% of its original level. Conversely, no action was taken in the Waitemata and the population there has since undergone rapid radiation throughout the harbour, with new satellite populations now detected in Whangarei, Tauranga and Nelson (G. Inglis, NIWA, pers. comm. 2014).

Should incursion eradication be attempted, population monitoring during the process must be continued for sufficient time to ensure that eradication has in fact been achieved. Also, while there may be some initial enthusiasm for eradication attempts, it can be more difficult to find funding for ongoing programmes. This was a hard-learned lesson for the Animal Health Board (now TBfree New Zealand) in its efforts to control bovine TB via reduction of possum numbers. The re-emergence of TB in livestock in the 1980s and early 1990s occurred directly because of reduced funding for control (Hutchings et al. 2013).

In undertaking monitoring and surveillance, opportunities abound for Citizen Science in empowering local enthusiasts to look after local areas and waterways. For example, botanical societies have engaged with wilding pine removal and old man's beard (C. vitalba) control. A recent survey assessed whether there were ongoing monitoring activities being carried out around New Zealand that could be harnessed for a national surveillance programme for ‘new to New Zealand’ species. The results identified a widespread and diverse range of habitats that were monitored and sampled, based on a diversity of technical skills and taxonomic capabilities that could be harnessed for national surveillance programmes (MAF 2011).

Need for specialist taxonomic and ecological expertise

Biosystematics provides the conceptual framework for understanding the relationships between species and is underpinned by national collections of biological specimens. The names of organisms and their phylogenetic connections provide a critical entry point into databases and knowledge with which to discriminate native from non-native organisms.

Improving the ability of border biosecurity and pest managers to distinguish native from exotic species requires significant attention covering a number of considerations. The first is the need for a good understanding of the New Zealand native flora and fauna. For many groups of organisms, the discovery and documentation of the native biota is incomplete and scant data are often distributed across multiple databases in various agencies. Also, often there is no adequate reference material or descriptions of key groups of native organisms related to potential exotic pests. Second, identification requires that the target organism can be recognised as being of interest. This demands that front-line staff are very familiar with the local flora and fauna and able to spot unusual specimens. Third is the requirement for specialists with a broad knowledge of particular groups of organisms of interest. They must be familiar not only with New Zealand species, but also be able to use their associated networks offshore, so that they can rapidly undertake comparative analyses to determine if a target organism is new to New Zealand. Across the board, New Zealand lacks such critical skills (Lester et al. 2014). Overseas expertise is already employed to identify organisms in a number of groups and, although modern communication technology is definitely assisting, there is no substitute for expertise in the country.

Conclusions

This contribution has sought to highlight the inescapable importance of effective pest management in New Zealand under changing circumstances. Considerations include accelerating globalisation, the ongoing arrival of invasive species, declining effectiveness and acceptability of existing controls, intensifying land use and climate change. Accentuating this is the growing demand by international trading partners for quality assurance. Doing nothing about these trends and conditions is not an option because of New Zealand's unique dependence on its high-quality and often unusual environment, combined with the need to maintain its reputation for primary products that are utterly safe, residue-free and produced to the highest ethical standards.

New Zealand's natural environment poses unique challenges, particularly regarding the control of vertebrate pests. Likewise, pest management in aquatic environments presents its own set of difficulties, not least because of a lack of political and public awareness probably arising from a general lack of visibility of pest impacts compared with what goes on in terrestrial ecosystems.

Future efforts could make better use of the wealth of existing data on pest abundance and of the tools used to achieve understanding of responses of native biota resulting from medium- to long-term pest management interventions conducted by agencies such as the DOC. This is very much in the spirit of whole-of-system approaches to pest management in New Zealand.

The inherent limitations of existing pest management approaches underline the requirement for discovery, new technologies and ongoing refinement of existing methods. In particular, there needs to be an overall move away from the use of pesticides to control systems based on good biological understanding. Such a shift would extend beyond the study of individual organisms to ecosystem function and susceptibility to invasion by exotic species. All of this calls for a commensurate increase in expertise. With this, it will be necessary to engage early with the public over novel pest control tools and strategies, or risk losing the battle for pest control.

To complement such developments within scientific research, there are real opportunities for more citizen involvement, for example, in areas such as biosecurity surveillance. Here, New Zealand is fortunate in having a motivated population concerned about the quality of its environment.

Finally, the necessity for increasingly sophisticated pest management and biosecurity surveillance requires improved integration of pest management science with central and local government agencies. This is an area where excellent progress has already been made but there is no room for complacency.

Acknowledgements

The authors would like to thank the following for their valuable input in commenting on and reviewing the paper: Paul Champion (NIWA), Phil Cowan (Landcare Research), Roberta Farrell (University of Waikato), Simon Fowler (Landcare Research), Philip Hulme (Bio-Protection Research Centre, Lincoln University), Graeme Inglis (NIWA), Paul Livingstone (TBfree New Zealand), Susan Timmins (DOC), Barry Scott (Bio-Protection Research Centre, Massey University), Steve Wratten (Bio-Protection Research Centre, Lincoln University), Stephen Hartley (University of Victoria), Chris Baddeley (Ministry for Primary Industries) who kindly compiled much of the pest impact information used in this study and two anonymous reviewers. Thanks also to Christine Bezar (Landcare Research), Dawn McMillan (AgResearch) and Lois McKay (AgResearch) for editorial contributions. Finally, the authors thank Dr Marc Rands of the Royal Society of New Zealand for the encouragement and persistence that was needed to get this review completed.